Citation
A mixed method investigation of predictors of pain and disability in patients with lumbar spinal stenosis

Material Information

Title:
A mixed method investigation of predictors of pain and disability in patients with lumbar spinal stenosis
Creator:
Hammerich, Amy Stone ( author )
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
1 electronic file (237 pages). : ;

Subjects

Subjects / Keywords:
Spinal canal -- Stenosis ( lcsh )
Pain perception ( lcsh )
Pain -- Psychological aspects ( lcsh )
Pain -- Measurement ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
This study identified characteristics associated with pain and disability in LSS at entry to treatment and factors contributing to long-term reduction of pain and disability. It also revealed how patients with LSS appraise the stressors of pain and disability and use coping resources. This study used qualitative and quantitative methods to address the dearth of rea=search in the LSS literature regarding factors associated with severity of pain and disability upon entry to treatment and the influence of coping resources on post-treatment outcomes. A cohort of 34 patients was evaluated to understand the relative contribution of perceived health, self-efficacy, and social support on the stressors of pain and disability using a combination of bivariate and multivariate regression analyses and patient interviews. Several key finding emerged from this research. First, there as an inverse relationship between perceived physical and mental health and severity of disability upon entry to treatment. In interviews, patients with LSS described a variety of sources contributing to stressors of pain and disability including physical limitations, reduced confidence, lack of control, vulnerability, mental health concerns, reduced social participation, frustrations at needing support, and financial limitations. Better perceived physical health had the greatest contribution to reduced levels of pain and disability post-treatment. In interviews, a variety of behaviors were described that improved coping including a greater understanding of limitation and needs a long with beliefs for pushing through and remaining positive. In line with the Stress and Coping Model, patients with stronger coping behaviors and more resources reduced the severity and improved outcomes of pain and disability associated with LSS. There is an increased need for research improving the quality of life in later years particularly when faced with a chronic disabling condition such as LSS. Early identification of those at risk for higher pain and disability as well as support for positive coping behaviors post-treatment can help patients with LSS maintain a higher level of participation in society. The factors uncovered in this study serve to inform translational research, clinical practice, and policy recommendations, which may ultimately add to treatment and resources for reducing pain and disability associated with LSS.
Thesis:
Thesis (Ph.D.)--University of Colorado Denver. Health and behavioral sciences
Bibliography:
Includes bibliographic references.
General Note:
Department of Health and Behavioral Sciences
Statement of Responsibility:
by Amy Stone Hammerich.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
891656209 ( OCLC )
ocn891656209

Downloads

This item has the following downloads:


Full Text
A MIXED METHOD INVESTIGATION OF PREDICTORS OF PAIN AND DISABILITY IN
PATIENTS WITH LUMBAR SPINAL STENOSIS.
by
AMY STONE HAMMERICH
B.S., Miami University, 1992
M.S.P.T., Regis University, 2000
D.P.T., Regis University, 2003
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Health and Behavioral Sciences
2013


This thesis for the Doctor of Philosophy degree by
Amy Stone Hammerich
has been approved for the
Health and Behavioral Sciences Program
by
Deborah Main, Chair
Susan Dreisbach, Advisor
Susan Scherer
Joshua Cleland
November 1, 2013


Hammerich, Amy Stone (Ph.D., Health and Behavioral Sciences)
A Mixed Method Investigation of Predictors of Pain and Disability in Patients with Lumbar
Spinal Stenosis
Thesis directed by Assistant Research Professor Susan L. Dreisbach.
ABSTRACT
This study identified characteristics associated with pain and disability in LSS at entry to
treatment and factors contributing to long-term reduction of pain and disability. It also revealed
how patients with LSS appraise the stressors of pain and disability and use coping resources. This
study used qualitative and quantitative methods to address the dearth of research in the LSS
literature regarding factors associated with severity of pain and disability upon entry to treatment
and the influence of coping resources on post-treatment outcomes. A cohort of 34 patients was
evaluated to understand the relative contribution of perceived health, self-efficacy, and social
support on the stressors of pain and disability using a combination of bivariate and multivariate
regression analyses and patient interviews.
Several key findings emerged from this research. First, there was an inverse relationship
between perceived physical and mental health and severity of disability upon entry to treatment.
In interviews, patients with LSS described a variety of sources contributing to stressors of pain
and disability including physical limitations, reduced confidence, lack of control, vulnerability,
mental health concerns, reduced social participation, frustrations at needing support, and financial
limitations. Better perceived physical health had the greatest contribution to reduced levels of
pain and disability post-treatment. In interviews, a variety of behaviors were described that
improved coping including a greater understanding of limitations and needs along with beliefs for
pushing through and remaining positive. In line with the Stress and Coping Model, patients with
stronger coping behaviors and more resources reduced the severity and improved outcomes of
pain and disability associated with LSS.


There is an increased need for research improving the quality of life in later years
particularly when faced with a chronic disabling condition such as LSS. Early identification of
those at risk for higher pain and disability as well as support for positive coping behaviors post-
treatment can help patients with LSS maintain a higher level of participation in society. The
factors uncovered in this study serve to inform translational research, clinical practice, and policy
recommendations, which may ultimately add to treatment and resources for reducing pain and
disability associated with LSS.
The form and content of this abstract are approved. I recommend its publication.
Approved: Susan L. Dreisbach
IV


ACKNOWLEDGMENTS
I would like to thank the faculty at Regis University School of Physical Therapy who
gave me the support and flexibility to follow my passions. In addition, I would like to thank the
Health and Behavioral Science Department for challenging my preconceptions, testing my limits,
and giving me new perspectives on health and health care. First and foremost in the Health and
Behavioral Science Department, I would like to thank Dr. Susan Dreisbach for her mentorship
and teaching me the art of qualitative research. I would also like to thank Dr. Debbi Main for
showing me how to define the most important questions that need to be answered.
In addition, I would like to thank Dr. Josh Cleland for his mentorship over the years.
With this mentorship, I have appreciated his persistence, patience, and sense of exploration as we
continue to identify and understand the complex needs of those with the chronic condition of
lumbar spinal stenosis. I would also like to thank Dr. Julie Whitman who was there in the
beginning and continues to be a constant force in this research. Moreover, I would like to thank
Dr. Sue Scherer for all of her past and continued mentorship in academic, research, and clinical
endeavors. Most of all, I would like to thank my husband, Scott who was my strongest and most
resilient support in every adventure during my PhD process.
v


TABLE OF CONTENTS
CHAPTER
I. BACKGROUND AND SIGNIFICANCE........................................................1
Low Back Pain......................................................................1
Lumbar Spinal Stenosis (LSS)...................................................1
Definition, pathoanatomy, and clinical presentation of LSS..............2
Incidence and impact of LSS.............................................2
Limitations in the current literature for LSS...........................4
Rationale...........................................................................4
Research Question and Specific Aims............................................5
Research Goals.................................................................7
II. REVIEW 01 THE LITERATURE...........................................................8
Pain and Disablement in LSS........................................................8
Previous Literature: Psychosocial Factors......................................9
Previous Literature: General Health Status....................................11
Previous Literature: Research Design in LSS...................................13
Previous Literature: Summary..................................................15
Theoretic Perspectives.............................................................15
Integrating Theory in LSS.....................................................15
Theoretical Framework.........................................................16
Stress and Coping Model.......................................................16
Cognitive appraisal....................................................18
Personal and situational factors.......................................18
Different appraisals...................................................19
Coping.................................................................20
Application of the Stress and Coping Model to LBP......................21


Managing Stressors: Pain
22
Biopsychosocial Models of Pain.....................................................22
Stress-Diathesis Model of Pain.....................................................24
Pain appraisals............................................................25
Pain beliefs...............................................................25
Managing Stressors: Disability..........................................................26
Disablement Model..................................................................26
Expanded Disablement Model.................................................27
Social model of disability.................................................29
ICF Framework......................................................................29
Functioning ...............................................................30
Contextual factors.........................................................30
Application of disablement models in EBP...................................33
Resource Theories of Stress and Coping..................................................33
Social Cognitive Theory and the Construct of Self-Efficacy ........................34
Self-efficacy..............................................................35
Application of self-efficacy...............................................37
Application of self-efficacy during treatment..............................38
Social Relationships and Health....................................................39
Social support.............................................................40
Application of social support in health....................................42
General Health Status..............................................................44
Appraisal of health........................................................45
Application of self-rated health...........................................46
Theoretical Summary.....................................................................46
Gaps Filled........................................................................47
vii


III. EXPERIMENTAL DESIGN AND METHODS
48
Research Design.......................................................................48
Subset of the RCT................................................................50
Mixed Method Design..............................................................51
Data Collection..................................................................53
Sampling and participants................................................53
Subject randomization, baseline examination, and treatment procedures....53
Subjects.................................................................55
Research Question and Specific Aims..................................................58
Measures and Constructs..........................................................59
Measuring pain: Numeric Pain Rating Scale (NPRS).........................59
Measuring disability: Oswestry Disability Index (ODI)....................59
Measuring self-efficacy: Low Back Activity Confidence Scale (LoBACS).... 60
Measuring social support: Medical Outcomes Survey Social Support Scale
(MOS SSS)................................................................60
Measuring perceived health status: Medical Outcomes Survey Short Form 36
(SI-36)..................................................................61
Data Collection Aims 1 and 2: Quantitative.......................................63
Data Analysis Aims 1 and 2: Quantitative.........................................64
Bivariate regression analysis............................................65
Multiple regression analysis.............................................65
Sample size and power analysis...........................................67
Data Collection Aim 3: Qualitative...............................................68
Data Analysis Aim 3: Qualitative and Combined Methods............................68
IV. RESULTS.............................................................................71
Descriptive Results..................................................................71
Dependent Variables..............................................................73
viii


Rating of pain...........................................................73
Rating of disability.....................................................73
Independent Variables............................................................76
Aim 1: Entry to Treatment.............................................................78
Aim 1: Hypothesis Testing A & B..................................................78
Bivariate Regression Analysis: Disability and Pain at Entry to Treatment.........79
Aim 1: Hypothesis Testing C & D..................................................80
Multiple Regression Analysis: Disability and Pain at Entry to Treatment..........81
Aim 2: Post-Treatment.................................................................83
Aim 2: Hypothesis Testing A & B..................................................83
Bivariate Regression Analysis: Disability and Pain Post-Treatment................84
Aim 2: Hypothesis Testing C & D..................................................85
Multiple Regression Analysis: Disability and Pain Post-Treatment.................86
Aim 3: Qualitative and Mixed Methods Results..........................................88
Stressors: Pain and Disability...................................................88
Entry to Treatment: Greater Severity of Pain and Disability......................89
Physical limitations.....................................................90
Lacking confidence in tasks and activities...............................91
Lacking control..........................................................93
Reduction in social activities...........................................94
Feelings of vulnerability................................................95
Poor mental state........................................................96
Frustration about needing support........................................97
Financial limitations....................................................98
Post-Treatment: Improving Outcomes of Pain and Disability.......................101
Greater Confidence......................................................102
IX


Knowing limitations
103
Learning how to manage..................................................104
Push through attitude...................................................106
Having support..........................................................108
Positive outlook........................................................Ill
Quantitative and Qualitative: Combined Results.......................................113
V. DISCUSSION..........................................................................116
Key Findings: Quantitative Summary...................................................116
Key Findings: Combined Summary.......................................................119
Entry to Treatment: Profde for Increased Pain and Disability....................120
Perceived physical health...............................................120
Perceived mental health.................................................124
Social support..........................................................125
Self-efficacy...........................................................126
Entry to Treatment: Emergent Theme..............................................128
Financial limitations...................................................128
Post-Treatment: Improving Outcomes of Pain and Disability.......................128
Perceived physical health...............................................128
Self-efficacy...........................................................131
Social support..........................................................132
Post-Treatment: Emergent Themes.................................................134
Push through attitude...................................................134
Positive outlook........................................................135
VI. CONCLUSIONS........................................................................136
Limitations.........................................................................137
Significance and Future Directions...................................................142
x


REFERENCES..................................................................144
APPENDIX....................................................................168
A. Manual of Standard Operating Procedures (MSOP)....................168
B. IRB Approvals.....................................................183
C. Informed Consent Form.............................................186
D. Study Questionnaires..............................................200
E. Interview Guide...................................................221
xi


LIST OF TABLES
TABLE
3.1 Inclusion and Exclusion Criteria for the Nonsurgical RCT with Patients with LSS.... 54
3.2 Data Collection Summary........................................................ 64
4.1 Demographic Characteristics of the Sample of Patients with LSS................. 72
4.2 Pain Rating in Patients with Lumbar Spinal Stenosis Pre and Post-Treatment.... 73
4.3 Disability Rating in Patients with Lumbar Spinal Stenosis Pre and Post-Treatment.... 74
4.4 Means, Standard Deviations, and Paired T-test for Disability and Pain.......... 75
4.5 Correlation Analysis Results for Disability.....................................75
4.6 Correlation Analysis Results for Pain.......................................... 76
4.7 Levels of Self-efficacy and Social Support at Entry and Post-Treatment..........77
4.8 Means, Standard Deviations and Paired T-test of Predictor Variables of Self-efficacy,
Social Support, Physical Health and Mental Health...............................77
4.9 Means, Standard Deviations and Paired T-test for Predictor Variables of Subscales of
Self- Efficacy and Subscales of Social Support................................. 78
4.10 Bivariate Linear Regression Results for Greater Disability at Entry to
Treatment.......................................................................80
4.11 Bivariate Linear Regression Results for Higher Pain at Entry to Treatment...... 80
4.12 Model 1: All Variables Contribute to Greater Disability at Entry to
Treatment.......................................................................81
4.13 Model 2: All Variables Contribute to Higher Pain at Entry to Treatment..........82
4.14 Model 3: Two Variables Optimally Contribute to Higher Pain at Entry to
Treatment.......................................................................82
4.15 Model 4: All Variables Contribute to Higher Pain at Entry to Treatment..........83
4.16 Bivariate Linear Regression Results for Predicting Reduced Disability Post-
Treatment.......................................................................85
4.17 Bivariate Linear Regression Results for Predicting Reduced Pain Post-
Treatment........................................................................85
xii


4.18 Model 1: All Variables Contribute to Prediction for Reduced Disability Post-
Treatment...................................................................... 86
4.19 Model 2: Two Variables Have Clinical Relevance in Predicting Reduced Disability Post-
Treatment...................................................................... 87
4.20 Model 3: Two Variables Have Clinical Relevance in Contributing to Prediction for
Reduced Pain Post-Treatment.................................................... 87
xiii


LIST OF FIGURES
FIGURE
2.1 Stress and Coping Model and stressful health conditions........................ 17
2.2 Biopsychosocial model of pain.................................................. 23
2.3 The Stress-Diathesis Model of pain and disability.............................. 24
2.4 The modified Disablement Model................................................. 27
2.5 International Classification of Functioning, Disability, and Health framework.. 32
2.6 Social support interactions with Stress and Coping Model....................... 41
3.1 Design of the randomized clinical trial for non-surgical treatment of LSS...... 49
3.2 Longitudinal, cohort study design using mixed methods..........................51
3.3 Subject enrollment, consent, treatment retention, and follow-up................57
3.4 Summary of SF-36 subscales and domains.........................................62
4.1 Oswestry Disability Index score at entry to treatment in patients with LSS.....74
4.2 Physical, mental, social and environmental limitations contributing to stressor of
pain and disability............................................................ 90
4.3 Factors related to financial limitations....................................... 100
4.4 Factors related to improved outcomes........................................... 102
4.5 Factors increasing severity of pain and disability at entry to treatment....... 114
4.6 Factors mediating the reduction of pain and disability post-treatment..........115
xiv


LIST OF ABBREVIATIONS
ADLs BDI COMIRB ESI EXSE FSE ICF IRB LBP LoBACS LSS MCID MCS MOS SSS MR I MrOS NPRS NPRSbb NPRStl OA ODI PCS PT RA RCT SCT SF-36 VAS WHO Activities of Daily Living Beck Depression Index Colorado Multiple Institutional Review Board Epidural steroid injection Exercise Self-Efficacy subscale of the LoBACS Functional Self-Efficacy subscale of the LoBACS International Classification of Functioning Institutional Review Board Low back pain Lower Back Activity Confidence Scale Lumbar spinal stenosis Minimal Clinically Important Difference Mental Component Summary of the SF-36 Medical Outcomes Survey Social Support Scale Magnetic resonance imaging Osteoporotic Fractures in Men Study Numeric Pain Rating Scale Numeric Pain Rating Scale back and buttock subscale Numeric Pain Rating Scale thigh and leg subscale Osteoarthritis Oswestry Disability Index Physical Component Summary of the SF-36 Physical therapy Rheumatoid arthritis Randomized clinical trial Social Cognitive Theory Short Form 36 from the Medical Outcomes Survey Visual Analog Scale World Health Organization
xv


CHAPTER I
BACKGROUND AND SIGNIFICANCE
Low Back Pain
Low back pain (LBP) occurs at epidemic proportions in the United States, affecting 60-
90% of individuals during their lifetimes.1 Total healthcare expenditures in the U.S. incurred by
patients with LBP are 90 billion dollars annually, 60% higher than in individuals without LBP.2
Next to the common cold, the complaint of LBP is the most common reason individuals visit a
physicians office.34 Approximately 40% of individuals have experienced LBP for a day or more
within the last 12 months, with 40% of them having consulted their primary care provider.
Because of the substantial impact of LBP on healthcare, there is a need to identify effective
management strategies to reduce the pain and disability associated with diagnoses of LBP.
Lumbar Spinal Stenosis
A US national survey of physician visits identified back pain as the most commonly
reported musculoskeletal symptom and third most frequently reported symptom in patients over
75 years of age.5 Lumbar spinal stenosis (LSS) is a subgroup of LBP that clinicians can identify
through age and symptom related factors. LSS is a prevalent and disabling musculoskeletal
condition in the aging population that often results in substantial stress and physical burden for
individuals with the disorder, and is associated with significant healthcare costs.6"8 Physicians
diagnose LSS in an estimated 13-14% of patients who seek help from a specialty physician, and
3-4% who seek care from a general practitioner for LBP.8,9 Elderly patients with LSS as a source
of their back and leg pain suffer from significant functional disability,1011 and many eventually
seek surgical intervention. The Osteoporotic Fractures in Men Study (MrOS) examined a multi-
center community cohort of 5995 men, 66 years and older, and found 65% of the men had at least
1 episode of LBP in the past year and 26% had clinically relevant LBP which occurred more than
1


3 times in the prior year at a pain level of moderate to severely bothersome.12 In addition, the
authors found that approximately 10% of the group with clinically relevant LBP had symptoms
that were similar to those with LSS.12
Definition, pathoanatomy, and clinical presentation of LSS. Degenerative lumbar
spinal stenosis (LSS) is defined as a focal narrowing of the spinal canal, although there are
varying opinions about the precise amount of narrowing that must occur before the canal is
considered stenotic. LSS is classified as an arthritic condition. Postaccini and other scholars apply
the general term spinal stenosis to 3 root compression mechanisms alone or in combination: (1)
disk protrusion or herniation, (2) osteotic overgrowth into the spinal canal or the foramina
through which the roots pass laterally, and (3) vertebral slippage or spondylolisthesis.13"16
Because degenerative changes are the predominant etiology of LSS, affected individuals
tend to be older, typically at least 50 years of age, with a prolonged history of low back pain.17'19
Leg pain, either unilateral or bilateral, is also reported in about 90% of patients seeking medical
attention.19 Chronic compression of the spinal nerve roots can also lead to sensory, reflex and/or
strength changes in the lower extremities. Fifty percent of patients report these neurological
changes.20 Acute cauda equina syndrome is rare, but clinicians have reported it.21
The hallmark finding of patients with LSS is the postural-dependency of the symptoms.
Patients have reported that symptoms typically worsen with standing or walking, and can be
relieved or diminished by sitting or bending forward.22"24Neurogenic claudication frequently
accompanies LSS, defined by clinicians as poorly localized pain, paresthesias, and cramping in
one or both lower extremities that is of a neurological origin, brought on by walking and relieved
by sitting.25 Walking tolerance can become substantially limited in patients with LSS, with pain
and reduction in walking capacity often cited as a reason for seeking medical intervention.25-27
Incidence and impact of LSS. Although there is literature on the burden of disease of
LBP in general, only 1 study has addressed the societal burden of disease related to functional
loss and disability specifically for patients with spinal stenosis. A Swedish study by Johnsson,28
2


that defined spinal stenosis as a canal of 11 mm or less, described the annual incidence of LSS
among patients referred to orthopedic departments as approximately 50 per 100,000 inhabitants.
This 1995 population study of 2 regions in Sweden reported that, with a LSS incidence of about
50 per 100,000, between 42% and 58% of these patients had claudication (leg pain or weakness
upon standing or walking). From these data, Johnsson calculated the incidence as about 25 per
100,000 inhabitants for spinal-stenosis-associated claudication. More severe stenosis can result in
cauda equina syndrome, which is characterized by the loss of sexual function and urinary and/or
fecal incontinence. This same study reported that cauda equina syndrome had an incidence of less
than 1 per 100,000. If untreated, clinicians commonly believe severe LSS has the potential to lead
to symptoms that become permanent and unresponsive to medical or surgical treatment.111329 34
Review articles233536 and textbooks3237 provide some evidence that patients with
symptomatic spinal stenosis typically have chronic LBP and pain and weakness in the legs that
limits functional tasks such as standing and walking to brief durations and short distances. These
physical impairments place stressors on the individual, particularly limitations on the ability to
carry out self-supporting, daily activities as well as work, social, and recreational activities, that
would be defined as sources leading to disability in the person with LSS.
Patients with LSS often report decreases in physical health.38"40 Decreased physical health
either real or perceived can impact an individual in many ways. Overall, it is known that those
who have decreased physical health reduce or eliminate participation in daily living, social, and
recreational activities. A lack of activity has been shown to lead to other stressors such as obesity
and general physical deterioration that may eventually result in further disability with the onset of
cardiovascular and other serious health problems.41 Activity restrictions may also lead to low self-
confidence, fear or avoidance behaviors, depression and other psychological problems that further
restrict the person with LSS from participation in activities of daily living (ADLs), recreational
activities, social activities, and community functions.
3


Limitations in the current literature for LSS. While the substantial societal impact of
LSS is apparent, there continue to be controversies in the literature regarding the effectiveness of
both surgical and non-surgical management of LSS.5n1' 4+47 Suboptimal research designs along
with other factors such as incomplete reporting of patient characteristics and unmeasured
psychosocial factors that may influence coping behaviors in managing the stressors of pain and
disability account for some of the gaps in current research on LSS. The interaction between the
stressors of pain and limited physical function, limited participation in activities, and altered
patient behaviors in relationship to intrapersonal and interpersonal coping factors is complex and
has been poorly studied in patients with LSS. This research study fills an important gap in the
literature by using a novel approach to evaluate the relative contribution of self-efficacy, social
support, and self-reported physical and mental health when managing the stressors of pain and
disability in LSS.
Rationale
The Institute of Medicine has recognized LBP as atop 15 priority condition, calling for
health care organizations to develop new evidence-based care process models.48 Current practice
guidelines for LBP in primary care recommend a stepped care approach with an initial treatment
of education and advice that focuses on remaining active throughout care. For a subgroup of
patients with LBP diagnosed with LSS, evidence shows success for conservative treatment that
includes informational education, advice for physical activity, epidural steroid injections and
physical therapy.49"51 Identifying the multiple factors resulting in heightened pain and disability is
the first goal in clinical management. Identifying factors for successful clinical outcomes is the
next goal. Success in treatment is measured by reduction in pain and disability for this subgroup
of patients with LSS. Several studies have indicated that psychosocial factors such as depression
and fear avoidance behaviors are negative influences on the stressors of pain and disability and
result in declines in treatment success in LBP in general.52'54 Studies have shown that depression
4


and fear avoidance behaviors correlate with reduced participation in functional daily tasks
indicating higher risk for disability. Studies have also shown that depression and fear avoidance
scores can improve during LBP management and result in positive changes in patient outcomes of
pain and disability.5556,51 Overall, research has demonstrated that using an enhanced or
multimodal team approach to treatment results in improved patient outcomes for LBP.51,57
Although the general LBP literature can provide some insight into management of LSS,
identifying factors contributing to higher pre-treatment pain and disability as well as determining
what factors help improve treatment outcomes for patients with LSS continues to be incomplete.
Due to a lack of reporting and a limited understanding of what influences the stressors of
pain and disability, there are gaps in LSS literature about the influence of psychosocial factors on
the severity and outcomes in LSS. Rarely has LSS research examined the intrapersonal and
interpersonal influences in this chronic condition. No other study to date has examined the
relative contribution of self-efficacy and social support in predicting the severity and outcomes of
pain and disability in patients with LSS. As the population continues to age and the provision of
medical care for patients with LSS requires considerable healthcare resources, it is important to
identify and understand factors that impact pain and disability and thereby improve successful
treatment of patients suffering from this spinal disorder.
Research Question and Specific Aims
The current study addressed the following research question: What is the relative
contribution of self-efficacy, social support, perceived physical health, and perceived mental
health to pain and disability in individuals with LSS? Three Specific Aims guided the
investigation of this overarching research question.
5


Aim 1: To determine the relative contribution of self-efficacy, social support, self-rated
physical health, and self-rated mental health to the severity of pain and disability among
patients with LSS upon entry to treatment of LSS.
H1A: Lower levels of self-efficacy, social support, perceived physical health and
perceived mental health will each independently predict higher disability at entry to
treatment.
H1B: Lower levels of self-efficacy, social support, perceived physical health and
perceived mental health will each independently predict higher pain at entry to treatment.
H1C: Lower levels of self-efficacy, social support, perceived physical health and
perceived mental health will together predict higher disability at entry to treatment.
HID: Lower levels of self-efficacy, social support, perceived physical health and
perceived mental health will together predict higher pain at entry to treatment.
Aim 2: To determine the relative contribution of self-efficacy, social support, self-rated
physical health, and self-rated mental health to the outcomes of pain and disability among
patients with LSS in the year following treatment of LSS.
H2A: Higher levels of self-efficacy, social support, perceived physical health and
perceived mental health will each independently predict lower disability after treatment
of LSS.
H2B: Higher levels of self-efficacy, social support, perceived physical health and
perceived mental health will each independently predict lower pain after treatment of
LSS.
H2C: Higher levels of self-efficacy, social support, perceived physical health and
perceived mental health will together predict lower disability after treatment of LSS.
H2D: Higher levels of self-efficacy, social support, perceived physical health and
perceived mental health will together predict lower pain after treatment of LSS.
6


Aim 3: Explore how self-efficacy, social support, perceived physical health and perceived
mental health contribute to a patients lived experience with LSS and influence meanings of
pain and disability.
Research Goals
The goal of this dissertation research project was to determine the relative contribution of
self-efficacy, social support, and self-rated physical and mental health to pain and disability in
patients with LSS upon entry to treatment and long term post-treatment. This goal requires a
greater understanding of the role of self-efficacy, social support, and self-rated health beyond
what currently exists in the literature for patients with LSS. This project also sought to uncover
additional unknown factors that contribute to the stressors of pain and disability and identify any
additional coping resources that might be available to patients with LSS. These research findings
will improve the understanding of how stressors, particularly pain and disability, are appraised in
a population of individuals with LSS. In addition, this research will help identify the role of self-
efficacy, social support, self-rated physical health and self-rated mental health in coping with LSS
and determine their contribution in predicting outcomes of pain and disability. Results from this
study will help direct clinical management and future research in this patient population.
7


CHAPTER II
REVIEW OF THE LITERATURE
Pain and Disablement in LSS
Research illustrates that the LSS condition results in significant stressors for individuals
with LSS in the form of pain and disability.5859 Research also confirms that after treatment, many
with LSS continue to have multiple stressors related to pain and disability with basic activities
such as standing and walking.60,61 Similar to those living with chronic conditions such as cancer,
osteoarthritis, and diabetes, it is likely that persons with LSS experience challenges with
psychological, social, medical, and financial limitations in addition to basic physical functioning
limitations.59,62 65
It is clear that somatic symptoms related to body structure or function has the most
recognition in the current literature for LSS. Low back pain, leg pain and spinal canal narrowing
are the top 3 reported factors contributing to disability in the LSS population.6,39,58,60,66 Severity
of LSS is almost exclusively determined by pain and spinal canal narrowing on imaging and are
the most recognized factors for surgical decision making.67,68 However, imaging of spinal canal
narrowing has not been found to be a reliable way to identify severity in LSS.61,69,70 Also,
problems with other somatic factors in addition to pain have been reported including neurogenic
claudication, neurological changes, decreased lumbar spine motion particularly in extension, and
decreased range of motion of most of the joints and soft tissues in the lower extremity and
trunk.40,51
Alterations in activity due to pain and weakness are frequently reported factors
contributing to disablement in LSS. The top 3 reported factors related to changes in activities or
participation includes limitations in walking and standing but increased activity of sitting.6,59,60,66
These 3 activity/participation factors have become part of the clinical diagnosis for LSS.39 Other
8


functional problems seem to be present such as stair climbing or general activities of daily living
(ADLs) but these have been widely under reported.
Identification and measurement of pain and disability in patients with LSS has been
almost exclusively documented by physical examination or patient self-report. Condition-specific
instruments have been used often in the LSS literature to indicate the severity of disability by
quantification with measures like the Oswestry Disability Index (ODI) and the Roland-Morris
Disability Questionnaire.7172 In addition, more generic health status instruments have been
frequently used to report general health in LSS such as the Medical Outcomes Study Short Form
36 (SF-36) and the Nottingham Health Profile.7374 Pain specific instruments such as the Numeric
Pain Rating Scale (NPRS), the Visual Analog Scale (VAS) and other dimension specific
instruments such as the Beck Depression Index (BDI) have also been used to identify specific
issues in patients with LSS.75"78 All of these condition-specific, general, and dimension specific
instruments have collectively been used to address the stress related experience of pain and
disability in patients with LSS although with little standardization in the use and comparison
among studies.
Previous Literature: Psychosocial Factors
Previous research on patients with arthritis and other health conditions has identified that
persistent psychosocial needs can decrease the effectiveness of medical treatment, general health
status and quality of life while increasing health care costs.647980 Although unidentified in the
LSS literature, unaddressed personal and social needs may contribute to reduced compliance with
treatment and follow-up recommendations, diminished self-care, and reduced overall health
management.81 Similar to other health conditions, the daily lives of patients with LSS likely
involve unanticipated challenges. Patients with LSS must leam to navigate life with the stress of
chronic LBP, cope with continued side effects of reduced functioning, experience continued
9


limitations to their physical and social abilities, manage through decreased activities and abilities,
and face fears about recurrence or worsening of their condition.
The American College of Physicians and the American Pain Society published clinical
guidelines that recommend clinicians conduct a focused history and physical exam with patients
with LBP. This focused history is recommended to include assessment of psychosocial factors
which predict risk for chronic disabling back pain.82 Psychosocial factors and emotional distress
have been found to be stronger predictors of LBP outcomes when compared to physical exam
findings or severity of pain alone.83 Psychosocial factors that predict poorer LBP outcomes
include presence of depression, passive coping strategies, higher disability levels, or
somatization.83,84 Psychosocial factors often delay recovery and identification can help to target
interventions. It is recommended that patients with chronic LBP who do not improve with self-
care options would benefit from non-pharmacological therapy including interdisciplinary
rehabilitation, exercise therapy, spinal manipulation, and cognitive-behavioral therapy.82
Only a few studies have examined the presence of psychosocial factors and the
relationship to severity of pain and disability in patients with LSS. Moreover, only a few have
examined the association between psychosocial factors and outcomes of treatment for patients
with LSS. Depression, low sense of coherence, life dissatisfaction, and fear or avoidance
behaviors have all been described to be present in baseline examination of some individuals with
Lss77-85-86 havc been linked to worse treatment outcomes in LSS.65,78,87,88 Specifically,
Wood et al86 identified that patients with LSS have higher fear of reinjury and activity avoidance.
Individuals with higher depression scores also had greater disability at baseline76 and had worse
surgical outcomes60,77 in the treatment of LSS. In one study, Turner et al76 found that lower
depression after surgical treatment was related to reduced disability scores but not pain scores.
However, Katz et al89 found that depression scores pre-treatment were associated with higher pain
scores 6 months post-surgical intervention. In another study by Katz et al,90 higher depression
10


scores were also associated with poorer patient satisfaction post-surgery in bivariate but not
multivariate analysis.
Several studies examined the relationship between life dissatisfaction and other clinical
baseline measures and post-surgical outcomes in LSS.76788588 Sinikallio et al78 found that 25%
of the pre-operative patients with LSS reported life dissatisfaction. In the general population, life
dissatisfaction has been recorded in 13% of the healthy population and in 25% of those with
illnesses.91 Therefore, patients with LSS have more life dissatisfaction than the general population
but are similar to patients with other chronic illnesses. Sinikallio et al85 also found that
dissatisfied patients with LSS had lower coping resources, elevated alexithymia, and elevated
depression scores. Two follow up studies by Sinikallio et al7788 found that both pre-operative and
post-operative life dissatisfaction was independently correlated with disability and sense of
coherence, a measure of coping. Pain, however, was not associated with life dissatisfaction either
pre-operatively or post-operatively.
Patients with LSS remain a vulnerable population who often find themselves in need of
general information, guidance, and social support.92 Only 1 study by Katz et al90 examined social
support in patients with LSS using a single question: 1-5 Likert scale related to the amount of
contact with friends and family, ranging from daily contact to no contact in the past month. The
results of this study did not find significant correlations between pre-operative contact amounts
with friends and family and post-operative outcomes of pain from surgery for LSS.90
Previous Literature: General Health Status
General health status and health related quality of life has been minimally investigated in
the population of patients with LSS. General or self-rated health measures have often been related
to quality of life. Battie et al93 examined health related quality of life and comorbidities associated
with LSS and noted that individuals with LSS have health related quality of life issues 4 times
that of the general population. Gunzburg et al94 identified that patients with LSS have high
11


baseline rates of non-organic symptoms reflecting the potential significance of illness behaviors
in this population. In addition, these authors found that better pre-operative self-rated health
(scale: excellent, good, fair, poor) was associated with better post-surgical outcomes of walking
capacity, symptom severity, and treatment satisfaction.
In the general LBP literature, Fanuele et al10 investigated the Medical Outcomes Survey
SF-36 physical component scores (PCS) in a prospective sample of 17,774 patients with spinal
disorders. They found that the patients with LBP had a mean PCS score of 30.4 9.95 (SD)
compared with 50.0 10.00 for the general United States population. This indicates that the PCS
score is greatly reduced in patients with spinal disorders in general and is similar to other patient
population PCS scores such as chronic heart failure (31.0), Chronic Obstructive Pulmonary
Disorder (33.9), Systemic lupus erythematosus (37.1), cancer (38.4), primary total hip
arthroplasty (29.0), primary total knee arthroplasty (32.6), and glenohumeral degenerative joint
disease (35.2).10
Individuals with LSS often self-report worse physical health when compared to the
population at large. In the MrOS study, Vogt et allz found that men with LBP symptoms similar
to LSS had SF-12 self-reported scores for the physical component summary (PCS) of 38.3 out of
100, ranking them at the 14th percentile for physical health compared to the US population with a
mean ranking of 50th percentile.12,73 In addition, patients with LSS enrolled in the National Spine
Network registry had PCS scores of 32.92 out of 100, ranking them in the 8th percentile for
physical health compared to the US population with an mean ranking of 50th percentile.10 Patients
in an RCT investigating both surgical and non-surgical intervention found patients with LSS to
have mean SF-36 PCS values between 34.3 and 35.4 22 (SD) compared with 50.0 10.00 for
the general United States population indicating similarly low PCS scores comparable to the other
studies sited.38 Although there appears to be an association between LSS and self-rating of
physical health, this relationship and its impact on pain and disability is not clearly understood.
12


Pahl et al95 examined the impact of health status in 4,442 patients with LBP including a
subpopulation of patients with LSS. Although not specific to LSS, they found that self-ratings of
physical health measured by the PCS, ranked patients with LBP lower than normative values for
matched age groups. Pahl et al also found that individuals over 40 years old with LBP also had
poorer mental health status and declines in vitality or energy and other social functioning domains
compared to their matched age group norms. Overall, this evidence illustrates that both perceived
physical and mental health are likely contributing factors to the outcomes of pain and disability in
LSS and warrant further investigation.
Previous Literature: Research Design in LSS
Literature in LSS almost exclusively uses quantitative methods for determining health
care outcomes, providing minimal depth of understanding about sources for and appraisal of
stressors as well as coping resources available to those with LSS. Standardized measurements
cannot capture the complete experience of pain and disability with LSS. Qualitative methods can
provide a more detailed account of a patients experience with a stressful health condition like
LSS and provide insight into the coping resources available to those individuals. These insights
can help direct health care resources for treatment of individuals with LSS.
Although several authors in the current literature indicate they used qualitative methods,
these authors all describe qualitative assessment by a clinician for MRI imaging, disc protrusion
measurement, and nerve root involvement. Two authors described qualitative assessment as self-
report of the number of analgesic medications used during the study,96 and self-report of
limitation in walking distance (yes/no) along with the self-report of limitation in ADLs performed
(yes/no).97 Two additional studies mentioned the use of patient interviews for measurement of
post-operative outcomes. One of these studies, Sharma et al97 used 2 questions to inquire if the
patient was satisfied with the outcomes of surgery (yes/no), and any significant limitation in
ADLs (yes/no) although the methodology is unclear. The other study, Slosar et al98 interviewed
13


patients using questions from a standardized outcomes measurement tool but added no additional
information from these interviews beyond the survey questions. Orlin et al" used a novel
approach in preoperative interviews asking 209 patients with LBP (50 with LSS) 8 questions
about bladder, anal and sexual function and dermatome sensitivity in the only study published on
urinary voiding dysfunction with patients undergoing lumber spine surgery from both lumbar disc
lesions and spinal stenosis. Results from this study illustrate that 68% of these patients
undergoing lumbar spine surgery have baseline urinary voiding dysfunction. This study clearly
illustrates that patients with LSS have other concerns other than just standing and walking
difficulty.
Only 1 additional study by Sekiguchi et al100 actually used a more comprehensive
qualitative methodology with patients with LSS in designing a new questionnaire to rate
symptoms of LSS based on quality of life constructs. In this study, a team of 2 orthopedic
specialists and 1 psychometric specialist devised several constructs of the new scale and
operationalized the constructs as they related to the overall measurement of quality of life in
patient with LSS. Next, they asked 10 patients with LSS (7 men, 3 women) divided into 2 focus
groups to discuss their pain, disabilities and difficulties with ADLs. Finally, 1 group was asked to
confirm their understanding of the meaning of the items of the scale and if it appropriately
described what was bothering the patient. The results confirmed specific quality of life issues for
patients with LSS that include: 1) limitations on going out, 2) sleeping problems, 3) burden
imposed by hospital visits and treatment, 4) urination problems, 5) walking problems, 6)
limitations in everyday activities, and 7) emotional burdens. This study adds unique clinical
implications that behavioral concerns are important to patients with LSS. Unfortunately this
symptom scale has not been tested nor is it used in clinic or in LSS research.
14


Previous Literature: Summary
While the substantial societal impact of LSS is apparent, there are controversies in the
literature regarding the identification of subgroups according to patient characteristics,
effectiveness of treatment, and general management in LSS.5n15' 4+47 Suboptimal research
designs also account for some of the gaps in current research on LSS, along with other factors
such as incomplete reporting of many patient characteristics, poorly measured psychosocial
factors that may influence treatment as well as outcomes of treatment, and limited research design
leading to a lack of understanding of the larger issues of pain and disablement in this population.
Rarely has research in this area evaluated the contribution of intrapersonal and interpersonal
factors such as self-efficacy and social support or explored the meanings of self-rated health in
understanding the experience of pain and disability in the management of LSS.
Theoretical Perspectives
Integrating Theory in LSS
Like people with other medical conditions, individuals with chronic and recurring LSS
may need to adjust their habits and lifestyles while trying to maintain basic physical, social,
vocational, and recreational activities. To manage the condition of LSS, patients must try to
understand the nature of their problem, create self-care strategies for dealing with pain flare-ups,
manage and attempt to overcome functional problems, and identify and utilize available support
and resources wisely. This process underscores the behavioral component of coping with LSS. To
date the general LSS literature lacks a theoretical foundation for this behavioral component,
particularly in comparison to research on other areas of chronic disease.101"106 Therefore, this
study was developed from a theoretical perspective to better understand the traditionally
biomedical characterized condition of LSS. This study was guided by the Stress and Coping
15


Model107 to identify predictors associated with 2 main sources for stress in individuals with LSS:
pain and disability. In addition, this study was designed using mixed methodology with the aim to
uncover the appraisals and coping mechanisms identified and used by individuals with this
disabling condition.
Theoretical Framework
The present study used the underlying theoretical concepts in the Stress and Coping
Model107 to define both stress and coping as well as identify the relationship of these concepts in
the condition of LSS. Two primary stressors have been identified in previous LSS literature: pain
and disability. Differing experiences with LSS result in different appraisals of the stressors of
pain and disability. Individuals with LSS also have different outcomes related to the experience
of pain and disability. Two guiding theories will be used to understand the experience of pain and
disability in patients with LSS including the Stress-Diathesis Model108109 and the ICF Framework
of Health and Disablement.108110 Finally, the Stress and Coping Model107 will be used to identify
what coping behaviors and resources preserve well-being in patients with LSS while facing
stressful experiences with this chronic condition. Specifically, the study will explore the construct
self-efficacy proposed as a personal resource in LSS. The concept of self-efficacy originates from
Banduras work in Social Cognitive Theory.111112 A second construct, social support, will be
examined as a social resource for health as identified by Lazarus and Folkman in the Stress and
Coping Model.107 and expanded in health application by Schwarzer and Leppin.113 Finally, the
constructs of self-rated physical and mental health will also be explored from the perspective of
the Stress and Coping Model.107
Stress and Coping Model
Within the Stress and Coping Model,107 stress involves the relationship between an
individual and his or her environment. This relationship or transaction between individuals and
16


their environment indicates that stress is more than an external stimulation or specific pattern of
physiological, behavioral, or subjective reactions (see Figure 2.1). This relationship also views 2
processes as mediators within the person-environment transaction: cognitive appraisal and coping
effort. Cognitive appraisals and coping efforts are influenced by moderators such personal and
situation factors that result in the outcome of adaptation of the individual on many levels
impacting their health.
Adaptation
*emotional well-being,
funcitonal status, health
behaviors
, t A
1
Moderators
*coping styles
*coping behaviors
*optimism
information seeking
*social support
T
Meaning-
Based Coping
*positive emotion,
beliefs, events
*positive reappraisal
*revised goals

Figure 2.1. Stress and Coping Model and stressful health conditions. Adapted from Glanz
114
17


Cognitive appraisal. The cognitive appraisal or evaluation by the individual is
paramount to determining if the stress is threatening to his/her well-being or surpassing his/her
resources. Upon appraising a stressor, the theory asserts that people engage in coping, defined as
fluctuating behavioral exertions, in an effort to manage that stressor. The cognitive appraisal
process helps an individual determine both the controllability and availability of coping resources
identified as necessary to manage the stressor(s).107
Cognitive appraisal is a necessary component of dealing with a stressor. It accounts for
the different ways in which individuals react to similar events. Health conditions such as those
resulting in pain provocation and disability like LSS can cause stress in nearly every person, yet
people vary in their reactions and interpretations of the same event and condition.107 A health
condition is a general stressor but the different ways individuals respond to the same health
condition can depend on their cognitive appraisals. This variability in cognitive appraisals can
change a persons level of vulnerability during a stressful health condition. Vulnerability is
closely related to appraisal since vulnerability increases as a person appraises that he/she has
reduced coping resources available. Vulnerability reflects the inability of individuals to withstand
adverse impacts from multiple stressors to which they are exposed. It can be associated with a
pattern of thought that is believed to predispose the individual to psychological problems and
feelings of hopelessness. In addition, the variability of individual appraisals as well as his/her
responses helps to explain why some individuals experience similar health conditions but have
differing quality, intensity, and duration of physical, social, and emotional outcomes.
Personal and situational factors. Two types of factors influence the cognitive appraisal
process: personal factors and situational factors.107 Personal factors consist of the personal values
that motivate individuals to make certain decisions and beliefs that give the individual a personal
sense of control. Therefore, at the individual-level, commitments and beliefs are all part of an
individuals process of appraisal. Commitments are expressions of what is important to a person
and can be related to vulnerability. The deeper a persons commitment, the greater potential for
18


threat, but also the greater the push toward ameliorative action and hope. Beliefs are also
important in determining how a person evaluates a stressful event or health condition. Beliefs of
personal control over situations can relate to how an individual believes internal self-responses to
situations can be controlled. General control beliefs relate to the extent that the person believes
the health related outcomes can be controlled.
Another influence to the appraisal process is situational factors which play a critical role
in determining the external controllability of the stressor and what ameliorative action can be
taken.107 Situational factors can include predictability and uncertainty, temporal and life course
factors, and ambiguity. Because patients with LSS are often in their later stages of life, aging
related concerns are a component of situational factors in LSS. Situational factors can also
influence how and to what extent the stressor can be managed. Maximum uncertainty is often
extremely stressful for an individual experiencing a health condition like LSS. The uncertainty
can have immobilizing effects on anticipatory and actual coping processes and can cause mental
confusion in the individual. Overall, it is important to evaluate both situational factors and
personal factors in order to understand what ultimately influences the level of stress related to
pain and disability in patients with LSS.
Different appraisals. While differences between individuals under similar
circumstances are inevitable due to various personal and situational factors, the Stress and Coping
Model emphasizes that all individuals evaluating a stressor undergo a cognitive appraisal process
involving primary appraisal, secondary appraisal and/or reappraisal.107 Primary appraisals involve
assessment of the magnitude and significance of a stressor or traumatic event. During primary
appraisal, the individual will assess the actual harm, loss, threat, or challenge that must be
encountered with a stressful health condition. When an individual first experiences a health
condition or a reoccurrence of a chronic condition like with LSS, primary appraisal takes place.
Individuals with LSS are often in the process of primary appraisal when they seek out medical
care and treatment.
19


Secondary appraisal refers to an individuals assessment of the degree to which the
stressor or traumatic event can be controlled and the available coping resources.107 Secondary
appraisals are a judgment about what might and can be done in the situation.107 Secondary
appraisals include an evaluation about whether a coping option will accomplish what it is
supposed to do as well as the consequences in the context of other internal and/or external
demands and constraints. Appraisals of controllability of a health condition can be stress-reducing
if the outcome is believed to be controllable or that one has the coping resources to manage the
outcomes. However, appraisals of controllability can also heighten threat and give rise to negative
emotions and beliefs about coping if control and resources are diminished. For example,
individuals with LSS will use secondary appraisals when their chronic LBP has reoccurred and
their activities have been limited. This in turn motivates some to seek treatment to control their
pain and disability as part of the coping process if they have the necessary access and resources.
Reappraisal is the final feature of the appraisal process which entails an altered or revised
version of a previous appraisal. Reappraisals can occur multiple times for reasons such as
changes in the environment or the health condition.107 Reappraisals can also occur when an
individual has gone through the cognitive coping process and has altered the assessment of the
available coping resources. Because the complete process of appraisal is dynamic, a patient with a
chronic disease such as LSS is likely to appraise and reappraise the stressors of pain and disability
and respective coping resources before, during, and after treatment multiple times and with each
successive chronic episode.
Coping. Coping is intimately related to the concept of cognitive appraisal and the
person-environment relationship. Coping involves the cognitive and behavioral efforts to address
external and internal demands on the person experiencing a stressful encounter. Coping can either
be focused on changing the person-environment problem behind the stress, or be directed toward
changing the appraisal of the situation. Moreover, coping can be focused toward trying to reduce
a negative emotional state of the situation.
20


There are 2 types of coping research in the literature: trait oriented and state oriented.
Trait oriented coping research aims to identify individuals whose coping behaviors are
diminished due to the demands of the stressful encounter. Early identification of individuals who
have reduced coping behaviors could help target certain factors and resources needed to improve
coping behaviors. For example in patients with LSS, early identification of individuals who have
negative appraisals and reduced coping can help direct clinicians to increase cognitive-behavioral
efforts to improve patient beliefs, motivation, and participation in treatment. State oriented coping
on the other hand, centers on understanding which coping strategies an individual actually
chooses and uses during stressful events. State oriented coping explores the relationship between
coping strategies and the outcome variables. Variables of interest include coping efficiency,
emotional reactions, and variables of adaptive outcome (e.g., health status) that accompanies
certain coping efforts. In patients with LSS, understanding coping behaviors in relationship to
health outcomes can be important to identify and prioritize health care resources for individuals
with this chronic condition. In general, state oriented coping research strategies tend to lay the
foundation for resource allocation and targeted programs to improve outcomes. This study
identifies a combination of both trait and state oriented coping by determining coping behaviors
at entry to treatment and actual coping efforts during the management period by individuals with
LSS. The overall intention in exploring the coping resources and behaviors of individuals with
LSS is to create a foundation for early identification of coping strategies and resource allocation
for future LSS treatment programs.
Application of the Stress and Coping Model to LBP. There are no current studies
using Lazarus and Folkmans Stress and Coping Model107 in the treatment of LSS. Previous
studies on LBP and other chronic diseases have emphasized the importance of appraisal of the
stressor, typically conceptualized as general challenges with pain management, normal function,
and coping resources (or perception of resources) available to allow for adjustment to life during
and after treatment completion.115"117 This dissertation study defined specific stressors, pain and
21


disability, within the global stressor of the health condition of LSS, as reflected in the Specific
Aims. This dissertation study aimed to further understand the stressors of pain and disability in
LSS to better identify and understand coping behaviors and resources appraised by individuals
with LSS in order to manage their pain and disability.
Managing Stressors: Pain
The experience of pain is often described as stressful by an individual and identified as a
source of stress for individuals in previous literature.118 In recognizing the role of pain as a
stressor, it is necessary to explore and expand upon an understanding of the factors contributing
to an individuals experience of pain. The traditional biomedical model of pain dates back
hundreds of years when pain was understood to be a primarily sensory experience resulting from
the stimulation of noxious sensory receptors usually from physical damage or injury.119 This
theory of pain describes primarily nociceptive pain, defined as pain elicited when sensory
receptors specialized to sense mechanical, thermal, or chemical pain react when stimulated past a
sensory threshold. This simplistic view suggests that pain only comes from specific physical
pathology and is often called the biomedical model of pain. It does not take into account how pain
is experienced by the individual, involving additional psychological, social and behavioral
mechanisms of injury and illness. Due to the narrow scope of the biomedical model of pain, it is
often criticized for being reductionist and exclusionary.120 In the condition of LSS for example,
patients with LSS often experience pain that is unrelated or only partly due to their radiographic
severity or lumbar segmental level of pathology,95121 illustrating that their pain experience is
greater than the pure pathophysiologic processes involved.
Biopsychosocial Models of Pain
To encompass a broader view of pain, Turk and Flor 120 described a biopsychosocial
approach to pain which addresses many of the shortfalls found in traditional biomedical models.
22


In this approach, it is recognized that the experience of pain is the reciprocal and dynamic
interaction of biological, psychological, and social factors. It is based on the concept that the
experience of pain arises from illness behavior although it is initiated and/or has contributions
from nociceptive pain. Illness behavior is a tenn used to describe the ways in which given
symptoms may be differently perceived, evaluated, and acted or not acted upon by different kinds
of persons.122 Illness behavior is believed to be a dynamic process that allows for the role of the
biological, psychological and social factors to change in chronic conditions and as the condition
evolves.123
In the 1980s, Waddell et al124 applied the construct of illness behavior to LBP (see
Figure 2.2). Their view was that persons with chronic LBP experienced illness behaviors and not
just nocioceptive pain. This view represented a broader biopsychosocial model of pain stemming
from physiological impairment but with broader cognition, affective and social factors resulting
in the experience of pain perceived and reported by individuals with chronic LBP.124
Figure 2.2. Biopsychosocial model of pain. Cross sectional representation of the Glasgow
model representing the role of fear-avoidance beliefs in chronic low back pain and disability.
Adapted from Waddell et al.52
23


Stress-Diathesis Model of Pain
Asmundson and Wright125 conceptualized a stress-diathesis model of pain and disability
that stems from the biopsychosocial framework and can be applied to chronic pain conditions
resulting in disability (see Figure 2.3). It is this evolved theory of pain that describes how an
initial physical pathology or injury is recognized as nociception and leads to a cycle of appraisal
of the stress and coping factors needed to manage the health encounter. Also key to this model is
the construct of diathesis which is described as a pre-dispositional vulnerability of the individual.
This vulnerability as demonstrated in the model can be influenced by both internal and external
factors. Along the pathway there are physical, cognitive, social, cultural and behavioral processes
that contribute to the cycle and associate with die pain-specific distress. The combination of the
distress and the diathesis contributes to a larger cycle of pain resulting in apprehension,
avoidance, and overall disability (see Figure 2.3).
Figure 2.3. The Stress-Diathesis Model of pain and disability. An integrated stress-diadiesis
model of chronic pain adapted from Asmundson et al.125
24


Pain appraisals. A key part of the stress-diathesis model is the issue of appraisal. Pain
appraisal refers to the meaning ascribed to pain by an individual.126 In accordance with the Stress
and Coping Model,127 a distinction can be made between primary appraisal of pain in terms of it
being threatening, benign, or irrelevant and secondary appraisal which evaluates the
controllability of pain and one's coping resources. Primary appraisal of a threat or harm/loss is
indication of a stressor. Therefore, pain as described as threatening by the stress-diathesis model
is inherently stressful. The degree of the stress depends on the other factors of vulnerability,
social and cultural influence, and assessment of coping resources.
Pain beliefs. Pain beliefs and therefore reports of the pain experience develop both
during the lifetime of an individual as well as throughout the duration of a chronic illness such as
with LSS. Appraisal and beliefs about pain can have a big impact on an individual's affective and
behavioral response to pain. If a pain signal is interpreted as harmful and appraised as a great
threat, then it may be perceived as intense and unpleasant and may evoke escape or avoidance
behavior. An example of this is the rating of pain associated with cancer described as more
unpleasant than labor pain, even though the actual rating of intensity is equivalent.128
Studies show that pain appraisal and beliefs are also determinants of adjustment to
chronic LBP.129 After treatment, a large majority of patients with LBP believed that a wrong
movement could have serious negative consequences for their back and further associated this
belief with reduced activity levels and increased disability.130 Using the Stress-Diathesis Model, it
is reasonable to expect that through the appraisal process, there are larger biopsychosocial
influences that impact perception and a patients quantification of pain. The combination of actual
physical limitations, environmental resources and barriers, and perceived physical, mental, and
emotional characteristics influencing actual behaviors by the individual may in turn impact a
persons quantification of pain.
25


Managing Stressors: Disability
As we see in the Stress-Diathesis Model, pain is only one factor in a more complex set of
factors resulting from pain but contributing to disability. Pain sensations, appraisal of the threat
from the pain, vulnerability factors and even social and cognitive factors all contribute to the
individuals experience of somatic pain in this model. An individuals responses and beliefs
related to pain can play a large role in how he/she acts or fails to act when participating in many
of lifes activities and therefore can magnify his/her level of disability.
Understanding disability has been an evolving process in health and health care literature.
Multiple models have been proposed that have developed the complex understanding we have
today about disability and the factors that contribute to disability. Understanding this complex
process of disability is an important component of this research study as this study sought to
uncover factors beyond somatic or physical factors that relate to disability in the population of
patients with USS.
Disablement Model
Over the years, multiple frameworks have been developed to explain a broader concept of
disability. One of the earliest frameworks for disability was proposed in the 1960s by the
sociologist Saad Nagi called the Disablement Model that illustrates a disease pathway that is
still used by health care professionals (see Figure 2.4). The Disablement Model describes a
pathway comprised of 4 inter-related but separate constructs that contribute in a linear fashion to
disability.131 The model starts with pathology as the underlying disease condition that eventually
leads to impairments, functional limitations, and disability. Impairments are abnormalities in
structure or function of the bodys systems, which can either be associated with pathology or
remain after the pathology has resolved. Functional limitations are limitations in a persons
ability to perform tasks and ADUs such as self-care. Therefore, limitation in performance reduces
the function of the individual. Finally, disability, as described by Nagi, is the pattern of behavior
26


that starts with pathology but develops with chronic impairments and functional limitations (see
Figure 2.4).
Figure 2.4. The modified Disablement Model. Representation of the pathway from pathology to
disability with environment contributing to both functional limitations and overall disability.
Adapted from Saad Nagis disablement model.131
Expanded Disablement Model. A component of this model that is extraneous to the
biomedical pathway of the original Nagi model is the environment. The Institute of Medicine
revised its model to include disability as a function of the interaction of individuals with the
influence of social and physical environments. As a result, the revised Disablement Model
describes environment as including the natural environment, the built environment, culture, the
economic system, the political system, and psychological factors.132 The less supportive the
environment is to an individual from both a physical and social perspective, the greater the
resulting disability. It is important to identify this transition in health literature as this initial
recognition of environment in disability marked a turning point for health care in understanding a
more complex disablement concept. This expanded disablement concept is particularly important
to this dissertation study as the focus was to examine factors pertaining to the personal and social
environment as they interact to define disability in patients with LSS.
Because the traditional Disablement Model illustrates a linear biomedical framework, it is
unclear to what degree the ability or disability of an individual is a direct result of the disease
process versus the contributing environment. Moreover, the ability to manage functional
27


limitations and disability often depends on the personal and social resources available to the
individual, but these resources are only grossly reflected by the environment in this model. When
using this model, it is difficult to direct resources or even identify specific environmental factors
that impact disability.
Verbrugge and Jette133 expanded on the Disablement Model and conceptualized
additional factors that may affect the pathway from pathology to function. Included in this
expansion was the concept of risk factors which are predisposing factors that could affect either
the presence or severity of the impairments, functional limitations or disability in the individual.
These factors could include medical care, external supportive actions or systems, activity
accommodations, or what is described as intra-individual factors such as lifestyle, behavior
changes, and psychosocial attributes such as coping strategies. These factors are of particular
interest to this research study as they identify that specific behavioral and social factors may
interact during the coping process and contribute to disablement in patients with LSS.
Moreover, Verbrugge and colleagues133134 proposed that life activities include much
more than those of the traditional medical model which included only ADLs. In this new
definition, life activities could be grouped in 3 categories: obligatory, committed, and
discretionary activities. Obligatory activities are those required for survival and self-sufficiency,
including activities such as personal care, sleep, rest, walking and local transportation traditional
referred to as ADLs. Committed activities include principle productive roles, such as paid work,
and household management, such as food preparation, repairs, yard maintenance, shopping, and
child or elder care. Discretionary activities are free-time activities that could include socializing
with family and friends, entertainment away from home, hobbies, physical recreation, volunteer
service, religious activities or other adult educational activities. Until this new conceptualization
of the Disablement Model, the majority of disability research had only focused on obligatory
activities and few committed activities but had ignored discretionary activities as important in the
disablement process.133135 These concepts of expanded activity are important to this dissertation
28


study as they identify multiple areas of activity that are important in individuals lives that when
disrupted have the potential to result in disability. This dissertation study looked closely at
limitations in activity beyond obligatory activities of daily living recognizing that social and
emotional behaviors and roles can be impacted to a greater degree during committed and
discretionary activities.
Social model of disability. An alternative model of disability, called the social model of
disability offered an opposing view that disability is primarily a socially created problem and not
an attribute of the individual.136 In this model, an unaccommodating physical environment creates
the problems of disability through negative attitudes and other negative features of the
environment. Although, the social model of disability was never intended to be an all-
encompassing model of disability, it illustrates the need for reframing the idea of disability from
the biomedical model to include issues of social interaction and social values that to contribute to
disablement. This dissertation study has emphasized that the role of social function in the lives of
patient with LSS is critical to fully understand the coping behaviors and resources of these
individuals.
ICF Framework
To combine all of these models and contributions to the concept of disability, a new
model was developed by the World Health Organization (WHO) and the Committee on a US
National Agenda for the Prevention of Disabilities. This new framework called the World Health
Organizations International Classification of Functioning, Disability and Health (WHO-ICF)
elucidates the larger biopsychosocial components of disablement. Known more commonly as
ICF, this framework is a universal classification system to define and measure health and
disability. It describes health and health-related domains from body, individual, and societal
perspectives by means of 2 domains: Functioning and Contextual Factors. Each domain can be
used to describe different changes and the impact of those changes that occur in the individual
29


and his/her usual environment due to health related states. The domains characterize different
factors that interact with the individual to facilitate or hinder functioning and/or disability.137
Changes to different factors often result in a decrement of health and thereby can be used to
describe disability in the individual.
Functioning. Similar to the Disablement Model, the ICF has components that
encompass pathology, impairment and functioning. These are seen in the domain called
Functioning. The functioning domain includes 4 components called body function, body
structure, activity, and participation. Body functions are defined as the physiological and
psychological functions of body systems. Body structures are defined as the anatomical parts of
the body. Problems in both body function and structure are together usually termed impairments
by health professionals and defined as deviations or loss of structure or function. The other 2
components of activity and participation encompass a larger concept of functioning. Activity and
participation are similar concepts although each has a slightly different definition. Activity is
defined as the action or task performance by an individual. Participation represents the
involvement in life activities. Problems in an individuals ability to participate in activities will
result in participation restrictions and limit actual activity. For individuals with LSS, activities
that are limited may include activities such as walking, climbing steps, or lifting objects. When a
person with LSS cannot perform activities such as walking or stairs, it often precludes his/her
participation in many social, recreational, and daily chores that require the functional ability and
the confidence to execute these activities.
Contextual factors. As previously seen in the Disablement Model, the ICF describes the
environment as a contributor to disability in a larger domain called Contextual Factors.
Contextual factors in the ICF are classified as either personal or environmental factors that
influence health through functioning and disability. As previously described, environmental
factors are defined as exerting an influence on external domains that in turn influence a persons
functioning and disability. Environmental factors refer to all aspects of the external or extrinsic
30


world that form the context of an individuals personal life that have an impact on a persons
functioning. Environmental factors can be both facilitators that can assist a persons functioning
or barriers that hinder or limit a persons functioning. Because of the inclusion of environmental
factors in the original Disablement Model, literature related to environmental factors in relation to
disability has been extensive. In addition, the contribution of the social model of disability in the
literature has added increased emphasis that persons with disabilities have reduced access to full
participation in life and society.138
The other type of contextual factor, personal factors, can include anything that is an
internal influence on a persons functioning and disability. This includes individual characteristics
like age, sex, social status, and life experiences.139 Personal factors have been described in the
literature as comprising 2 broad categories of demographic information and personality traits.140
Personality traits are of particular interest to the present dissertation research study. These can
include personality traits such as coping styles, lifestyle, habits, social background, past
experiences, self-esteem, self-advocacy, and other psychological assets that are not directly
related to the health condition. These personality traits can influence how a person responds to
possible limitations secondary to disability. Some individuals may have difficulty with
assertiveness or have a negative outlook, regardless of the health condition and corresponding
disability. Others may have a sense of positive optimism or self-determination that persists
regardless of the disability. Therefore, personal factors are those attributes within the person that
are not caused by or have nothing to do with the disabling condition but may influence the
persons overall disability. Researchers have described these factors as those traits that would still
be there even if the health condition and disability were suddenly gone. The recognition of
personal factors in the ICF framework illustrates that a person is more than simply the sum of
his/her physical functioning. The ICF framework and its components are displayed in Figure 2.5.
31


Health Condition
(disease or disorder)
Xif X
* ______________________ *
*
Figure 2.5. International Classification of Functioning, Disability and Health framework.
Representation of the different factors that interact and contribute to disruption of function and
increase overall disablement. Adapted from the WHO ICF framework141
Contextual factors of the person and the environment interact and modify the impact of
changes to body functions, body structures, activities, and participation of the individual with a
health condition. The degree of impact these factors have in relation to the health condition help
to describe the degree of disability in an individual with the health condition. The greater the
challenges to the persons functions or contextual factors, the greater the potential physical and
psychological stress that is placed on the individual. Applying the Stress and Coping Model, if
health related factors of the person or the environment are appraised as taxing or exceeding the
individuals resources or well-being, the health state will be viewed as stressful and therefore
disabling to the individual. Contextual factors are of particular interest to this research study as
32


they identify both person and environmental factors that are potential moderators in the
relationship between coping and outcomes of LSS. The ICF as a culmination of defining
disablement provides a framework to identify both intrapersonal and interpersonal factors that
may be interacting to exacerbate or diminish disability in patients with LSS.
Application of disablement models in LBP. The models of disablement described
above serve to illustrate the complex and multifactorial nature of disability as currently
understood by clinicians, researchers, and the public health community. As previously identified,
LBP is a significant cause of disability in our society. Waddell in 1980 revolutionized the clinical
perspective of LBP by illustrating that there were many nonorganic and behavioral components to
individuals experience of LBP.142,143 The work to revise the concept of disablement along with
continued research in LBP has resulted in a more complex understanding of the typical spectrum
of problems of functioning in patients with nonspecific LBP. A variety of studies have used the
ICF framework to assist in understanding the complexity of nonspecific LBP.144"146 Psychosocial
and environmental risk factors have been identified in the development and management of
nonspecific chronic LBP.52,124'l43'141148 Although not focused on the condition of LSS, these
studies show the complexity of chronic LBP beyond somatic symptoms and support the
multifactorial disablement models in LBP. Using the Stress and Coping Model in combination
with the ICF framework and the Stress-Diathesis Model, respective coping processes and
resources can be better identified that describe the experience of pain and disablement in patients
with LSS. Ultimately, these theoretical contributions serve to support the goal of understanding
and improving the allocation of resources to reduce the consequences of disability in persons with
LSS.
Resource Theories of Stress and Coping
In order to fully understand the experiences of pain and disablement in LSS, it is
important to investigate whether and how intrapersonal and interpersonal factors interact to
33


influence the severity and the clinical outcomes of LSS. Intrapersonal and interpersonal
influences would be classified as part of the contextual factor category in the ICF framework;
namely personal and environmental factors. To further understand how personal and
environmental factors contribute to disablement in a chronic condition like LSS, theoretical
constructs have been identified at the personal and environmental in a number of resource
theories of stress and coping. In other health conditions, social and personal constructs have been
proposed to serve as coping resources that address stressful person-environmental transactions. It
is unknown what social and personal constructs are important in coping by patients with LSS.
Four specific constructs are explored in this research study including the personal constructs of
self-efficacy, self-rated physical health, self-rated mental health, and the social construct of social
support.
Social Cognitive Theory and the Construct of Self-Efficacy
At the clinical level, it may be important to identify risk factors for heightened pain and
disability and whether individual influences affect or alter outcomes in identical treatment
protocols for patients with LSS. There is evidence that individual factors related to stress, coping,
and cognitive mechanisms can influence aging-related health in general,149 although currently no
evidence exists on the role of these individual factors on the severity or the health outcomes of
pain and disability in patients with LSS.
Social Cognitive Theory (SCT),112150 originally called Social Learning Theory by Miller
and Dollard and expanded by Bandura,112151152 addresses the psychosocial aspects of individual
health behavior as well as methods of promoting behavioral change. According to SCT,
individuals possess a self-system that enables them to exert control over their thoughts, feelings,
motivations, and actions.152 SCT is a theory of human behavior that emphasizes the importance of
personal beliefs as key components of control and personal agency. In this theory, individuals are
viewed both as products and as producers of their own environments and social systems.
34


SCT hypothesizes that peoples beliefs in their capabilities mediate how they behave in
situations such as participating in treatment for a health condition. Treatment provides an
environment in which there are opportunities for physical and psychosocial support by a health
professional, family, or friends which can in turn affect their beliefs about their capabilities. Both
external support and treatment itself may help to improve an individuals beliefs about his/her
capacity to perform functional tasks. At this time, it is unknown how personal beliefs or external
support can influence appraisals and coping behaviors in a patient with LSS.
Self-efficacy. The construct of self-efficacy is a key concept within the SCT framework.
Self-efficacy is an intrapersonal factor defined as a persons belief in ones capacity to organize
and execute the courses of action required to produce given attainments.153 It is a context-
specific assessment of ones competence to perform a specific task or range of tasks in a given
domain. Bandura153 describes self-efficacy as influencing 4 areas: 1) the choices that are made,
2) the effort put forth in task specific roles, 3) the time one persists when there are obstacles, and
4) ones feelings of confidence in performing specific tasks in specific situations.
Depending on what is being managed, the tasks over which personal influence is
exercised may entail regulation of one's motivation, thought processes, affective states and
actions, or changing environmental conditions. Self-efficacy beliefs are sensitive to these
contextual factors. Therefore, they differ from other expectancy beliefs in that self-efficacy
judgments are more task and situation specific and individuals make use of these judgments in
reference to the type of goal.150154155 Self-efficacy can refer to sub-skills required to organize
actions that are governed by broader self-regulatory skills such as knowing how to diagnose task
demands or constructing and evaluating alternative strategies. Possessing these self-regulatory
sub-skills can permit individuals to improve their performances across varied activities
particularly when faced with sensations of pain during activity performance. Self-efficacy can
generalize across skills when commonalities are cognitively structured across activities. For
example, physical exercises and motor skills practiced in health care treatments may transfer to
35


other activities performed in daily living as their confidence to perform these types of tasks
increase.
People make and shape self-efficacy beliefs in the context of performing a specific goal
or task. Theoretically, self-efficacy beliefs might influence a patients experiences of pain,
functional limitations, and disability with LSS. A patient in treatment practicing functional
activities like walking could increase their task specific self-efficacy for walking just by
practicing and as a result, increase their confidence and activity outside of treatment. Moreover,
a patient who enters treatment with high self-efficacy for functional tasks might have even greater
belief that he/she could organize and execute all of the actions needed to successfully complete
tasks required in treatment. Theoretically, a patient with higher self-efficacy could have
heightened beliefs about accomplishing tasks that include positive health behaviors such as
attending regularly scheduled clinical appointments, following medical instructions, executing a
series of home and clinic exercises, and performing positive avoidance behaviors to minimize
pain and discomfort. Overall, high self-efficacy beliefs lead to setting challenging individual
goals and maintaining strong commitment to these goals that ultimately could affect outcomes of
treatment.
In addition, researchers think self-efficacy beliefs in part determine outcome
expectations.153156 Individuals who expect success in a particular task often produce successful
outcomes in that task. The opposite is also true of those who lack such confidence in task
performance; those who doubt their success will produce failed outcomes. Patients undertaking
tasks such as those found in treatment would have similar anticipatory outcome expectations
about performing the specific tasks required for the process of treatment and transfer these beliefs
to similar skills outside of treatment.
Issues of self-efficacy in the elderly population often center on reappraisals and
misappraisals of their capabilities. Because biological conceptions of aging focus extensively on
declining abilities, many physical functions can decrease as people grow older and require
36


reappraisals of self-efficacy for activities in which biological functions have been significantly
affected. When a therapist teaches an older adult to use his/her intellectual and physical
capabilities, his/her improvement in cognitive and physical functioning more than offsets the
average decrement in performance over 2 decades.153 Therefore, elderly persons who invest the
necessary effort can function at higher levels comparable to some younger adults. Although never
studied in previous LSS research, it is possible that perceived self-efficacy may affect the level of
involvement in activities and theoretically mitigate the decline seen from chronic disease
processes such as LSS.
Application of self-efficacy. What is unknown is if self-efficacy can maintain or
improve in the presence of pain and disability perceived by the patient with LSS. It is known that
pain can impede or prevent patients from participating in an activity because of the actual
physical sensations of pain, the fear and or avoidance behaviors that preempt participation, and
the cognitive appraisals or reappraisals that are made after activities have repeatedly resulted in
painful experiences. Reductions in function can also lower an individuals self-confidence, and as
a result continue to lower the level of activity in which an individual participates in daily life.
In laboratory experiments, self-efficacy beliefs predict tolerance levels to pain.157 From a
physical perspective of pain, perceived self-efficacy has been shown to affect the body's opioid
and immune systems.158 In patients with chronic pain disorders related to LBP, self-efficacy
positively affects physical and psychological functioning.159 Evidence shows that self-efficacy
influences pain and function after acute physical interventions like surgery.160,161 Prospective
studies in patients who underwent orthopedic surgery demonstrated that high self-efficacy before
the start of treatment and larger increases over the course of treatment speed recovery and predict
better long-term outcomes.162"164 It is possible that individuals with high self-efficacy may be
more motivated to engage in health promoting behaviors and adhere better to treatment
recommendations because they have higher performance success expectancies. In addition, it is
believed that people with high self-efficacy are less likely to give up an activity when faced with
37


barriers such as pain and are therefore less likely to become trapped in the negative spiral of
activity avoidance, physical deconditioning, loss of social supports and depression.
Application of self-efficacy during treatment. In rehabilitation literature, self-efficacy
is measured by using task-specific constructs relevant to the chronic disease process and specific
rehabilitation treatments being studied. Research indicates that task specific self-efficacy can
improve during the treatment process. Scherer and Schmieder165 demonstrated improved task
specific self-efficacy with patients who participated in treatment for dyspnea due to chronic
obstructive pulmonary disease. Patients who completed educational and exercise training had
significant increases in their self-efficacy scores to manage or avoid breathing difficulty. In
addition, patients increased their self-efficacy expectations for exercise endurance.165 Carlson et
al166 showed that cardiac rehabilitation treatment involving physical exercise improved patients
post-treatment self-efficacy beliefs for independent exercise. Another study by Jeng and Braun167
found that those with greater success in functional outcomes from cardiac rehabilitation treatment
correlated with higher exercise specific self-efficacy scores. Finally, Rejeski et al168 found that
when patients undergo aerobic and strength training exercise for knee osteoarthritis (OA) those
who actively engaged in physical exercises compared to controls had increased self-efficacy
outcomes for stair climbing, a task-specific limitation for patients with knee OA.
Research has demonstrated that self-efficacy is also an important factor in improved
coping and psychological outcomes.169"171 Taal et al103 surveyed patients with rheumatoid arthritis
(RA) to determine their level of self-efficacy in relationship to function, pain, and disability when
coping with RA. Higher self-efficacy scores correlated with improved coping abilities
independent of pain, disease status and functional abilities. Moreover, Strahl et al172 found that
higher self-efficacy levels in patients with arthritis were predictive of better outcomes related to
psychological functioning.
An investigation of self-efficacy for patients with chronic LBP found that initial self-
efficacy beliefs predicted functional abilities. In a study by Lackner and Carosella173 that included
38


patients with chronic LBP, 100 patients rated their confidence to perform load lifting tasks before
any examination or treatment. Next, these patients underwent a subsequent physical exam and
physical performance test for lifting loads. The results indicated a significant association between
self-efficacy beliefs and lifting higher loads and higher physical performance. Lackner and
Carosella did not find an association between pain perceptions or measures of psychological
distress and the physical performance measures.
Overall, these studies indicate that self-efficacy is an important construct in health care
management. Two of the studies167173 found significant relationships that demonstrate self-
efficacy beliefs have an impact on the outcomes of pain and disability for patients with chronic
disease processes. In addition, no studies have examined the impact of self-efficacy beliefs during
treatment of patients with LSS. Although the role of self-efficacy in the treatment process
warrants more investigation, these studies indicate that heightened patient beliefs regarding the
ability to perform specific tasks might also raise the functional, psychological, and overall coping
behaviors of patients during and post-treatment and therefore help to decrease overall perceptions
of pain and disability. Although the role of self-efficacy is unknown in LSS, this research study
sought to investigate the impact of self-efficacy in the appraisal of pain and disability and coping
behaviors of patients with LSS.
Social Relationships and Health
In addition to individual level factors such as self-efficacy, there is considerable evidence
that interpersonal factors play a role in disease prevention, disease management, and health
outcomes. At this time, it is unknown the role of interpersonal factors play in the health and
management of patients with LSS. In general, researchers studying aging have determined that
people with social support and social ties, regardless of their source, live longer than people who
are isolated.174 People with a close network of ties with other people maintain better health, resist
disease and deal more successfully with problems they encounter.174,175 People who are socially
39


isolated and have fewer social relationships have been found to have mortality rates 2 to 5 times
higher than those with good social relationships, regardless of gender, race, ethnic background
and socioeconomic status.176177
There are several terms used to describe social relationships as they relate to health.
Social integration refers to the existence of social ties.176 Social network describes the
interconnectedness of social relationships that surround an individual.177 Together, social
integration and social networks describe the existence and quantity of social connections that an
individual has available. Social support on the other hand is a concept that illustrates not only the
existence of a social relationship but also the positive function of that social connection and its
potential to play a supportive role in an individuals health.113,178 Social support is defined as the
degree to which a persons basic social needs are met through interaction with other people.179 It
includes tangible and intangible resources, emotional support, informational and instructional
support, as well as a persons perception of assistance in times of need. When measuring
supportive functions in health, measurement should include support that one believes is available
if needed (perceived support) or functions that one reports and is recently provided (received
support).113,177 This research study focused on identifying both perceived and received support in
patients with LSS.
Social support. The role of social support has evolved in health research and is believed
to be a social construct that protects and preserves well-being of an individual in the face of a
stressful encounter such as a negative health condition.177 Social support is believed to contribute
to global self-evaluations of health.180 In addition, social support serves as a social resource that
can assist an individual in his/her capability to cope with stress.107 Even though social support
comes from an external source, this external support is believed to be able to help protect people
from the adverse effects of stress (Figure 2.6). In the stress and coping literature, social support is
a coping resource that can alter beliefs and commitments that can in turn alter how one appraises
a stressful situation.107
40


When considering the role of social support, stressors are believed to act on 2 different
pathways. One pathway is through appraisals of available social support (Figure 2.6). The
perception that social support is available may protect people from the adverse effects of stressors
by leading them to appraise or interpret stressfiil situations less negatively. For example, a patient
with LSS with high perceived social support might interpret his/her pain and disability less
negatively than a person with low perceived social support and therefore approach the process of
treatment and self-management differently. Hypothetically, patients interpretations of support
may moderate their experience of pain and disability before, during and after the treatment
process.
Figure 2.6. Social support interactions with Stress and Coping Model. Interactions between
perceived support with appraisals as well as received support with coping to modify the
relationship of a stressor on health.
The other pathway social support is believed to assist in coping is by actual supportive
actions, or received support. Received social support is the actual assistance by others during
stress.181182 Received support is believed to interact and alter a persons coping abilities in times
of stress. This improved coping due to increased received support is theorized to reduce the
41


impact of stress on a persons health.177 When a person can reduce the impact of stress on his/her
health, it is theorized that this could be measured through improved health outcomes such as
reduced pain and/or disability. A way clinicians have measured social support is by quantifying
the frequency or number of supportive actions a person reports as well as his/her perception of the
quality of that received support.183 Retrospectively identifying the frequency and quality of
support received during a specific time period of treatment, identifying the supportive actions,
and rating how this support was helpful or unhelpful are all ways that patients might report
received social support for a health condition.
Application of social support in treatment. Evidence exists on the importance of
overall social support in the treatment of chronic diseases, particularly in the treatment of
cardiovascular disease184185 and stroke.186"189 These studies have shown that improvements in
physical functioning and psychological adjustment in older adults in cardiovascular and stroke
recovery were greater among those with higher sources of social support.185'191 Moreover, other
studies have shown that social support from all sources positively correlated with increased
perceived general health and quality of life after stroke.192"194 Researchers have also shown lack of
social support in general to be associated with negative health consequences including suicidal
thoughts,195 depression,104 length of hospital stay,196 discharge to rehabilitation and nursing home
facilities,197 and reduced physical functioning.198,199
There is additional evidence that patients with a variety of chronic disease processes
benefit from social support from all sources. In one study, Littlefield et al200 found patients with
Type II Diabetes Mellitus had improved function with higher total social support scores but worse
function with higher depression scores. In another study, Yates201 surveyed patients with
cardiovascular disease who reported that both emotional support and tangible aid from health care
providers and the patients spouses were important sources of support for coping with their
disease process and improving overall physical recovery. Additionally, Gulick202 conducted
research involving patients with multiple sclerosis and found that higher total social support
42


scores correlated positively with ADL function scores and inversely correlated with depression
scores.
Research specifically investigating arthritis found that social support does mediate the
response to the treatment process and function of the patient. In patients with RA, Taal et al103
found tangible or instrumental support by caregivers corresponded positively with improved
patient reported health status and participation in daily physical activities. However, emotional
support did not relate to patients reported health status for this sub-group of patients with RA.
Weinberger et al203 illustrated in research involving patients with OA with concurrent functional
limitations that both older age and lower levels of tangible support from caregivers, friends, or
family associated directly with greater physical disability. Moreover, greater psychological
disability was associated in patients with OA with lower levels of emotional support from all
sources.
Little research has directly examined the association between treatment for LBP and
social support. Masters et al183 investigated whether patients received social support in the
treatment process for LBP. Retrospectively, the researchers asked patients who participated in
treatment for LBP to indicate which sources of social support they received and whether this
support was helpful or unhelpful. Out of the 50 patients surveyed, 43% indicated that they
received tangible support from a physical therapist and physician that was helpful. Thirty-three
percent of the patients indicated they received helpful emotional support from physical therapists,
family, and friends. Twenty-three percent of the patients additionally reported that they received
helpful informational support from physical therapists. However, 50% of the patients also
reported that they received unhelpful emotional support from their family and friends during the
treatment process. Additionally, 37% reported receiving unhelpful informational support and 10%
reported receiving unhelpful tangible support from their physicians during the treatment process.
Unfortunately, the researchers did not evaluate the relationship of support to other general health,
physical function or psychosocial measures. In addition, the researchers did not evaluate the
43


relationship of social support to the outcomes of pain and disability in treatment for LBP. Overall,
this study illustrates that social support sources are available and are identified as both helpful
and unhelpful in treatment for LBP. Yet understanding how social support impacts the experience
of pain and disability in LBP, particularly in LSS, continues to be lacking in the literature.
General Health Status
Self-rated health is another way to identify a patients perspective of his/her health
condition and may contribute to poor coping behaviors as well as improvement in health
outcomes. Self-ratings of health are used frequently in health literature and have shown to be
predictors of morbidity and mortality.204 Self-rated health has also been proven sensitive to
improvements and declines in many health related outcomes.205 206 Self-rated health measures
assess subjective well-being related to physical and mental domains of health. These measures are
used extensively in clinical trials and in health services research where they have shown evidence
of reliability, validity, and responsiveness.207 Yet, researchers have only a limited understanding
of how these self-evaluations are reached or their implications for health behaviors. In addition,
few studies have examined changes in self-rated health over time and in specific health conditions
such as LSS.
The perception of health status during a negative health experience is believed to be just
as relevant to health outcomes as the actual somatic state of the individual. Even though most
medical treatment still focuses on a persons actual physical state as it relates to pain and illness,
perceived health may largely contribute to a persons well-being and function. Cross-sectional
studies have typically shown that general self-rated health is closely related to the experience of
physical symptoms.205208 Moreover, a 5 year study by Gold et al209 of health outcomes in the
community determined that both functional ability and self-rated health were independent
predictors of health outcomes. Those with initial low functional scores also had greater decline in
self-rated health overtime reflecting the physical component in self-evaluations of health.
44


Appraisal of health. Self-rated health, both physical and mental health, is likely to be
part of the appraisal process when faced with a stressful health condition. Health perceptions may
also be part of coping processes when an individual faces a specific illness. When an individual
has a beher perception of their overall health, they have a greater potential for these health beliefs
to bolster positive coping behaviors during illness experiences. It has been shown that older
adults overall perception of health can be predictive of their use of active coping strategies to
deal with age-related health challenges.210 Research has also shown that those who dont have a
positive health perception may engage in sick role behaviors that can lead to self-destructive
behaviors during illness.122 One reason for this is that self-rated health is believed to be a
reflection of an enduring self-concept of a healthy or unhealthy person.211 Therefore this self-
concept of health may act as a moderator when engaging in positive or negative illness behaviors.
Variance in physical measures and outcomes have also been shown to be related to
respondents previous self-rated health.211 Longitudinal studies have illustrated that those with
initial lower self-rated health predict poor functional ability and increased health care utilization
over time.206 212 Negative self-rated health has been found to predict long term disability and
health decline in the general population.212 Change in self-rated health have been shown to
coincide with long-term changes in physical health, mental health, perceived social support, and
performance related behaviors.211 Moreover, self-rated health has been shown to vary according
to whether respondents intended to improve specific health-related behaviors in the future.211
These findings suggest that self-rated health is both a current assessment of one's health status,
similar to a self-concept, and a reflection of efforts to achieve relatively important health-related
goals.
During illness, perceived health status may be determined in part by an individuals level
of emotion and distress during illness. Individuals have been found to experience different
physical bodily responses based on their differing emotional states.213214 Some previous research
has indicated that psychological function and perceived mental health are related to functional
45


status.215 Using the Stress-Diathesis Model, vulnerability or diathesis is an integral component in
the experience of pain and disability. Therefore, it is probable that components of perceived
psychological function, or mental health, can also have proximate influences on the appraisal
process and indirectly influence health outcomes.
Application of self-rated health. Measurement of perceived physical and mental health
has been performed routinely in the health literature by several standardized tools.73,209215,216
These general health surveys have been designed to measure overall self-rated health with a broad
range of questions covering a variety of aspects of physical and mental health. It is commonly felt
that the usefulness of general measures is in their ability to allow comparisons among patients
with the same condition as well as between patients with different conditions. In some cases,
general health measures may be able to identify unsuspected issues from a diagnostic group that
would be highlighted as scores deviating from population or disease specific norms. These self-
rated indicators likely take into account the patients physical sensations such as bodily pain, the
patients comparison of what he/she can perform in his/her daily life as compared to previously or
as compared to his/her peers, as well as his/her psychological and social perception of
functioning. Overall, understanding how a patient perceives his/her general physical and mental
health may add to the larger role of intrapersonal and interpersonal level influences on outcomes
of pain and disability.
Theoretical Summary
Improvements in levels of pain and disability are routinely monitored and measured in
treatment programs; however the influences mediating the severity and outcomes of these clinical
variables are not fully understood. This study examined if personal factors at an individual or
intrapersonal level, specifically self-efficacy, and social factors at an interpersonal level,
specifically social support, contribute to a greater understanding of the experience of pain and
disability in patients with LSS. Because there is some evidence that individual factors such as
46


stress, coping, and cognitive mechanisms can influence aging-related health in general,149 it was
important to include a measure of perceived physical health and perceived mental health to the
understanding of the experience of pain and disability. Although self-efficacy and social support
have been previously studied in stroke, cardiovascular disease, arthritis and LBP, the contribution
and association of self-efficacy and social support to understanding the experience of pain and
disability in patients with LSS has not been studied.
Gaps Filled
The goal of this study was to understand how the influences of intrapersonal and
interpersonal factors contribute to the experience of pain and disability in LSS. This study
addressed multiple gaps in the LSS literature. Gaps in the LSS literature include a broader
understanding of the experience of pain and disablement in this population, understanding
appraisals and coping used by patients with LSS, and the use of mixed methods research. This
study aimed to fill some of the gaps in the literature by evaluating the experience of pain and
disability, particularly identifying the contribution of self-efficacy, social support, and perceived
health status in management of patients with LSS.
This study tested existing measures of self-efficacy, social support, perceived physical
health, and perceived mental health in a unique population before and after clinical treatment.
This study added to a body of literature on how coping behaviors operate in the context of health
and disease. Ultimately, this study aimed to identify factors not recognized in the current
literature that contribute to the experience of pain and disability in patients with LSS. By using a
mixed methods research design, a more complete picture was developed that will inform clinical
practice as well as future research and policy recommendations for this patient population. As
information is revealed, future research will need to determine how to best identify, monitor, and
possibly intervene using concepts of self-efficacy, social support, general health perceptions, and
any novel factors uncovered during the study period.
47


CHAPTER III
EXPERIMENTAL DESIGN AND METHODS
Research Design
The study used a subset of a larger data set from an ongoing randomized clinical trial
(RCT). This study used a longitudinal, cohort design with mixed quantitative and qualitative
methods.217 The subset of patients with LSS studied participated in 1 of 2 non-surgical treatment
programs (Figure 3.1). Patients were evaluated upon entry to treatment and the time period
between 6 months and one year post-treatment. Treatment effects were not evaluated in this sub-
study and will be evaluated upon completion of the larger RCT. The sample in this subset is
similar to the larger population of patients with LSS.218 All patients in this subset and in the larger
RCT are at a similar stage of chronic disease as they have been diagnosed with LSS upon entry to
the study and display signs and symptoms consistent with the diagnosis of LSS.217 All patients in
this subset and in the larger RCT have been randomized to treatment including epidural steroid
injection (ESI) and low back education, with half the subjects receiving additional physical
therapy (PT) (Figure 3.1). Both ESI and PT represent the standard of care for current clinical
practice. Although both non-surgical treatment options have been found to be helpful, at this time
neither have been found to be superior in determining clinical outcomes of LSS.219
48


Diagnosis of LSS: MD
exam and imaging
Consent, Exam and
Baseline Surveys

Randomization
%
ESI + Back Book
ESI + PT + Back Book
8-10 PT Treatment
Figure 3.1. Design of the randomized clinical trial for non-surgical treatment of LSS.
Research design illustrating baseline procedure, randomization, and treatment groups.
Within the larger RCT, there are opportunities to answer many questions. However, the
focus of this dissertation research was to investigate if behavioral and social factors such as self-
efficacy, social support, and self-rated health had an impact on standard clinical intake measures
of pain and disability. In addition, this dissertation research sought to identify the role of self-
efficacy, social support, and self-rated health in predicting positive outcomes of pain and
disability in patients with LSS long-term. Understanding treatment effects is an important
component in management of patients with LSS; however, unless we look closer at behavioral
and social factors in the management process, we cant draw complete or sound conclusions
regarding patient severity or outcomes, hi addition, understanding the role of behavioral and
social factors in the long-term management of LSS can help direct clinical resources to address
the specific needs of the patient. Future studies will need to test the findings from this study in an
49


intervention trial that uses strategies which target significant influencing behavioral and social
factors in order to improve outcomes in patients with LSS.
Subset of the RCT
For this study, a subset of the study population from the RCT was selected for both
quantitative, repeated measures survey and qualitative open-ended interviews. A cohort study
design involves participants who meet specified criteria such as a specific health outcome or
membership in a group of interest.217 In this case, persons with LSS who enter conservative
treatment are the persons of interest and the entry point to treatment and the outcomes of their
treatment related to pain and disability scores are the outcomes of interest.
All patients participated in self-report surveys of general health, condition specific pain
and disability as well as specific intrapersonal and interpersonal domain measures of self-
efficacy, social support, and self-rated health at entry to treatment and at long-term follow-up.
Follow-up surveys were mailed at approximately 6 months post entry to treatment to 22
participants who completed treatment. Twenty subjects that completed treatment and the survey
packets were included in the post-treatment measurement. The individual response time for
follow up of post-treatment measurement in the 20 subjects was between 6 and 12 months. The
same 20 subjects who completed treatment and follow up surveys participated in a 30-50 minute
interview. All subjects received the same non-surgical treatment protocol for low back education
and epidural steroid injection (ESI), with approximately half of the subjects receiving additional
physical therapy treatment (ESI+PT). All procedures for the RCT treatment protocols are located
in the Manual of Standard Operating Procedures (MOSP, Appendix A). Figure 3.2 depicts the
study design.
50



1 1 ln Combined terpretatic 1 )n i
L J
Data Collection
Data Analysis
Results and
Interpretation
Figure 3.2. Longitudinal, cohort study design using mixed methods. The quantitative analysis
sample consisted of 34 patients at entry to treatment and 20 patients at follow-up. The qualitative
interview sample consisted of the same 20 participants at follow-up. All participants were part of
a subset of the ongoing RCT sample of patients.
Mixed Method Design
Mixed methods approach to inquiry was selected over other study designs because it
enables the researcher to capture both empirical measures and patients accounts of the lived
experience of pain and disability with the chronic condition of LSS. By combining quantitative
and qualitative data collection techniques, mixed methods allows for a rich, complex, and more
complete understanding of the phenomenon under study.220 Mixed methods utilizes the strengths
of both quantitative and qualitative methods to offset the weaknesses inherent in each individual
method and enhances validity of the findings.221 For the chronic condition of LSS, previous
literature has not fully explored the meanings of pain and disablement. Although a narrow and
focused understanding of pain and disability in patients with LSS is present in the current
51


literature, the addition of qualitative research to the quantitative data provides different
perspectives on the experience of both pain and disability.
This mixed methods study employed a semi-convergent parallel design, capturing
quantitative data at entry to treatment but then both quantitative and qualitative data at relatively
the same time period post-treatment, as shown in Figure 3.2.222 Quantitative methods were used
to explore questions and test hypotheses regarding the factors associated with pain and disability
in LSS. Qualitative methods were used to enhance and expand upon the quantitative findings,
give the patients a voice, and explore the nuances and meanings of pain and disability related to
the lived experience of LSS. The purpose of the semi-convergent design is to obtain different but
complementary data on the same topic to pull together the strengths of both quantitative and
qualitative methods.223 This design is often used to triangulate findings by directly comparing
and contrasting quantitative statistical results with qualitative findings for corroboration and
validation purposes as well as to develop a more complete understanding of a phenomenon,222 in
this case, the experience of pain and disability in individuals with LSS.
Finally, the mixed method approach allows for multiple worldviews in research related to
LSS. Since a large portion of the research to date in LSS is hypothesis driven and uses a single
method, the primarily focus has been a paradigm of positivism. The addition of other worldviews
including paradigms from both post-positivism and constructivism in understanding pain and
disability in this population is warranted. Since a single worldview currently exists in the
literature investigating LSS but the pain and disability literature uses multiple worldviews, a
pragmatism worldview was used in this study. Pragmatism is a set of ideas that draws on diverse
approaches and uses both objective and subjective knowledge. According to Teddlie and
Tashakkori,224 using a pragmatism approach in mixed methods places the research question as the
primary focus rather than the method or the philosophical worldview that underlies the method.
The pragmatism approach is well suited to answer the studys research question by combining 2
52


types of data sources to uncover a more complete story about the experience of pain and disability
in patients with LSS.
Data Collection
In order to corroborate and compare the information from both quantitative and
qualitative data sources, this study included data from the same 20 participants at post-treatment.
Parallel data collection questions were conducted across both the quantitative and qualitative data
sources. Equal weight was given to both quantitative and qualitative data. In addition, results of
the longitudinal data collection was conducted using repeated measures surveys at entry to
treatment (n= 34) and post-treatment (n=20) and corroborated with in-depth, open-ended
interviews (n=20) collected post-treatment from the RCT titled Epidural Steroid Injection versus
Epidural Steroid Injection and Manual Physical Therapy and Exercise in the Management of
Lumbar Spinal Stenosis; a Randomized Clinical Trial. (NCT00786981). IRB approval was
completed by COMIRB and Hawkins Foundation (see Appendix B).
Sampling and participants. Figure 3.3 provides a visual representation of the process
for sampling. The population for the study included individuals 50 years of age and older with a
diagnosis of LSS in Colorado and South Carolina who entered conservative treatment for this
chronic condition. The study used a subset of 43 of the total 80 subjects planned to be enrolled
into the larger RCT using convenience sampling of individuals who sought treatment either at the
Spine Center at the University of Colorado or the Steadman Hawkins Clinic of the Carolinas.
These sites were located in 2 different geographic areas in the United States and represented a
broader view of the population of LSS.
Subject randomization, baseline examination, and treatment procedures. Patients
who sought medical care at the 2 clinical sites, met the inclusion criteria (see Table 3.1), and
consented to participate were enrolled in the RCT. All patients were required to have health care
insurance coverage (private insurance, Medicare, or Medicaid) to cover the standard treatment
53


protocols designed for the RCT which included standard treatment with ESI and PT. All patients
in the study had a diagnosis of LSS, had symptoms of neurogenic claudication or radicular pain,
and were candidates for non-surgical treatment. For the purposes of this study, LSS was defined
as evidence of narrowing of the spinal canal and/or intervertebral foramen demonstrated on
diagnostic imaging studies (MRI or CT scan), and clinical signs and symptoms consistent with a
diagnosis of LSS including patient-reported reduction in walking tolerance due to pain and/or
cramping in the leg(s), defined as neurogenic claudication.
Table 3.1. Inclusion and Exclusion Criteria for the Nonsurgical RCT with Patients with LSS
Inclusion Criteria
1. Age greater than or equal to 50 years
2. MRI findings consistent with lateral foraminal and/or central lumbar spinal stenosis (evidence
of compression of lumbar spinal nerve root(s) by degenerative lesions of the facet joint, disc,
and/or ligamentum flavum)
3. Chief complaint of pain in the low back, buttock and/or lower extremity (s)
4. Presence of symptoms consistent with neurogenic claudication (numbness, tingling, cramping,
and downward radiating pain which is not exacerbated with biking, uphill ambulation, and
lumbar flexion but is not alleviated with standing; or does not include signs of pulselessness,
paralysis, or pallor).
5. Rates sitting as a better position with respect to symptom severity than standing or walking
6. Lives within one hour of a research site
7. Can attend 10 regular physical therapy appointments spread over 10 weeks and 4 examination
appointments (baseline, end of treatment, 6 months, and 1 year)
8. Sufficient English reading and language skills and mental capability to complete self-report
assessment questionnaires
9. No contraindications to MRI_______________________________________________________________
Exclusion Criteria
1. Severe vascular, pulmonary or coronary artery disease which limits ambulation (as determined
by the referring physician or the therapist)
2. Other orthopedic conditions or physical impairments of unrelated nature which would limit
ambulation or prevent the subject from fully participating in any other aspect of the
rehabilitation exercises (as determined by the referring physician or the therapist)
3. Previous spinal surgery that included fusion of two or more vertebrae
4. History of spinal tumors, spinal infection, or lumbar vertebral fractures other than spondylolisis
or spondylolisthesis
5. Signs/symptoms suggestive of potential non-benign or pathologic condition as the origin of
symptoms
6. Presence of any absolute contraindications to sub-maximal treadmill testing per the American
College of Sports Medicine (ACSM) standards.
7. Epidural steroid injection within the last 365 days.__________________________________________
54


A physician, resident, or physicians assistant indicated whether the patient met the
inclusion criteria to participate in the RCT. In addition, the physician or physician assistant
explained the study, the risks and benefits for patients, and asked eligible patients who were
interested in participating to sign 3 copies of the informed consent form (Appendix C). The
physician, resident, or physician assistant, performed the general baseline examination. After the
baseline physical examination, patients completed paper and pen self-reported questionnaires
(Appendix D).
The RCT employed a random number generator to establish randomization lists prior to
the initiation of the RCT. After all baseline procedures were completed including consent,
physical exam, and baseline surveys, the patients were randomized to 1 of 2 treatment groups:
epidural steroid injection only or epidural steroid injection plus physical therapy. The procedures
for baseline examination, randomization, and group treatment procedures were standardized so
that all patients in each treatment group and at each site received the same care (see MOSP:
Appendix A).
Subjects. As of October 2012, 43 subjects were enrolled into the RCT. Of these 43
subjects, 12 subjects dropped out and 1 subject died after baseline enrollment. Of those 13
subjects who dropped out, only 9 had completed baseline paperwork, 11 had no or partial
treatment, and 2 had completed treatment but were lost to follow up (see Figure 3.3). Eight of the
43 subjects were still in the process of treatment at the end of data collection for this sub-study on
10/31/2012. Two subjects who had completed treatment were not a part of the patient interviews.
In total 34 subjects completed baseline information and were included in the sub-study data at
entry to treatment. Twenty of these same 34 subjects completed follow-up surveys and a phone
interviews during the post-treatment phase (see Figure 3.3). All quantitative data were collected
and kept in a locked storage cabinet and were entered into a secure data base in a password
protected computer. All quantitative data were only accessible to the research investigators of the
55


RCT. Participants who completed baseline evaluation, treatment, and completed the follow up
surveys between 6 and 12 months after enrolling in the study, received $50 in remuneration.
A total of 20 subjects who completed baseline evaluation, treatment, and follow up
surveys were invited and verbally agreed by phone to an interview session consisting of a series
of open-ended questions. Of these 20 subjects, 12 were in the ESI only group and 8 were in the
ESI plus PT group. All interviews lasted approximately 30-50 minutes and were conducted by
phone. Interviews were audio-recorded during the phone conversation and transcribed verbatim
into word documents along with field notes made after the interview. Interview and field note
data were kept in a password protected computer. All data collected during the interview were
kept confidential and coded to the subjects number given during the RCT randomization
procedures. Patients who completed an interview received an additional $25 in appreciation for
their time, effort, and sharing their experiences with LSS. Figure 3.3 illustrates subjects who
participated in baseline evaluation and subsequent follow up surveys and phone interview post-
treatment.
56


Figure 3.3. Subject enrollment, consent, treatment retention, and follow-up.
57


Research Question and Specific Aims
The current study addressed the following research question: What is the relative
contribution of self-efficacy, social support, perceived physical health and perceived mental
health to pain and disability in individuals with LSS? Three Specific Aims guided the
investigation of this overarching research question.
Aim 1: To determine the relative contribution of self-efficacy, social support, self-rated
physical health, and self-rated mental health to the severity of pain and disability among
patients with LSS upon entry to treatment of LSS.
H1A: Lower levels of self-efficacy, social support, perceived physical health and
perceived mental health will each independently predict higher disability at entry to
treatment.
H1B: Lower levels of self-efficacy, social support, perceived physical health and
perceived mental health will each independently predict higher pain at entry to treatment.
H1C: Lower levels of self-efficacy, social support, perceived physical health and
perceived mental health will together predict higher disability at entry to treatment.
HID: Lower levels of self-efficacy, social support, perceived physical health and
perceived mental health will together predict higher pain at entry to treatment.
Aim 2: To determine the relative contribution of self-efficacy, social support, self-rated
physical health, and self-rated mental health to the outcomes of pain and disability among
patients with LSS in the year following treatment of LSS.
H2A: Higher levels of self-efficacy, social support, perceived physical health and
perceived mental health will each independently predict lower disability after treatment
of LSS.
58


H2B: Higher levels of self-efficacy, social support, perceived physical health and
perceived mental health will each independently predict lower pain after treatment of
LSS.
H2C: Higher levels of self-efficacy, social support, perceived physical health and
perceived mental health will together predict lower disability after treatment of LSS.
H2D: Higher levels of self-efficacy, social support, perceived physical health and
perceived mental health will together predict lower pain after treatment of LSS.
Aim 3: Explore how self-efficacy, social support, perceived physical health and perceived
mental health contribute to a patients lived experience with LSS and influence meanings of
pain and disability.
Measures and Constructs
Measuring pain: Numeric Pain Rating Scale (NPRS). The Numeric Pain Rating Scale
(NPRS) asked subjects to rate their pain on an 11-point scale. The anchors were 0 and 10, with 0
representing No Pain and 10 representing Worst Pain Imaginable. Subjects rated their
minimum, average, and greatest pain intensity during the last 48 hours for their thigh and leg
symptoms (NPRStl) and their low back and/or buttock pain (NPRSbb). This method of recording
pain has demonstrated only fair reliability.225 Older patients tend to prefer the NPRS and tend to
exhibit fewer errors than with the use of the visual analog scale.226227 Whitman et al225 have
shown the minimal clinically important difference (MCID) for the NPRS in a patient population
with LSS to be between 1.25 and 1.5 respectively. Additionally, the body diagram was used to
assess the distribution of symptoms by categorizing the location of symptoms as low back,
buttock, thigh, and/or leg (distal to knee).
Measuring disability: Oswestry Disability Index (ODI). The modified ODI is a
region-specific disability scale focusing on the lower back and leg regions for patients with LBP.
59


The ODI consists of 10 items addressing different aspects of function, each scored from 0-5 with
higher values representing greater disability. The modified ODI was used as the primary outcome
measure of disability. Whitman et al225 found that the modified ODI exhibits excellent test-retest
reliability in patients with LSS (ICC, 0.84). They also found the modified ODI to be more
responsive (AUC = 0.84) than the Spinal Stenosis Scale symptom and function subscales (AUC =
0.74 and 0.75 respectively). The authors identified the minimum clinically important difference in
a population receiving non-surgical management for LSS as 5% points which means the change
in ODI scores must be at least 6 points to be clinically significant in patients with LSS.
Measuring self-efficacy: Low Back Activity Confidence Scale (LoBACS). The
LoBACS228 assesses confidence in performance of key functional activities and regular exercise
in patients with histories of LBP and surgical interventions. Self-efficacy literature, LBP
literature, and clinicians observations were all used to develop the LoBACS. Seven items pertain
to self-efficacy for functional activity: self-confidence for carrying, lifting, pushing, sitting,
standing, walking, and stair climbing. Three items address self-efficacy for self-regulation of
back health, and 5 items relate to confidence for regular exercise. The 2 scales of the LoBACS
have good test-retest reliability with Functional Self-Efficacy (FSE): ICC, 0.92 and Exercise Self-
Efficacy (EXSE): ICC, 0.75. The FSE subscale is also strongly negatively correlated with
disability using the ODI: r = -0.86, p<.0001. The EXSE subscale is moderately correlated with
disability using the ODI: r = -0.28, p = .044. Both subscales are correlated to subjective quality of
life measures (FSE: r = 0.55, p<.0001; EXSE: r= 0.35, p = .010).
Measuring social support: Medical Outcomes Survey Social Support Scale (MOS
SSS). The MOS SSS is a brief, multidimensional, self-report survey developed for patients with
chronic conditions in the Medical Outcomes Study.229 This survey was designed to be
comprehensive addressing different dimensions of social support. Multi-trait scaling analyses
support the dimensionality of 4 functional support scales (emotional/informational, tangible,
affectionate, and positive social interaction) and the construction of an overall functional social
60


support index. These support measures are distinct from structural measures of social support and
from other related health measures. The support measures are reliable (alpha >0.91), fairly stable
over time, and illustrate construct validity.
Measuring perceived health status: Medical Outcomes Survey Short Form 36 (SF-
36). General health surveys have been designed to measure overall self-rated health with a broad
range of questions covering a variety of aspects of physical and mental health. The Medical
Outcomes Survey SF-36 is one such multi-purpose, general health survey which yields
psychometrically-based physical and mental health summary measures. In multiple studies, the
SF-36 has been used for general and specific populations evaluating health outcomes, comparing
the relative burden of diseases, and differentiating the perceived health benefits produced by a
wide range of different treatments. The content validity of the SF-36 has been compared to that of
O'sn 0*21 _
other widely used generic health surveys. The SF-36 was judged to be the most widely
evaluated generic patient assessed health outcome measure in a study published in the British
Medical Journal232 In addition, the SF-36 is widely used in studies of patients with LBP and has
well established psychometric properties.73'233 235
The SF-36 is 36-item, generic self-report measure of perceived health status covering 8
domains and 2 subscales. The 8 domains include physical functioning, role limitation as a result
of physical health problems, role limitations as a result of emotional health problems,
energy/fatigue, emotional well-being, social functioning, pain and general health perceptions (see
Figure 3.4).73,236 Each subscale and domain is scored from 0 (poor health) to 100 (optimal
health). The 8 domains are designed to capture the 2 major components of health, physical and
mental, contained in the 2 subscales.
61


Physical
Function
S
PCS
Physical health
subscale
Role-Physical
Bodily Pain
General Health
Mental Health
Role-Emotional
Social
Functioning
Vitality
Figure 3.4. Summary of SF-36 subscales and domains. Adapted from Ware237
The 2 subscales provide a more concise measure of overall physical and mental health
and are called the Physical Component Summary (PCS) and Mental Component Summary
(MCS) respectively. These summary scales use the items on the SF-36 with the PCS heavily
weighting physical measures and the MCS heavily weighting mental health measures. Reliability
coefficients for the physical and mental summary scores exceed 0.90 in previous studies.237 The
PCS and MCS are adjusted by the population mean and standard deviation to produce norm-
based scores with a common mean of 50 and standard deviation of 10. Thus, any score below 50
represents a decrement from nonnal health and functioning.
The physical health subscale or Physical Component Summary (PCS) comprises half of
the SF-36 items which include domains of perceived physical function, role-physical, bodily pain,
62


and general health (see Figure 3.4). It is important to note that the PCS does not actually measure
physical performance, ability or function. Rather it is a tool to indicate the patients perception of
physical function, physical role in his/her life, his/her interpretation of body pain, and his/her
overall indication of general health.
The mental health subscale or Mental Component Summary (MCS) comprises the other
half of the SF-36 items which include domains of vitality or energy, social functioning, role-
emotional, and mental health (see Figure 3.4). The mental health subscale in its name is
misleading as it represents dimensions of intrapersonal energy and emotional well-being but also
interpersonal interaction and social and emotional function. Again, it is important to note that this
scale represents the patients perceived levels of energy, emotional roles, social function, and
mental health. Moreover, it is important to note that the mental health subscale is does not
represent a tool for psychiatric diagnosis of a mental health condition. The mental health subscale
has been used to screen for psychiatric disorders such as depression,238 but is not as such a
diagnostic tool for any mental disorder.
Data Collection Aims 1 and 2: Quantitative
Quantitative data for Aims 1 and 2 were collected on 34 subjects upon entry into
treatment and 20 of these 34 subjects who completed post-treatment measurements. Baseline and
follow-up measures included demographics and information related to pain, disability, perceived
physical health, perceived mental health, self-efficacy for low back related tasks, and social
support. A summary of the quantitative data collection is below (Table 3.2). Copies of the surveys
are located in Appendix D.
63


Table 3.2. Data Collection Summary. Summary of quantitative data collection at entry to
treatment and post-treatment during long-term follow up period of 6-12 months after entry to
stud\ .___________________________________________________________________________
Entry to Treatment Post-Treatment
Demographics X
NPRS X X
ODI X X
SF-36 X X
LoBACS X X
MOS SSS X X
Data Analysis Aims 1 and 2: Quantitative
Prior to addressing specific Aims 1 and 2, the data were examined using descriptive
statistics to describe demographics of the sample using grouped frequency distributions and
measures of central tendency and variability at entry to treatment collected from the physical
exam and the baseline survey instruments (n=34). All survey data at entry to treatment and post-
treatment were converted to index scores and scored according to the literature for each
measurement tool. For each survey index score, parametric analysis was most appropriate in
meeting statistical assumptions for linearity, normally distributed errors, and uncorrelated errors.
Parametric data analysis has been recommended in the literature as preferable as long as
assumptions of linearity are met since parametric data analysis is more robust with small sample
sizes when compared to non-parametric analysis.239
Paired t-tests were performed to determine if there were pre to post differences between
demographics, survey measures of the dependent variables of pain and disability, and
independent variables of self-efficacy, social support, perceived physical health, perceived mental
health, and all the subscales of both self-efficacy and social support during the study period. Next,
associations between the independent variables and dependent variables using Pearsons
correlation coefficient were identified. Correlation analysis was conducted between the
independent variables of self-efficacy, social support, perceived physical health, perceived mental
health, and subscales of both self-efficacy and social support in association with the dependent
64


variables of pain and disability at entry to treatment and post-treatment. Correlations between
predictors of greater than .70 can lead to multicollinearity while correlations below .20 between
predictor and outcomes may not support the assumption of a linear relationship.239 Therefore, the
assumptions of linearity, normally distributed errors, and uncorrelated errors were checked and
met before further analysis. In addition, tests for mulitcollinearity were performed and checked
during multiple regression analysis procedures and reported.
Bivariate regression analysis. Next, a simple or bivariate linear regression was
conducted to investigate whether the individual variables of self-efficacy, social support,
perceived physical health, and perceived mental health independently predicted the dependent
variables of pain and disability in patients with LSS. These preliminary analyses were used to
determine the degree of individual association between the independent variables of self-efficacy,
social support and general measures of perceived physical and mental health on the dependent
variables of pain and disability in patients with LSS. Because these variables have largely been
unexplored in the LSS literature, understanding the individual contribution or prediction of the
criterion variables of pain and disability was an important first step.
Multiple regression analysis. It was estimated that each of the independent variables of
self-efficacy, social support, perceived physical health, and/or perceived mental health status may
not individually represent a strong association with pain and disability but collectively they could
contribute to a proportion of variability in the dependent variables. Because the independent
variables have not been explored together in the current literature for LSS, a further step of using
a multivariate analysis procedure of multiple regression further tested these relationships in the
clinical phenomena of pain and disability in patients with LSS. Multiple regression is supported
for analysis of theoretical components of constructs particularly when the purpose is to better
understand the factors associated with it.239 The theoretical constructs of self-efficacy measured
by the LoBACS self-efficacy scale and social support measured by the MOS SSS social support
scale along with perceived physical and mental health status measured by the SF-36 physical
65


health subscale (PCS) and mental health subscale (MCS) were all modeled a priori to determine
what combination, if any, of these constructs contributed to pain and disability at entry to
treatment and post-treatment in patients with LSS. The literature states 2 main purposes of the
modeling approach in multiple regression analysis: (1) to achieve best single predictor model, or
(2) to achieve an explanatory approach for developing new insights, exploring potentially
causal/explanatory relationships and designing future research.240 Statistical literature has stated
that the second approach to achieve explanation must necessarily simultaneously consider all
possible models.240241 The second approach to developing new insights and exploring causal
relationships underlies the multiple regression methods in this study.
The simultaneous multiple regression method or enter selection procedure for SPSS
version 19 (IBM SPSS Inc., Chicago IL) using underlying theory to enter all independent
variables in the model was chosen to test how the combination of self-efficacy, social support,
and perceived physical and mental health all contribute to the understanding of pain and
disability. This procedure tests all of the variables in a model at both time points of data
collection to determine their group association to the dependent variables of pain and disability. A
final step was taken if statistical significance was not found or if multicollinearity was found in
the group model. This final step tested the most parsimonious combination of all variables using a
backwards procedure in which the SPSS analysis program searches for the best fit with the
largest R2 and most significant F value.
A key methodological point is that all models were guided a priori by the existing
theoretical framework of self-efficacy and social support in a novel approach to test the
contribution of these factors in combination with the measures of general perceived physical
health and perceived mental health in the modeling process. Because there is no literature
comparing these variables in a similar sample of patients with LSS, there is no prior indication
about which variables will create the best prediction equation. In addition, since there is a small
set of predictors being tested, simultaneous regression is the best method to use to test the group
66


of predictors.242 The models for Aims 1 and 2 include the combinations of self-efficacy, social
support, perceived physical health, and perceived mental health to achieve the most clinically
relevant and parsimonious combination predicting the dependent variables of pain and disability.
Only significant model combinations are reported.
Since this is a relatively small sample with only 34 observed cases at entry to treatment
and only 20 of the original 34 observed cases followed post-treatment, testing the sample with all
4 independent variables reached the minimum recommendations of observations to variables
(5:1) cited in the literature for multiple regression analysis.239 An a priori alpha level of p= 0.05
was set to determine significance. The adjusted R2 value, reflecting the goodness of fit of the
linear model adjusted for the number of independent variables in the equation was calculated for
each model. The significance using the F statistic for each model was determined. Standardized
beta coefficients for each variable in the model were calculated and the significance of each was
determined under the null hypothesis that the coefficient was not different from zero.
Sample size and power analysis. It was anticipated that entering the independent
variables self-efficacy, social support, perceived physical health, and perceived mental health into
the multivariate models would result in a moderate effect size (0.30).243 Using an a priori sample
size calculator for multiple regression considering 4 independent variables, an alpha level equal to
0.05, and a desired power of 80%, a sample size of 42 subjects would be required. Therefore, 34
subjects at entry and 20 at follow-up was determined unlikely to provide adequate power and
therefore increased the risk of a type II error on detection of the true impact of the independent
variables in their contribution to pain and disability. However with a large effect size (0.50) as
reported by Cohen,243 a sample size of 27 subjects would be adequate from the 34 observations at
entry to treatment but still less than what was available in the follow-up data set of 20
observations. Caution is warranted in interpretation due to the potential of committing a type II
error with the small sample size. Therefore, a post hoc analysis was run on all models to help
identify the magnitude of the observed data. Because of the minimal evidence in the literature
67


regarding the contribution of self-efficacy, social support, and self-rated health in understanding
the severity and outcomes of pain and disability in LSS, this study provided a foundation for
future research and provided direction to further explore these variables with a larger data set.
Data Collection Aim 3: Qualitative
The focus of the interview was to explore the meanings of pain and disability in a sample
of patients with LSS. This focus guided the open ended questions to probe the details and depth
of identified factors of self-efficacy and social support using the theoretical constructs discussed
in Chapter 2. In addition, a broader sense of patient identified factors such as general health
perception was explored. Data collected in Aim 3 consisted of patient interviews and field notes
that were recorded at post-treatment, the time period of 6 to 12 month from entry into the study.
The quantitative data from Aims 1 and 2 were therefore supplemented by patient descriptions of
physical, behavioral, and psychosocial factors specifically self-efficacy, social support, and self-
reported health, as they related to perception of pain and disability upon entry into treatment and
post-treatment. Patient narratives were transcribed and coded by the researcher to characterize the
patients perceived status at entry to treatment and post-treatment related to pain and disability.
The interview explored the lived experience with LSS. Specifically, questions explored the
patients self-confidence for back related tasks as well as patients social support sources and if
these support sources were perceived as helpful or non-helpful during treatment and post-
treatment. The interview guide is included in Appendix E.
Data analysis Aim 3: Qualitative and Combined Methods
Identification, confirmation, or contradiction of factors related to self-efficacy, social
support, and general health status and their contribution to the understanding of pain and
disability were identified through analysis of qualitative interviews (n=20). Interviews were
conducted using an iterative process in which the researcher continued to interview subjects until
68


no new themes or factors related to the experience of living with LSS, experience of treatment, or
the concepts of self-efficacy, social support, perceived physical health, perceived mental health,
pain, and disability were revealed.
Interview transcripts were analyzed to identify any factors, including self-efficacy and
social support that contributed to the patients perceived pain and disability upon entry into
treatment and post-treatment. Verbatim transcripts from interviews were imported into the
qualitative software package Atlas .ti v6 (GmbH; Berlin, Germany), and preliminary themes were
identified and coded. In the first phase, the focus was on discovery and comprehension,244
whereby segments of data in a quote, sentence, or paragraph were coded with words or phrases
that identified a particular theoretical constructs related to general health status, self-efficacy,
social support, pain, and/or disability. These codes were complemented by de novo coding
which represents new information.245,246 These initial codes reflected quotes using the
informants original words, and captured the essential meaning of a concept, idea, or
description.246 Each transcript was read for code specific phrases until no new codes were
identified in the data.
While the coding process began with the deductive process of breaking the narrative
down into small coded phrases, it ended with the inductive process of recognizing thematic
commonalities and generating observations which described typical behavior patterns. Major
concepts or domains in the data were used that linked back to groups of statements or quotes.
Domains related to pain and disability were compared and contrasted and relationships between
them were identified and recorded on the network diagram of Atlas.ti v6. This process focused on
synthesizing the data to define and clarify the essential concepts related to pain and disability
revealed by the informants.
In the final stage of data analysis, theories were incorporated that built working
explanations about the patterns observed in the data. The Stress and Coping Model247 was used as
the underlying theory for understanding the appraisals of the stressors of pain and disability and
69


coping processes and resources identified by patients with LSS. Additionally, the Stress-Diathesis
Model125 representing a biopsychosocial approach to pain was also used to identify the experience
of pain. The WHO ICF framework248 was used as the overarching theoretical framework of
disability. Finally, the Social Cognitive Theory112,153 and the Stress and Coping Model were used
to interpret data pertaining to the constructs of self-efficacy, social support, general perceived
health status, and any other uncovered coping resources. A diagram that outlines the association
between beliefs and behaviors related to pain and disability was developed in the final combined
methods procedures to illustrate if and how factors of self-efficacy, social support, general health
perception as well as other uncovered factors contribute to the experience of pain and disability in
patients with LSS.
70


CHAPTER IV
RESULTS
Descriptive Results
Complete data sets on 34 patients upon entry to treatment and the remaining 20 patients
post-treatment are represented below in the demographic analysis of the subset of patients with
LSS. Demographic characteristics were reported with mean and standard deviations for interval
data and with frequencies for categorical data in Table 4.1. The average age of the patients was
68 years at entry to treatment with only a slight increase to 71 years for the 20 patients remaining
at follow up. These ages represent the typical age group found in patients with LSS. There were
no statistical differences between the entry to treatment sample of 34 patients and the remaining
20 patients at post-treatment except for prior history of leg pain where 76% in the entry to
treatment sample had a history of leg pain but 90% of the remaining 20 subjects had previous leg
pain. Across the study sample, approximately 55% of the patients were male and 45% were
female at the 2 time points. The race distribution in the sample was predominately white (70%)
with over half of the sample married with a spouse (55%). Only a quarter of the patients currently
worked in some capacity with the approximately half of patients retired, although 15% of the
patients were permanently unable to work due to work or disease related medical disability status.
All patients in the sample came to treatment complaining of LBP with approximately
95% complaining of pain into buttock and leg(s). Nearly all patients rated their pain best when
sitting (95%), however a few reported sitting caused increased pain which conflicts with the
clinical diagnostic criteria although it is possible depending on acuity and severity that they had
pain in any position.249 The average pain duration upon entry to the study for this current episode
of LBP was 12 months prior to entry, with most patients reporting a prior history of LBP (90%).
71


Table 4.1. Demographic Characteristics of the Sample of Patients with LSS
Demographics Entry to Rehab (n=34) Post-Rehab (n=20) Sig.
Age 68.382 (8.714) 70.737 (8.881) 0.193
Gender male 18 (53%) 11 (55%) 0.921
Gender female 16 (47%) 9 (45%) 0.921
Race White 24 (73%) 14 (70%) 0.304
Race Black 5 (15%) 4 (20%) 0.304
Race Hispanic 1 (3%) 1 (5%) 0.304
Race Asian 2 (6%) 0 (0%) 0.304
Race Pacific Islander 1 (3%) 0 (0%) 0.304
Resting HR (bpm) 78.409 (17.220) 77.588 (16.915) 0.873
Resting BP -SBP (mmHg) 129.462 (21.424) 128.850 (23.623) 0.821
Resting BP DBP (mmHg) 78.360 (11.665) 79.105 (12.679) 0.271
BMI 29.615 (4.880) 29.977 (5.152) 0.741
Single 2 (6%) 1 (5%) 0,441
Married 21 (62%) 11 (55%) 0.441
Divorced 4 (12%) 2 (10%) 0.441
Widowed 7 (21%) 6 (30%) 0.441
Smoke now 3 (9%) 1 (5%) 0,340
Smoked over 100 cigarettes in lifetime 17 (50%) 12 (60%) 0.186
Low back pain 34 (100%) 20 (100%) 1.000
Pain beyond buttock 33 (97%) 19 (95%) 0.227
Current episode LBP (months) 14.348 (19.905) 12.467 (9.372) 0.227
Prior history LBP 29 (85%) 19 (95%) 0.080
Prior history leg pain and numbness 26 (76%) 18 (90%) 0,010
Physical activity <3 days/week 16 (47%) 9 (45%) 0.829
Physical activity 3-4 days/week 8 (24%) 5 (25%) 0.829
Physical activity 5 days/week 5 (15%) 4 (20%) 0.829
No physical activity 3 (9%) 2 (10%) 0.829
Education high school 11 (32%) 5 (25%) 0.450
Education some college 9 (26%) 6 (30%) 0.450
Education graduated college 5 (15%) 3 (15%) 0.450
Education some post graduate 2 (6%) 2 (10%) 0.450
Education completed post graduate 7 (21%) 4 (20%) 0.450
Work full time 6(18%) 4 (20%) 0.386
Work part time 1 (3%) 1 (5%) 0.386
Work lermanentlT unable 5(15%) 3 (15%) 0.386
Work retired 16 (47%) 10 (50%) 0.386
Work Homemaker ^5 (15%)^^H ^M2 (10%)^^ k'386i
*p<05, **p<01, ***p<001
72


Dependent Variables
Rating of pain. Patients rated their pain at entry to treatment at varying levels of
intensity. Mean score on the NPRS for the sample at entry to treatment was 5.27 (2.57) which
represent a moderate level of pain (4-6/10). At entry to treatment, 46% of patients rated their
current pain in the high category, 29% rated their current pain in the moderate category, 23%
rated their current pain in the low category, and 3% or 1 patient rated he had no pain (see Table
4.2). Irrespective of treatment condition, pain scores decreased over the study period with mean
pain score for the sample at entry to treatment 5.27 (2.57) and mean post-treatment score 4.14
(2.52) indicating reduced pain of 1.13 points post-treatment although this was not significant (see
Table 4.4). Post-treatment, only 25% of patients rated their current pain in the high category (2 in
ESI group, 3 in ESI+PT group), 25% in the moderate category (4 in ESI group, 1 in ESI+PT
group), 40% in the low category (4 in ESI group, 4 in ESI+PT group), and 10% with no pain (2 in
ESI group, 0 in ESI+PT group) (see Table 4.2).
Table 4.2. Pain Rating in Patients with Lumbar Spinal Stenosis Pre and Post-Treatment
Pain Ratin 2 No Pain % RBI % Moderate (4-6) % %
Entry to Treatment (n=34) 1 3% 8 24% 10 29% 15 44%
Post-Treatment (n=20) 2 10% 8 40% 5 25% 5 25%
Rating of disability. Patients at entry to treatment all reported higher levels of disability
as measured by the ODI. The mean disability score for the patients at entry to treatment was
40.34 out of 100 (13.89) which indicate borderline high disability (see Table 4.4). Patients
reported their overall disability at entry to treatment as 61% in the moderate disability category,
26% in the severe disability category, and 12% in the crippled category indicating they were
extremely limited in all daily activities (see Table 4.3 and Figure 4.1). Irrespective of treatment
condition, mean disability scores significantly decreased post-treatment indicating improved
function and reduced disability from entry to treatment 40.34 (13.89) to post treatment 29.29
73


(17.06)(see Table 4.4). Post-treatment, 25% of the patient reported disability in the low category
(3 in ESI group, 2 in ESI+PT group), 40% in the moderate category (5 in ESI group, 3 in ESI+PT
group), 25% in the severe category (3 in ESI group, 2 in ESI+PT group), and 10% in the crippled
category (0 in ESI group, 2 in ESI+PT group)(see Table 4.3).
Table 4.3. Disability Rating in Patients with Lumbar Spinal Stenosis Pre and Post-
Treatment
Disability Ratin g Low (1-19) % Moderate (20-40) % Severe (41-60) % Crippled (61+) %
Entry to Treatment (n=34) 0 0% 21 61% 9 26% 4 12%
Post-Treatment (n=20) 5 25% 8 40% 5 25% 2 10%
Oswestry Disability Index Scores
n=34
0-20: Minimal disability
20-40: Moderate disability
41-60: Severe disability
61-80: Crippled
Figure 4.1. Oswestry Disability Index score at entry to treatment in patients with LSS. The
majority of patients fall into the moderate disability category with severe disability next highest
and crippling disability third highest category. No patients fell into minimal or no disability
categories.
Mean scores for both disability and pain positively improved from entry to treatment to
post-treatment however only the change in disability from entry to post-treatment was significant
(p=.04) (see Table 4.4). In addition, a higher pain rating at entry to treatment was significantly
correlated with a higher total disability score as indicated by the Oswestry Disability Index
74


p=.444 (p=.016)(n=34). After treatment, patients with a lower pain rating were significantly
correlated with having reduced disability scores p =.610 (p=.006)(n=20).
Table 4.4. Means, Standard Deviations, and Paired T-test for Disability and Pain_______
Variable M SD Si:
Disability
Entry to Treatment (n=34) 40.338 13.891
Post-Treatment (n=20) 29.286 17.056 0.040*
Pain
Entry to Treatment (n=34) 5.265 2.574
Post-Treatment (n=20) 4.143 2.516 0.18
*p<05, **p<01, ***p<001
Table 4.5 summarizes the correlation analysis results for disability. The independent
variables of self-efficacy p=-.319 (p<.05), social support p=-.343 (p<.05), physical health status
p=-.610 (p<.001), and mental health status p=-.444 (p<01) are negatively and significantly
correlated with disability upon entry to treatment, indicating that those with lower self-efficacy,
lower social support, worse perceived physical health, and worse perceived mental health tend to
have higher disability scores.
Only the variables of self-efficacy and social support were significantly correlated with
disability post-treatment.243 Patients who increased or improved their self-efficacy scores were
significantly more likely to reduce their disability scores p=-.620 (p<.01). However, social
support had a significant and positive correlation with disability p=.389 (p<.001), indicating that
those with reduced social support had an overall reduction in disability post-treatment.
Table 4.5. Correlation Analysis Results for Disability
Variable Entry to Treatment Post-Treatment
Disability
Self-Efficacy -0.319* -0.620**
Social Support -0.343* 0.389**
Perceived Physical Health -0.610*** -0.886
Perceived Mental Health *p<05, **p<01, ***p<001 -0.444** -0.085
75


Table 4.6 summarizes the correlation analyses for pain. Only the independent variable of
perceived physical health had a significant negative association with pain upon entry to treatment
p=-.550 (p<.05), indicating that those with higher pain scores upon entry to treatment had a worse
perceived physical health status. The functional self-efficacy subscale p=-.332 (p<.05) and
perceived physical health p=-.631 (p<.01) were both significantly and inversely correlated with
pain post-treatment. This would indicate that a patient who had higher functional self-efficacy
and better perceived physical health long-term tended to have lower pain scores.
Table 4.6. Correlation Analysis Results for Pain
Variable Entry to Treatment Post-Treatment
Pain
Self-efficacy subscales
Functional Self-efficacy -.420 -0.332*
Self-Regulation Self-efficacy -.256 -0.082
Exercise Self-efficacy Social Support subscales -.190 0.014
Emotional Support -.205 -0.004
Tangible Support -.411 -0.0253
Affectionate Support -0.089 0.214
Positive Social Interaction -.179 0.007
SF-36 subscales
Perceived Physical Health -.550* -0.631**
Perceived Mental Health .038 0.012
*p<05, **p<01, ***p<001
Independent Variables
Upon entry to treatment, 16 of the 20 subjects who were followed post-treatment had
moderate to high levels of self-efficacy. Four patients had low levels of self-efficacy, below 50
out of 100. Of those 4, 2 patients continued to have low self-efficacy but 2 increased their self-
efficacy above 50. Additionally, 2 patients reduced their self-efficacy over the study period
(Table 4.7). Regarding social support, 15 of the 20 subjects had moderate to high levels of
perceived support. Five patients had low perceived support entering the study. Over the study
period, 4 of the 5 patients with low social support increased their perceived support scores over
50%. However 1 patient continued to report low social support at follow-up. In addition, 1 patient
76


reduced her perceived support scores below 50 out of 100 during the follow up period (Table
4.7).
Table 4.7. Levels of Self-Efficacy and Social Support At Entry and Post-Treatment
Entry to Treatment Mod-High >50 Entry to Treatment Low<50 Post-Treatment Mod-High >50 Post- Treatment Low <50
Self-efficacy 16 4 16 4* *2 continued low *2 improved *2 reduced
Social support 15 5 18 2* *1 continued low *1 improved *1 reduced
Tables 4.8 and 4.9 show the means and standard deviations for the independent variables
of self-efficacy, social support, perceived physical health, perceived mental health, and the
subscales of both self-efficacy and social support at the time points of entry to treatment and post-
treatment. All of the mean survey scores for the sample positively changed or improved from
entry to treatment to post-treatment. The only significant differences of score improvement were
seen entry to post-treatment between the subscale measures of functional self-efficacy (p=.043),
emotional support (p=.043) and positive social interaction (p=.040) (see Table 4.8 and 4.9).
Table 4.8. Means, Standard Deviations and Paired T-test of Predictor Variables of Self-
efficacy, Social Support, Physical Health and Mental Health________________________
Variable Mean SD Sig
Total Self-Efficacy
Entry to Rehab (n=34) 59.519 20.484
Post-Rehab (n=20) 70.393 21.805 0.266
Total Social Support
Entry to Rehab (n=34) 72.753 23.356
Post-Rehab (n=20) 81.674 18.376 0.106
| Physical Health Status
Entry to Rehab (n=34) 33.768 8.228
Post-Rehab (n=20) 41.171 8.903 0.102
Mental Health Status
Entry to Rehab (n=34) 47.721 10.948
Post-Rehab (n=20) 52.01 8.702 0.073
*p< 10; **p<05; ***p<01
77


Table 4.9. Means, Standard Deviations, and Paired T-test for Predictor Variables of
Subscales of Self-Efficacy and Subscales of Social Support_____________________________
Variable Mean SD Sig.
Functional Self-efficacy I
Entry to Rehab (n=34) 41.899 27.383
Post-Rehab (n=20) 53.698 32.21 0.043*
Self Regulation Self-efficacy
Entry to Rehab (n=34) 66.667 27.377
Post-Rehab (n=20) 76.481 23.556 0.639
Exercise Self-efficacy
Entry to Rehab (n=34) 76.059 20.606
Post-Rehab (n=20) 80.889 18.939 0.612
Emotional Support
Entry to Rehab (n=34) 71.232 22.961
Post-Rehab (n=20) 84.375 18.006 0.043*
Tangible Support
Entry to Rehab (n=34) 71.838 30.048
Post-Rehab (n=20) 82.721 21.025 0.407
Affectionate Support
Entry to Rehab (n=34) 79.865 28.52
Post-Rehab (n=20) 90.741 15.095 0.363
Positive Social Interaction
Entry to Rehab (n=34) 74.755 29.764
Post-Rehab (n=20) 89.815 15.539 0.040*
*p<10; **p<05; ***p<01
Aim 1: Entry to Treatment
Aim 1: To determine the relative contribution of self-efficacy, social support, self-rated
physical health, and self-rated mental health to the severity of pain and disability among
patients with LSS upon entry to treatment of LSS.
Aim 1: Hypothesis Testing A & B
H1A: Lower levels of self-efficacy, social support, perceived physical health and
perceived mental health will each independently predict higher disability at entry to
treatment.
H1B: Lower levels of self-efficacy, social support, perceived physical health and
perceived mental health will each independently predict higher pain at entry to treatment.
78


Bivariate Regression Analysis: Pain and Disability at Entry to Treatment
A perception of worse physical health significantly predicted greater disability at entry to
treatment, F(l,32) = 18.935, p<.001, adjusted R2 = .352. Upon entry to treatment, the beta
weights presented in Table 4.10 indicate when perceived physical health scores decrease or get
worse by 1 unit, disability scores increase or get worse by 1.209 units. This relationship between
perceived physical health and disability had adequate post-hoc observed power >80% and
supports hypothesis H1A, therefore the null hypothesis can be rejected (see Table 4.10).
A perception of worse mental health significantly predicted greater disability at entry to
treatment, F(l,32) = 7.837, p<.01, adjusted R2 = .197. The beta weights presented in Table 4.10
indicate when perceived mental health scores decrease or get worse by 1 unit, disability scores
increase or get worse by .563 units. This relationship supports hypothesis H1A and the post-hoc
observed power is 81%; therefore the null hypothesis can be rejected.
Lower social support significantly predicted greater disability at entry to treatment,
F(l,32) = 4.253, p<.05, adjusted R2 = .10. The beta weights, presented in Table 4.10 indicate
when social support scores decrease by 1 unit, disability scores increase or get worse by .204
units. This relationship supports hypothesis H1A. However, using a post-hoc observed power
analysis, this relationship was underpowered at 30%. No other significant associations were
found between the individual independent variables of self-efficacy, social support, perceived
physical health, and perceived mental health and the dependent variable of disability upon entry
to treatment.
79


Table 4.10. Bivariate Linear Regression Results for Greater Disability at Entry to
Treatment
Disability R2 Adjusted R2 F B SEB B2 Sig
Entry to Treatment
Self-efficacy 0.102 0.073 3.617 -0.216 0.114 -0.319 0.066
Social Support 0.117 0.10 4.253 -0.204 0.099 -0.343 0.047*
Physical Health 0.372 0.352 18.935 -1.029 0.237 -0.610 0.000***
Mental Health 0.197 0.172 7.837 -0.563 0.201 -0.444 0.009**
*p<05,**p<01,: ***p<001
Using the self-efficacy subscales, only the subscale of functional self-efficacy
independently and significantly predicted greater pain at entry to treatment, F(l,32) =4.824,
p<.05, adjusted R2 = .104. The beta weights presented in Table 4.11 illustrate when functional
self-efficacy scores decrease or get worse by 1 unit, pain scores increase or get worse by .034
units. This relationship supports the hypothesis H1B however using post-hoc observed power
analysis this was underpowered at 52%.
Table 4.11. Bivariate Linear Regression Results for Higher Pain at Entry to Treatment
Pain R2 Adjusted R2 F B SEB B2 Sig
Entry to Treatment
Functional self-efficacy 0.131 0.104 4.824 -0.034 0.015 -0.362 0.035*
Self-regulation self- efficacy 0.012 -0.019 0.374 -0.01 0.017 -0.108 0.545
Exercise self-efficacy 0.012 -0.019 0.4 -0.014 0.022 -0.111 0.531
Emotional support 0.022 -0.008 0.724 0.017 0.02 0.149 0.401
Tangible support 0.011 -0.02 0.356 -0.009 0.015 -0.105 0.555
Affectionate support 0.016 -0.015 0.521 -0.011 0.016 -0.127 0.476
Positive Social Interaction 0.000 -0.031 0.004 0.001 0.015 0.011 0.942
Physical Health 0.070 0.041 2.415 -0.083 0.053 -0.265 0.130
Mental Health 0.007 -0.024 0.232 0.02 0.041 0.085 0.633
*p<05,**p<01,***p<001
Aim 1: Hypothesis Testing C & D
H1C: Lower levels of self-efficacy, social support, perceived physical health and
perceived mental health will together predict higher disability at entry to treatment.
80


HID: Lower levels of self-efficacy, social support, perceived physical health and
perceived mental health will together predict higher pain at entry to treatment.
Multiple Regression Analysis: Disability and Pain at Entry to Treatment
With self-efficacy, social support, perceived physical health and perceived mental health
in the model, the combinations of all 4 variables significantly predicted higher disability at entry
to treatment, F (4, 29) = 6.920, p<.001, R2=.488, with only perceived physical and mental health
significantly contributing to the prediction. The adjusted R2 value was .418 indicating that 42% of
the variance in disability at entry to treatment was explained by the model. The beta weights,
presented in Table 4.12, suggest that worse perceived physical health and worse perceived mental
health were the only significant contributors to predicting greater disability at entry to treatment.
Post-hoc observed power analysis revealed that this relationship was adequately powered at 99%
and there was no multicollinearity in the data. This relationship supports hypothesis H1C that a
worse perception of physical health and a worse perception of mental health predict greater
disability at entry to treatment.
Table 4.12. Model 1: All Variables Contribute to Greater Disability at Entry to Treatment
B SEB B2
Self-Efficacy -0.115 0.102 -0.17
Social Support 0.009 0.097 0.015
Physical Health -0.829 0.242 -0.491*
Mental Health -0.399 0.188 -0.315*
Constant 93.636 10.617
Note. R2=.488; F(4,29) = 6.920, p<001
*p<05, **p<01, ***p<001
With 4 variables of self-efficacy, social support, perceived physical health and perceived
mental health in the model, the only significant model using subscales of self-efficacy in
predicting greater pain at entry to treatment included functional self-efficacy, total social support,
perceived physical health and perceived mental health, F (4, 29) = 2.717, p<.05, R2=.273, with
81


only the variable of functional self-efficacy significantly contributing to the prediction. The
adjusted R2 value was .172 indicating that 17% of the variance in pain was explained by the
model. The beta weights, presented in Table 4.13, suggest that reduced functional self-efficacy
predicted greater pain at entry to treatment supporting the hypothesis HID. Post-hoc observed
power analysis revealed that this relationship was underpowered at 76% and there was
multicollinearity found in the data between social support and mental health.
Table 4.13. Model 2: All Variables Contribute to Higher Pain at Entry to Treatment
B SEB B2
Functional Self-Efficacy -0.046 0.018 -0.489*
Social Support 0.038 0.022 0.341
Physical Health -0.082 0.053 -0.262
Mental Health 0.027 0.041 0.114
Constant 5.995 2.251
Note. R2=.273; F(4,29) = 2.717, p<05
*p<05, **p<01, ***p<001
Using the backward entry method, only the 2 variables of functional self-efficacy and
social support were found to be most parsimonious in the model, significantly predicted greater
pain at entry to treatment, F (2, 31) = 4.080, p<05, R2=.208, with only functional self-efficacy
significantly contributing to the prediction. The adjusted R2 value was .157 indicating that 16% of
the variance in pain was explained by the model. Post-hoc observed power analysis revealed that
this relationship was under powered at 74% but there was no multicollinearity in the data. The
beta weights, presented in Table 4.14, suggest that having reduced functional self-efficacy
contributes to predicting greater pain at entry to treatment supporting hypothesis HID.
Table 4.14. Model 3: Two Variables Optimally Contribute to Higher Pain at Entry to
Treatment
B SEB B2
Functional Self-Efficacy -0.05 0.018 -0.532*
Social Support 0.036 0.021 0.326
Constant 4.796 1.344
Note. R2=.208; F(2,31) = 4.080, p<05
*p<.05, **p<01, ***p<001
82


With 4 variables in the model, the only combination of subscales of social support that
significantly predict greater pain at entry to treatment was the model with emotional support, total
self-efficacy, perceived physical health, and perceived mental health, F (4, 29) = 2.739, p<.05,
R2=.274, with total self-efficacy significantly contributing to the prediction. The adjusted R2
value was .174 indicating that 17% of the variance in pain was explained by the model. Post-hoc
observed power analysis revealed that this relationship was under powered at 75% but there was
no multicollinearity in the data. The beta weights, presented in Table 4.15, suggest that reduced
total self-efficacy predicts greater pain at entry to treatment supporting hypothesis HID.
Table 4.15. Model 4: All Variables Contribute to Higher Pain at Entry to Treatment
B SEB B2
Emotional Support 0.039 0.02 0.351
Self-Efficacy -0.049 0.021 -0.391*
Physical Health -0.093 0.054 -0.298
Mental Health 0.024 0.039 0.102
Constant 7.388 2.366
Note. R2=.274; F(4,29) = 2.739, p<05
*p<.05, **p<01, ***p<001
Aim 2: Post-Treatment
Aim 2: To determine the relative contribution of self-efficacy, social support, self-rated
physical health, and self-rated mental health to the outcomes of pain and disability among
patients with LSS in the year following treatment of LSS.
Aim 2: Hypothesis Testing A & B
H2A: Higher levels of self-efficacy, social support, perceived physical health and
perceived mental health will each independently predict lower disability after treatment
of LSS.
83


H2B: Higher levels of self-efficacy, social support, perceived physical health and
perceived mental health will each independently predict lower pain after treatment of
LSS.
Bivariate Regression Analysis: Pain and Disability Post-Treatment
The association between self-efficacy, social support, perceived physical health, and
perceived mental health in relationship to pain and disability was followed up in 20 of the
original 34 patients with LSS post-treatment. Higher total self-efficacy significantly predicted
reduced disability post-treatment, F(l, 18) =14.942, p<.01, adjusted R2 = .423. The beta weights,
presented in Table 4.18 indicate when total self-efficacy scores increased by 1 unit, disability
scores decreased by .484 units. Using a post-hoc observed power analysis, this relationship has
adequate power at 95%. In addition, this relationship supports hypothesis H2A and the null
hypothesis can be rejected for the post-treatment period.
Better perceived physical health significantly predicted reduced disability post-treatment
F( 1,17) = 55.458, p<.05, adjusted R2 = .765. At post-treatment, when perceived physical health
scores improved or increased by 1 unit, disability scores decreased by 1.681 units. This
relationship between better perceived physical health and reduced disability had adequate post-
hoc observed power 99% and supports hypothesis H2A, therefore the null hypothesis can be
rejected (see Table 4.16). No other significant associations were found between the individual
independent variables of total self-efficacy, total social support, perceived physical health and
mental health and the dependent variable of disability post-treatment.
84


Table 4.16. Bivariate Linear Regression Results for Predicting Reduction of Disability Post-
Treatment
Disability R2 Adjusted R2 F B SEB B2 Sig
Post-Treatment
Self-efficacy 0.454 0.423 14.942 -0.484 0.125 -0.673 0.001**
Social Support 0.000 -0.055 0.003 -0.012 0.215 -0.014 0.954
Physical Health 0.765 0.752 55.458 -1.681 0.226 -0.875 0.000***
Mental Health 0.003 -0.056 0.049 -0.1 0.452 -0.054 0.828
*p<05,**p<01,***p<001
Only better perceived physical health independently and significantly predicted reduced
pain post-treatment, F(l,7) = 6.931, p<.05, adjusted R2 = .259. The beta weights, presented in
Table 4.17 indicate when perceived physical health scores improve or increased by 1 unit, pain
scores decreased by 157 units. This relationship supports hypothesis H2B that better perceived
physical health can predict a reduction in pain however post-hoc observed power analysis reveals
lack of power at 72%.
Table 4.17. Bivariate Linear Regression Results for Predicting Reduced Pain Post-
Treatment
Pain R2 Adjusted R2 F B SEB B2 Sig
Post-Treatment
Functional self-efficacy 0.177 0.125 3.436 -0.034 0.018 -0.42 0.082
Self-regulation self- efficacy 0.065 0.01 1.189 -0.025 0.023 -0.256 0.291
Exercise self-efficacy 0.036 -0.02 0.639 -0.023 0.028 -0.19 0.435
Emotional support 0.042 -0.022 0.659 -0.027 0.033 -0.205 0.43
Tangible support 0.169 0.117 3.252 -0.04 0.022 -0.411 0.09
Affectionate support 0.008 -0.05 0.137 -0.012 0.032 -0.089 0.716
Positive Social Interaction 0.032 -0.025 0.564 -0.027 0.036 -0.179 0.463
Physical Health 0.302 0.259 6.931 -0.157 0.06 -0.55 0.018*
Mental Health 0.001 -0.061 0.023 0.01 0.068 0.038 0.88
*p<05,**p<01,***p<001
Aim 2: Hypothesis Testing C & D
H2C: Higher levels of self-efficacy, social support, perceived physical health and
perceived mental health will together predict lower disability after treatment of LSS.
85


Full Text

PAGE 1

A MIXED METHOD INVESTIGATION OF PREDICTORS OF PAIN AND DISABILITY IN PATIENTS WITH LUMBAR SPINAL STENOSIS. by AMY STONE HAMMERICH B.S., Miami University, 1992 M.S.P.T., Regis University, 2000 D.P.T., Regis University, 2003 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Doctor of Philosophy Health and Behavioral Sciences 2013

PAGE 2

ii This thesis for the Doctor of Philosophy degree by Amy Stone Hammerich has been approved for the Health and Behavioral Sciences Program by Deborah Main, Chair Susan Dreisbach, Advisor Susan Scherer Joshua Cleland November 1, 2013

PAGE 3

iii Hammerich, Amy Stone (Ph.D., Health and Behavioral Sciences) A Mixed Method Investigation of Predictors of Pain and Disability in Patients with Lumbar Spinal Stenosis Thesis directed by Assistant Research Professor Sus an L. Dreisbach. ABSTRACT This study identified characteristics associated wi th pain and disability in LSS at entry to treatment and factors contributing to long-term red uction of pain and disability. It also revealed how patients with LSS appraise the stressors of pai n and disability and use coping resources. This study used qualitative and quantitative methods to address the dearth of research in the LSS literature regarding factors associated with severi ty of pain and disability upon entry to treatment and the influence of coping resources on post-treat ment outcomes. A cohort of 34 patients was evaluated to understand the relative contribution o f perceived health, self-efficacy, and social support on the stressors of pain and disability usi ng a combination of bivariate and multivariate regression analyses and patient interviews. Several key findings emerged from this research. Fi rst, there was an inverse relationship between perceived physical and mental health and se verity of disability upon entry to treatment. In interviews, patients with LSS described a variet y of sources contributing to stressors of pain and disability including physical limitations, redu ced confidence, lack of control, vulnerability, mental health concerns, reduced social participatio n, frustrations at needing support, and financial limitations. Better perceived physical health had t he greatest contribution to reduced levels of pain and disability post-treatment. In interviews, a variety of behaviors were described that improved coping including a greater understanding o f limitations and needs along with beliefs for pushing through and remaining positive. In line wit h the Stress and Coping Model, patients with stronger coping behaviors and more resources reduce d the severity and improved outcomes of pain and disability associated with LSS.

PAGE 4

iv There is an increased need for research improving t he quality of life in later years particularly when faced with a chronic disabling co ndition such as LSS. Early identification of those at risk for higher pain and disability as wel l as support for positive coping behaviors posttreatment can help patients with LSS maintain a hig her level of participation in society. The factors uncovered in this study serve to inform tra nslational research, clinical practice, and policy recommendations, which may ultimately add to treatm ent and resources for reducing pain and disability associated with LSS. The form and content of this abstract are approved. I recommend its publication. Approved: Susan L. Dreisbach

PAGE 5

v ACKNOWLEDGMENTS I would like to thank the faculty at Regis Universi ty School of Physical Therapy who gave me the support and flexibility to follow my pa ssions. In addition, I would like to thank the Health and Behavioral Science Department for challe nging my preconceptions, testing my limits, and giving me new perspectives on health and health care. First and foremost in the Health and Behavioral Science Department, I would like to than k Dr. Susan Dreisbach for her mentorship and teaching me the art of qualitative research. I would also like to thank Dr. Debbi Main for showing me how to define the most important questio ns that need to be answered. In addition, I would like to thank Dr. Josh Cleland for his mentorship over the years. With this mentorship, I have appreciated his persis tence, patience, and sense of exploration as we continue to identify and understand the complex nee ds of those with the chronic condition of lumbar spinal stenosis. I would also like to thank Dr. Julie Whitman who was there in the beginning and continues to be a constant force in t his research. Moreover, I would like to thank Dr. Sue Scherer for all of her past and continued m entorship in academic, research, and clinical endeavors. Most of all, I would like to thank my hu sband, Scott who was my strongest and most resilient support in every adventure during my PhD process.

PAGE 6

vi TABLE OF CONTENTS CHAPTER I. BACKGROUND AND SIGNIFICANCE .................... ................................................... .......... 1 Low Back Pain ..................................... ................................................... ............................... 1 Lumbar Spinal Stenosis (LSS) ..................... ................................................... ................. 1 Definition, pathoanatomy, and clinical presentation of LSS .............................. 2 Incidence and impact of LSS ....................... ................................................... .... 2 Limitations in the current literature for LSS ..... .................................................. 4 Rationale ......................................... ................................................... ...................................... 4 Research Question and Specific Aims .............. ................................................... ............ 5 Research Goals .................................... ................................................... ......................... 7 II. REVIEW OF THE LITERATURE ...................... ................................................... ................. 8 Pain and Disablement in LSS ....................... ................................................... ....................... 8 Previous Literature: Psychosocial Factors .. ................................................... .................. 9 Previous Literature: General Health Status ........ ................................................... ......... 11 Previous Literature: Research Design in LSS ....... ................................................... ...... 13 Previous Literature: Summary ............... ................................................... ..................... 15 Theoretic Perspectives ............................ ................................................... ............................ 15 Integrating Theory in LSS ......................... ................................................... ................. 15 Theoretical Framework ............................ ................................................... ................... 16 Stress and Coping Model ........................... ................................................... ................. 16 Cognitive appraisal ............................. ................................................... ........... 18 Personal and situational factors .................. ................................................... ... 18 Different appraisals .............................. ................................................... .......... 19 Coping ............................................ ................................................... ................ 20 Application of the Stress and Coping Model to LBP.. ...................................... 21

PAGE 7

vii Managing Stressors: Pain .......................... ................................................... ........................ 22 Biopsychosocial Models of Pain .................... ................................................... ............. 22 Stress-Diathesis Model of Pain ................... ................................................... ................ 24 Pain appraisals ................................... ................................................... ............ 25 Pain beliefs ...................................... ................................................... ............... 25 Managing Stressors: Disability..................... ................................................... ..................... 26 Disablement Model ................................. ................................................... .................... 26 Expanded Disablement Model ...................... ................................................... 27 Social model of disability ........................ ................................................... ...... 29 ICF Framework ..................................... ................................................... ...................... 29 Functioning .................................... ................................................... ............... 30 Contextual factors ................................ ................................................... .......... 30 Application of disablement models in LBP .......... ............................................ 33 Resource Theories of Stress and Coping ............ ................................................... ............... 33 Social Cognitive Theory and the Construct of SelfEfficacy ....................................... 34 Self-efficacy ..................................... ................................................... .............. 35 Application of self-efficacy ...................... ................................................... ..... 37 Application of self-efficacy during treatment ..... .............................................. 38 Social Relationships and Health .................. ................................................... ............... 39 Social support .................................... ................................................... ............ 40 Application of social support in health ........... .................................................. 42 General Health Status ............................. ................................................... .................... 44 Appraisal of health ............................... ................................................... .......... 45 Application of self-rated health .................. ................................................... ... 46 Theoretical Summary ............................... ................................................... ........................... 46 Gaps Filled ...................................... ................................................... ............................ 47

PAGE 8

viii III. EXPERIMENTAL DESIGN AND METHODS .............. ............................................... 48 Research Design ................................... ................................................... ............................... 48 Subset of the RCT ................................. ................................................... ...................... 50 Mixed Method Design ............................... ................................................... ................. 51 Data Collection .................................. ................................................... ......................... 53 Sampling and participants ......................... ................................................... ..... 53 Subject randomization, baseline examination, and tr eatment procedures ........ 53 Subjects .......................................... ................................................... ................ 55 Research Question and Specific Aims ............... ................................................... ............... 58 Measures and Constructs .......................... ................................................... .................. 59 Measuring pain: Numeric Pain Rating Scale (NPRS) .. .................................... 59 Measuring disability: Oswestry Disability Index (OD I) ................................... 59 Measuring self-efficacy: Low Back Activity Confid ence Scale (LoBACS) .... 60 Measuring social support: Medical Outcomes Survey S ocial Support Scale (MOS SSS) ......................................... ................................................... ........... 60 Measuring perceived health status: Medical Outcomes Survey Short Form 36 (SF-36) ........................................... ................................................... ................ 61 Data Collection Aims 1 and 2: Quantitative ....... ................................................... ........ 63 Data Analysis Aims 1 and 2: Quantitative ......... ................................................... ......... 64 Bivariate regression analysis ................... ................................................... ...... 65 Multiple regression analysis .................... ................................................... ...... 65 Sample size and power analysis .................. ................................................... ... 67 Data Collection Aim 3: Qualitative ............... ................................................... ............. 68 Data Analysis Aim 3: Qualitative and Combined Meth ods ........................................... 68 IV. RESULTS ....................................... ................................................... ................................... 71 Descriptive Results ............................... ................................................... ............................. 71 Dependent Variables .............................. ................................................... ..................... 73

PAGE 9

ix Rating of pain .................................... ................................................... ............ 73 Rating of disability............................... ................................................... .......... 73 Independent Variables ............................. ................................................... ................... 76 Aim 1: Entry to Treatment ......................... ................................................... ......................... 78 Aim 1: Hypothesis Testing A & B .................. ................................................... ............ 78 Bivariate Regression Analysis: Disability and Pain at Entry to Treatment ................... 79 Aim 1: Hypothesis Testing C & D .................. ................................................... ............ 80 Multiple Regression Analysis: Disability and Pain a t Entry to Treatment .................... 81 Aim 2: Post-Treatment ............................. ................................................... ........................... 83 Aim 2: Hypothesis Testing A & B .................. ................................................... ............ 83 Bivariate Regression Analysis: Disability and Pain Post-Treatment ............................. 84 Aim 2: Hypothesis Testing C & D .................. ................................................... ............ 85 Multiple Regression Analysis: Disability and Pain Post-Treatment .............................. 86 Aim 3: Qualitative and Mixed Methods Results ...... ................................................... ........... 88 Stressors: Pain and Disability ................... ................................................... .................. 88 Entry to Treatment: Greater Severity of Pain and Di sability ......................................... 89 Physical limitations ............................ ................................................... ............ 90 Lacking confidence in tasks and activities ...... .................................................. 91 Lacking control ................................. ................................................... ............. 93 Reduction in social activities .................. ................................................... ....... 94 Feelings of vulnerability ....................... ................................................... ......... 95 Poor mental state ............................... ................................................... ............. 96 Frustration about needing support ............... ................................................... .. 97 Financial limitations ........................... ................................................... ........... 98 Post-Treatment: Improving Outcomes of Pain and Dis ability ..................................... 101 Greater Confidence .............................. ................................................... ........ 102

PAGE 10

x Knowing limitations ............................. ................................................... ....... 103 Learning how to manage .......................... ................................................... .... 104 Push through attitude ........................... ................................................... ........ 106 Having support .................................. ................................................... ........... 108 Positive outlook ................................ ................................................... ........... 111 Quantitative and Qualitative: Combined Results ... ................................................... .......... 113 V. DISCUSSION ..................................... ................................................... .............................. 116 Key Findings: Quantitative Summary ................ ................................................... ............... 116 Key Findings: Combined Summary .................... ................................................... .............. 119 Entry to Treatment: Profile for Increased Pain and Disability ..................................... 1 20 Perceived physical health ....................... ................................................... ...... 120 Perceived mental health ........................... ................................................... .... 124 Social support .................................... ................................................... .......... 125 Self-efficacy ................................... ................................................... .............. 126 Entry to Treatment: Emergent Theme ............... ................................................... ....... 128 Financial limitations ............................. ................................................... ....... 128 Post-Treatment: Improving Outcomes of Pain and Dis ability ..................................... 128 Perceived physical health ....................... ................................................... ...... 128 Self-efficacy ..................................... ................................................... ............ 131 Social support .................................. ................................................... ............ 132 Post-Treatment: Emergent Themes .................. ................................................... ......... 134 Push through attitude ............................. ................................................... ...... 134 Positive outlook ................................ ................................................... ........... 135 VI. CONCLUSIONS ................................... ................................................... .......................... 136 Limitations ....................................... ................................................... ................................. 137 Significance and Future Directions ................ ................................................... ................... 142

PAGE 11

xi REFERENCES ........................................ ................................................... .............................. 144 APPENDIX .......................................... ................................................... .................................. 168 A. Manual of Standard Operating Procedures (MSOP) .... ......................................... 168 B. IRB Approvals ..................................... ................................................... ............... 183 C. Informed Consent Form ............................. ................................................... ........ 186 D. Study Questionnaires .............................. ................................................... ............ 200 E. Interview Guide ................................... ................................................... ............... 221

PAGE 12

xii LIST OF TABLES TABLE 3.1 Inclusion and Exclusion Criteria for the Nonsurgica l RCT with Patients with LSSÂ…. 54 3.2 Data Collection Summary............................ ..............Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â… 64 4.1 Demographic Characteristics of the Sample of Patien ts with LSSÂ…Â…Â…Â…Â…Â…Â…. 72 4.2 Pain Rating in Patients with Lumbar Spinal Stenosis Pre and Post-TreatmentÂ…Â…Â… 73 4.3 Disability Rating in Patients with Lumbar Spinal St enosis Pre and Post-TreatmentÂ…. 74 4.4 Means, Standard Deviations, and Paired T-test for D isability and PainÂ…Â…Â…Â…Â…Â… 75 4.5 Correlation Analysis Results for DisabilityÂ…Â…Â…Â…Â…Â…Â…Â… Â…Â…Â…Â…Â…Â…Â…Â…Â… 75 4.6 Correlation Analysis Results for PainÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â… Â…Â…Â…Â…Â… 76 4.7 Levels of Self-efficacy and Social Support at Entry and Post-TreatmentÂ…Â…Â…Â….Â…. 77 4.8 Means, Standard Deviations and Paired T-test of Pre dictor Variables of Self-efficacy, Social Support, Physical Health and Mental HealthÂ…Â… Â…Â…Â…Â…Â…Â…Â…Â…Â…...Â…Â… 77 4.9 Means, Standard Deviations and Paired T-test for Pr edictor Variables of Subscales of SelfEfficacy and Subscales of Social SupportÂ…Â…Â…Â…Â… Â…Â…Â…Â…Â…Â…Â…Â….Â…Â… 78 4.10 Bivariate Linear Regression Results for G reater Disability at Entry to TreatmentÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…..Â…Â…Â….. 80 4.11 Bivariate Linear Regression Results for Highe r Pain at Entry to TreatmentÂ…Â…Â…Â… 80 4.12 Model 1: All Variables Contribute to Grea ter Disability at Entry to TreatmentÂ…......................................... ................................................... ........................ 81 4.13 Model 2: All Variables Contribute to High er Pain at Entry to Treatment Â…Â…Â…Â…Â…. 82 4.14 Model 3: Two Variables Optimally Contribu te to Higher Pain at Entry to TreatmentÂ…Â…Â…Â…Â…..................................... ................................................... .............82 4.15 Model 4: All Variables Contribute to High er Pain at Entry to TreatmentÂ…Â…Â…Â…Â…...83 4.16 Bivariate Linear Regression Results for P redicting Reduced Disability PostTreatmentÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…..Â…Â….. Â….85 4.17 Bivariate Linear Regression Results for P redicting Reduced Pain PostTreatment.Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…..Â…..... Â…..85

PAGE 13

xiii 4.18 Model 1: All Variables Contribute to Pred iction for Reduced Disability PostTreatmentÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…....Â… 86 4.19 Model 2: Two Variables Have Clinical Rele vance in Predicting Reduced Disability PostTreatmentÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…...Â…. 87 4.20 Model 3: Two Variables Have Clinical Rele vance in Contributing to Prediction for Reduced Pain Post-TreatmentÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…..Â…Â…Â…Â…Â…Â…Â…Â…Â…..Â… 87

PAGE 14

xiv LIST OF FIGURES FIGURE 2.1 Stress and Coping Model and stressful health condit ions......Â…Â…Â…Â…..Â…Â…..Â…Â… 17 2.2 Biopsychosocial model of painÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…...Â…Â…Â…Â… Â… 23 2.3 The Stress-Diathesis Model of pain and disabilityÂ…Â… Â…Â…Â…Â…Â…Â…...Â…Â…Â…Â…. 24 2.4 The modified Disablement ModelÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…...Â…Â…. Â…. 27 2.5 International Classification of Functioning, Disabi lity, and Health framework....... 32 2.6 Social support interactions with Stress and Coping ModelÂ…Â…Â…Â…..Â…Â…Â…Â…Â…. 41 3.1 Design of the randomized clinical trial for non-sur gical treatment of LSSÂ…Â….Â…... 49 3.2 Longitudinal, cohort study design using mixed metho dsÂ…Â…Â…Â…Â…Â…Â…Â…Â…..Â….. 51 3.3 Subject enrollment, consent, treatment retention, a nd follow-upÂ…Â…..Â…Â…Â…Â…..Â… 57 3.4 Summary of SF-36 subscales and domainsÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â… Â….Â….... 62 4.1 Oswestry Disability Index score at entry to treatme nt in patients with LSSÂ….Â…Â…..74 4.2 Physical, mental, social and environmental limitati ons contributing to stressor of pain and disabilityÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â….Â…Â…. 90 4.3 Factors related to financial limitationsÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â… Â…Â…Â…Â…Â…Â….. 100 4.4 Factors related to improved outcomesÂ…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â…Â… .Â…Â…. 102 4.5 Factors increasing severity of pain and disability at entry to treatment Â…Â…Â…Â….Â… 114 4.6 Factors mediating the reduction of pain and disabil ity post-treatment Â…Â…Â…Â…Â…... 115

PAGE 15

xv LIST OF ABBREVIATIONS ADLs Activities of Daily Living BDI Beck Depression Index COMIRB Colorado Multiple Institutional Review Board ESI Epidural steroid injection EXSE Exercise Self-Efficacy subscale of the LoBACS FSE Functional Self-Efficacy subscale of the LoBAC S ICF International Classification of Functioning IRB Institutional Review Board LBP Low back pain LoBACS Lower Back Activity Confidence Scale LSS Lumbar spinal stenosis MCID Minimal Clinically Important Difference MCS Mental Component Summary of the SF-36 MOS SSS Medical Outcomes Survey Social Support Sca le MRI Magnetic resonance imaging MrOS Osteoporotic Fractures in Men Study NPRS Numeric Pain Rating Scale NPRSbb Numeric Pain Rating Scale back and buttock s ubscale NPRStl Numeric Pain Rating Scale thigh and leg sub scale OA Osteoarthritis ODI Oswestry Disability Index PCS Physical Component Summary of the SF-36 PT Physical therapy RA Rheumatoid arthritis RCT Randomized clinical trial SCT Social Cognitive Theory SF-36 Short Form 36 from the Medical Outcomes Sur vey VAS Visual Analog Scale WHO World Health Organization

PAGE 16

1CHAPTER I BACKGROUND AND SIGNIFICANCE Low Back Pain Low back pain (LBP) occurs at epidemic proportions in the United States, affecting 6090% of individuals during their lifetimes.1 Total healthcare expenditures in the U.S. incurre d by patients with LBP are 90 billion dollars annually, 60% higher than in individuals without LBP.2 Next to the common cold, the complaint of LBP is th e most common reason individuals visit a physicianÂ’s office.3, 4 Approximately 40% of individuals have experienced LBP for a day or more within the last 12 months, with 40% of them having consulted their primary care provider. Because of the substantial impact of LBP on healthc are, there is a need to identify effective management strategies to reduce the pain and disabi lity associated with diagnoses of LBP. Lumbar Spinal Stenosis A US national survey of physician visits identified back pain as the most commonly reported musculoskeletal symptom and third most fre quently reported symptom in patients over 75 years of age.5 Lumbar spinal stenosis (LSS) is a subgroup of LBP that clinicians can identify through age and symptom related factors. LSS is a p revalent and disabling musculoskeletal condition in the aging population that often result s in substantial stress and physical burden for individuals with the disorder, and is associated wi th significant healthcare costs.6-8 Physicians diagnose LSS in an estimated 13-14% of patients who seek help from a specialty physician, and 3-4% who seek care from a general practitioner for LBP.8, 9 Elderly patients with LSS as a source of their back and leg pain suffer from significant functional disability,10, 11 and many eventually seek surgical intervention. The Osteoporotic Fract ures in Men Study (MrOS) examined a multicenter community cohort of 5995 men, 66 years and o lder, and found 65% of the men had at least 1 episode of LBP in the past year and 26% had clini cally relevant LBP which occurred more than

PAGE 17

2 3 times in the prior year at a pain level of modera te to severely bothersome.12 In addition, the authors found that approximately 10% of the group w ith clinically relevant LBP had symptoms that were similar to those with LSS.12 Definition, pathoanatomy, and clinical presentation of LSS Degenerative lumbar spinal stenosis (LSS) is defined as a focal narrowi ng of the spinal canal, although there are varying opinions about the precise amount of narrow ing that must occur before the canal is considered stenotic. LSS is classified as an arthri tic condition. Postaccini and other scholars apply the general term “spinal stenosis” to 3 root compre ssion mechanisms alone or in combination: (1) disk protrusion or herniation, (2) osteotic overgro wth into the spinal canal or the foramina through which the roots pass laterally, and (3) ver tebral slippage or spondylolisthesis.13-16 Because degenerative changes are the predominant et iology of LSS, affected individuals tend to be older, typically at least 50 years of ag e, with a prolonged history of low back pain.17-19 Leg pain, either unilateral or bilateral, is also r eported in about 90% of patients seeking medical attention.19 Chronic compression of the spinal nerve roots can also lead to sensory, reflex and/or strength changes in the lower extremities. Fifty pe rcent of patients report these neurological changes.20 Acute cauda equina syndrome is rare, but clinician s have reported it.21 The hallmark finding of patients with LSS is the po stural-dependency of the symptoms. Patients have reported that symptoms typically wors en with standing or walking, and can be relieved or diminished by sitting or bending forwar d.22-24 Neurogenic claudication frequently accompanies LSS, defined by clinicians as poorly lo calized pain, paresthesias, and cramping in one or both lower extremities that is of a neurolog ical origin, brought on by walking and relieved by sitting.25 Walking tolerance can become substantially limited in patients with LSS, with pain and reduction in walking capacity often cited as a reason for seeking medical intervention. 25-27 Incidence and impact of LSS. Although there is literature on the burden of disea se of LBP in general, only 1 study has addressed the soci etal burden of disease related to functional loss and disability specifically for patients with spinal stenosis. A Swedish study by Johnsson,28

PAGE 18

3 that defined spinal stenosis as a canal of 11 mm or less, described the annual incidence of LSS among patients referred to orthopedic departments a s approximately 50 per 100,000 inhabitants. This 1995 population study of 2 regions in Sweden r eported that, with a LSS incidence of about 50 per 100,000, between 42% and 58% of these patien ts had claudication (leg pain or weakness upon standing or walking). From these data, Johnsso n calculated the incidence as about 25 per 100,000 inhabitants for spinal-stenosis-associated claudication. More severe stenosis can result in cauda equina syndrome, which is characterized by th e loss of sexual function and urinary and/or fecal incontinence. This same study reported that c auda equina syndrome had an incidence of less than 1 per 100,000. If untreated, clinicians common ly believe severe LSS has the potential to lead to symptoms that become permanent and unresponsive to medical or surgical treatment.11, 13, 29-34 Review articles23, 35, 36 and textbooks32, 37 provide some evidence that patients with symptomatic spinal stenosis typically have chronic LBP and pain and weakness in the legs that limits functional tasks such as standing and walkin g to brief durations and short distances. These physical impairments place stressors on the individ ual, particularly limitations on the ability to carry out self-supporting, daily activities as well as work, social, and recreational activities, that would be defined as sources leading to disability i n the person with LSS. Patients with LSS often report decreases in physica l health.38-40 Decreased physical health either real or perceived can impact an individual i n many ways. Overall, it is known that those who have decreased physical health reduce or elimin ate participation in daily living, social, and recreational activities. A lack of activity has bee n shown to lead to other stressors such as obesity and general physical deterioration that may eventua lly result in further disability with the onset of cardiovascular and other serious health problems.41 Activity restrictions may also lead to low selfconfidence, fear or avoidance behaviors, depression and other psychological problems that further restrict the person with LSS from participation in activities of daily living (ADLs), recreational activities, social activities, and community functi ons.32, 42, 43

PAGE 19

4 Limitations in the current literature for LSS. While the substantial societal impact of LSS is apparent, there continue to be controversies in the literature regarding the effectiveness of both surgical and non-surgical management of LSS.5, 11, 19, 44-47 Suboptimal research designs along with other factors such as incomplete reporting of patient characteristics and unmeasured psychosocial factors that may influence coping beha viors in managing the stressors of pain and disability account for some of the gaps in current research on LSS. The interaction between the stressors of pain and limited physical function, li mited participation in activities, and altered patient behaviors in relationship to intrapersonal and interpersonal coping factors is complex and has been poorly studied in patients with LSS. This research study fills an important gap in the literature by using a novel approach to evaluate th e relative contribution of self-efficacy, social support, and self-reported physical and mental heal th when managing the stressors of pain and disability in LSS. Rationale The Institute of Medicine has recognized LBP as a t op 15 priority condition, calling for health care organizations to develop new evidence-b ased care process models.48 Current practice guidelines for LBP in primary care recommend a step ped care approach with an initial treatment of education and advice that focuses on remaining a ctive throughout care. For a subgroup of patients with LBP diagnosed with LSS, evidence show s success for conservative treatment that includes informational education, advice for physic al activity, epidural steroid injections and physical therapy.49-51 Identifying the multiple factors resulting in heig htened pain and disability is the first goal in clinical management. Identifying factors for successful clinical outcomes is the next goal. Success in treatment is measured by redu ction in pain and disability for this subgroup of patients with LSS. Several studies have indicate d that psychosocial factors such as depression and fear avoidance behaviors are negative influence s on the stressors of pain and disability and result in declines in treatment success in LBP in g eneral.52-54 Studies have shown that depression

PAGE 20

5 and fear avoidance behaviors correlate with reduced participation in functional daily tasks indicating higher risk for disability. Studies have also shown that depression and fear avoidance scores can improve during LBP management and result in positive changes in patient outcomes of pain and disability.55, 56,51 Overall, research has demonstrated that using an e nhanced or multimodal team approach to treatment results in im proved patient outcomes for LBP.51, 57 Although the general LBP literature can provide som e insight into management of LSS, identifying factors contributing to higher pre-trea tment pain and disability as well as determining what factors help improve treatment outcomes for pa tients with LSS continues to be incomplete. Due to a lack of reporting and a limited understand ing of what influences the stressors of pain and disability, there are gaps in LSS literatu re about the influence of psychosocial factors on the severity and outcomes in LSS. Rarely has LSS re search examined the intrapersonal and interpersonal influences in this chronic condition. No other study to date has examined the relative contribution of self-efficacy and social s upport in predicting the severity and outcomes of pain and disability in patients with LSS. As the po pulation continues to age and the provision of medical care for patients with LSS requires conside rable healthcare resources, it is important to identify and understand factors that impact pain an d disability and thereby improve successful treatment of patients suffering from this spinal di sorder. Research Question and Specific Aims The current study addressed the following research question: What is the relative contribution of self-efficacy, social support, perc eived physical health, and perceived mental health to pain and disability in individuals with L SS? Three Specific Aims guided the investigation of this overarching research question

PAGE 21

6 Aim 1: To determine the relative contribution of se lf-efficacy, social support, self-rated physical health, and self-rated mental health to th e severity of pain and disability among patients with LSS upon entry to treatment of LSS. H1A: Lower levels of self-efficacy, social support, perc eived physical health and perceived mental health will each independently pre dict higher disability at entry to treatment. H1B : Lower levels of self-efficacy, social support, pe rceived physical health and perceived mental health will each independently pre dict higher pain at entry to treatment. H1C: Lower levels of self-efficacy, social support, per ceived physical health and perceived mental health will together predict highe r disability at entry to treatment. H1D: Lower levels of self-efficacy, social support, per ceived physical health and perceived mental health will together predict highe r pain at entry to treatment. Aim 2: To determine the relative contribution of s elf-efficacy, social support, self-rated physical health, and self-rated mental health to th e outcomes of pain and disability among patients with LSS in the year following treatment o f LSS. H2A: Higher levels of self-efficacy, social support, per ceived physical health and perceived mental health will each independently pre dict lower disability after treatment of LSS. H2B: Higher levels of self-efficacy, social support, pe rceived physical health and perceived mental health will each independently pre dict lower pain after treatment of LSS. H2C: Higher levels of self-efficacy, social support, pe rceived physical health and perceived mental health will together predict lower disability after treatment of LSS. H2D : Higher levels of self-efficacy, social support, p erceived physical health and perceived mental health will together predict lower pain after treatment of LSS.

PAGE 22

7 Aim 3: Explore how self-efficacy, social support, p erceived physical health and perceived mental health contribute to a patientÂ’s lived exper ience with LSS and influence meanings of pain and disability. Research Goals The goal of this dissertation research project was to determine the relative contribution of self-efficacy, social support, and self-rated physi cal and mental health to pain and disability in patients with LSS upon entry to treatment and long term post-treatment. This goal requires a greater understanding of the role of self-efficacy, social support, and self-rated health beyond what currently exists in the literature for patient s with LSS. This project also sought to uncover additional unknown factors that contribute to the s tressors of pain and disability and identify any additional coping resources that might be available to patients with LSS. These research findings will improve the understanding of how stressors, pa rticularly pain and disability, are appraised in a population of individuals with LSS. In addition, this research will help identify the role of selfefficacy, social support, self-rated physical healt h and self-rated mental health in coping with LSS and determine their contribution in predicting outc omes of pain and disability. Results from this study will help direct clinical management and futu re research in this patient population.

PAGE 23

8CHAPTER II REVIEW OF THE LITERATURE Pain and Disablement in LSS Research illustrates that the LSS condition results in significant stressors for individuals with LSS in the form of pain and disability.58, 59 Research also confirms that after treatment, many with LSS continue to have multiple stressors relate d to pain and disability with basic activities such as standing and walking.60, 61 Similar to those living with chronic conditions suc h as cancer, osteoarthritis, and diabetes, it is likely that per sons with LSS experience challenges with psychological, social, medical, and financial limit ations in addition to basic physical functioning limitations.59, 62-65 It is clear that somatic symptoms related to body s tructure or function has the most recognition in the current literature for LSS. Low back pain, leg pain and spinal canal narrowing are the top 3 reported factors contributing to disa bility in the LSS population.6, 39, 58, 60, 66 Severity of LSS is almost exclusively determined by pain and spinal canal narrowing on imaging and are the most recognized factors for surgical decision m aking.67, 68 However, imaging of spinal canal narrowing has not been found to be a reliable way t o identify severity in LSS.61, 69, 70 Also, problems with other somatic factors in addition to pain have been reported including neurogenic claudication, neurological changes, decreased lumba r spine motion particularly in extension, and decreased range of motion of most of the joints and soft tissues in the lower extremity and trunk.40, 51 Alterations in activity due to pain and weakness ar e frequently reported factors contributing to disablement in LSS. The top 3 repor ted factors related to changes in activities or participation includes limitations in walking and s tanding but increased activity of sitting.6, 59, 60, 66 These 3 activity/participation factors have become part of the clinical diagnosis for LSS.39 Other

PAGE 24

9 functional problems seem to be present such as stai r climbing or general activities of daily living (ADLs) but these have been widely under reported. Identification and measurement of pain and disabili ty in patients with LSS has been almost exclusively documented by physical examinati on or patient self-report. Condition-specific instruments have been used often in the LSS literat ure to indicate the severity of disability by quantification with measures like the Oswestry Disa bility Index (ODI) and the Roland-Morris Disability Questionnaire.71, 72 In addition, more generic health status instrument s have been frequently used to report general health in LSS suc h as the Medical Outcomes Study Short Form 36 (SF-36) and the Nottingham Health Profile.73, 74 Pain specific instruments such as the Numeric Pain Rating Scale (NPRS), the Visual Analog Scale ( VAS) and other dimension specific instruments such as the Beck Depression Index (BDI) have also been used to identify specific issues in patients with LSS.75-78 All of these condition-specific, general, and dime nsion specific instruments have collectively been used to address the stress related experience of pain and disability in patients with LSS although with littl e standardization in the use and comparison among studies. Previous Literature: Psychosocial Factors Previous research on patients with arthritis and ot her health conditions has identified that persistent psychosocial needs can decrease the effe ctiveness of medical treatment, general health status and quality of life while increasing health care costs.64, 79, 80 Although unidentified in the LSS literature, unaddressed personal and social nee ds may contribute to reduced compliance with treatment and follow-up recommendations, diminished self-care, and reduced overall health management.81 Similar to other health conditions, the daily live s of patients with LSS likely involve unanticipated challenges. Patients with LSS must learn to navigate life with the stress of chronic LBP, cope with continued side effects of re duced functioning, experience continued

PAGE 25

10 limitations to their physical and social abilities, manage through decreased activities and abilities, and face fears about recurrence or worsening of the ir condition. The American College of Physicians and the American Pain Society published clinical guidelines that recommend clinicians conduct a focu sed history and physical exam with patients with LBP. This focused history is recommended to in clude assessment of psychosocial factors which predict risk for chronic disabling back pain.82 Psychosocial factors and emotional distress have been found to be stronger predictors of LBP ou tcomes when compared to physical exam findings or severity of pain alone.83 Psychosocial factors that predict poorer LBP outco mes include presence of depression, passive coping stra tegies, higher disability levels, or somatization.83, 84 Psychosocial factors often delay recovery and iden tification can help to target interventions. It is recommended that patients with chronic LBP who do not improve with selfcare options would benefit from non-pharmacological therapy including interdisciplinary rehabilitation, exercise therapy, spinal manipulati on, and cognitive-behavioral therapy.82 Only a few studies have examined the presence of ps ychosocial factors and the relationship to severity of pain and disability in patients with LSS. Moreover, only a few have examined the association between psychosocial facto rs and outcomes of treatment for patients with LSS. Depression, low sense of coherence, life dissatisfaction, and fear or avoidance behaviors have all been described to be present in baseline examination of some individuals with LSS77, 85, 86 and have been linked to worse treatment outcomes i n LSS.65, 78, 87, 88 Specifically, Wood et al86 identified that patients with LSS have higher fear of reinjury and activity avoidance. Individuals with higher depression scores also had greater disability at baseline76 and had worse surgical outcomes60, 77 in the treatment of LSS. In one study, Turner et a l76 found that lower depression after surgical treatment was related to reduced disability scores but not pain scores. However, Katz et al89 found that depression scores pre-treatment were as sociated with higher pain scores 6 months post-surgical intervention. In anot her study by Katz et al,90 higher depression

PAGE 26

11 scores were also associated with poorer patient sat isfaction post-surgery in bivariate but not multivariate analysis. Several studies examined the relationship between l ife dissatisfaction and other clinical baseline measures and post-surgical outcomes in LSS .76, 78, 85, 88 Sinikallio et al78 found that 25% of the pre-operative patients with LSS reported lif e dissatisfaction. In the general population, life dissatisfaction has been recorded in 13% of the hea lthy population and in 25% of those with illnesses.91 Therefore, patients with LSS have more life dissat isfaction than the general population but are similar to patients with other chronic illn esses. Sinikallio et al85 also found that dissatisfied patients with LSS had lower coping res ources, elevated alexithymia, and elevated depression scores. Two follow up studies by Sinikal lio et al77, 88 found that both pre-operative and post-operative life dissatisfaction was independent ly correlated with disability and sense of coherence, a measure of coping. Pain, however, was not associated with life dissatisfaction either pre-operatively or post-operatively.77, 78, 88 Patients with LSS remain a vulnerable population wh o often find themselves in need of general information, guidance, and social support.92 Only 1 study by Katz et al90 examined social support in patients with LSS using a single questio n: 1-5 Likert scale related to the amount of contact with friends and family, ranging from daily contact to no contact in the past month. The results of this study did not find significant corr elations between pre-operative contact amounts with friends and family and post-operative outcomes of pain from surgery for LSS.90 Previous Literature: General Health Status General health status and health related quality of life has been minimally investigated in the population of patients with LSS. General or sel f-rated health measures have often been related to quality of life. Battie et al93 examined health related quality of life and comorb idities associated with LSS and noted that individuals with LSS have h ealth related quality of life issues 4 times that of the general population. Gunzburg et al94 identified that patients with LSS have high

PAGE 27

12 baseline rates of non-organic symptoms reflecting t he potential significance of illness behaviors in this population. In addition, these authors foun d that better pre-operative self-rated health (scale: excellent, good, fair, poor) was associated with better post-surgical outcomes of walking capacity, symptom severity, and treatment satisfact ion. In the general LBP literature, Fanuele et al10 investigated the Medical Outcomes Survey SF-36 physical component scores (PCS) in a prospect ive sample of 17,774 patients with spinal disorders. They found that the patients with LBP ha d a mean PCS score of 30.4 9.95 (SD) compared with 50.0 10.00 for the general United S tates population. This indicates that the PCS score is greatly reduced in patients with spinal di sorders in general and is similar to other patient population PCS scores such as chronic heart failure (31.0), Chronic Obstructive Pulmonary Disorder (33.9), Systemic lupus erythematosus (37.1), cancer (38.4), primary total hip arthroplasty (29.0), primary total knee arthroplast y (32.6), and glenohumeral degenerative joint disease (35.2).10 Individuals with LSS often self-report worse physic al health when compared to the population at large. In the MrOS study, Vogt et al12 found that men with LBP symptoms similar to LSS had SF-12 self-reported scores for the physi cal component summary (PCS) of 38.3 out of 100, ranking them at the 14th percentile for physical health compared to the US population with a mean ranking of 50th percentile.12, 73 In addition, patients with LSS enrolled in the Nat ional Spine Network registry had PCS scores of 32.92 out of 100 ranking them in the 8th percentile for physical health compared to the US population with an mean ranking of 50th percentile.10 Patients in an RCT investigating both surgical and non-surgi cal intervention found patients with LSS to have mean SF-36 PCS values between 34.3 and 35.4 22 (SD) compared with 50.0 10.00 for the general United States population indicating similarly low PCS scores comparable to the other studies sited.38 Although there appears to be an association betwee n LSS and self-rating of physical health, this relationship and its impact o n pain and disability is not clearly understood.

PAGE 28

13 Pahl et al95 examined the impact of health status in 4,442 pati ents with LBP including a subpopulation of patients with LSS. Although not sp ecific to LSS, they found that self-ratings of physical health measured by the PCS, ranked patient s with LBP lower than normative values for matched age groups. Pahl et al also found that indi viduals over 40 years old with LBP also had poorer mental health status and declines in vitalit y or energy and other social functioning domains compared to their matched age group norms. Overall, this evidence illustrates that both perceived physical and mental health are likely contributing factors to the outcomes of pain and disability in LSS and warrant further investigation. Previous Literature: Research Design in LSS Literature in LSS almost exclusively uses quantitat ive methods for determining health care outcomes, providing minimal depth of understan ding about sources for and appraisal of stressors as well as coping resources available to those with LSS. Standardized measurements cannot capture the complete experience of pain and disability with LSS. Qualitative methods can provide a more detailed account of a patientÂ’s expe rience with a stressful health condition like LSS and provide insight into the coping resources a vailable to those individuals. These insights can help direct health care resources for treatment of individuals with LSS. Although several authors in the current literature indicate they used qualitative methods, these authors all describe qualitative assessment b y a clinician for MRI imaging, disc protrusion measurement, and nerve root involvement. Two author s described qualitative assessment as selfreport of the number of analgesic medications used during the study,96 and self-report of limitation in walking distance (yes/no) along with the self-report of limitation in ADLs performed (yes/no).97 Two additional studies mentioned the use of patien t interviews for measurement of post-operative outcomes. One of these studies, Shar ma et al97 used 2 questions to inquire if the patient was satisfied with the outcomes of surgery (yes/no), and any significant limitation in ADLs (yes/no) although the methodology is unclear. The other study, Slosar et al98 interviewed

PAGE 29

14 patients using questions from a standardized outcom es measurement tool but added no additional information from these interviews beyond the survey questions. Orlin et al99 used a novel approach in preoperative interviews asking 209 pati ents with LBP (50 with LSS) 8 questions about bladder, anal and sexual function and dermato me sensitivity in the only study published on urinary voiding dysfunction with patients undergoin g lumber spine surgery from both lumbar disc lesions and spinal stenosis. Results from this stud y illustrate that 68% of these patients undergoing lumbar spine surgery have baseline urina ry voiding dysfunction. This study clearly illustrates that patients with LSS have other conce rns other than just standing and walking difficulty. Only 1 additional study by Sekiguchi et al100 actually used a more comprehensive qualitative methodology with patients with LSS in d esigning a new questionnaire to rate symptoms of LSS based on quality of life constructs In this study, a team of 2 orthopedic specialists and 1 psychometric specialist devised s everal constructs of the new scale and operationalized the constructs as they related to t he overall measurement of quality of life in patient with LSS. Next, they asked 10 patients wit h LSS (7 men, 3 women) divided into 2 focus groups to discuss their pain, disabilities and diff iculties with ADLs. Finally, 1 group was asked to confirm their understanding of the meaning of the i tems of the scale and if it appropriately described what was bothering the patient. The resul ts confirmed specific quality of life issues for patients with LSS that include:1) limitations on go ing out, 2) sleeping problems, 3) burden imposed by hospital visits and treatment, 4) urinat ion problems, 5) walking problems, 6) limitations in everyday activities, and 7) emotiona l burdens. This study adds unique clinical implications that behavioral concerns are important to patients with LSS. Unfortunately this symptom scale has not been tested nor is it used in clinic or in LSS research.

PAGE 30

15 Previous Literature: Summary While the substantial societal impact of LSS is app arent, there are controversies in the literature regarding the identification of subgroup s according to patient characteristics, effectiveness of treatment, and general management in LSS.5, 11, 19, 44-47 Suboptimal research designs also account for some of the gaps in curren t research on LSS, along with other factors such as incomplete reporting of many patient charac teristics, poorly measured psychosocial factors that may influence treatment as well as out comes of treatment, and limited research design leading to a lack of understanding of the larger is sues of pain and disablement in this population. Rarely has research in this area evaluated the cont ribution of intrapersonal and interpersonal factors such as self-efficacy and social support or explored the meanings of self-rated health in understanding the experience of pain and disability in the management of LSS. Theoretical Perspectives Integrating Theory in LSS Like people with other medical conditions, individu als with chronic and recurring LSS may need to adjust their habits and lifestyles whil e trying to maintain basic physical, social, vocational, and recreational activities. To manage the condition of LSS, patients must try to understand the nature of their problem, create self -care strategies for dealing with pain flare-ups, manage and attempt to overcome functional problems, and identify and utilize available support and resources wisely. This process underscores the behavioral component of coping with LSS. To date the general LSS literature lacks a theoretical foundation for this behavioral component, particularly in comparison to research on other are as of chronic disease.101-106 Therefore, this study was developed from a theoretical perspective to better understand the traditionally biomedical characterized condition of LSS. This stu dy was guided by the Stress and Coping

PAGE 31

16 Model107 to identify predictors associated with 2 main sour ces for stress in individuals with LSS: pain and disability. In addition, this study was de signed using mixed methodology with the aim to uncover the appraisals and coping mechanisms identi fied and used by individuals with this disabling condition. Theoretical Framework The present study used the underlying theoretical c oncepts in the Stress and Coping Model107 to define both stress and coping as well as identi fy the relationship of these concepts in the condition of LSS. Two primary stressors have be en identified in previous LSS literature: pain and disability. Differing experiences with LSS resu lt in different appraisals of the stressors of pain and disability. Individuals with LSS also have different outcomes related to the experience of pain and disability. Two guiding theories will b e used to understand the experience of pain and disability in patients with LSS including the Stres s-Diathesis Model108, 109 and the ICF Framework of Health and Disablement.108, 110 Finally, the Stress and Coping Model107 will be used to identify what coping behaviors and resources preserve well-b eing in patients with LSS while facing stressful experiences with this chronic condition. Specifically, the study will explore the construct self-efficacy proposed as a personal resource in LS S. The concept of self-efficacy originates from BanduraÂ’s work in Social Cognitive Theory.111, 112 A second construct, social support, will be examined as a social resource for health as identif ied by Lazarus and Folkman in the Stress and Coping Model.107 and expanded in health application by Schwarzer an d Leppin.113 Finally, the constructs of self-rated physical and mental health will also be explored from the perspective of the Stress and Coping Model.107 Stress and Coping Model Within the Stress and Coping Model,107 stress involves the relationship between an individual and his or her environment. This relatio nship or transaction between individuals and

PAGE 32

17 their environment indicates that stress is more tha n an external stimulation or specific pattern of physiological, behavioral, or subjective reactions (see Figure 2.1). This relationship also views 2 processes as mediators within the person-environmen t transaction: cognitive appraisal and coping effort. Cognitive appraisals and coping efforts are influenced by moderators such personal and situation factors that result in the outcome of ada ptation of the individual on many levels impacting their health. Figure 2.1. Stress and Coping Model and stressful h ealth conditions. Adapted from Glanz114 n rr nrnrnn nrrnnr rrr nnr rn r n rr r nrnn rn nrn rn n nrr nn r !nr nrr rrnrrn nrrn rnrn rnn "r #rn rrnnrn

PAGE 33

18 Cognitive appraisal The cognitive appraisal or evaluation by the ind ividual is paramount to determining if the stress is threateni ng to his/her well-being or surpassing his/her resources. Upon appraising a stressor, the theory asserts that people engage in coping, defined as fluctuating behavioral exertions, in an effort to m anage that stressor. The cognitive appraisal process helps an individual determine both the cont rollability and availability of coping resources identified as necessary to manage the stressor(s).107 Cognitive appraisal is a necessary component of dea ling with a stressor. It accounts for the different ways in which individuals react to si milar events. Health conditions such as those resulting in pain provocation and disability like L SS can cause stress in nearly every person, yet people vary in their reactions and interpretations of the same event and condition.107 A health condition is a general stressor but the different w ays individuals respond to the same health condition can depend on their cognitive appraisals. This variability in cognitive appraisals can change a personÂ’s level of vulnerability during a s tressful health condition. Vulnerability is closely related to appraisal since vulnerability in creases as a person appraises that he/she has reduced coping resources available. Vulnerability r eflects the inability of individuals to withstand adverse impacts from multiple stressors to which th ey are exposed. It can be associated with a pattern of thought that is believed to predispose t he individual to psychological problems and feelings of hopelessness. In addition, the variabi lity of individual appraisals as well as his/her responses helps to explain why some individuals exp erience similar health conditions but have differing quality, intensity, and duration of physi cal, social, and emotional outcomes. Personal and situational factors Two types of factors influence the cognitive app raisal process: personal factors and situational factors.107 Personal factors consist of the personal values that motivate individuals to make certain decisions and beliefs that give the individual a personal sense of control. Therefore, at the individual-leve l, commitments and beliefs are all part of an individualÂ’s process of appraisal. Commitments are expressions of what is important to a person and can be related to vulnerability. The deeper a p ersonÂ’s commitment, the greater potential for

PAGE 34

19 threat, but also the greater the push toward amelio rative action and hope. Beliefs are also important in determining how a person evaluates a s tressful event or health condition. Beliefs of personal control over situations can relate to how an individual believes internal self-responses to situations can be controlled. General control belie fs relate to the extent that the person believes the health related outcomes can be controlled. Another influence to the appraisal process is situa tional factors which play a critical role in determining the external controllability of the stressor and what ameliorative action can be taken.107 Situational factors can include predictability and uncertainty, temporal and life course factors, and ambiguity. Because patients with LSS a re often in their later stages of life, aging related concerns are a component of situational fac tors in LSS. Situational factors can also influence how and to what extent the stressor can b e managed. Maximum uncertainty is often extremely stressful for an individual experiencing a health condition like LSS. The uncertainty can have immobilizing effects on anticipatory and a ctual coping processes and can cause mental confusion in the individual. Overall, it is importa nt to evaluate both situational factors and personal factors in order to understand what ultima tely influences the level of stress related to pain and disability in patients with LSS. Different appraisals While differences between individuals under similar circumstances are inevitable due to various persona l and situational factors, the Stress and Coping Model emphasizes that all individuals evaluating a stressor undergo a cognitive appraisal process involving primary appraisal, secondary appraisal an d/or reappraisal.107 Primary appraisals involve assessment of the magnitude and significance of a s tressor or traumatic event. During primary appraisal, the individual will assess the actual ha rm, loss, threat, or challenge that must be encountered with a stressful health condition. When an individual first experiences a health condition or a reoccurrence of a chronic condition like with LSS, primary appraisal takes place. Individuals with LSS are often in the process of pr imary appraisal when they seek out medical care and treatment.

PAGE 35

20 Secondary appraisal refers to an individualÂ’s asses sment of the degree to which the stressor or traumatic event can be controlled and t he available coping resources.107 Secondary appraisals are a judgment about what might and can be done in the situation.107 Secondary appraisals include an evaluation about whether a co ping option will accomplish what it is supposed to do as well as the consequences in the c ontext of other internal and/or external demands and constraints. Appraisals of controllabil ity of a health condition can be stress-reducing if the outcome is believed to be controllable or th at one has the coping resources to manage the outcomes. However, appraisals of controllability ca n also heighten threat and give rise to negative emotions and beliefs about coping if control and re sources are diminished. For example, individuals with LSS will use secondary appraisals when their chronic LBP has reoccurred and their activities have been limited. This in turn mo tivates some to seek treatment to control their pain and disability as part of the coping process i f they have the necessary access and resources. Reappraisal is the final feature of the appraisal p rocess which entails an altered or revised version of a previous appraisal. Reappraisals can o ccur multiple times for reasons such as changes in the environment or the health condition.107 Reappraisals can also occur when an individual has gone through the cognitive coping pr ocess and has altered the assessment of the available coping resources. Because the complete pr ocess of appraisal is dynamic, a patient with a chronic disease such as LSS is likely to appraise a nd reappraise the stressors of pain and disability and respective coping resources before, during, and after treatment multiple times and with each successive chronic episode. Coping Coping is intimately related to the concept of c ognitive appraisal and the person-environment relationship. Coping involves th e cognitive and behavioral efforts to address external and internal demands on the person experie ncing a stressful encounter. Coping can either be focused on changing the person-environment probl em behind the stress, or be directed toward changing the appraisal of the situation. Moreover, coping can be focused toward trying to reduce a negative emotional state of the situation.

PAGE 36

21 There are 2 types of coping research in the literat ure: trait oriented and state oriented. Trait oriented coping research aims to identify ind ividuals whose coping behaviors are diminished due to the demands of the stressful enco unter. Early identification of individuals who have reduced coping behaviors could help target cer tain factors and resources needed to improve coping behaviors. For example in patients with LSS, early identification of individuals who have negative appraisals and reduced coping can help dir ect clinicians to increase cognitive-behavioral efforts to improve patient beliefs, motivation, and participation in treatment. State oriented coping on the other hand, centers on understanding which c oping strategies an individual actually chooses and uses during stressful events. State ori ented coping explores the relationship between coping strategies and the outcome variables. Variab les of interest include coping efficiency, emotional reactions, and variables of adaptive outc ome (e.g., health status) that accompanies certain coping efforts. In patients with LSS, under standing coping behaviors in relationship to health outcomes can be important to identify and pr ioritize health care resources for individuals with this chronic condition. In general, state ori ented coping research strategies tend to lay the foundation for resource allocation and targeted pro grams to improve outcomes. This study identifies a combination of both trait and state or iented coping by determining coping behaviors at entry to treatment and actual coping efforts dur ing the management period by individuals with LSS. The overall intention in exploring the coping resources and behaviors of individuals with LSS is to create a foundation for early identificat ion of coping strategies and resource allocation for future LSS treatment programs. Application of the Stress and Coping Model to LBP There are no current studies using Lazarus and FolkmanÂ’s Stress and Coping Model107 in the treatment of LSS. Previous studies on LBP and other chronic diseases have emph asized the importance of appraisal of the stressor, typically conceptualized as general chall enges with pain management, normal function, and coping resources (or perception of resources) a vailable to allow for adjustment to life during and after treatment completion.115-117 This dissertation study defined specific stressors pain and

PAGE 37

22 disability, within the global stressor of the healt h condition of LSS, as reflected in the Specific Aims. This dissertation study aimed to further unde rstand the stressors of pain and disability in LSS to better identify and understand coping behavi ors and resources appraised by individuals with LSS in order to manage their pain and disabili ty. Managing Stressors: Pain The experience of pain is often described as stress ful by an individual and identified as a source of stress for individuals in previous litera ture.118 In recognizing the role of pain as a stressor, it is necessary to explore and expand upo n an understanding of the factors contributing to an individualÂ’s experience of pain. The traditio nal biomedical model of pain dates back hundreds of years when pain was understood to be a primarily sensory experience resulting from the stimulation of noxious sensory receptors usuall y from physical damage or injury.119 This theory of pain describes primarily nociceptive pain defined as pain elicited when sensory receptors specialized to sense mechanical, thermal, or chemical pain react when stimulated past a sensory threshold. This simplistic view suggests th at pain only comes from specific physical pathology and is often called the biomedical model of pain. It does not take into account how pain is experienced by the individual, involving additio nal psychological, social and behavioral mechanisms of injury and illness. Due to the narrow scope of the biomedical model of pain, it is often criticized for being reductionist and exclusi onary.120 In the condition of LSS for example, patients with LSS often experience pain that is unr elated or only partly due to their radiographic severity or lumbar segmental level of pathology,95, 121 illustrating that their pain experience is greater than the pure pathophysiologic processes in volved. Biopsychosocial Models of Pain To encompass a broader view of pain, Turk and Flor 120 described a biopsychosocial approach to pain which addresses many of the shortf alls found in traditional biomedical models.

PAGE 38

23 In this approach, it is recognized that the experie nce of pain is the reciprocal and dynamic interaction of biological, psychological, and socia l factors. It is based on the concept that the experience of pain arises from illness behavior alt hough it is initiated and/or has contributions from nociceptive pain. Illness behavior is a term u sed to describe the ways in which given symptoms may be differently perceived, evaluated, a nd acted or not acted upon by different kinds of persons.122 Illness behavior is believed to be a dynamic proce ss that allows for the role of the biological, psychological and social factors to cha nge in chronic conditions and as the condition evolves.123 In the 1980Â’s, Waddell et al124 applied the construct of illness behavior to LBP ( see Figure 2.2). Their view was that persons with chron ic LBP experienced illness behaviors and not just nocioceptive pain. This view represented a bro ader biopsychosocial model of pain stemming from physiological impairment but with broader cogn ition, affective and social factors resulting in the experience of pain perceived and reported by individuals with chronic LBP.124 Figure 2.2. Biopsychosocial model of pain. Cross sectional representation of the Glasgow model representing the role of fear-avoidance belie fs in chronic low back pain and disability. Adapted from Waddell et al.52 r $rnn!r n %nrnn rn !rrn

PAGE 39

24 Stress-Diathesis Model of Pain Asmundson and Wright125 conceptualized a stress-diathesis model of pain an d disability that stems from the biopsychosocial framework and c an be applied to chronic pain conditions resulting in disability (see Figure 2.3). It is thi s evolved theory of pain that describes how an initial physical pathology or injury is recognized as nociception and leads to a cycle of appraisal of the stress and coping factors needed to manage t he health encounter. Also key to this model is the construct of diathesis which is described as a pre-dispositional vulnerability of the individual. This vulnerability as demonstrated in the model can be influenced by both internal and external factors. Along the pathway there are physical, cogn itive, social, cultural and behavioral processes that contribute to the cycle and associate with the pain-specific distress. The combination of the distress and the diathesis contributes to a larger cycle of pain resulting in apprehension, avoidance, and overall disability (see Figure 2.3). Figure 2.3. The Stress-Diathesis Model of pain and disability. An integrated stress-diathesis model of chronic pain adapted from Asmundson et al.125 rn nn&rr 'n rrn r %n (r )n $rr "r r r !nrn *rn nr nn rr +rnn

PAGE 40

25 Pain appraisals A key part of the stress-diathesis model is the issue of appraisal. Pain appraisal refers to the meaning ascribed to pain by an individual.126 In accordance with the Stress and Coping Model,127 a distinction can be made between primary appraisa l of pain in terms of it being threatening, benign, or irrelevant and second ary appraisal which evaluates the controllability of pain and one's coping resources. Primary appraisal of a threat or harm/loss is indication of a stressor. Therefore, pain as descri bed as threatening by the stress-diathesis model is inherently stressful. The degree of the stress d epends on the other factors of vulnerability, social and cultural influence, and assessment of co ping resources. Pain beliefs Pain beliefs and therefore reports of the pain e xperience develop both during the lifetime of an individual as well as thr oughout the duration of a chronic illness such as with LSS. Appraisal and beliefs about pain can have a big impact on an individual's affective and behavioral response to pain. If a pain signal is in terpreted as harmful and appraised as a great threat, then it may be perceived as intense and unp leasant and may evoke escape or avoidance behavior. An example of this is the rating of pain associated with cancer described as more unpleasant than labor pain, even though the actual rating of intensity is equivalent.128 Studies show that pain appraisal and beliefs are al so determinants of adjustment to chronic LBP.129 After treatment, a large majority of patients with LBP believed that a wrong movement could have serious negative consequences f or their back and further associated this belief with reduced activity levels and increased d isability.130 Using the Stress-Diathesis Model, it is reasonable to expect that through the appraisal process, there are larger biopsychosocial influences that impact perception and a patientÂ’s q uantification of pain. The combination of actual physical limitations, environmental resources and b arriers, and perceived physical, mental, and emotional characteristics influencing actual behavi ors by the individual may in turn impact a personÂ’s quantification of pain.

PAGE 41

26 Managing Stressors: Disability As we see in the Stress-Diathesis Model, pain is on ly one factor in a more complex set of factors resulting from pain but contributing to dis ability. Pain sensations, appraisal of the threat from the pain, vulnerability factors and even socia l and cognitive factors all contribute to the individual’s experience of somatic pain in this mod el. An individual’s responses and beliefs related to pain can play a large role in how he/she acts or fails to act when participating in many of life’s activities and therefore can magnify his/ her level of disability. Understanding disability has been an evolving proce ss in health and health care literature. Multiple models have been proposed that have develo ped the complex understanding we have today about disability and the factors that contrib ute to disability. Understanding this complex process of disability is an important component of this research study as this study sought to uncover factors beyond somatic or physical factors that relate to disability in the population of patients with LSS. Disablement Model Over the years, multiple frameworks have been devel oped to explain a broader concept of disability. One of the earliest frameworks for disa bility was proposed in the 1960’s by the sociologist Saad Nagi called the “Disablement Model ” that illustrates a disease pathway that is still used by health care professionals (see Figure 2.4). The Disablement Model describes a pathway comprised of 4 inter-related but separate c onstructs that contribute in a linear fashion to disability.131 The model starts with pathology as the underlying disease condition that eventually leads to impairments, functional limitations, and d isability. Impairments are abnormalities in structure or function of the body’s systems, which can either be associated with pathology or remain after the pathology has resolved. Functiona l limitations are limitations in a person’s ability to perform tasks and ADLs such as self-care Therefore, limitation in performance reduces the function of the individual. Finally, disabilit y, as described by Nagi, is the pattern of behavior

PAGE 42

27 that starts with pathology but develops with chroni c impairments and functional limitations (see Figure 2.4). Figure 2.4. The modified Disablement Model. Representation of the pathway from pathology to disability with environment contributing to both fu nctional limitations and overall disability. Adapted from Saad NagiÂ’s disablement model.131 Expanded Disablement Model A component of this model that is extraneous to the biomedical pathway of the original Nagi model is th e environment. The Institute of Medicine revised its model to include disability as a functi on of the interaction of individuals with the influence of social and physical environments. As a result, the revised Disablement Model describes environment as including the natural envi ronment, the built environment, culture, the economic system, the political system, and psycholo gical factors.132 The less supportive the environment is to an individual from both a physica l and social perspective, the greater the resulting disability. It is important to identify t his transition in health literature as this initial recognition of environment in disability marked a t urning point for health care in understanding a more complex disablement concept. This expanded dis ablement concept is particularly important to this dissertation study as the focus was to exam ine factors pertaining to the personal and social environment as they interact to define disability i n patients with LSS. Because the traditional Disablement Model illustrat es a linear biomedical framework, it is unclear to what degree the ability or disability of an individual is a direct result of the disease process versus the contributing environment. Moreo ver, the ability to manage functional $rn ) ,n %n r

PAGE 43

28 limitations and disability often depends on the per sonal and social resources available to the individual, but these resources are only grossly re flected by the environment in this model. When using this model, it is difficult to direct resourc es or even identify specific environmental factors that impact disability. Verbrugge and Jette133 expanded on the Disablement Model and conceptualiz ed additional factors that may affect the pathway from pathology to function. Included in this expansion was the concept of risk factors which are predisposing factors that could affect either the presence or severity of the impairments, functi onal limitations or disability in the individual. These factors could include medical care, external supportive actions or systems, activity accommodations, or what is described as “intra-indi vidual factors” such as lifestyle, behavior changes, and psychosocial attributes such as coping strategies. These factors are of particular interest to this research study as they identify th at specific behavioral and social factors may interact during the coping process and contribute t o disablement in patients with LSS. Moreover, Verbrugge and colleagues133, 134 proposed that life activities include much more than those of the traditional medical model wh ich included only ADLs. In this new definition, life activities could be grouped in 3 c ategories: obligatory, committed, and discretionary activities. Obligatory activities are those required for survival and self-sufficiency, including activities such as personal care, sleep, rest, walking and local transportation traditional referred to as ADLs. Committed activities include p rinciple productive roles, such as paid work, and household management, such as food preparation, repairs, yard maintenance, shopping, and child or elder care. Discretionary activities are f ree-time activities that could include socializing with family and friends, entertainment away from ho me, hobbies, physical recreation, volunteer service, religious activities or other adult educat ional activities. Until this new conceptualization of the Disablement Model, the majority of disabilit y research had only focused on obligatory activities and few committed activities but had ign ored discretionary activities as important in the disablement process.133, 135 These concepts of expanded activity are important to this dissertation

PAGE 44

29 study as they identify multiple areas of activity t hat are important in individuals lives that when disrupted have the potential to result in disabilit y. This dissertation study looked closely at limitations in activity beyond obligatory activitie s of daily living recognizing that social and emotional behaviors and roles can be impacted to a greater degree during committed and discretionary activities. Social model of disability An alternative model of disability, called the s ocial model of disability offered an opposing view that disability is primarily a socially created problem and not an attribute of the individual.136 In this model, an unaccommodating physical environ ment creates the problems of disability through negative attitud es and other negative features of the environment. Although, the social model of disabili ty was never intended to be an allencompassing model of disability, it illustrates th e need for reframing the idea of disability from the biomedical model to include issues of social in teraction and social values that to contribute to disablement. This dissertation study has emphasized that the role of social function in the lives of patient with LSS is critical to fully understand th e coping behaviors and resources of these individuals. ICF Framework To combine all of these models and contributions to the concept of disability, a new model was developed by the World Health Organizatio n (WHO) and the Committee on a US National Agenda for the Prevention of Disabilities. This new framework called the World Health OrganizationÂ’s International Classification of Func tioning, Disability and Health (WHO-ICF) elucidates the larger biopsychosocial components of disablement. Known more commonly as ICF, this framework is a universal classification s ystem to define and measure health and disability. It describes health and health-related domains from body, individual, and societal perspectives by means of 2 domains: Functioning and Contextual Factors. Each domain can be used to describe different changes and the impact o f those changes that occur in the individual

PAGE 45

30 and his/her usual environment due to health related states. The domains characterize different factors that interact with the individual to facili tate or hinder functioning and/or disability.137 Changes to different factors often result in a decr ement of health and thereby can be used to describe disability in the individual. Functioning Similar to the Disablement Model, the ICF has compo nents that encompass pathology, impairment and functioning. Th ese are seen in the domain called “Functioning”. The functioning domain includes 4 co mponents called body function, body structure, activity, and participation. Body functi ons are defined as the physiological and psychological functions of body systems. Body struc tures are defined as the anatomical parts of the body. Problems in both body function and struct ure are together usually termed impairments by health professionals and defined as deviations o r loss of structure or function. The other 2 components of activity and participation encompass a larger concept of functioning. Activity and participation are similar concepts although each ha s a slightly different definition. Activity is defined as the action or task performance by an ind ividual. Participation represents the involvement in life activities. Problems in an indi vidual’s ability to participate in activities will result in participation restrictions and limit actu al activity. For individuals with LSS, activities that are limited may include activities such as wal king, climbing steps, or lifting objects. When a person with LSS cannot perform activities such as w alking or stairs, it often precludes his/her participation in many social, recreational, and dai ly chores that require the functional ability and the confidence to execute these activities. Contextual factors As previously seen in the Disablement Model, the ICF describes the environment as a contributor to disability in a lar ger domain called “Contextual Factors”. Contextual factors in the ICF are classified as eit her personal or environmental factors that influence health through functioning and disability As previously described, environmental factors are defined as exerting an influence on ext ernal domains that in turn influence a person’s functioning and disability. Environmental factors r efer to all aspects of the external or extrinsic

PAGE 46

31 world that form the context of an individualÂ’s pers onal life that have an impact on a personÂ’s functioning. Environmental factors can be both faci litators that can assist a personÂ’s functioning or barriers that hinder or limit a personÂ’s functio ning. Because of the inclusion of environmental factors in the original Disablement Model, literatu re related to environmental factors in relation to disability has been extensive. In addition, the con tribution of the social model of disability in the literature has added increased emphasis that person s with disabilities have reduced access to full participation in life and society.138 The other type of contextual factor, personal facto rs, can include anything that is an internal influence on a personÂ’s functioning and di sability. This includes individual characteristics like age, sex, social status, and life experiences.139 Personal factors have been described in the literature as comprising 2 broad categories of demo graphic information and personality traits.140 Personality traits are of particular interest to th e present dissertation research study. These can include personality traits such as coping styles, l ifestyle, habits, social background, past experiences, self-esteem, self-advocacy, and other psychological assets that are not directly related to the health condition. These personality traits can influence how a person responds to possible limitations secondary to disability. Some individuals may have difficulty with assertiveness or have a negative outlook, regardles s of the health condition and corresponding disability. Others may have a sense of positive opt imism or self-determination that persists regardless of the disability. Therefore, personal f actors are those attributes within the person that are not caused by or have nothing to do with the di sabling condition but may influence the personÂ’s overall disability. Researchers have descr ibed these factors as those traits that would still be there even if the health condition and disabilit y were suddenly gone. The recognition of personal factors in the ICF framework illustrates t hat a person is more than simply the sum of his/her physical functioning. The ICF framework and its components are displayed in Figure 2.5.

PAGE 47

32 Figure 2.5. International Classification of Functio ning, Disability and Health framework. Representation of the different factors that intera ct and contribute to disruption of function and increase overall disablement. Adapted from the WHO ICF framework141 Contextual factors of the person and the environmen t interact and modify the impact of changes to body functions, body structures, activit ies, and participation of the individual with a health condition. The degree of impact these factor s have in relation to the health condition help to describe the degree of disability in an individu al with the health condition. The greater the challenges to the personÂ’s functions or contextual factors, the greater the potential physical and psychological stress that is placed on the individu al. Applying the Stress and Coping Model, if health related factors of the person or the environ ment are appraised as taxing or exceeding the individualÂ’s resources or well-being, the health st ate will be viewed as stressful and therefore disabling to the individual. Contextual factors are of particular interest to this research study as "r -nrnrnr. !r ) -rn. -n. rnn. r')n rn)n r)n

PAGE 48

33 they identify both person and environmental factors that are potential moderators in the relationship between coping and outcomes of LSS. Th e ICF as a culmination of defining disablement provides a framework to identify both i ntrapersonal and interpersonal factors that may be interacting to exacerbate or diminish disabi lity in patients with LSS. Application of disablement models in LBP The models of disablement described above serve to illustrate the complex and multifact orial nature of disability as currently understood by clinicians, researchers, and the publ ic health community. As previously identified, LBP is a significant cause of disability in our soc iety. Waddell in 1980 revolutionized the clinical perspective of LBP by illustrating that there were many nonorganic and behavioral components to individualÂ’s experience of LBP.142, 143 The work to revise the concept of disablement alon g with continued research in LBP has resulted in a more co mplex understanding of the typical spectrum of problems of functioning in patients with nonspec ific LBP. A variety of studies have used the ICF framework to assist in understanding the comple xity of nonspecific LBP.144-146 Psychosocial and environmental risk factors have been identified in the development and management of nonspecific chronic LBP.52, 124, 143, 147, 148 Although not focused on the condition of LSS, thes e studies show the complexity of chronic LBP beyond s omatic symptoms and support the multifactorial disablement models in LBP. Using the Stress and Coping Model in combination with the ICF framework and the Stress-Diathesis Mod el, respective coping processes and resources can be better identified that describe th e experience of pain and disablement in patients with LSS. Ultimately, these theoretical contributio ns serve to support the goal of understanding and improving the allocation of resources to reduce the consequences of disability in persons with LSS. Resource Theories of Stress and Coping In order to fully understand the experiences of pai n and disablement in LSS, it is important to investigate whether and how intraperso nal and interpersonal factors interact to

PAGE 49

34 influence the severity and the clinical outcomes of LSS. Intrapersonal and interpersonal influences would be classified as part of the conte xtual factor category in the ICF framework; namely personal and environmental factors. To furth er understand how personal and environmental factors contribute to disablement in a chronic condition like LSS, theoretical constructs have been identified at the personal and environmental in a number of resource theories of stress and coping. In other health cond itions, social and personal constructs have been proposed to serve as coping resources that address stressful person-environmental transactions. It is unknown what social and personal constructs are important in coping by patients with LSS. Four specific constructs are explored in this resea rch study including the personal constructs of self-efficacy, self-rated physical health, self-rat ed mental health, and the social construct of socia l support. Social Cognitive Theory and the Construct of Self-E fficacy At the clinical level, it may be important to ident ify risk factors for heightened pain and disability and whether individual influences affect or alter outcomes in identical treatment protocols for patients with LSS. There is evidence that individual factors related to stress, coping, and cognitive mechanisms can influence aging-relate d health in general,149 although currently no evidence exists on the role of these individual fac tors on the severity or the health outcomes of pain and disability in patients with LSS. Social Cognitive Theory (SCT),112, 150 originally called Social Learning Theory by Miller and Dollard and expanded by Bandura,112, 151, 152 addresses the psychosocial aspects of individual health behavior as well as methods of promoting beh avioral change. According to SCT, individuals possess a self-system that enables them to exert control over their thoughts, feelings, motivations, and actions.152 SCT is a theory of human behavior that emphasizes the importance of personal beliefs as key components of control and p ersonal agency. In this theory, individuals are viewed both as products and as producers of their o wn environments and social systems.

PAGE 50

35 SCT hypothesizes that people’s beliefs in their cap abilities mediate how they behave in situations such as participating in treatment for a health condition. Treatment provides an environment in which there are opportunities for ph ysical and psychosocial support by a health professional, family, or friends which can in turn affect their beliefs about their capabilities. Both external support and treatment itself may help to i mprove an individual’s beliefs about his/her capacity to perform functional tasks. At this time, it is unknown how personal beliefs or external support can influence appraisals and coping behavio rs in a patient with LSS. Self-efficacy The construct of self-efficacy is a key concept within the SCT framework. Self-efficacy is an intrapersonal factor defined as a person’s belief in one’s capacity to “organize and execute the courses of action required to produ ce given attainments.”153 It is a contextspecific assessment of one’s competence to perform a specific task or range of tasks in a given domain. Bandura153 describes self-efficacy as influencing 4 areas: 1) the choices that are made, 2) the effort put forth in task specific roles, 3) the time one persists when there are obstacles, and 4) one’s feelings of confidence in performing speci fic tasks in specific situations. Depending on what is being managed, the tasks over which personal influence is exercised may entail regulation of one's motivation thought processes, affective states and actions, or changing environmental conditions. Self -efficacy beliefs are sensitive to these contextual factors. Therefore, they differ from oth er expectancy beliefs in that self-efficacy judgments are more task and situation specific and individuals make use of these judgments in reference to the type of goal.150, 154, 155 Self-efficacy can refer to sub-skills required to organize actions that are governed by broader self-regulator y skills such as knowing how to diagnose task demands or constructing and evaluating alternative strategies. Possessing these self-regulatory sub-skills can permit individuals to improve their performances across varied activities particularly when faced with sensations of pain dur ing activity performance. Self-efficacy can generalize across skills when commonalities are cog nitively structured across activities. For example, physical exercises and motor skills practi ced in health care treatments may transfer to

PAGE 51

36 other activities performed in daily living as their confidence to perform these types of tasks increase. People make and shape self-efficacy beliefs in the context of performing a specific goal or task. Theoretically, self-efficacy beliefs might influence a patientÂ’s experiences of pain, functional limitations, and disability with LSS. A patient in treatment practicing functional activities like walking could increase their task s pecific self-efficacy for walking just by practicing and as a result, increase their confiden ce and activity outside of treatment. Moreover, a patient who enters treatment with high self-effic acy for functional tasks might have even greater belief that he/she could organize and execute all o f the actions needed to successfully complete tasks required in treatment. Theoretically, a pati ent with higher self-efficacy could have heightened beliefs about accomplishing tasks that i nclude positive health behaviors such as attending regularly scheduled clinical appointments following medical instructions, executing a series of home and clinic exercises, and performing positive avoidance behaviors to minimize pain and discomfort. Overall, high self-efficacy b eliefs lead to setting challenging individual goals and maintaining strong commitment to these go als that ultimately could affect outcomes of treatment. In addition, researchers think self-efficacy belief s in part determine outcome expectations.153, 156 Individuals who expect success in a particular tas k often produce successful outcomes in that task. The opposite is also true o f those who lack such confidence in task performance; those who doubt their success will pro duce failed outcomes. Patients undertaking tasks such as those found in treatment would have s imilar anticipatory outcome expectations about performing the specific tasks required for th e process of treatment and transfer these beliefs to similar skills outside of treatment. Issues of self-efficacy in the elderly population o ften center on reappraisals and misappraisals of their capabilities. Because biolog ical conceptions of aging focus extensively on declining abilities, many physical functions can de crease as people grow older and require

PAGE 52

37 reappraisals of self-efficacy for activities in whi ch biological functions have been significantly affected. When a therapist teaches an older adult t o use his/her intellectual and physical capabilities, his/her improvement in cognitive and physical functioning more than offsets the average decrement in performance over 2 decades.153 Therefore, elderly persons who invest the necessary effort can function at higher levels comp arable to some younger adults. Although never studied in previous LSS research, it is possible th at perceived self-efficacy may affect the level of involvement in activities and theoretically mitigat e the decline seen from chronic disease processes such as LSS. Application of self-efficacy What is unknown is if self-efficacy can maintain or improve in the presence of pain and disability perc eived by the patient with LSS. It is known that pain can impede or prevent patients from participat ing in an activity because of the actual physical sensations of pain, the fear and or avoida nce behaviors that preempt participation, and the cognitive appraisals or reappraisals that are m ade after activities have repeatedly resulted in painful experiences. Reductions in function can als o lower an individualÂ’s self-confidence, and as a result continue to lower the level of activity in which an individual participates in daily life. In laboratory experiments, self-efficacy beliefs pr edict tolerance levels to pain.157 From a physical perspective of pain, perceived self-effica cy has been shown to affect the body's opioid and immune systems.158 In patients with chronic pain disorders related to LBP, self-efficacy positively affects physical and psychological funct ioning.159 Evidence shows that self-efficacy influences pain and function after acute physical i nterventions like surgery.160, 161 Prospective studies in patients who underwent orthopedic surger y demonstrated that high self-efficacy before the start of treatment and larger increases over th e course of treatment speed recovery and predict better long-term outcomes.162-164 It is possible that individuals with high self-eff icacy may be more motivated to engage in health promoting behavi ors and adhere better to treatment recommendations because they have higher performanc e success expectancies. In addition, it is believed that people with high self-efficacy are le ss likely to give up an activity when faced with

PAGE 53

38 barriers such as pain and are therefore less likely to become trapped in the negative spiral of activity avoidance, physical deconditioning, loss o f social supports and depression. Application of self-efficacy during treatment In rehabilitation literature, self-efficacy is measured by using task-specific constructs relev ant to the chronic disease process and specific rehabilitation treatments being studied. Research indicates that task specific self-efficacy can improve during the treatment process. Scherer and Schmieder165 demonstrated improved task specific self-efficacy with patients who participat ed in treatment for dyspnea due to chronic obstructive pulmonary disease. Patients who complet ed educational and exercise training had significant increases in their self-efficacy scores to manage or avoid breathing difficulty. In addition, patients increased their self-efficacy ex pectations for exercise endurance.165 Carlson et al166 showed that cardiac rehabilitation treatment invol ving physical exercise improved patientÂ’s post-treatment self-efficacy beliefs for independen t exercise. Another study by Jeng and Braun167 found that those with greater success in functional outcomes from cardiac rehabilitation treatment correlated with higher exercise specific self-effic acy scores. Finally, Rejeski et al168 found that when patients undergo aerobic and strength training exercise for knee osteoarthritis (OA) those who actively engaged in physical exercises compared to controls had increased self-efficacy outcomes for stair climbing, a task-specific limita tion for patients with knee OA. Research has demonstrated that self-efficacy is als o an important factor in improved coping and psychological outcomes.169-171 Taal et al103 surveyed patients with rheumatoid arthritis (RA) to determine their level of self-efficacy in r elationship to function, pain, and disability when coping with RA. Higher self-efficacy scores correla ted with improved coping abilities independent of pain, disease status and functional abilities. Moreover, Strahl et al172 found that higher self-efficacy levels in patients with arthri tis were predictive of better outcomes related to psychological functioning. An investigation of self-efficacy for patients with chronic LBP found that initial selfefficacy beliefs predicted functional abilities. In a study by Lackner and Carosella173 that included

PAGE 54

39 patients with chronic LBP, 100 patients rated their confidence to perform load lifting tasks before any examination or treatment. Next, these patients underwent a subsequent physical exam and physical performance test for lifting loads. The re sults indicated a significant association between self-efficacy beliefs and lifting higher loads and higher physical performance. Lackner and Carosella did not find an association between pain perceptions or measures of psychological distress and the physical performance measures. Overall, these studies indicate that self-efficacy is an important construct in health care management. Two of the studies167, 173 found significant relationships that demonstrate s elfefficacy beliefs have an impact on the outcomes of pain and disability for patients with chronic disease processes. In addition, no studies have exa mined the impact of self-efficacy beliefs during treatment of patients with LSS. Although the role o f self-efficacy in the treatment process warrants more investigation, these studies indicate that heightened patient beliefs regarding the ability to perform specific tasks might also raise the functional, psychological, and overall coping behaviors of patients during and post-treatment and therefore help to decrease overall perceptions of pain and disability. Although the role of selfefficacy is unknown in LSS, this research study sought to investigate the impact of self-efficacy i n the appraisal of pain and disability and coping behaviors of patients with LSS. Social Relationships and Health In addition to individual level factors such as sel f-efficacy, there is considerable evidence that interpersonal factors play a role in disease p revention, disease management, and health outcomes. At this time, it is unknown the role of interpersonal factors play in the health and management of patients with LSS. In general, resear chers studying aging have determined that people with social support and social ties, regardl ess of their source, live longer than people who are isolated.174 People with a close network of ties with other peo ple maintain better health, resist disease and deal more successfully with problems th ey encounter.174, 175 People who are socially

PAGE 55

40 isolated and have fewer social relationships have b een found to have mortality rates 2 to 5 times higher than those with good social relationships, r egardless of gender, race, ethnic background and socioeconomic status.176, 177 There are several terms used to describe social rel ationships as they relate to health. Social integration refers to the existence of socia l ties.176 Social network describes the interconnectedness of social relationships that sur round an individual.177 Together, social integration and social networks describe the existe nce and quantity of social connections that an individual has available. Social support on the ot her hand is a concept that illustrates not only the existence of a social relationship but also the pos itive function of that social connection and its potential to play a supportive role in an individua lÂ’s health.113, 178 Social support is defined as the degree to which a personÂ’s basic social needs are m et through interaction with other people.179 It includes tangible and intangible resources, emotion al support, informational and instructional support, as well as a personÂ’s perception of assist ance in times of need. When measuring supportive functions in health, measurement should include support that one believes is available if needed (perceived support) or functions that one reports and is recently provided (received support).113, 177 This research study focused on identifying both pe rceived and received support in patients with LSS. Social support The role of social support has evolved in health research and is believed to be a social construct that protects and preserve s well-being of an individual in the face of a stressful encounter such as a negative health condi tion.177 Social support is believed to contribute to global self-evaluations of health.180 In addition, social support serves as a social res ource that can assist an individual in his/her capability to c ope with stress.107 Even though social support comes from an external source, this external suppor t is believed to be able to help protect people from the adverse effects of stress (Figure 2.6). In the stress and coping literature, social support i s a coping resource that can alter beliefs and commit ments that can in turn alter how one appraises a stressful situation.107

PAGE 56

41 When considering the role of social support, stress ors are believed to act on 2 different pathways. One pathway is through appraisals of ava ilable social support (Figure 2.6). The perception that social support is available may pro tect people from the adverse effects of stressors by leading them to appraise or interpret stressful situations less negatively. For example, a patient with LSS with high perceived social support might i nterpret his/her pain and disability less negatively than a person with low perceived social support and therefore approach the process of treatment and self-management differently. Hypothet ically, patientsÂ’ interpretations of support may moderate their experience of pain and disabilit y before, during and after the treatment process. Figure 2.6. Social support interactions with Stress and Coping Model. Interactions between perceived support with appraisals as well as receiv ed support with coping to modify the relationship of a stressor on health. The other pathway social support is believed to ass ist in coping is by actual supportive actions, or received support. Received social suppo rt is the actual assistance by others during stress.181, 182 Received support is believed to interact and alter a personÂ’s coping abilities in times of stress. This improved coping due to increased re ceived support is theorized to reduce the rnn n "r rrr +rrr

PAGE 57

42 impact of stress on a personÂ’s health.177 When a person can reduce the impact of stress on h is/her health, it is theorized that this could be measured through improved health outcomes such as reduced pain and/or disability. A way clinicians ha ve measured social support is by quantifying the frequency or number of supportive actions a per son reports as well as his/her perception of the quality of that received support.183 Retrospectively identifying the frequency and quali ty of support received during a specific time period of t reatment, identifying the supportive actions, and rating how this support was helpful or unhelpfu l are all ways that patients might report received social support for a health condition. Application of social support in treatment Evidence exists on the importance of overall social support in the treatment of chronic diseases, particularly in the treatment of cardiovascular disease184, 185 and stroke.186-189 These studies have shown that improvements in physical functioning and psychological adjustment i n older adults in cardiovascular and stroke recovery were greater among those with higher sourc es of social support.185-191 Moreover, other studies have shown that social support from all sou rces positively correlated with increased perceived general health and quality of life after stroke.192-194 Researchers have also shown lack of social support in general to be associated with neg ative health consequences including suicidal thoughts,195 depression,104 length of hospital stay,196 discharge to rehabilitation and nursing home facilities,197 and reduced physical functioning.198, 199 There is additional evidence that patients with a v ariety of chronic disease processes benefit from social support from all sources. In on e study, Littlefield et al200 found patients with Type II Diabetes Mellitus had improved function wit h higher total social support scores but worse function with higher depression scores. In another study, Yates201 surveyed patients with cardiovascular disease who reported that both emoti onal support and tangible aid from health care providers and the patientsÂ’ spouses were important sources of support for coping with their disease process and improving overall physical reco very. Additionally, Gulick202 conducted research involving patients with multiple sclerosis and found that higher total social support

PAGE 58

43 scores correlated positively with ADL function scor es and inversely correlated with depression scores. Research specifically investigating arthritis found that social support does mediate the response to the treatment process and function of t he patient. In patients with RA, Taal et al103 found tangible or instrumental support by caregiver s corresponded positively with improved patient reported health status and participation in daily physical activities. However, emotional support did not relate to patientsÂ’ reported health status for this sub-group of patients with RA. Weinberger et al203 illustrated in research involving patients with OA with concurrent functional limitations that both older age and lower levels of tangible support from caregivers, friends, or family associated directly with greater physical di sability. Moreover, greater psychological disability was associated in patients with OA with lower levels of emotional support from all sources. Little research has directly examined the associati on between treatment for LBP and social support. Masters et al183 investigated whether patients received social supp ort in the treatment process for LBP. Retrospectively, the res earchers asked patients who participated in treatment for LBP to indicate which sources of soci al support they received and whether this support was helpful or unhelpful. Out of the 50 pat ients surveyed, 43% indicated that they received tangible support from a physical therapist and physician that was helpful. Thirty-three percent of the patients indicated they received hel pful emotional support from physical therapists, family, and friends. Twenty-three percent of the pa tients additionally reported that they received helpful informational support from physical therapi sts. However, 50% of the patients also reported that they received unhelpful emotional sup port from their family and friends during the treatment process. Additionally, 37% reported recei ving unhelpful informational support and 10% reported receiving unhelpful tangible support from their physicians during the treatment process. Unfortunately, the researchers did not evaluate the relationship of support to other general health, physical function or psychosocial measures. In addi tion, the researchers did not evaluate the

PAGE 59

44 relationship of social support to the outcomes of p ain and disability in treatment for LBP. Overall, this study illustrates that social support sources are available and are identified as both helpful and unhelpful in treatment for LBP. Yet understandi ng how social support impacts the experience of pain and disability in LBP, particularly in LSS, continues to be lacking in the literature. General Health Status Self-rated health is another way to identify a pati entÂ’s perspective of his/her health condition and may contribute to poor coping behavio rs as well as improvement in health outcomes. Self-ratings of health are used frequentl y in health literature and have shown to be predictors of morbidity and mortality.204 Self-rated health has also been proven sensitive t o improvements and declines in many health related ou tcomes.205, 206 Self-rated health measures assess subjective well-being related to physical an d mental domains of health. These measures are used extensively in clinical trials and in health s ervices research where they have shown evidence of reliability, validity, and responsiveness.207 Yet, researchers have only a limited understanding of how these self-evaluations are reached or their implications for health behaviors. In addition, few studies have examined changes in self-rated hea lth over time and in specific health conditions such as LSS. The perception of health status during a negative health experien ce is believed to be just as relevant to health outcomes as the actual somati c state of the individual. Even though most medical treatment still focuses on a personÂ’s actua l physical state as it relates to pain and illness, perceived health may largely contribute to a person Â’s well-being and function. Cross-sectional studies have typically shown that general self-rate d health is closely related to the experience of physical symptoms.205, 208 Moreover, a 5 year study by Gold et al209 of health outcomes in the community determined that both functional ability a nd self-rated health were independent predictors of health outcomes. Those with initial l ow functional scores also had greater decline in self-rated health over time reflecting the physical component in self-evaluations of health.

PAGE 60

45 Appraisal of health Self-rated health, both physical and mental heal th, is likely to be part of the appraisal process when faced with a str essful health condition. Health perceptions may also be part of coping processes when an individual faces a specific illness. When an individual has a better perception of their overall health, th ey have a greater potential for these health belief s to bolster positive coping behaviors during illness experiences. It has been shown that older adultsÂ’ overall perception of health can be predict ive of their use of active coping strategies to deal with age-related health challenges.210 Research has also shown that those who donÂ’t have a positive health perception may engage in sick role behaviors that can lead to self-destructive behaviors during illness.122 One reason for this is that self-rated health is b elieved to be a reflection of an enduring self-concept of a healthy or unhealthy person.211 Therefore this selfconcept of health may act as a moderator when engag ing in positive or negative illness behaviors. Variance in physical measures and outcomes have als o been shown to be related to respondentsÂ’ previous self-rated health.211 Longitudinal studies have illustrated that those w ith initial lower self-rated health predict poor functi onal ability and increased health care utilization over time.206, 212 Negative self-rated health has been found to predi ct long term disability and health decline in the general population.212 Change in self-rated health have been shown to coincide with long-term changes in physical health, mental health, perceived social support, and performance related behaviors.211 Moreover, self-rated health has been shown to vary according to whether respondents intended to improve specific health-related behaviors in the future.211 These findings suggest that self-rated health is bo th a current assessment of one's health status, similar to a self-concept, and a reflection of effo rts to achieve relatively important health-related goals. During illness, perceived health status may be dete rmined in part by an individualÂ’s level of emotion and distress during illness. Individuals have been found to experience different physical bodily responses based on their differing emotional states.213, 214 Some previous research has indicated that psychological function and perce ived mental health are related to functional

PAGE 61

46 status.215 Using the Stress-Diathesis Model, vulnerability or diathesis is an integral component in the experience of pain and disability. Therefore, i t is probable that components of perceived psychological function, or mental health, can also have proximate influences on the appraisal process and indirectly influence health outcomes. Application of self-rated health Measurement of perceived physical and mental hea lth has been performed routinely in the health literatu re by several standardized tools.73, 209, 215, 216 These general health surveys have been designed to measure overall self-rated health with a broad range of questions covering a variety of aspects of physical and mental health. It is commonly felt that the usefulness of general measures is in their ability to allow comparisons among patients with the same condition as well as between patients with different conditions. In some cases, general health measures may be able to identify uns uspected issues from a diagnostic group that would be highlighted as scores deviating from popul ation or disease specific norms. These selfrated indicators likely take into account the patie ntÂ’s physical sensations such as bodily pain, the patientÂ’s comparison of what he/she can perform in his/her daily life as compared to previously or as compared to his/her peers, as well as his/her ps ychological and social perception of functioning. Overall, understanding how a patient perceives his /her general physical and mental health may add to the larger role of intrapersonal and interpersonal level influences on outcomes of pain and disability. Theoretical Summary Improvements in levels of pain and disability are r outinely monitored and measured in treatment programs; however the influences mediatin g the severity and outcomes of these clinical variables are not fully understood. This study exa mined if personal factors at an individual or intrapersonal level, specifically self-efficacy, and social fact ors at an interpersonal level, specifically social support, contribute to a greate r understanding of the experience of pain and disability in patients with LSS. Because there is some evidence that individual factors such as

PAGE 62

47 stress, coping, and cognitive mechanisms can influe nce aging-related health in general,149 it was important to include a measure of perceived physica l health and perceived mental health to the understanding of the experience of pain and disabil ity. Although self-efficacy and social support have been previously studied in stroke, cardiovascu lar disease, arthritis and LBP, the contribution and association of self-efficacy and social support to understanding the experience of pain and disability in patients with LSS has not been studie d. Gaps Filled The goal of this study was to understand how the in fluences of intrapersonal and interpersonal factors contribute to the experience of pain and disability in LSS. This study addressed multiple gaps in the LSS literature. Gaps in the LSS literature include a broader understanding of the experience of pain and disable ment in this population, understanding appraisals and coping used by patients with LSS, an d the use of mixed methods research. This study aimed to fill some of the gaps in the literat ure by evaluating the experience of pain and disability, particularly identifying the contributi on of self-efficacy, social support, and perceived health status in management of patients with LSS. This study tested existing measures of self-efficac y, social support, perceived physical health, and perceived mental health in a unique pop ulation before and after clinical treatment. This study added to a body of literature on how cop ing behaviors operate in the context of health and disease. Ultimately, this study aimed to identi fy factors not recognized in the current literature that contribute to the experience of pai n and disability in patients with LSS. By using a mixed methods research design, a more complete pict ure was developed that will inform clinical practice as well as future research and policy reco mmendations for this patient population. As information is revealed, future research will need to determine how to best identify, monitor, and possibly intervene using concepts of self-efficacy, social support, general health perceptions, and any novel factors uncovered during the study period

PAGE 63

48CHAPTER III EXPERIMENTAL DESIGN AND METHODS Research Design The study used a subset of a larger data set from a n ongoing randomized clinical trial (RCT). This study used a longitudinal, cohort desig n with mixed quantitative and qualitative methods.217 The subset of patients with LSS studied participat ed in 1 of 2 non-surgical treatment programs (Figure 3.1). Patients were evaluated upon entry to treatment and the time period between 6 months and one year post-treatment. Treat ment effects were not evaluated in this substudy and will be evaluated upon completion of the larger RCT. The sample in this subset is similar to the larger population of patients with L SS.218 All patients in this subset and in the larger RCT are at a similar stage of chronic disease as th ey have been diagnosed with LSS upon entry to the study and display signs and symptoms consistent with the diagnosis of LSS.217 All patients in this subset and in the larger RCT have been randomi zed to treatment including epidural steroid injection (ESI) and low back education, with half t he subjects receiving additional physical therapy (PT) (Figure 3.1). Both ESI and PT represen t the standard of care for current clinical practice. Although both non-surgical treatment opti ons have been found to be helpful, at this time neither have been found to be superior in determini ng clinical outcomes of LSS.219

PAGE 64

49 Figure 3.1. Design of the randomized clinical trial for non-surgical treatment of LSS. Research design illustrating baseline procedure, ra ndomization, and treatment groups. Within the larger RCT, there are opportunities to a nswer many questions. However, the focus of this dissertation research was to investig ate if behavioral and social factors such as selfefficacy, social support, and self-rated health had an impact on standard clinical intake measures of pain and disability. In addition, this dissertat ion research sought to identify the role of selfefficacy, social support, and self-rated health in predicting positive outcomes of pain and disability in patients with LSS long-term. Understa nding treatment effects is an important component in management of patients with LSS; howev er, unless we look closer at behavioral and social factors in the management process, we ca nÂ’t draw complete or sound conclusions regarding patient severity or outcomes. In addition understanding the role of behavioral and social factors in the long-term management of LSS c an help direct clinical resources to address the specific needs of the patient. Future studies w ill need to test the findings from this study in an %nn,/% r' nr' !nrrrn +0 $1&1! 234&&rr $1!

PAGE 65

50 intervention trial that uses strategies which targe t significant influencing behavioral and social factors in order to improve outcomes in patients wi th LSS. Subset of the RCT For this study, a subset of the study population fr om the RCT was selected for both quantitative, repeated measures survey and qualitat ive open-ended interviews. A cohort study design involves participants who meet specified cri teria such as a specific health outcome or membership in a group of interest.217 In this case, persons with LSS who enter conservat ive treatment are the persons of interest and the entry point to treatment and the outcomes of their treatment related to pain and disability scores are the outcomes of interest. All patients participated in self-report surveys of general health, condition specific pain and disability as well as specific intrapersonal an d interpersonal domain measures of selfefficacy, social support, and self-rated health at entry to treatment and at long-term follow-up. Follow-up surveys were mailed at approximately 6 mo nths post entry to treatment to 22 participants who completed treatment. Twenty subjec ts that completed treatment and the survey packets were included in the post-treatment measure ment. The individual response time for follow up of post-treatment measurement in the 20 s ubjects was between 6 and 12 months. The same 20 subjects who completed treatment and follow up surveys participated in a 30-50 minute interview. All subjects received the same non-surgi cal treatment protocol for low back education and epidural steroid injection (ESI), with approxim ately half of the subjects receiving additional physical therapy treatment (ESI+PT). All procedure s for the RCT treatment protocols are located in the Manual of Standard Operating Procedures (MOS P, Appendix A). Figure 3.2 depicts the study design.

PAGE 66

51 Figure 3.2. Longitudinal, cohort study design using mixed metho ds. The quantitative analysis sample consisted of 34 patients at entry to treatme nt and 20 patients at follow-up. The qualitative interview sample consisted of the same 20 participa nts at follow-up. All participants were part of a subset of the ongoing RCT sample of patients. Mixed Method Design Mixed methods approach to inquiry was selected over other study designs because it enables the researcher to capture both empirical me asures and patientsÂ’ accounts of the lived experience of pain and disability with the chronic condition of LSS. By combining quantitative and qualitative data collection techniques, mixed m ethods allows for a rich, complex, and more complete understanding of the phenomenon under stud y.220 Mixed methods utilizes the strengths of both quantitative and qualitative methods to off set the weaknesses inherent in each individual method and enhances validity of the findings.221 For the chronic condition of LSS, previous literature has not fully explored the meanings of p ain and disablement. Although a narrow and focused understanding of pain and disability in pat ients with LSS is present in the current %nn +rnn $rr 5r % nn 5r % nn 5r % $rrn 5r % +r 5r % ) rn r $rr %r

PAGE 67

52 literature, the addition of qualitative research to the quantitative data provides different perspectives on the experience of both pain and dis ability. This mixed methods study employed a semi-convergent parallel design, capturing quantitative data at entry to treatment but then bo th quantitative and qualitative data at relatively the same time period post-treatment, as shown in Fi gure 3.2.222 Quantitative methods were used to explore questions and test hypotheses regarding the factors associated with pain and disability in LSS. Qualitative methods were used to enhance an d expand upon the quantitative findings, give the patients a voice, and explore the nuances and meanings of pain and disability related to the lived experience of LSS. The purpose of the sem i-convergent design is to obtain different but complementary data on the same topic to pull togeth er the strengths of both quantitative and qualitative methods.223 This design is often used to triangulate findings by directly comparing and contrasting quantitative statistical results wi th qualitative findings for corroboration and validation purposes as well as to develop a more co mplete understanding of a phenomenon,222 in this case, the experience of pain and disability in individuals with LSS. Finally, the mixed method approach allows for multi ple worldviews in research related to LSS. Since a large portion of the research to date in LSS is hypothesis driven and uses a single method, the primarily focus has been a paradigm of positivism. The addition of other worldviews including paradigms from both post-positivism and c onstructivism in understanding pain and disability in this population is warranted. Since a single worldview currently exists in the literature investigating LSS but the pain and disab ility literature uses multiple worldviews, a pragmatism worldview was used in this study. Pragma tism is a set of ideas that draws on diverse approaches and uses both objective and subjective k nowledge. According to Teddlie and Tashakkori,224 using a pragmatism approach in mixed methods place s the research question as the primary focus rather than the method or the philoso phical worldview that underlies the method. The pragmatism approach is well suited to answer th e studyÂ’s research question by combining 2

PAGE 68

53 types of data sources to uncover a more complete st ory about the experience of pain and disability in patients with LSS. Data Collection In order to corroborate and compare the information from both quantitative and qualitative data sources, this study included data from the same 20 participants at post-treatment. Parallel data collection questions were conducted a cross both the quantitative and qualitative data sources. Equal weight was given to both quantitativ e and qualitative data. In addition, results of the longitudinal data collection was conducted usin g repeated measures surveys at entry to treatment (n= 34) and post-treatment (n=20) and cor roborated with in-depth, open-ended interviews (n=20) collected post-treatment from the RCT titled Epidural Steroid Injection versus Epidural Steroid Injection and Manual Physical Ther apy and Exercise in the Management of Lumbar Spinal Stenosis; a Randomized Clinical Trial (NCT00786981). IRB approval was completed by COMIRB and Hawkins Foundation (see App endix B). Sampling and participants Figure 3.3 provides a visual representation of the process for sampling. The population for the study included individuals 50 years of age and older with a diagnosis of LSS in Colorado and South Carolina who entered conservative treatment for this chronic condition. The study used a subset of 43 of the total 80 subjects planned to be enrolled into the larger RCT using convenience sampling of i ndividuals who sought treatment either at the Spine Center at the University of Colorado or the S teadman Hawkins Clinic of the Carolinas. These sites were located in 2 different geographic areas in the United States and represented a broader view of the population of LSS. Subject randomization, baseline examination, and tr eatment procedures Patients who sought medical care at the 2 clinical sites, me t the inclusion criteria (see Table 3.1), and consented to participate were enrolled in the RCT. All patients were required to have health care insurance coverage (private insurance, Medicare, or Medicaid) to cover the standard treatment

PAGE 69

54 protocols designed for the RCT which included stand ard treatment with ESI and PT. All patients in the study had a diagnosis of LSS, had symptoms o f neurogenic claudication or radicular pain, and were candidates for non-surgical treatment. For the purposes of this study, LSS was defined as evidence of narrowing of the spinal canal and/or intervertebral foramen demonstrated on diagnostic imaging studies (MRI or CT scan), and cl inical signs and symptoms consistent with a diagnosis of LSS including patient-reported reducti on in walking tolerance due to pain and/or cramping in the leg(s), defined as neurogenic claud ication. Table 3.1. Inclusion and Exclusion Criteria for the Nonsurgical RCT with Patients with LSS Inclusion Criteria 1. Age greater than or equal to 50 years 2. MRI findings consistent with lateral foraminal and/ or central lumbar spinal stenosis (evidence of compression of lumbar spinal nerve root(s) by de generative lesions of the facet joint, disc, and/or ligamentum flavum) 3. Chief complaint of pain in the low back, buttock an d/or lower extremity(s) 4. Presence of symptoms consistent with neurogenic cla udication (numbness, tingling, cramping, and downward radiating pain which is not exacerbate d with biking, uphill ambulation, and lumbar flexion but is not alleviated with standing; or does not include signs of pulselessness, paralysis, or pallor). 5. Rates sitting as a better position with respect to symptom severity than standing or walking 6. Lives within one hour of a research site 7. Can attend 10 regular physical therapy appointments spread over 10 weeks and 4 examination appointments (baseline, end of treatment, 6 months, and 1 year) 8. Sufficient English reading and language skills and mental capability to complete self-report assessment questionnaires 9. No contraindications to MRI Exclusion Criteria 1. Severe vascular, pulmonary or coronary artery disea se which limits ambulation (as determined by the referring physician or the therapist) 2. Other orthopedic conditions or physical impairments of unrelated nature which would limit ambulation or prevent the subject from fully partic ipating in any other aspect of the rehabilitation exercises (as determined by the refe rring physician or the therapist) 3. Previous spinal surgery that included fusion of two or more vertebrae 4. History of spinal tumors, spinal infection, or lumb ar vertebral fractures other than spondylolisis or spondylolisthesis 5. Signs/symptoms suggestive of potential non-benign o r pathologic condition as the origin of symptoms 6. Presence of any absolute contraindications to sub-m aximal treadmill testing per the American College of Sports Medicine (ACSM) standards. 7. Epidural steroid injection within the last 365 days

PAGE 70

55 A physician, resident, or physicianÂ’s assistant ind icated whether the patient met the inclusion criteria to participate in the RCT. In a ddition, the physician or physician assistant explained the study, the risks and benefits for pat ients, and asked eligible patients who were interested in participating to sign 3 copies of the informed consent form (Appendix C). The physician, resident, or physician assistant, perfor med the general baseline examination. After the baseline physical examination, patients completed p aper and pen self-reported questionnaires (Appendix D). The RCT employed a random number generator to estab lish randomization lists prior to the initiation of the RCT. After all baseline proc edures were completed including consent, physical exam, and baseline surveys, the patients w ere randomized to 1 of 2 treatment groups: epidural steroid injection only or epidural steroid injection plus physical therapy. The procedures for baseline examination, randomization, and group treatment procedures were standardized so that all patients in each treatment group and at ea ch site received the same care (see MOSP: Appendix A). Subjects As of October 2012, 43 subjects were enrolled into the RCT. Of these 43 subjects, 12 subjects dropped out and 1 subject die d after baseline enrollment. Of those 13 subjects who dropped out, only 9 had completed base line paperwork, 11 had no or partial treatment, and 2 had completed treatment but were l ost to follow up (see Figure 3.3). Eight of the 43 subjects were still in the process of treatment at the end of data collection for this sub-study on 10/31/2012. Two subjects who had completed treatmen t were not a part of the patient interviews. In total 34 subjects completed baseline information and were included in the sub-study data at entry to treatment. Twenty of these same 34 subject s completed follow-up surveys and a phone interviews during the post-treatment phase (see Fig ure 3.3). All quantitative data were collected and kept in a locked storage cabinet and were enter ed into a secure data base in a password protected computer. All quantitative data were only accessible to the research investigators of the

PAGE 71

56 RCT. Participants who completed baseline evaluation treatment, and completed the follow up surveys between 6 and 12 months after enrolling in the study, received $50 in remuneration. A total of 20 subjects who completed baseline evalu ation, treatment, and follow up surveys were invited and verbally agreed by phone t o an interview session consisting of a series of open-ended questions. Of these 20 subjects, 12 were in the ESI only group and 8 were in the ESI plus PT group. All interviews lasted approximat ely 30-50 minutes and were conducted by phone. Interviews were audio-recorded during the ph one conversation and transcribed verbatim into word documents along with field notes made aft er the interview. Interview and field note data were kept in a password protected computer. A ll data collected during the interview were kept confidential and coded to the subjectÂ’s number given during the RCT randomization procedures. Patients who completed an interview re ceived an additional $25 in appreciation for their time, effort, and sharing their experiences w ith LSS. Figure 3.3 illustrates subjects who participated in baseline evaluation and subsequent follow up surveys and phone interview posttreatment.

PAGE 72

57 6rn"nr rr*7894 rr)n*784: r*79; n34<94<38 n) rr*799 rr)n*78; r*7= n34<94<38 r+&n34<94<38*7;9 !nrrnnrnnrnrr nr'nr r*7;4 %rnr rrnrrnnrnnr *79 $rrnrr< n0*7; +0rrr*792 $/*784$1&/*732 %rrnrr 38$2$1& *rrr*78 rnrrrrr rr %rr nrrnnrnnr<0r *78 Figure 3.3. Subject enrollment, consent, treatment retention, and f ollow-up.

PAGE 73

58 Research Question and Specific Aims The current study addressed the following research question: What is the relative contribution of self-efficacy, social support, perc eived physical health and perceived mental health to pain and disability in individuals with L SS? Three Specific Aims guided the investigation of this overarching research question Aim 1: To determine the relative contribution of se lf-efficacy, social support, self-rated physical health, and self-rated mental health to th e severity of pain and disability among patients with LSS upon entry to treatment of LSS. H1A: Lower levels of self-efficacy, social support, perc eived physical health and perceived mental health will each independently pre dict higher disability at entry to treatment. H1B : Lower levels of self-efficacy, social support, pe rceived physical health and perceived mental health will each independently pre dict higher pain at entry to treatment. H1C: Lower levels of self-efficacy, social support, per ceived physical health and perceived mental health will together predict highe r disability at entry to treatment. H1D: Lower levels of self-efficacy, social support, per ceived physical health and perceived mental health will together predict highe r pain at entry to treatment. Aim 2: To determine the relative contribution of s elf-efficacy, social support, self-rated physical health, and self-rated mental health to th e outcomes of pain and disability among patients with LSS in the year following treatment o f LSS. H2A: Higher levels of self-efficacy, social support, per ceived physical health and perceived mental health will each independently pre dict lower disability after treatment of LSS.

PAGE 74

59 H2B: Higher levels of self-efficacy, social support, pe rceived physical health and perceived mental health will each independently pre dict lower pain after treatment of LSS. H2C: Higher levels of self-efficacy, social support, pe rceived physical health and perceived mental health will together predict lower disability after treatment of LSS. H2D : Higher levels of self-efficacy, social support, p erceived physical health and perceived mental health will together predict lower pain after treatment of LSS. Aim 3: Explore how self-efficacy, social support, p erceived physical health and perceived mental health contribute to a patient’s lived exper ience with LSS and influence meanings of pain and disability. Measures and Constructs Measuring pain: Numeric Pain Rating Scale (NPRS) The Numeric Pain Rating Scale (NPRS) asked subjects to rate their pain on an 11-p oint scale. The anchors were 0 and 10, with 0 representing “No Pain” and 10 representing “Worst P ain Imaginable.” Subjects rated their minimum, average, and greatest pain intensity durin g the last 48 hours for their thigh and leg symptoms (NPRStl) and their low back and/or buttock pain (NPRSbb). This method of recording pain has demonstrated only fair reliability.225 Older patients tend to prefer the NPRS and tend to exhibit fewer errors than with the use of the visua l analog scale.226, 227 Whitman et al225 have shown the minimal clinically important difference ( MCID) for the NPRS in a patient population with LSS to be between 1.25 and 1.5 respectively. A dditionally, the body diagram was used to assess the distribution of symptoms by categorizing the location of symptoms as low back, buttock, thigh, and/or leg (distal to knee). Measuring disability: Oswestry Disability Index (OD I) The modified ODI is a region-specific disability scale focusing on the lo wer back and leg regions for patients with LBP.

PAGE 75

60 The ODI consists of 10 items addressing different a spects of function, each scored from 0-5 with higher values representing greater disability. The modified ODI was used as the primary outcome measure of disability. Whitman et al225 found that the modified ODI exhibits excellent tes t-retest reliability in patients with LSS (ICC, 0.84). They also found the modified ODI to be more responsive (AUC = 0.84) than the Spinal Stenosis Sc ale symptom and function subscales (AUC = 0.74 and 0.75 respectively). The authors identified the minimum clinically important difference in a population receiving non-surgical management for LSS as 5% points which means the change in ODI scores must be at least 6 points to be clini cally significant in patients with LSS. Measuring self-efficacy: Low Back Activity Confiden ce Scale (LoBACS) The LoBACS228 assesses confidence in performance of key function al activities and regular exercise in patients with histories of LBP and surgical inte rventions. Self-efficacy literature, LBP literature, and clinicians’ observations were all u sed to develop the LoBACS. Seven items pertain to self-efficacy for functional activity: self-conf idence for carrying, lifting, pushing, sitting, standing, walking, and stair climbing. Three items address self-efficacy for self-regulation of back health, and 5 items relate to confidence for r egular exercise. The 2 scales of the LoBACS have good test-retest reliability with Functional S elf-Efficacy (FSE): ICC, 0.92 and Exercise SelfEfficacy (EXSE): ICC, 0.75. The FSE subscale is als o strongly negatively correlated with disability using the ODI: r = –0.86, p<.0001. The E XSE subscale is moderately correlated with disability using the ODI: r = –0.28, p = .044. Both subscales are correlated to subjective quality of life measures (FSE: r = 0.55, p<.0001; EXSE: r= 0.3 5, p = .010). Measuring social support: Medical Outcomes Survey S ocial Support Scale (MOS SSS) The MOS SSS is a brief, multidimensional, self-repo rt survey developed for patients with chronic conditions in the Medical Outcomes Study.229 This survey was designed to be comprehensive addressing different dimensions of so cial support. Multi-trait scaling analyses support the dimensionality of 4 functional support scales (emotional/informational, tangible, affectionate, and positive social interaction) and the construction of an overall functional social

PAGE 76

61 support index. These support measures are distinct from structural measures of social support and from other related health measures. The support mea sures are reliable (alpha >0.91), fairly stable over time, and illustrate construct validity. Measuring perceived health status: Medical Outcomes Survey Short Form 36 (SF36) General health surveys have been designed to meas ure overall self-rated health with a broad range of questions covering a variety of aspects of physical and mental health. The Medical Outcomes Survey SF-36 is one such multi-purpose, ge neral health survey which yields psychometrically-based physical and mental health s ummary measures. In multiple studies, the SF-36 has been used for general and specific popula tions evaluating health outcomes, comparing the relative burden of diseases, and differentiatin g the perceived health benefits produced by a wide range of different treatments. The content val idity of the SF-36 has been compared to that of other widely used generic health surveys.230, 231 The SF-36 was judged to be the most widely evaluated generic patient assessed health outcome m easure in a study published in the British Medical Journal .232 In addition, t he SF-36 is widely used in studies of patients with LBP and has well established psychometric properties.73, 233-235 The SF-36 is 36-item, generic self-report measure o f perceived health status covering 8 domains and 2 subscales. The 8 domains include phys ical functioning, role limitation as a result of physical health problems, role limitations as a result of emotional health problems, energy/fatigue, emotional well-being, social functi oning, pain and general health perceptions (see Figure 3.4).73, 236 Each subscale and domain is scored from 0 (poor he alth) to 100 (optimal health). The 8 domains are designed to capture the 2 major components of health, physical and mental, contained in the 2 subscales.

PAGE 77

62 Figure 3.4. Summary of SF-36 subscales and domains. Adapted from Ware237 The 2 subscales provide a more concise measure of o verall physical and mental health and are called the Physical Component Summary (PCS) and Mental Component Summary (MCS) respectively. These summary scales use the it ems on the SF-36 with the PCS heavily weighting physical measures and the MCS heavily wei ghting mental health measures. Reliability coefficients for the physical and mental summary sc ores exceed 0.90 in previous studies.237 The PCS and MCS are adjusted by the population mean and standard deviation to produce normbased scores with a common mean of 50 and standard deviation of 10. Thus, any score below 50 represents a decrement from “normal” health and fun ctioning. The physical health subscale or Physical Component Summary (PCS) comprises half of the SF-36 items which include domains of perceived physical function, role-physical, bodily pain, )9: nr nnr n ) +rn ?rr"r rr nnr r"r +r ) (

PAGE 78

63 and general health (see Figure 3.4). It is importan t to note that the PCS does not actually measure physical performance, ability or function. Rather i t is a tool to indicate the patientÂ’s perception of physical function, physical role in his/her life, h is/her interpretation of body pain, and his/her overall indication of general health. The mental health subscale or Mental Component Summ ary (MCS) comprises the other half of the SF-36 items which include domains of vi tality or energy, social functioning, roleemotional, and mental health (see Figure 3.4). The mental health subscale in its name is misleading as it represents dimensions of intrapers onal energy and emotional well-being but also interpersonal interaction and social and emotional function. Again, it is important to note that this scale represents the patientÂ’s perceived levels of energy, emotional roles, social function, and mental health. Moreover, it is important to note th at the mental health subscale is does not represent a tool for psychiatric diagnosis of a men tal health condition. The mental health subscale has been used to screen for psychiatric disorders s uch as depression,238 but is not as such a diagnostic tool for any mental disorder. Data Collection Aims 1 and 2: Quantitative Quantitative data for Aims 1 and 2 were collected on 34 subjects upon entry into treatment and 20 of these 34 subjects who completed post-treatment measurements. Baseline and follow-up measures included demographics and inform ation related to pain, disability, perceived physical health, perceived mental health, self-effi cacy for low back related tasks, and social support. A summary of the quantitative data collect ion is below (Table 3.2). Copies of the surveys are located in Appendix D.

PAGE 79

64 Table 3.2. Data Collection Summary Summary of quantitative data collection at entry to treatment and post-treatment during long-term follo w up period of 6-12 months after entry to study. Entry to Treatment Post-Treatment Demographics X NPRS X X ODI X X SF-36 X X LoBACS X X MOS SSS X X Data Analysis Aims 1 and 2: Quantitative Prior to addressing specific Aims 1 and 2, the data were examined using descriptive statistics to describe demographics of the sample u sing grouped frequency distributions and measures of central tendency and variability at ent ry to treatment collected from the physical exam and the baseline survey instruments (n=34). Al l survey data at entry to treatment and posttreatment were converted to index scores and scored according to the literature for each measurement tool. For each survey index score, para metric analysis was most appropriate in meeting statistical assumptions for linearity, norm ally distributed errors, and uncorrelated errors. Parametric data analysis has been recommended in th e literature as preferable as long as assumptions of linearity are met since parametric d ata analysis is more robust with small sample sizes when compared to non-parametric analysis.239 Paired t-tests were performed to determine if there were pre to post differences between demographics, survey measures of the dependent vari ables of pain and disability, and independent variables of self-efficacy, social supp ort, perceived physical health, perceived mental health, and all the subscales of both self-efficacy and social support during the study period. Next, associations between the independent variables and dependent variables using PearsonÂ’s correlation coefficient were identified. Correlatio n analysis was conducted between the independent variables of self-efficacy, social supp ort, perceived physical health, perceived mental health, and subscales of both self-efficacy and soc ial support in association with the dependent

PAGE 80

65 variables of pain and disability at entry to treatm ent and post-treatment. Correlations between predictors of greater than .70 can lead to multicol linearity while correlations below .20 between predictor and outcomes may not support the assumpti on of a linear relationship.239 Therefore, the assumptions of linearity, normally distributed erro rs, and uncorrelated errors were checked and met before further analysis. In addition, tests for mulitcollinearity were performed and checked during multiple regression analysis procedures and reported. Bivariate regression analysis Next, a simple or bivariate linear regression wa s conducted to investigate whether the individual var iables of self-efficacy, social support, perceived physical health, and perceived mental hea lth independently predicted the dependent variables of pain and disability in patients with L SS. These preliminary analyses were used to determine the degree of individual association betw een the independent variables of self-efficacy, social support and general measures of perceived ph ysical and mental health on the dependent variables of pain and disability in patients with L SS. Because these variables have largely been unexplored in the LSS literature, understanding the individual contribution or prediction of the criterion variables of pain and disability was an i mportant first step. Multiple regression analysis It was estimated that each of the independent va riables of self-efficacy, social support, perceived physical h ealth, and/or perceived mental health status may not individually represent a strong association wit h pain and disability but collectively they could contribute to a proportion of variability in the de pendent variables. Because the independent variables have not been explored together in the cu rrent literature for LSS, a further step of using a multivariate analysis procedure of multiple regre ssion further tested these relationships in the clinical phenomena of pain and disability in patien ts with LSS. Multiple regression is supported for analysis of theoretical components of construct s particularly when the purpose is to better understand the factors associated with it.239 The theoretical constructs of self-efficacy measur ed by the LoBACS self-efficacy scale and social suppor t measured by the MOS SSS social support scale along with perceived physical and mental heal th status measured by the SF-36 physical

PAGE 81

66 health subscale (PCS) and mental health subscale (M CS) were all modeled a priori to determine what combination, if any, of these constructs contr ibuted to pain and disability at entry to treatment and post-treatment in patients with LSS. The literature states 2 main purposes of the modeling approach in multiple regression analysis: (1) to achieve best single predictor model, or (2) to achieve an explanatory approach for developi ng new insights, exploring potentially causal/explanatory relationships and designing futu re research.240 Statistical literature has stated that the second approach to achieve explanation mus t necessarily simultaneously consider all possible models.240, 241 The second approach to developing new insights and exploring causal relationships underlies the multiple regression met hods in this study. The simultaneous multiple regression method or “ent er” selection procedure for SPSS version 19 (IBM SPSS Inc., Chicago IL) using underl ying theory to enter all independent variables in the model was chosen to test how the c ombination of self-efficacy, social support, and perceived physical and mental health all contri bute to the understanding of pain and disability. This procedure tests all of the variabl es in a model at both time points of data collection to determine their group association to the dependent variables of pain and disability. A final step was taken if statistical significance wa s not found or if multicollinearity was found in the group model. This final step tested the most pa rsimonious combination of all variables using a “backwards” procedure in which the SPSS analysis pr ogram searches for the best fit with the largest R2 and most significant F value. A key methodological point is that all models were guided a priori by the existing theoretical framework of self-efficacy and social s upport in a novel approach to test the contribution of these factors in combination with t he measures of general perceived physical health and perceived mental health in the modeling process. Because there is no literature comparing these variables in a similar sample of pa tients with LSS, there is no prior indication about which variables will create the best predicti on equation. In addition, since there is a small set of predictors being tested, simultaneous regres sion is the best method to use to test the group

PAGE 82

67 of predictors.242 The models for Aims 1 and 2 include the combinatio ns of self-efficacy, social support, perceived physical health, and perceived m ental health to achieve the most clinically relevant and parsimonious combination predicting th e dependent variables of pain and disability. Only significant model combinations are reported. Since this is a relatively small sample with only 3 4 observed cases at entry to treatment and only 20 of the original 34 observed cases follo wed post-treatment, testing the sample with all 4 independent variables reached the minimum recomme ndations of observations to variables (5:1) cited in the literature for multiple regressi on analysis.239 An a priori alpha level of p= 0.05 was set to determine significance. The adjusted R2 value, reflecting the goodness of fit of the linear model adjusted for the number of independent variables in the equation was calculated for each model. The significance using the F statistic for each model was determined. Standardized beta coefficients for each variable in the model we re calculated and the significance of each was determined under the null hypothesis that the coeff icient was not different from zero. Sample size and power analysis. It was anticipated that entering the independent variables self-efficacy, social support, perceived physical health, and perceived mental health into the multivariate models would result in a moderate effect size (0.30).243 Using an a priori sample size calculator for multiple regression considering 4 independent variables, an alpha level equal to 0.05, and a desired power of 80%, a sample size of 42 subjects would be required. Therefore, 34 subjects at entry and 20 at follow-up was determine d unlikely to provide adequate power and therefore increased the risk of a type II error on detection of the true impact of the independent variables in their contribution to pain and disabil ity. However with a large effect size (0.50) as reported by Cohen,243 a sample size of 27 subjects would be adequate fro m the 34 observations at entry to treatment but still less than what was ava ilable in the follow-up data set of 20 observations. Caution is warranted in interpretatio n due to the potential of committing a type II error with the small sample size. Therefore, a post hoc analysis was run on all models to help identify the magnitude of the observed data. Becaus e of the minimal evidence in the literature

PAGE 83

68 regarding the contribution of self-efficacy, social support, and self-rated health in understanding the severity and outcomes of pain and disability in LSS, this study provided a foundation for future research and provided direction to further e xplore these variables with a larger data set. Data Collection Aim 3: Qualitative The focus of the interview was to explore the meani ngs of pain and disability in a sample of patients with LSS. This focus guided the open en ded questions to probe the details and depth of identified factors of self-efficacy and social s upport using the theoretical constructs discussed in Chapter 2. In addition, a broader sense of patie nt identified factors such as general health perception was explored. Data collected in Aim 3 co nsisted of patient interviews and field notes that were recorded at post-treatment, the time peri od of 6 to12 month from entry into the study. The quantitative data from Aims 1 and 2 were theref ore supplemented by patient descriptions of physical, behavioral, and psychosocial factors spec ifically self-efficacy, social support, and selfreported health, as they related to perception of p ain and disability upon entry into treatment and post-treatment. Patient narratives were transcribed and coded by the researcher to characterize the patientÂ’s perceived status at entry to treatment an d post-treatment related to pain and disability. The interview explored the lived experience with LS S. Specifically, questions explored the patientÂ’s self-confidence for back related tasks as well as patientÂ’s social support sources and if these support sources were perceived as helpful or non-helpful during treatment and posttreatment. The interview guide is included in Appen dix E. Data analysis Aim 3: Qualitative and Combined Metho ds Identification, confirmation, or contradiction of f actors related to self-efficacy, social support, and general health status and their contri bution to the understanding of pain and disability were identified through analysis of qual itative interviews (n=20). Interviews were conducted using an iterative process in which the r esearcher continued to interview subjects until

PAGE 84

69 no new themes or factors related to the experience of living with LSS, experience of treatment, or the concepts of self-efficacy, social support, perc eived physical health, perceived mental health, pain, and disability were revealed. Interview transcripts were analyzed to identify any factors, including self-efficacy and social support that contributed to the patient’s pe rceived pain and disability upon entry into treatment and post-treatment. Verbatim transcripts from interviews were imported into the qualitative software package Atlas.ti v6 (GmbH; Ber lin, Germany), and preliminary themes were identified and coded. In the first phase, the focus was on discovery and comprehension,244 whereby segments of data in a quote, sentence, or p aragraph were coded with words or phrases that identified a particular theoretical constructs related to general health status, self-efficacy, social support, pain, and/or disability. These code s were complemented by ‘de novo’ coding which represents new information.245, 246 These initial codes reflected quotes using the informants’ original words, and captured the essent ial meaning of a concept, idea, or description.246 Each transcript was read for code specific phrases until no new codes were identified in the data. While the coding process began with the deductive p rocess of breaking the narrative down into small coded phrases, it ended with the in ductive process of recognizing thematic commonalities and generating observations which des cribed typical behavior patterns. Major concepts or domains in the data were used that link ed back to groups of statements or quotes. Domains related to pain and disability were compare d and contrasted and relationships between them were identified and recorded on the network di agram of Atlas.ti v6. This process focused on synthesizing the data to define and clarify the ess ential concepts related to pain and disability revealed by the informants. In the final stage of data analysis, theories were incorporated that built working explanations about the patterns observed in the dat a. The Stress and Coping Model247 was used as the underlying theory for understanding the apprais als of the stressors of pain and disability and

PAGE 85

70 coping processes and resources identified by patien ts with LSS. Additionally, the Stress-Diathesis Model125 representing a biopsychosocial approach to pain wa s also used to identify the experience of pain. The WHO ICF framework248 was used as the overarching theoretical framework of disability. Finally, the Social Cognitive Theory112, 153 and the Stress and Coping Model were used to interpret data pertaining to the constructs of s elf-efficacy, social support, general perceived health status, and any other uncovered coping resou rces. A diagram that outlines the association between beliefs and behaviors related to pain and d isability was developed in the final combined methods procedures to illustrate if and how factors of self-efficacy, social support, general health perception as well as other uncovered factors contr ibute to the experience of pain and disability in patients with LSS.

PAGE 86

71CHAPTER IV RESULTS Descriptive Results Complete data sets on 34 patients upon entry to tre atment and the remaining 20 patients post-treatment are represented below in the demogra phic analysis of the subset of patients with LSS. Demographic characteristics were reported with mean and standard deviations for interval data and with frequencies for categorical data in T able 4.1. The average age of the patients was 68 years at entry to treatment with only a slight i ncrease to 71 years for the 20 patients remaining at follow up. These ages represent the typical age group found in patients with LSS. There were no statistical differences between the entry to tre atment sample of 34 patients and the remaining 20 patients at post-treatment except for prior hist ory of leg pain where 76% in the entry to treatment sample had a history of leg pain but 90% of the remaining 20 subjects had previous leg pain. Across the study sample, approximately 55% of the patients were male and 45% were female at the 2 time points. The race distribution in the sample was predominately white (70%) with over half of the sample married with a spouse (55%). Only a quarter of the patients currently worked in some capacity with the approximately half of patients retired, although 15% of the patients were permanently unable to work due to wor k or disease related medical disability status. All patients in the sample came to treatment compla ining of LBP with approximately 95% complaining of pain into buttock and leg(s). Ne arly all patients rated their pain best when sitting (95%), however a few reported sitting cause d increased pain which conflicts with the clinical diagnostic criteria although it is possibl e depending on acuity and severity that they had pain in any position.249 The average pain duration upon entry to the study for this current episode of LBP was 12 months prior to entry, with most pati ents reporting a prior history of LBP (90%).

PAGE 87

72 Table 4.1. Demographic Characteristics of the Sampl e of Patients with LSS Demographics Entry to Rehab (n=34) Post-Rehab (n=20) Sig. Age 68.382 (8.714) 70.737 (8.881) 0.193 Gender male 18 (53%) 11 (55%) 0.921 Gender female 16 (47%) 9 (45%) 0.921 Race White 24 (73%) 14 (70%) 0.304 Race Black 5 (15%) 4 (20%) 0.304 Race – Hispanic 1 (3%) 1 (5%) 0.304 Race – Asian 2 (6%) 0 (0%) 0.304 Race – Pacific Islander 1 (3%) 0 (0%) 0.304 Resting HR (bpm) 78.409 (17.220) 77.588 (16.915) 0.873 Resting BP -SBP (mmHg) 129.462 (21.424) 128.850 (23.623) 0.821 Resting BP DBP (mmHg) 78.360 (11.665) 79.105 (12.679) 0.271 BMI 29.615 (4.880) 29.977 (5.152) 0.741 Single 2 (6%) 1 (5%) 0.441 Married 21 (62%) 11 (55%) 0.441 Divorced 4 (12%) 2 (10%) 0.441 Widowed 7 (21%) 6 (30%) 0.441 Smoke now 3 (9%) 1 (5%) 0.340 Smoked over 100 cigarettes in lifetime 17 (50%) 12 (60%) 0.186 Low back pain 34 (100%) 20 (100%) 1.000 Pain beyond buttock 33 (97%) 19 (95%) 0.227 Current episode LBP (months) 14.348 (19.905) 12.467 (9.372) 0.227 Prior history LBP 29 (85%) 19 (95%) 0.080 Prior history leg pain and numbness 26 (76%) 18 (90%) 0.010 Physical activity <3 days/week 16 (47%) 9 (45%) 0.829 Physical activity 3-4 days/week 8 (24%) 5 (25%) 0.829 Physical activity 5 days/week 5 (15%) 4 (20%) 0.829 No physical activity 3 (9%) 2 (10%) 0.829 Education – high school 11 (32%) 5 (25%) 0.450 Education – some college 9 (26%) 6 (30%) 0.450 Education – graduated college 5 (15%) 3 (15%) 0.450 Education – some post graduate 2 (6%) 2 (10%) 0.450 Education – completed post graduate 7 (21%) 4 (20%) 0.450 Work – full time 6 (18%) 4 (20%) 0.386 Work – part time 1 (3%) 1 (5%) 0.386 Work – permanently unable 5 (15%) 3 (15%) 0.386 Work – retired 16 (47%) 10 (50%) 0.386 Work – Homemaker 5 (15%) 2 (10%) 0.386 *p<.05, **p<.01, ***p<.001

PAGE 88

73 Dependent Variables Rating of pain. Patients rated their pain at entry to treatment at varying levels of intensity. Mean score on the NPRS for the sample at entry to treatment was 5.27 (2.57) which represent a moderate level of pain (4-6/10). At ent ry to treatment, 46% of patients rated their current pain in the high category, 29% rated their current pain in the moderate category, 23% rated their current pain in the low category, and 3 % or 1 patient rated he had no pain (see Table 4.2). Irrespective of treatment condition, pain sco res decreased over the study period with mean pain score for the sample at entry to treatment 5.2 7 (2.57) and mean post-treatment score 4.14 (2.52) indicating reduced pain of 1.13 points posttreatment although this was not significant (see Table 4.4). Post-treatment, only 25% of patients ra ted their current pain in the high category (2 in ESI group, 3 in ESI+PT group), 25% in the moderate category (4 in ESI group, 1 in ESI+PT group), 40% in the low category (4 in ESI group, 4 in ESI+PT group), and 10% with no pain (2 in ESI group, 0 in ESI+PT group) (see Table 4.2). Table 4.2. Pain Rating in Patients with Lumbar Spin al Stenosis Pre and Post-Treatment Pain Rating No Pain % Low (1-3) % Moderate (4-6) % High (7-10) % Entry to Treatment (n=34) 1 3% 8 24% 10 29% 15 44% Post-Treatment (n=20) 2 10% 8 40% 5 25% 5 25% Rating of disability Patients at entry to treatment all reported high er levels of disability as measured by the ODI. The mean disability score f or the patients at entry to treatment was 40.34 out of 100 (13.89) which indicate borderline high disability (see Table 4.4). Patients reported their overall disability at entry to treat ment as 61% in the moderate disability category, 26% in the severe disability category, and 12% in t he crippled category indicating they were extremely limited in all daily activities (see Tabl e 4.3 and Figure 4.1). Irrespective of treatment condition, mean disability scores significantly dec reased post-treatment indicating improved function and reduced disability from entry to treat ment 40.34 (13.89) to post treatment 29.29

PAGE 89

74 (17.06)(see Table 4.4). Post-treatment, 25% of the patient reported disability in the low category (3 in ESI group, 2 in ESI+PT group), 40% in the mod erate category (5 in ESI group, 3 in ESI+PT group), 25% in the severe category (3 in ESI group, 2 in ESI+PT group), and 10% in the crippled category (0 in ESI group, 2 in ESI+PT group)(see Ta ble 4.3). Table 4.3. Disability Rating in Patients with Lumba r Spinal Stenosis Pre and PostTreatment Disability Rating Low (1-19) % Moderate (20-40) % Severe (41-60) % Crippled (61+) % Entry to Treatment (n=34) 0 0% 21 61% 9 26% 4 12% Post-Treatment (n=20) 5 25% 8 40% 5 25% 2 10% Figure 4.1. Oswestry Disability Index score at entr y to treatment in patients with LSS. The majority of patients fall into the moderate disabil ity category with severe disability next highest and crippling disability third highest category. No patients fell into minimal or no disability categories. Mean scores for both disability and pain positively improved from entry to treatment to post-treatment however only the change in disabilit y from entry to post-treatment was significant (p=.04) (see Table 4.4). In addition, a higher pain rating at entry to treatment was significantly correlated with a higher total disability score as indicated by the Oswestry Disability Index :3@ 8:@ 38@ nnrn 484/n 84;4/rrn ;3:4/rrrn :324/r 79;

PAGE 90

75 =.444 (p=.016)(n=34). After treatment, patients wit h a lower pain rating were significantly correlated with having reduced disability scores =.610 (p=.006)(n=20). Table 4.4. Means, Standard Deviations, and Paired T-test for D isability and Pain Variable M SD Sig. Disability Entry to Treatment (n=34) 40.338 13.891 Post-Treatment (n=20) 29.286 17.056 0.040* Pain Entry to Treatment (n=34) 5.265 2.574 Post-Treatment (n=20) 4.143 2.516 0.18 *p<.05, **p<.01, ***p<.001 Table 4.5 summarizes the correlation analysis resul ts for disability. The independent variables of self-efficacy =-.319 (p<.05), social support =-.343 (p<.05), physical health status =-.610 (p<.001), and mental health status =-.444 (p<.01) are negatively and significantly correlated with disability upon entry to treatment, indicating that those with lower self-efficacy, lower social support, worse perceived physical heal th, and worse perceived mental health tend to have higher disability scores. Only the variables of self-efficacy and social supp ort were significantly correlated with disability post-treatment.243 Patients who increased or improved their self-effi cacy scores were significantly more likely to reduce their disabilit y scores =-.620 (p<.01). However, social support had a significant and positive correlation with disability =.389 (p<.001), indicating that those with reduced social support had an overall reduction in disabil ity post-treatment. Table 4.5. Correlation Analysis Results for Disabil ity Variable Entry to Treatment Post-Treatment Disability Self-Efficacy -0.319* -0.620** Social Support -0.343* 0.389** Perceived Physical Health -0.610*** -0.886 Perceived Mental Health -0.444** -0.085 *p<.05, **p<.01, ***p<.001

PAGE 91

76 Table 4.6 summarizes the correlation analyses for p ain. Only the independent variable of perceived physical health had a significant negativ e association with pain upon entry to treatment =-.550 (p<.05), indicating that those with higher p ain scores upon entry to treatment had a worse perceived physical health status. The functional se lf-efficacy subscale =-.332 (p<.05) and perceived physical health =-.631 (p<.01) were both significantly and inversel y correlated with pain post-treatment. This would indicate that a pat ient who had higher functional self-efficacy and better perceived physical health long-term tend ed to have lower pain scores. Table 4.6. Correlation Analysis Results for Pain Variable Entry to Treatment Post-Treatment Pain Self-efficacy subscales Functional Self-efficacy -.420 -0.332* Self-Regulation Self-efficacy -.256 -0.082 Exercise Self-efficacy -.190 0.014 Social Support subscales Emotional Support -.205 -0.004 Tangible Support -.411 -0.0253 Affectionate Support -0.089 0.214 Positive Social Interaction -.179 0.007 SF-36 subscales Perceived Physical Health -.550* -0.631** Perceived Mental Health .038 0.012 *p<.05, **p<.01, ***p<.001 Independent Variables Upon entry to treatment, 16 of the 20 subjects who were followed post-treatment had moderate to high levels of self-efficacy. Four pati ents had low levels of self-efficacy, below 50 out of 100. Of those 4, 2 patients continued to hav e low self-efficacy but 2 increased their selfefficacy above 50. Additionally, 2 patients reduced their self-efficacy over the study period (Table 4.7). Regarding social support, 15 of the 20 subjects had moderate to high levels of perceived support. Five patients had low perceived support entering the study. Over the study period, 4 of the 5 patients with low social support increased their perceived support scores over 50%. However 1 patient continued to report low soci al support at follow-up. In addition, 1 patient

PAGE 92

77 reduced her perceived support scores below 50 out o f 100 during the follow up period (Table 4.7). Table 4.7. Levels of Self-Efficacy and Social Suppo rt At Entry and Post-Treatment Entry to Treatment Mod-High >50 Entry to Treatment Low<50 Post-Treatment Mod-High >50 PostTreatment Low <50 Self-efficacy 16 4 16 4* *2 continued low *2 improved *2 reduced Social support 15 5 18 2* *1 continued low *1 improved *1 reduced Tables 4.8 and 4.9 show the means and standard devi ations for the independent variables of self-efficacy, social support, perceived physica l health, perceived mental health, and the subscales of both self-efficacy and social support at the time points of entry to treatment and posttreatment. All of the mean survey scores for the sa mple positively changed or improved from entry to treatment to post-treatment. The only sign ificant differences of score improvement were seen entry to post-treatment between the subscale m easures of functional self-efficacy (p=.043), emotional support (p=.043) and positive social inte raction (p=.040) (see Table 4.8 and 4.9). Table 4.8. Means, Standard Deviations and Paired T-test of Pre dictor Variables of Selfefficacy, Social Support, Physical Health and Menta l Health Variable Mean SD Sig Total Self-Efficacy Entry to Rehab (n=34) 59.519 20.484 Post-Rehab (n=20) 70.393 21.805 0.266 Total Social Support Entry to Rehab (n=34) 72.753 23.356 Post-Rehab (n=20) 81.674 18.376 0.106 Physical Health Status Entry to Rehab (n=34) 33.768 8.228 Post-Rehab (n=20) 41.171 8.903 0.102 Mental Health Status Entry to Rehab (n=34) 47.721 10.948 Post-Rehab (n=20) 52.01 8.702 0.073 *p<.10; **p<.05; ***p<.01

PAGE 93

78 Table 4.9. Means, Standard Deviations, and Paired T-test for P redictor Variables of Subscales of Self-Efficacy and Subscales of Social Support Variable Mean SD Sig. Functional Self-efficacy Entry to Rehab (n=34) 41.899 27.383 Post-Rehab (n=20) 53.698 32.21 0.043* Self Regulation Self-efficacy Entry to Rehab (n=34) 66.667 27.377 Post-Rehab (n=20) 76.481 23.556 0.639 Exercise Self-efficacy Entry to Rehab (n=34) 76.059 20.606 Post-Rehab (n=20) 80.889 18.939 0.612 Emotional Support Entry to Rehab (n=34) 71.232 22.961 Post-Rehab (n=20) 84.375 18.006 0.043* Tangible Support Entry to Rehab (n=34) 71.838 30.048 Post-Rehab (n=20) 82.721 21.025 0.407 Affectionate Support Entry to Rehab (n=34) 79.865 28.52 Post-Rehab (n=20) 90.741 15.095 0.363 Positive Social Interaction Entry to Rehab (n=34) 74.755 29.764 Post-Rehab (n=20) 89.815 15.539 0.040* *p<.10; **p<.05; ***p<.01 Aim 1: Entry to Treatment Aim 1: To determine the relative contribution of se lf-efficacy, social support, self-rated physical health, and self-rated mental health to th e severity of pain and disability among patients with LSS upon entry to treatment of LSS. Aim 1: Hypothesis Testing A & B H1A: Lower levels of self-efficacy, social support, perc eived physical health and perceived mental health will each independently pre dict higher disability at entry to treatment. H1B : Lower levels of self-efficacy, social support, pe rceived physical health and perceived mental health will each independently pre dict higher pain at entry to treatment.

PAGE 94

79 Bivariate Regression Analysis: Pain and Disability at Entry to Treatment A perception of worse physical health significantly predicted greater disability at entry to treatment, F(1,32) = 18.935, p<.001, adjusted R2 = .352. Upon entry to treatment, the beta weights presented in Table 4.10 indicate when perce ived physical health scores decrease or get worse by 1 unit, disability scores increase or get worse by 1.209 units. This relationship between perceived physical health and disability had adequa te post-hoc observed power >80% and supports hypothesis H1A, therefore the null hypothe sis can be rejected (see Table 4.10). A perception of worse mental health significantly p redicted greater disability at entry to treatment, F(1,32) = 7.837, p<.01, adjusted R2 = .197. The beta weights presented in Table 4.10 indicate when perceived mental health scores decrea se or get worse by 1 unit, disability scores increase or get worse by .563 units. This relations hip supports hypothesis H1A and the post-hoc observed power is 81%; therefore the null hypothesi s can be rejected. Lower social support significantly predicted greate r disability at entry to treatment, F(1,32) = 4.253, p<.05, adjusted R2 = .10. The beta weights, presented in Table 4.10 i ndicate when social support scores decrease by 1 unit, disa bility scores increase or get worse by .204 units. This relationship supports hypothesis H1A. H owever, using a post-hoc observed power analysis, this relationship was underpowered at 30% No other significant associations were found between the individual independent variables of self-efficacy, social support, perceived physical health, and perceived mental health and th e dependent variable of disability upon entry to treatment.

PAGE 95

80 Table 4.10. Bivariate Linear Regression Results for Greater Disability at Entry to Treatment Disability R 2 Adjusted R 2 F B SEB B2 Sig Entry to Treatment Self-efficacy 0.102 0.073 3.617 -0.216 0.114 -0.319 0.066 Social Support 0.117 0.10 4.253 -0.204 0.099 -0.343 0.047* Physical Health 0.372 0.352 18.935 -1.029 0.237 -0.610 0.000 *** Mental Health 0.197 0.172 7.837 -0.563 0.201 -0.444 0.009** *p<.05,**p<.01,***p<.001 Using the self-efficacy subscales, only the subscal e of functional self-efficacy independently and significantly predicted greater p ain at entry to treatment, F(1,32) =4.824, p<.05, adjusted R2 = .104. The beta weights presented in Table 4.11 i llustrate when functional self-efficacy scores decrease or get worse by 1 uni t, pain scores increase or get worse by .034 units. This relationship supports the hypothesis H1 B however using post-hoc observed power analysis this was underpowered at 52%. Table 4.11. Bivariate Linear Regression Results for Higher Pain at Entry to Treatment Pain R2 Adjusted R2 F B SEB B2 Sig Entry to Treatment Functional self-efficacy 0.131 0.104 4.824 -0.034 0 .015 -0.362 0.035* Self-regulation selfefficacy 0.012 -0.019 0.374 -0.01 0.017 -0.108 0.545 Exercise self-efficacy 0.012 -0.019 0.4 -0.014 0.022 -0.111 0.531 Emotional support 0.022 -0.008 0.724 0.017 0.02 0.149 0.401 Tangible support 0.011 -0.02 0.356 -0.009 0.015 -0. 105 0.555 Affectionate support 0.016 -0.015 0.521 -0.011 0.016 -0.127 0.476 Positive Social Interaction 0.000 -0.031 0.004 0.00 1 0.015 0.011 0.942 Physical Health 0.070 0.041 2.415 -0.083 0.053 -0.265 0.130 Mental Health 0.007 -0.024 0.232 0.02 0.041 0.085 0 .633 *p<.05,**p<.01,***p<.001 Aim 1: Hypothesis Testing C & D H1C: Lower levels of self-efficacy, social support, per ceived physical health and perceived mental health will together predict highe r disability at entry to treatment.

PAGE 96

81 H1D: Lower levels of self-efficacy, social support, per ceived physical health and perceived mental health will together predict highe r pain at entry to treatment. Multiple Regression Analysis: Disability and Pain a t Entry to Treatment With self-efficacy, social support, perceived phys ical health and perceived mental health in the model, the combinations of all 4 variables s ignificantly predicted higher disability at entry to treatment, F (4, 29) = 6.920, p<.001, R2=.488, with only perceived physical and mental heal th significantly contributing to the prediction. The a djusted R2 value was .418 indicating that 42% of the variance in disability at entry to treatment wa s explained by the model. The beta weights, presented in Table 4.12, suggest that worse perceiv ed physical health and worse perceived mental health were the only significant contributors to pr edicting greater disability at entry to treatment. Post-hoc observed power analysis revealed that this relationship was adequately powered at 99% and there was no multicollinearity in the data. Thi s relationship supports hypothesis H1C that a worse perception of physical health and a worse per ception of mental health predict greater disability at entry to treatment. Table 4.12. Model 1: All Variables Contribute to Gr eater Disability at Entry to Treatment B SEB B2 Self-Efficacy -0.115 0.102 -0.17 Social Support 0.009 0.097 0.015 Physical Health -0.829 0.242 -0.491* Mental Health -0.399 0.188 -0.315* Constant 93.636 10.617 Note. R 2 =.488; F(4,29) = 6.920, p<.001 *p<.05, **p<.01, ***p<.001 With 4 variables of self-efficacy, social support, perceived physical health and perceived mental health in the model, the only significant mo del using subscales of self-efficacy in predicting greater pain at entry to treatment inclu ded functional self-efficacy, total social support, perceived physical health and perceived mental heal th, F (4, 29) = 2.717, p<.05, R2=.273, with

PAGE 97

82 only the variable of functional self-efficacy signi ficantly contributing to the prediction. The adjusted R2 value was .172 indicating that 17% of the variance in pain was explained by the model. The beta weights, presented in Table 4.13, s uggest that reduced functional self-efficacy predicted greater pain at entry to treatment suppor ting the hypothesis H1D. Post-hoc observed power analysis revealed that this relationship was underpowered at 76% and there was multicollinearity found in the data between social support and mental health. Table 4.13. Model 2: All Variables Contribute to Hi gher Pain at Entry to Treatment B SEB B2 Functional Self-Efficacy -0.046 0.018 -0.489* Social Support 0.038 0.022 0.341 Physical Health -0.082 0.053 -0.262 Mental Health 0.027 0.041 0.114 Constant 5.995 2.251 Note. R 2 =.273; F(4,29) = 2.717, p<.05 *p<.05, **p<.01, ***p<.001 Using the “backward” entry method, only the 2 varia bles of functional self-efficacy and social support were found to be most parsimonious i n the model, significantly predicted greater pain at entry to treatment, F (2, 31) = 4.080, p<.0 5, R2=.208, with only functional self-efficacy significantly contributing to the prediction. The a djusted R2 value was .157 indicating that 16% of the variance in pain was explained by the model. Po st-hoc observed power analysis revealed that this relationship was under powered at 74% but ther e was no multicollinearity in the data. The beta weights, presented in Table 4.14, suggest that having reduced functional self-efficacy contributes to predicting greater pain at entry to treatment supporting hypothesis H1D. Table 4.14. Model 3: Two Variables Optimally Contri bute to Higher Pain at Entry to Treatment B SEB B2 Functional Self-Efficacy -0.05 0.018 -0.532* Social Support 0.036 0.021 0.326 Constant 4.796 1.344 Note. R 2 =.208; F(2,31) = 4.080, p<.05 *p<.05, **p<.01, ***p<.001

PAGE 98

83 With 4 variables in the model, the only combination of subscales of social support that significantly predict greater pain at entry to trea tment was the model with emotional support, total self-efficacy, perceived physical health, and perce ived mental health, F (4, 29) = 2.739, p<.05, R2=.274, with total self-efficacy significantly contr ibuting to the prediction. The adjusted R2 value was .174 indicating that 17% of the variance in pain was explained by the model. Post-hoc observed power analysis revealed that this relation ship was under powered at 75% but there was no multicollinearity in the data. The beta weights, presented in Table 4.15, suggest that reduced total self-efficacy predicts greater pain at entry to treatment supporting hypothesis H1D. Table 4.15. Model 4: All Variables Contribute to Hi gher Pain at Entry to Treatment B SEB B2 Emotional Support 0.039 0.02 0.351 Self-Efficacy -0.049 0.021 -0.391* Physical Health -0.093 0.054 -0.298 Mental Health 0.024 0.039 0.102 Constant 7.388 2.366 Note. R 2 =.274; F(4,29) = 2.739, p<.05 *p<.05, **p<.01, ***p<.001 Aim 2: Post-Treatment Aim 2: To determine the relative contribution of s elf-efficacy, social support, self-rated physical health, and self-rated mental health to th e outcomes of pain and disability among patients with LSS in the year following treatment o f LSS. Aim 2: Hypothesis Testing A & B H2A: Higher levels of self-efficacy, social support, per ceived physical health and perceived mental health will each independently pre dict lower disability after treatment of LSS.

PAGE 99

84 H2B: Higher levels of self-efficacy, social support, pe rceived physical health and perceived mental health will each independently pre dict lower pain after treatment of LSS. Bivariate Regression Analysis: Pain and Disability Post-Treatment The association between self-efficacy, social suppo rt, perceived physical health, and perceived mental health in relationship to pain and disability was followed up in 20 of the original 34 patients with LSS post-treatment. Highe r total self-efficacy significantly predicted reduced disability post-treatment, F(1,18) =14.942, p<.01, adjusted R2 = .423. The beta weights, presented in Table 4.18 indicate when total self-ef ficacy scores increased by 1 unit, disability scores decreased by .484 units. Using a post-hoc ob served power analysis, this relationship has adequate power at 95%. In addition, this relationsh ip supports hypothesis H2A and the null hypothesis can be rejected for the post-treatment p eriod. Better perceived physical health significantly pred icted reduced disability post-treatment F(1,17) = 55.458, p<.05, adjusted R2 = .765. At post-treatment, when perceived physical health scores improved or increased by 1 unit, disability scores decreased by 1.681 units. This relationship between better perceived physical heal th and reduced disability had adequate posthoc observed power 99% and supports hypothesis H2A, therefore the null hypothesis can be rejected (see Table 4.16). No other significant ass ociations were found between the individual independent variables of total self-efficacy, total social support, perceived physical health and mental health and the dependent variable of disabil ity post-treatment.

PAGE 100

85 Table 4.16. Bivariate Linear Regression Results for Predicting Reduction of Disability PostTreatment Disability R2 Adjusted R2 F B SEB B2 Sig Post-Treatment Self-efficacy 0.454 0.423 14.942 -0.484 0.125 -0.673 0.001** Social Support 0.000 -0.055 0.003 -0.012 0.215 -0.0 14 0.954 Physical Health 0.765 0.752 55.458 -1.681 0.226 -0.875 0.000*** Mental Health 0.003 -0.056 0.049 -0.1 0.452 -0.054 0.828 *p<.05,**p<.01,***p<.001 Only better perceived physical health independently and significantly predicted reduced pain post-treatment, F(1,7) = 6.931, p<.05, adjuste d R2 = .259. The beta weights, presented in Table 4.17 indicate when perceived physical health scores improve or increased by 1 unit, pain scores decreased by .157 units. This relationship supports hypothesis H2B that better perceived physical health can predict a reduction in pain how ever post-hoc observed power analysis reveals lack of power at 72%. Table 4.17. Bivariate Linear Regression Results for Predicting Reduced Pain PostTreatment Pain R2 Adjusted R2 F B SEB B2 Sig Post-Treatment Functional self-efficacy 0.177 0.125 3.436 -0.034 0.018 -0.42 0.082 Self-regulation selfefficacy 0.065 0.01 1.189 -0.025 0.023 -0.256 0.291 Exercise self-efficacy 0.036 -0.02 0.639 -0.023 0.028 -0.19 0.435 Emotional support 0.042 -0.022 0.659 -0.027 0.033 -0.205 0.43 Tangible support 0.169 0.117 3.252 -0.04 0.022 -0.411 0.09 Affectionate support 0.008 -0.05 0.137 -0.012 0.032 -0.089 0.716 Positive Social Interaction 0.032 -0.025 0.564 -0.027 0.036 -0.179 0.463 Physical Health 0.302 0.259 6.931 -0.157 0.06 -0.55 0.018* Mental Health 0.001 -0.061 0.023 0.01 0.068 0.038 0 .88 *p<.05,**p<.01,***p<.001 Aim 2: Hypothesis Testing C & D H2C: Higher levels of self-efficacy, social support, pe rceived physical health and perceived mental health will together predict lower disability after treatment of LSS.

PAGE 101

86 H2D : Higher levels of self-efficacy, social support, p erceived physical health and perceived mental health will together predict lower pain after treatment of LSS. Multiple Regression Analysis: Disability and Pain P ost-Treatment At post-treatment with all 4 variables in the mode l, the combinations of all 4 independent variables significantly predicted the reduction in disability, F (4, 14) = 13.268, p<.001, R2=.791, with only perceived physical health significantly c ontributing to the prediction. The adjusted R2 value was .732 indicating that 73% of the variance in disability was explained by the model. The beta weights, presented in Table 4.18, suggest that a perception of better physical health predicts reduced disability supporting hypothesis H2C. Posthoc observed power analysis revealed that this relationship was adequately powered at 99% and there was no multicollinearity in the data. Table 4.18. Model 1: All Variables Contribute to Pr ediction for Reduced Disability PostTreatment B SEB B2 Self-Efficacy -0.059 0.122 -0.078 Social Support 0.165 0.132 0.161 Physical Health -1.534 0.313 -0.798*** Mental Health -0.037 0.238 -0.02 Constant 84.673 17.473 Note. R 2 =.791; F(4,14) = 13.268, p<.001 *p<.05, **p<.01, ***p<.001 With 4 variables in the model, there was no combina tion using subscales of self-efficacy that significantly predicted pain post-treatment. U sing the “backward” entry method, the 2 variables of exercise self-efficacy and perceived p hysical health significantly predicted a reduction in pain post-treatment, F (2,15) = 3.784, p<.05, R2=.335, with perceived physical health significantly contributing to the prediction. The a djusted R2 value was .247 indicating that 25% of the variance in pain post-treatment was explained b y the model. Post-hoc observed power analysis revealed that this relationship was under powered at 74% but there was no multicollinearity in the data. The beta weights, pr esented in Table 4.19, suggest that better

PAGE 102

87 perceived physical health contributes to predicting reduced pain post-treatment supporting hypothesis H2D. Table 4.19. Model 2: Two Variables Have Clinical Re levance in Contributing to Prediction for Reduced Pain Post-Treatment B SEB B2 Exercise Self-Efficacy 0.025 0.03 0.19 Physical Health -0.173 0.063 -0.605* Constant 8.764 2.973 Note. R 2 =.335; F(2,15) = 3.784, p<.05 *p<.05, **p<.01, ***p<.001 With 4 variables in the model, there was no combina tion using subscales of social support that significantly predicted a reduction of pain post-treatment. Using the “backward” entry method, the 2 variables of emotional support and perceived physical health significantly predicted a reduction of pain post-treatment, F (2, 13) = 4.485, p<.05, R2=.408, with perceived physical health significantly contributing to the p rediction. The adjusted R2 value was .317 indicating that 32% of the variance in pain was exp lained by the model. Post-hoc observed power analysis revealed that this relationship was adequa tely powered at 87% and there was no multicollinearity in the data. The beta weights, pr esented in Table 4.20, suggest that better perceived physical health contributes to predicting a reduction of pain post-treatment which supports the hypothesis H2D. Table 4.20. Model 3: Two Variables Have Clinical Re levance in Contributing to Prediction for Reduced Pain Post-Treatment B SEB B2 Emotional Support -0.015 0.032 -0.102 Physical Health -0.191 0.064 -0.646* Constant 12.532 3.953 Note. R 2 =.408; F(2,13) = 4.485, p<.05 *p<.05, **p<.01, ***p<.001

PAGE 103

88 Aim 3: Qualitative and Combined Results Aim 3: Explore how self-efficacy, social support, p erceived physical health, and perceived mental health contribute to a patientÂ’s lived exper ience with LSS and influence meanings of pain and disability. The qualitative analysis was designed to complement the quantitative findings and to describe emergent themes that contribute to both th e severity and reduction of pain and disability in order to capture the lived experience of LSS. T he analysis expanded upon and contextualized the a priori factors of self-efficacy, social support, perceive d physical health and perceived mental health in addition to emergent themes uncove red during the interviews. Two theoretical frameworks informed the qualitative analysis. The Stress-Diathesis Model125 in conjunction with the Stress and Coping Model107 were used as an interpretive lens in analyzing the behaviors and perceptions of patients with LSS. Using the Stress and Coping Model, the qualitative interviews revealed multiple coping resources that were either absent upon entry to treatment when the patientsÂ’ pain and disa bility were at the highest or present posttreatment serving as coping resources to ameliorate the stressors of pain and disability. Many themes also complement the Stress-Diathesis Model s uch as feelings of vulnerability, anxiety, and the role of social influences on the patientsÂ’ experiences with LSS. In addition, the interviews uncovered themes that supported the roles of the a priori identified coping resources of selfefficacy, social support, perceived physical health and perceived mental health. Finally, several novel themes emerged during the interviews that rev ealed more about the condition of LSS than previously demonstrated in the literature. Stressors: Pain and Disability At entry to treatment, the quantitative data indica ted that all patients entered the study with some level of moderate to high pain in their l ow backs. The interviews also confirmed the patientsÂ’ experience of pain. In addition, all of t he patients in this study entered with moderate,

PAGE 104

89 severe or crippling levels of disability. Yet, some patients rated their pain and/or disability higher as well as described their experience as more inten se than others and therefore perceived a higher burden related to pain and disability. When patients were asked to describe what it is lik e to have LSS, all interviewees focused first on pain, particularly pain in the low back but also pain into legs with all activity. Patients noted that this pain had been on and off f or at least a year, if not many years. Pain was the primary reason patients stated they sought medi cal care. However, even though all patients had complaints of pain, not all patients had the sa me descriptions of their pain. For some patients, their pain was the direct sensation in th eir back, buttock or leg(s). For some this also included numbness or tingling sensations. For other s, their pain was tied to fatigue and weakness sensations. When probed further, meanings of pain o ften related to frustrations about not being able to perform activities they enjoyed, feelings o f helplessness, and feelings of a loss of control and choices in their daily lives. Across patient interviews, the experience of pain w as not always directly connected with their limitations in functional tasks and activitie s related to disability. When asked, the actual term “disability” did not have meaning for these in dividuals. The term disability was something that they believed related to other people who were really “bad off”: bed bound or incapable of doing anything. However, patients with LSS recogniz ed their limitations and described many frustrations related to their limitations in activi ties, but they did not associate these problems wit h having a disability or being disabled. The intervie ws revealed that the experience of disability and even pain was much more complex for these individua ls than pure physical losses as previously described by literature.39, 142, 250 Entry to Treatment: Greater Severity of Pain and Disability The overarching goal of the qualitative analysis wa s to elicit main factors that persisted to define patients’ experiences of pain and disability as well as identify sources of coping that were

PAGE 105

90 either present or absent at entry to treatment cont ributing to these stressors. Eight themes emerged to help explain why some patients appraised a highe r burden related to pain and disability: physical limitations, lacking confidence in tasks a nd activities, feeling a lack of control, reduced social participation, feelings of vulnerability, po or mental state, frustration about needing support, and financial limitations. Figure 4.2. Physical, mental, social and environmen tal limitations contributing to stressors of pain and disability. Physical limitations The most commonly reported theme in all the inte rviews was limitations in physical abilities. Most of these ph ysical limitations pertained to task specific activities related to the low back. Other physical limitations cited were more general in nature. In all instances, every patient expressed physical lim itations as primary to their experience with %n n,n rrrn ,n rr &n nrn )rr, ,n +rr )n *rr rrrr ,n )rrn (r )rr, rr r ,n ),n

PAGE 106

91 LSS. An example of one womanÂ’s explanation of her p hysical limitations focused on some specific low back related tasks but also included g eneral physical movement as well: I think getting up in the morning is just really ba d, getting moving, and sometimes getting up from a sitting position and then standing up for just a moment, it hurts a lot more. And I think, you know, doing about anything, just walki ng, getting up and the morning and standing from sitting position is the worse. (76-ye ar-old female) Another woman described her physical limitations re lated to recreational activities: Walking for example; running, you cant run. No way Walking, you are limited to a couple of blocks. It seems like when I would walk, it was putting a lot of pressure on my lower buttocks. (67-year-old female) Finally, one woman described her physical limitatio ns in terms of general reduced energy: I think it just kind of all went kind of like toget her with everything else. With the no energy and of course, having no energy, then you do nÂ’t do as much. When you donÂ’t do as much, then your back is at its best [laughs]. (7 6-year-old female) Lacking confidence in tasks and activities The next most commonly reported theme was lacking confidence in tasks and activities. Alt hough all patients expressed physical limitations, for some it was the uncertainty of the ir back pain and inability to respond to the pain and adjust for weakness and poor balance during spe cific activities that concerned them. The a priori identified coping source of self-efficacy was supported by thi s theme of lacking confidence in tasks and activities. Reduced belief that they c ould do specific tasks such as standing doing dishes, stair climbing, walking to the bathroom or to the mail box all were examples used by patients of tasks for which they had lower confiden ce upon entering treatment. These self-doubts illustrated patientsÂ’ reduced self-efficacy for the se low back related functional tasks. Moreover, their reduced self-efficacy expression also include d worry related to negative ramifications after the task performance. Even the simplest tasks of da ily living such as toileting were noted as problematic. One man described his decreased confid ence due to in walking short distances even to accomplish essential activities such as going to the bathroom: I mean I could get up from the chair and go to the bathroom and take care of business there. And then I had to get right back in my chair and then I was in excruciating pain by that time, from being on my feet. So I sat down and the pain kind of floated away, when I

PAGE 107

92 sat down again. It would take a period of time, I d on’t know, whether I would say seconds or minutes, a period of time for the pain t o dissipate from my leg after I sat back down and took the weight off it. So if going to the bathroom is so bad I figured I couldn’t do much of anything. I’d have to figure out how far I was to sitting down no matter what I did. (60-year-old male) In addition, other specific activities most frequen tly noted as reduced or eliminated due to lack of confidence in performing them included walk ing outside of their homes, shopping in large stores, stair usage, general house maintenance both inside and outside of the house, and lifting or carrying any object greater than about 5-10 lbs. Re creational participation tasks that were mentioned as given up were walking for exercise, fi tness classes, golf, and hiking and other outdoor activities. One man described his decreased confidence in participating in recreational activities: Yeah, I don’t do too much. A lot of the activities I like involve lots of walking, like hunting. I was going to try it this year, and it ha s been a year since I tried. I had to stop when my back got bad. But I guess, it has held me b ack. I’m just not sure I can do it and I am worried about making the back worse. Like workin g out: I used to work out a lot. I don’t do a whole lot of that since my back started up. (60-year-old male) Patients’ perceptions of worse physical health cont ributed to their reduction in activities and were echoed in most interviews. Fatigue and fea r of pain often contributed to the lack of confidence in their general physical abilities. In addition, since all of the patients with this condition were 50 years of age or older, age and ag e related declines in physical health interplayed with their overall perception or confid ence with physical abilities. Although some patients expressed that they expected some decrease s in energy and fatigue with aging, LSS made them feel older than their years. Due to these doub ts in general physical function, pain seemed to only add to their concerns when engaging in general activity. As a result, multiple patients expressed fear of not being physically able to do s omething. They were afraid to do something and then end up not being able to finish or get “st uck” somewhere such as a big box store or the mall. One woman described her decreased confidence to do grocery shopping in the following manner:

PAGE 108

93 Maybe I am just getting old, but 2 years ago, I was old and I think it doesnÂ’t happen quite that fast. You slow down for sure but this back pai n makes me feel old. By 2 or 3 in the afternoon, I really don't want to go to the grocery and walk back to the milk aisle. If I am in the produce and if I have to walk back to the mi lk, I donÂ’t think I can make it so I donÂ’t or I just hold my breathÂ… I hold it in. (63-year-ol d female) Lacking control The next theme that patients expressed related t o feelings of anxiety and a lack of control over their body and their pai n experience. Approximately half of the patients with the highest rating of pain and disability repo rted feeling a lack of control in some manner. This feeling of a lack of control illustrated the v ariability in actual pain experience for these patients. For these individuals, pain perceptions m anifested as something beyond just bodily pain and became anxiety expressions about not knowing wh at to expect when performing even the simplest tasks. These feelings of lacking control often combined with patientsÂ’ reduced confidence for performing tasks. This resulted in a general fear of going outside of the home or other safe areas due to the unknown quality of thei r condition. One woman stated: It didnÂ’t matter where I was, it [lower back] might go out anytime and then what? (69 year old female) This woman illustrated that her anguish was not ove r the pain per se but not knowing when and why the pain returned combined with her inability t o control the pain sensations. Her words show the physical nature of this feeling but also the me ntal component and anxiety produced by the lack of control. Similar sentiments were expressed by another female who described the unknown quality in relationship to her everyday activities: Well it is a little bit hard dealing with it. It se ems like it comes and goes, one day I can feel pretty good, and the next day I canÂ’t do anyth ing. It puts a limit on what I can do at work and that sort of thing, and so my house work i s piled up, a lot of scrubbing and dusting. I did that a couple of days ago, now yeste rday and today I am paying for it, I am stiff and sore and achy. (77-year-old female) When asked what her back was like before treatment and why she thought she felt that way, another woman stated: I didnÂ’t have any control. It does what it does. I could keep diary after diary and there would be no correlation in my mind. (63-year-old fe male)

PAGE 109

94 This woman is a good example of the uncertainty tha t comes with LSS. This uncertainty and lack of control created a large amount of anxiety in mos t patients which in turn led to a decrease in their overall participation in all types of activit ies. Anxiety is a key feature in the Stress-Diathes is Model as a factor contributing to pain and disabili ty. Heightened uneasiness and worry about their condition, pain, and inability to do or contr ol the outcome of activities are all potential reasons anxiety can lead from pain to disability in patients with LSS. Another woman described her anxiety about going to the grocery store: I didnÂ’t like going to the store by myself, doing t hings like bending over the steering wheel and then having to get out of the car by myse lf, so I would say ok when you go to the store let me know so I can go with you. I just didnÂ’t want to go to the store by myself. In case something happenedÂ… I just felt like I migh t get stuck. (66-year-old female) She managed her uncertainty by going grocery shoppi ng only if someone was going with her. In this way, she used support to help diminish the anx iety of being stuck somewhere without help or an ability to get herself out of the situation. Reduction in social activities Another related theme driven by several patients Â’ lack of confidence, lack of control, and/or anxiety about p articipating in activities is the resulting reduction of social activities. Social activities w ere seen in many cases as non-essential activity that could be readily eliminated from everyday life Stopping participating in social activities was one way that patients managed their feelings of los s of control and uncertainty. One woman who was in the highest pain and disability categories p rovided a good example of how those who perceived a worse experience made choices for a del iberate reduction in social activities: I don't make plans too many times now. I had to can cel some long stays. I have this German friend, she picks me up and we go out to lun ch every month. I had to cancel 4 or 5 times and she probably got mad at me but now she has pain herself and she knows what it is and she apologized. And I used to go to churc h every Sunday and now I can't go to church. This is, on the pew, it is hard, and when I cannot sit and listen to what is talked about, why do I go to church? So I stay home and wa tch it on TV, or pray by myself. (74year-old female) Social outings marked an added burden for this woma n and she no longer felt that she could participate in these activities with her back condi tion. Another man described similar sentiments:

PAGE 110

95 We used to like to get out more like on a light rai l and go down to the symphony or something like that. We are not doing that anymore. (72-year-old male) Finally, one woman described a reduction in social activities that seem to be connected largely to her change in emotional state. Throughout her inter view she expresses a variety of thoughts such as a loss of interest in social activities that she once enjoyed, feelings of frustration that she was limited in her social outings, and even sadness at the loss of her life as she once knew it. All of these emotions were tied to her decrease in social activities: I guess I am suffering from a lack of social activi ty but it is because I think I am usually peppier. I am sure I am not the same and I think th at shows to the people I am close to. (67-year-old female) Feelings of vulnerability Some patients also expressed mental health issue s related to hopelessness and vulnerability. Vulnerability was a stronger theme than expected across the patients with over 1/3 (7 out of 20) expressing fea rs related to vulnerability and six of these in the highest pain and disability categories. Of the seve n patients expressing concerns about vulnerability, four of these patients communicated a lack of perceived sources of support for many of their daily needs. When talking to the patients, most expressed thei r feelings of vulnerability as helplessness. Vulnerability is a key feature in Str ess-Diathesis Model as a factor contributing to pain and disability. Evidence of feelings of vulner ability was pervasive in all parts of patient’s lives. These feelings rendered them incapable to do even the simplest activities as expressed by one woman: I couldn’t stand up long enough to do anything and then I would just feel so much pain, it just affected my doing every day household work. Go ing to the store was very, very hard ‘specially when I just had to bend over and then I got scared because right now the area where I live in is not the best and I was afraid pe ople would see me like this and beat me up for something that I don’t even have. I was just that scared because I couldn’t stand up straight. I felt totally helpless. (66-year-old fe male)

PAGE 111

96 Many of the patients felt a variety of frustrations leading to reduced confidence about managing their daily lives. These feelings often interacted with perceptions of vulnerability. An example is when one woman echoed the sentiments of helplessnes s, frustration, and reduced confidence as she described how her helplessness put her in a pos ition that she must rely on her family and friends: Yes, I felt totally helpless. And I didnÂ’t like the fact that my girlfriends and my sister had to help me out a lot with things. It was bad, reall y bad. (53-year-old-female) Finally another woman described how her back proble ms left her completely vulnerable and unable to deal with flu related symptoms. She ended up needing her family to support her in multiple ways: At one time in the beginning it was so bad that I f elt that I had a touch of the flu, getting up in and out of the bed was very painful. The doct or had told me to come to the emergency room. I really had the flu so bad that I couldnÂ’t get up. I had diarrhea and vomit and I couldnÂ’t even get up from the bed to go to the bathroom. And my brother, I couldnÂ’t even get down the stairs, so he told me, h e said if you donÂ’t come down the stairs, I am going to kick in your door, so I sat o n the stairs and just kind-of scooted down the stairs and let my brother in so he could take m e to the hospital. I was so dehydrated by the time I did get to the hospital and totally h umiliated because my brother had to clean up all the mess. I never, ever want to go thr ough anything like that again. I felt terrible, the flu was bad and I felt worse because as I said, I just couldnÂ’t get out of the bed. (66-year-old-female) Poor mental state Only a few patients expressed concerns with menta l health and depression related to their back pain and disabilit y, however those few who did were in the highest categories for pain and disability. Stateme nts related to mental health ranged from specific mood changes to more general feelings of p oor mental state. One woman expressed how her LBP resulted in depression type symptoms: You feel depressed, you feel down. You are not happ y about how you feel, you always carry a little burden with you, a little sadness wi th you. Can you cope with it?Yes. Would you rather not? Yes. (63-year-old female) Two patients expressed their poor mental condition by revealing their previous thoughts of suicide when their low back problems were really ba d before they sought treatment. For example,

PAGE 112

97 one woman expressed how poor her mental state was b efore she sought help for her back condition: I donÂ’t know if I should tell you this or not but a fter I had fallen I was in so much pain, and I went to bed and the dog, he was never let on the bed because that was a cat territory. I went to bed at night, I was lying ther e and I was in so much pain, hurting so badly, I was down there and I thought to myself, I canÂ’t live like this, so I planned to get up, take the dog and the cat, get into the garage a nd hope the car didnÂ’t run out of gas before we all expired. But instead the little doggi e hopped on the bed, put his back to my back and I went to sleep and woke up the next morni ng, sometime during the night or so. He was gone of course. But somehow he knew that I w as in a great deal of pain, and he took his life into his hands hopping on the bed, an d he did that and my pain went away that night and after that I realized that wasnÂ’t wh at God intended me to do and I started looking for alternative treatments. Â… It was the on ly time in my life time I ever, ever contemplated suicide. I am 80 years old [in my eigh ties] and that is pretty good I might say. (87-year-old female) Another woman simply described how poor her mental state was before seeing the spine physician: I want to thank the people that really helped me, I really appreciated them, and you have no idea. Prior to going to Dr. A., I just didnÂ’t ca re anymore, I just wanted to die. (66year-old-female) Frustration about needing support One repeated theme that was uncovered was patientsÂ’ reluctance to ask for help from friends, family, neighbors, and even health care providers. Six individuals expressed that they were frustrated by the idea of needing help and were reluctant to ask for help from close family or friends or even health care professionals. Four others seemed to be frustrated that they could not avoid asking for help. Those who had to ask for help from different sources expressed feelings of f rustration tied to their feelings of vulnerability. For those individuals who wanted to avoid asking fo r help, some of their frustration revolved around the anxiety that they might need to ask for help at some point. This anxiety about needing support also confirmed their beliefs about vulnerability due to factors such as their age and their decline in abilities. A few individuals e ven discussed their resistance to giving in stating that they were not ready to be old, helpless, and d ependent.

PAGE 113

98 For those individuals who had experiences asking fo r help, their frustrations revolved around feelings of being a burden for those around them. A few stated they used to be the one who helped others, such as a mother or sibling, and therefore did not like the role reversal. The others stated their support sources did not have th e time or ability to support them but they did not have other options. One woman described how she ended up getting support from her family: I donÂ’t like people cleaning up my house, I do that for myself. I will let people do certain things for me, and that is so silly. My dad used to tell me pride is going to kill me one of these days. But I like to do things for myself. And whatever I could try to do for myself I would do and they would try to help me, but let me do it for myself because if I donÂ’t then I wonÂ’t be able to stay in my house. So they w ould tryÂ… I didnÂ’t want to move any of the furniture around, to help me get up and down the stairs. My sister, she just put her foot down, oh no, we are not doing that. We are goi ng to have to do that. I have to use the walker and I just wanted everything to stay the sam e, but she wouldnÂ’t let me. (66-yearold female) All of these factors led to support being more of a frustration and source of anxiety rather than a positive factor appraised by these patients. Financial limitations A novel theme that emerged related to financial limitations. Several individuals had reduced or no treatment whe n their current episode of LBP started due to financial reasons. For three patients, these financ ial reasons reflected limited or no insurance coverage for LBP treatment. Being underinsured or u ninsured resulted in a decrease in the amount of care or a delay in care for their LSS con dition. It is important to note that because all of the subjects in this study eventually received i nsurance and therefore enrolled in the RCT, the interviews only minimally reflect the issues of the underinsured and uninsured. However, one key individual had absolutely no insurance or financial means when he first began his current episode of LBP before being involved in the study. He descr ibed in his interview how a lack of insurance caused him delay in seeking medical care. He descri bed how he first tried to self-manage his condition for a while: I tried using some over-the-counter medicine, somet hing I got on my own. I wasn't prescribed any medication. I didn't have any insura nce and I couldn't afford to go to the doctor, $275 for a 10 minute visit. I did the best I could. That is the way it is, because I couldn't get medical care, so I didn't have any cho ice about it I just had to keep going along. (57-year-old-male)

PAGE 114

99 Then he described how he finally went into the Emer gency Department once his back pain got so bad that he couldnÂ’t cope: It got to the point to where I had to go to the ER, I couldn't walk. I went at 9:00 in the morning and I didn't get out until 11:30 at night. They started me out with liquid valium, then morphine and then topped it off with dilaudid. When they were finally convinced I wasn't there to do drugs, they moved me over until I ended up in the CIC (Colorado Indigent Care) program and it was very helpful to m e. Then my health got bad in another aspect because I have pulmonary disorders. I had to take social security early because I was at a point where there was nothing left. I had to go on early social security, so I could get the benefits, and start seeing doctors. (57-yea r-old male) In addition to lacking insurance, LSS resulted in f inancial implications related to work and earning an income. Due to their older age, many patients were retired but there were still a few who had job related impacts as a result of LSS. For those who were younger and had no source of retirement income, trying to keep a job a nd health insurance compounded their issues related to their LSS. One woman described how she t ried to keep her job as long as possible despite her LBP: I was able to deal with it for a while. I was able to work but my job required me to stand all day and it got to the point where I wasnÂ’t able to stand so I was losing a lot of time. Because it hurt so much I had to go home, so I was losing a lot of time at work. Because I couldnÂ’t stand I ended up losing my job. (53-year-o ld-female) This woman, similar to the 57-year-old man quoted a bove, not only couldnÂ’t work but also lost her health insurance when she lost her job. Both of these individuals were too young for Medicare, without other resources, and therefore bo th eventually ended up on Social Security disability benefits before entering treatment for t his study. Finally, about half of the patients cited other tan gible sources of support that they felt they needed but lacked due to diminished financial resources. These missing paid support sources included examples such as the inability to hire a d river to go to the grocery store as well as the inability to hire someone to help around the home i n cleaning, maintenance, or yard work. For a few individuals with highest levels of self-reporte d pain and/or disability, having financial

PAGE 115

100 limitations often kept them home bound and decrease d their options for self-management such as with exercise or seeking alternative care. One woma n described how her reduced financial resources resulted in her not participating in any exercise: What my doctor always tells me, walking is the best exercise. Or swimming. I can't afford to go to a place where you go swimming and I donÂ’t drive anymore. I used to swim a lot but now with me not driving anymore and the YMCA or whatever, that is too expensive I donÂ’t have the money to go there. (74-y ear-old female) As an emergent theme, the spectrum of financial lim itations that patients experienced with LSS was complex and contributed to differing l evels of pain and disability in this population. Age and age-related general health decl ine interacted with some of these financial limitations and increased need for help and assista nce. Others who were too young fell into a reverse age-related gap where they were not old eno ugh to get the support of financial resources such as Medicare. Many of the different issues expr essed concerning financial limitations cooccurred in the patients who expressed financial co ncerns as contributing to their health problems (see Figure 4.3). Figure 4.3. Factors related to financial limitation s. Patients with LSS describe sources of financial limitations related to greater pain and d isability pre-treatment. ) ,n $nr rr % rr rr rrn % 'rnr +rr r $nnrn % &n % ,rrn

PAGE 116

101 Post-Treatment: Improving Outcomes of Pain and Disa bility Improved outcomes of pain and disability are the ma in clinical goals of treatment for LSS. These goals for reduced pain and improved func tion were also clearly expressed by all patients during the interviews. However, not all pa tients reduced pain and improved function and therefore had outcomes of reduced disability. Out o f the 20 patients interviewed, 10 patients (8 ESI group, 2 ESI+PT group) had a self-reported redu ction of pain on the Numeric Pain Rating Scale during the course of treatment and the post-t reatment period (see Table 4.2). 13 patients (8 ESI group, 6 ESI+PT group) had a self-reported decr ease in disability on the Oswestry Disability Index during the treatment and post-treatment perio d (see Table 4.3). Interviews explored coping behaviors by these patients who were able to reduce pain and disability. For those who did have a reduction in pain and a de crease in disability, many still had some limitations in daily activities, social activi ties, and recreational activities. Some patients remarked that they were still limited compared to w hat they wanted to do. However, most of the patients noted they had been able to get back to do ing light housework, light yard work, walking in the community, using stairs more frequently, and going out to social events that did not include much strenuous activity and allowed for sitting dur ing the event. Patients used different strategies to balance activ ities with their pain levels. Some believed that to manage pain they had to eliminate all activities. Some believed to manage pain they had to modify how they performed certain activ ities. And others believed that they just needed to push through the pain. Regardless of the strategy used to manage their condition, all expressed that they would manage the repercussions of participation once they were home and could rest, take medications, or use other learned self-management techniques. The overarching goal of the qualitative analysis of the post-treatment period was to determine what factors contributed to diminished st ressors, namely pain and disability during the management period. Multiple themes emerged that pro vided insight into why some patients were able to reduce their pain and disability 6 to 12 mo nths after entering treatment. These themes

PAGE 117

102 included having greater confidence, knowing limitat ions, learning how to manage, having a push through attitude, having support, and having a posi tive outlook (see Figure 4.4). Figure 4.4. Factors related to improved outcomes. Patients with LSS describe six factors that contribute to improved outcomes of lower pain and r educed disability post-treatment. Greater confidence Improving confidence in both specific tasks and gen eral activities was a key theme reported by patients who had improv ement during the study period. Three big areas of improved confidence after treatment were r elated to physical function: walking tolerance and distance, stair tolerance, standing tolerance a nd time spent standing. Improving confidence in these key activities helped many increase their nor mal daily activities but also allowed them to increase their recreational and social activities t o some degree. For some, the return to basic ADLs was very satisfy ing and they did not aspire to do more. These individuals were confident and content with what they could do, though still limited. For others, having the improved performance of basi c ADLs did not satisfy them. For these individuals, some continued to either have goals fo r return to social or sport activity or had frustration about not returning to social or sport activity though they were more confident with many activities. When asked “what was the biggest f actor that contributed to your improvement?”, one man described the moment he reme mbered feeling more confident about doing basic everyday tasks that he could not do pre viously: +rr %n ?rr rr A ,n ,r r n & r " nr #

PAGE 118

103 Probably the biggest factor, that probably after I walked the steps in therapy and that really seemed to have a surprising positive impact on my being able to sit and walk and all that kind of stuff. That would be it. (60-year -old male, ESI+PT) One woman who had progressively increased all of he r activity discussed how she improved. In addition she commented on the goal setting she was doing with recreational activities: I definitely have done more and I monitored it, for the last year now. Just on different things that I would do, what I can do, what is good for me and what is bad for me. And actually, going up and down stairs is great for me. Because that does keep you moving. Walking I find that I am doing better walking too. And hopefully at least by fall I will be able to go back to walking. In the weekend I was do ing 5 and 10 miles, over the weekend before my back got bad. (67-year-old female, ESI+PT ) Knowing limitations Some patients were able to manage their condition by understanding their limitations in performing speci fic tasks and activities. This knowledge of their limits seemed to empower a number of individu als. Their self-determination was improved when they felt a degree of understanding of their l imitations with activities. One man described his abilities and limitations in this manner: Golf is what I can get out and do. I canÂ’t walk aro und a course because my back and my knee and the arthritis in my right foot. But I can play the game, I can ride in the cart, I can do well enough to not be terribly embarrassed. Ther e are people my age that could certainly beat the socks off of me, but the people I play with are roughly my skill level. It is more fun that way. (66-year-old male, ESI+PT) Another man described how he was able to manage out ings out of the home: When I do something I make myself take breaks. Stop ping to sit always helps. IÂ’ve sat on some interesting things at the grocery store. A pil e of bottled water. Once I sat on this box and it turned out to be potato chips and that w asnÂ’t a great idea. So thatÂ’s the way it goes, but we are doing ok. (72-year-old male, ESI+P T) Finally, one man illustrated his knowledge of his l imits by restricting where he goes and what he does: I donÂ’t go shopping any more than I have to, though And I try to stay away from big box stores. Because I donÂ’t really care to have to walk a mile in one of those big stores. IÂ’d rather go to a regular grocery store than a club or something. So I am, I guess you call that managing pain by managing where you go. (66-ye ar-old male, ESI+PT)

PAGE 119

104 Learning how to manage Some patients found a routine or strategy that h elped them manage their condition. For these patients, learnin g how to manage their daily lives gave them a sense of control that they previously lacked. Often participation in activities was managed with rest breaks, sitting, or doing specific stretches t hat they learned were successful and allowed them to manage any reoccurrence of pain or increase in f atigue. One man described how he had a routine to get his day going: I know if I want to reduce the pain, each morning, as I am getting up, it is best if I lay there in bed and do the stretching exercises before I get up. I know that is going to reduce the pain. So it is for my benefit to do that. So I do that. It doesnÂ’t eliminate the pain but it reduces the pain. (72-year-old male, ESI+PT) Medication and exercise were factors described acro ss most interviews as techniques used related to pain management. There was a consen sus among patients who had higher pain scores that medication was essential to management. Although many patients who lowered their reported pain and disability mentioned using medica tions at some point, only about 1/3 of the patients who had improved scores mentioned using re gular pain medication in their routine posttreatment. On the other hand, more than half of the patients who improved in pain and disability talked about exercise as helpful to manage their da ily condition. Some of the patients who described doing exercise learned how to do exercise s in physical therapy. Many patients mentioned doing exercise in the past and therefore tried to incorporate their previous exercise that they were familiar with. A few patients even create d exercises and stretches on their own from trial and error of different movements. An example of all of these is seen when one man talked about what he does to manage his condition includin g medication and exercise as well as different stretches he developed while golfing: I still take a medicine I donÂ’t know what it is num bs the numbness in my foot. I guess you could say it camouflages it. It doesnÂ’t stop it it just makes you donÂ’t know it. As opposed to an anti-inflammatory, which reduces the inflammation and reduces the pain by reducing the inflammation. But anyway, the exerc ises are certainly an important part of it, and some of it is exercises and some of it i s me going over to the fitness center and doing the lower back and the abs exercise. And ther e was one where you stand up and you pull one leg up behind you and stretch your thi gh. That one helps. There are, the

PAGE 120

105 main ones that you lay on your back, pulling your l egs up in the air, towards you to stretch out that way. Turning, laying, trying to ke ep your shoulders flat on the air and turning your hips so one leg is pulled over and the other leg out to the right or left, one leg stretching, sort of to stretch I guess your hip s or your side or something like that. When I am golfing, I will frequently either kneel d own and lean forwards to stretch my back or I will stand up straight and reach down and touch my toes and stay in that position for a few seconds to stretch it out. I kno w I need to do those kinds of things, and I know that is part of it, that unless I have back surgery I have no choice. Unless I want to live with more pain, then I continue doing those ki nds of exercises. (66-year-old male, ESI+PT) Another man explained how the physical therapist he lped him and how he added exercise to part of his daily management routine: I guess its probably the main benefit that came ou t of that [physical therapy treatment] was exercise, that I got from him, that I did in th e morning that made it a little easier for me to navigate during the day. If I didnÂ’t do these exercises, the system, the back, the hips, whatever, then I was even worse off... I gues s I still donÂ’t fully understand the relationship between the back and the stenosis, but there is also back pain per se and the exercises, the work that he did and the exercises t hat he gave me allowed me to feel, I donÂ’t know, feel better to be more able to get arou nd. (72-year-old male, ESI+PT) Patients often designed different strategies to man age their lives with LSS. One commonly discussed strategy was changing activities by spaci ng them out, taking rest breaks or changing positions, particularly by sitting. One man discuss ed how he learned to manage keeping his lawn mowed: I space things out to make it work. In other words I mow the front yard first so it looks good, and the next day I do the back yard, instead of doing it all in one day. (68-year-old male, ESI) Another woman talked about how she used trial and e rror to figure out what she could manage: I think I have learned by trial and error. Can I do this? Or maybe I better leave this alone, or maybe not so intense with what I am doing. I used to do a lot of garden work and would pick up these bags of dirt. T hatÂ’s one of my problems. I am picking up the heavy dog food regularly and in t he summer I was doing the dirt and the fertilizer and I think this was aggrav ating it so I learned I need to stop and say no, you need to wait and get some help with that, donÂ’t pick that up. (77year-old female, ESI) Finally, some patients described how changing focus or deciding on what was important and what they could ignore were helpful strategies to manage their daily lives. One woman described her

PAGE 121

106 strategy of ignoring or placing less emphasis on th ings that she used to focus on in her daily activities: I don’t feel confident that I can go up on a stool. I am tall so I can reach but if it is the corner on the window isn’t clean, I let it be. That ’s about it. I just cut out some things, play dumb, you know what I mean, just ignore it. Th at’s all. (73-year-old female, ESI) Push through attitude About half of the patients who improved in pain and disability scores described that they had a “grin and bear it” attitude that they attributed to managing their LSS condition. They noted that they were able to in crease their ability to do activities because they pushed through the pain, weakness, fear, or ot her factors that were keeping them inactive. Because of this attitude, many noted a return to fu nctional and recreational activities that they felt was important or that they enjoyed. One man describ ed this attitude as an internal drive learned from his grandmother: I think it is something internal. I just learn that I always do hard work, and I always see my grandmother in my mind. She lived to be almost 1 03 and she had arthritis real bad, she had the arthritis hands and she lived by hersel f until she was about 98 and she dealt with it, she didn’t take anything, she lived by her self until she fell and broke her hip. But I just see her and I knew she had to be in some kin d of pain because I had never seen her go to the doctor or take medicine. But I don’t know I just learned to deal with it. So yes, it is pain and it would be nice not to have it but I have also learned that if that is the way it is, it is the way it is. (60-year-old male, ESI) Another male patient described his motivation to pu sh through as something that was part of him and his personality. You always push on, because you are pushed from the inside. So you just keep living your life and doing what you are doing. Bend over, sit down, and get up and do it, just keep going. That is the point. (73-year-old male, E SI) For these individuals who pushed through the pain, doing activity was thought to be better than not doing activity despite pain. In addition, some felt they had to keep going because no one was available to help them in daily activities around t he home or outside the home. One woman with multiple comorbidities described improvement from t he actual treatment as well as the support she received from health care professionals. In thi s quote she described how her attitude to keep going got her through the worst part of the conditi on and still kept her going:

PAGE 122

107 I am a very strong person which is what everybody t ells me. Whether I am in excruciating pain or not, I still do my daily chore s. I don’t vacuum anymore. I wash my dishes like I used to and sometimes I stop in between and have to sit down, but I force myself. And a lot of times I cry but I will make my bed and sometimes I have to stop two or three times before I finish but I am determined I am going to do it. (74 year-old female, ESI) One woman described her need to push through as a n ecessity. Her ability to push herself was driven by her greater need to be independent. She s tated: What got me through was the fact that I wasn’t goin g to go stay with my son. And I knew that if I didn’t do better, I couldn’t stay by myse lf. And I wasn’t going to go stay with my son and his wife, that wasn’t, that’s out. And so t hat’s why I say, anything to make me feel better. And I am a very independent person; I want to do things for myself. I don’t like not being able to do certain things, I try to push myself, you know, I’ll just try to push myself, I am used to being by myself and I am used to doing things for myself. So it is embarrassing for me to ask my girlfriends and my sisters to help me. (66-year-old female, ESI+PT) Some individuals stated it was their nature to work hard and keep going regardless of the consequences. One man described his ability to push through as a result of his male pride and military mindset: We are at the category of male pride. So you keep d oing it. I was once hurt from a hard landing in a helicopter in Vietnam. Sitting on the floor, the floor in the helicopter was metal, to keep the bullets from getting through. An d it slammed to the ground and I was sitting on the floor. And I was just getting ready to retire: I was 25 years in the army. I felt it, and I banged my tailbone. I think that is when my back problems originally started. Shortly after that. So that gives you a rough idea. It went on for several years. But because of the male pride, you don’t ask for help. All my friends went to the chiropractor. No, you give them 40 dollars and all you get is the m to rub your back. The real impact, if you could get into the military mind, you never sti ck your arm out and have somebody help you. I would just keep doing. You never compla in, never let them see you sweat, that whole business. It never goes away. (73-yearold male, ESI) This “grin and bear it” attitude was not reported b y all patients. There were some who believed you had to do certain things, like exercis e, because it was good for you regardless of pain. This final thread that was seen in the theme for pushing through kept some patients exercising, doing chores, and other things they tho ught were essential despite pain. Therefore, having a push through attitude was not always helpf ul. Several patients who had higher levels of pain post-treatment discussed how they pushed throu gh their pain because of their beliefs that

PAGE 123

108 they should be exercising, moving, and completing o ther obligations. One woman described this belief like this: And I think if I was to give advice to anybody, the y need to keep moving and exercise. I donÂ’t care how much pain you are in, if you donÂ’t g et up and move around, you are going to be in more pain. That is my experience. And I do nÂ’t care how old you are, at least, you know I was given some good years before my back rea lly got bad and I was thankful, and like my Doctor said, thank goodness you were as act ive as you were, because if you werenÂ’t, you would be in much worse shape. And that Â’s why I would take their advice and keep going. (67-year-old female, ESI+PT) Having support Social support was one of the most disparate top ics across all patients. Most patients reported having some type of support during the study period. Some patients reported that support was helpful and important for management of their condition. Some expressed they did not want to ask for support and therefore did not feel support was important to manage their condition. Even a few described suppor t as unhelpful during the management period. However, a consistent theme emerged as patients who reduced pain and disability reported having support as a positive and helpful r esource during the post-treatment period. For those patients who did find that support positively contributed to their experience, social support was helpful to improve activities of daily living. One woman stated this as she saw having support as helpful motivation during her recovery: It does make a lot of difference. When you see that someone is in your corner, and someone is trying to help you, yes that makes me fe el very good about myself, it makes me try much harder. I would feel so bad sometimes, I didnÂ’t want to get out of my bed, and sometimes I didnÂ’t. I didnÂ’t want to do anythin g. I feel like I can do things now... It was really important to have the support.(66-year-o ld female, ESI+PT) When questioned, patients revealed perceived and re ceived support from a variety of sources including tangible support, informational s upport, positive social interaction, and emotional support. The most frequently cited support received across a ll subjects was tangible support. Examples of tangible support include someo ne performing an errand, a car ride to the

PAGE 124

109 store, or help doing some form of ADL. The provider of this support usually came from friends, family, neighbors, or hired individuals to do the s hopping or give a car ride. The next commonly discussed type of support was inf ormational support. Patients received information from a variety of individuals, although most informational support was reported to be received from health care profession als. When asked which health care professional provided support, there was a range of answers. Most patients noted they received informational support that was helpful from both th eir physician and physical therapist if they also had physical therapy treatment. Patients noted mixed amounts of education about their condition from providers with examples including a guideline of what activities they could do or precautions about what they should avoid. Some in b oth treatment groups reported feeling educated about their condition by their physician. Two patients with injection and physical therapy treatment felt that their physician did not take enough time to talk to them about their condition and what they could and couldnÂ’t do in th eir daily life. These two patients noted that the physical therapist did more to educate them abo ut their condition and provided more suggestions about activities than their physician o r other health care professionals. Three subjects with injection only treatment noted they received s ome of their information and education from the internet or other private pay sources such as a chiropractor or massage therapist. Overall, the majority of information came from health care provi ders rather than family, friends, other media or ancillary care services. One man noted he receiv ed support from family and friends, but support from medical professionals helped him impro ve his ability to do things and reduce his back pain: My friends and family have been helpful but particu larly I found help from medical people, from the orthopedic division at the VA, your division there at university doing the research, you were all contrib uting [to my getting better]. (80-year-old male, ESI) In contrast, 2 subjects particularly noted their di stress that their physician (a primary care physician and a spine physician) did not listen to them or acted cold and unsympathetic to their

PAGE 125

110 problems. These individuals expressed general despa ir about having any solution to their problems which seemed to be complicated by the issu e that they did not receive supportive gestures from their health care provider. However, these patients said they did receive very beneficial support from the physical therapist. Bot h patients described the physical therapist as providing motivation and emotional support as well as helped to educate the patients about their ability to advocate for their health and health car e services. One of these patients found a new physician who was reported to provide much higher l evels of support and in turn gave this individual more confidence and a better outlook for her condition overall. Most patients expressed that they did not want to b urden the people around them with complaints of their low back and/or leg pain. They did not want to complain about what they couldn’t do in their lives and a few stated that “o thers have it a lot worse off” noting they thought they should not complain about pain or lacking in a bility to do things. Only a few patients mentioned that they received emotional support from friends and occasionally a family member or even a health care professional. Those who were married did acknowledge that their spouses provided emotional support although a majority of p atients who were married also stated that they didn’t want to burden their spouse with complaints or emotional needs related to their LBP. For these patients receiving emotional support, it was important in order to feel “better” about their back problems as well as their daily life. One woma n reflected on her ability to ask her family members for support: If I needed, it was always there, I just had to ask I hate to ask all the time…you don't want to wear out a good thing, but they are there i f I need it. (69-year-old female, ESI) Even when asked directly, there were only a few pat ients who stated emotional support was important to their recovery. Friends were discu ssed by both female and male patients as indirect sources of support, but not specifically f or their LBP. Female patients more often than male patients did state friends were their main sou rces of emotional support even though some found that friends didn’t really understand what th ey were going through. For this reason, the few

PAGE 126

111 patients who talked about positive emotional suppor t related to their management of LSS were also patients who had higher levels of pain and dis ability post-treatment. One woman who lives in low income housing for people who are on disabil ity described how the people around have supported her: Well, living in this building, there are a lot peop le who are suffering. Many in wheelchairs. I have a good friend, named Randy; he lives two floors under me. He used to be a truck driver, but he had an acciden t driving his truck and messed up his lower back and he is in a wheelchair now, bu t he is huge, maybe 400 pounds. But we became good friends… He has always b een very supportive to me. Some of the people in the office that work here some of the staff, they are very supportive. My family, yes. Especially when I became a diabetic, like 8 years ago. My daughter in Connecticut…she has been my rock or whatever you call it. Whenever I feel really bad, I call her. An d then we have a few people in here who have back problems also, and we talk about it. One thing I have to say, I have a lot of people here who support me and I su pport them. (74-year-old female, ESI) All of these complexities captured by the patients’ voices helped to demonstrate the inconsistent role social support had in the lives o f those with LSS. As a result, support had both positive and negative connotations to these individ uals. A woman described this as she reflected the mixed role of support since she had it availabl e and helpful but doesn’t want to ask for it. She described these competing interests regarding suppo rt in this way: My friends, I don’t say too much to them, and that is probably because I don’t want them to think that I am getting old. But my son and my d aughter are pretty supportive, if we are going to do something, they make sure if I need to rest and my back is hurting, they are understandable. They understand, mom stay here, we will get this for you and of course I do not like to be waited on, so I once in a while let them do it, but mostly I am “no, no, I am fine, don’t baby me. I need to walk.” But they are really supportive, they understand, and of course I am their mother and the y don’t want to see me hurting. But I should be the one taking care of them. My younger b rother is pretty good, my older brother, he’s got a wife that has MS so I don’t go to too much with him, they don’t need to hear about my poor little back. But they are rea lly supportive, and my friends understand. They know. If you have a back problem, you have a back problem. We all have different problems going on. (67-year-old fema le, ESI+PT) Positive outlook Some patients seemed resigned to the fact that t hey could not and will not be able to do activities as they had done in th e past. These individuals seemed to re-define what activities they could do now and what they bel ieved they will be able to do in the future. For

PAGE 127

112 some this was a positive recognition of limitations in order to better self-manage their condition over the long term. Others actively engaged in goal setting, determined to return to some of the activities that they had to give up during their pe riod of back pain. One man described his positive attitude as his motivating factor: I am sure part of it is that I have a positive atti tude. I am not thinking about killing myself or any stuff like that. And my positive attitude co mes from having a good wife, good kids, being reasonably financially sound. I am reti red so I am not worried about losing my job. But I love to play golf. And that doesnÂ’t c ome without some pain. But I love to golf so much that I am willing to do all of that is necessary to enable me to get out there and play golf. (66-year-old male, ESI+PT) He continued on to describe his ability to push thr ough and manage pain due to his military training, his financial circumstances, and his fami ly. When asked about how he manages, he stated: First of all, yes I do deal with pain. But there ar e a lot of people that have a lot worse fates in life than I do. And I am retired from the Air Force, as a Colonel, that means that I am very fortunate that I have a pension that is bet ter than many other people have. That doesnÂ’t mean that I am rich, but it doesnÂ’t mean I am going to go into the poor house either. I am retired in a school district, I have s ocial security, my wifeÂ’s got a good job, and she is going to have a good pension. So there i s a lot to live for. I am not going to lose my house; I am not standing in the food line o r anything like that. In a very bad time in your country, my family and I are very fortunate And when you are fortunate, if no other thing good is going on, but if you are fortun ate like that, you have something to live for. I had kids late, so I donÂ’t have grandkids yet IÂ’d like to be a grandfather before I croak. So IÂ’ve got all kinds of things to look forw ard to. (66-year-old male, ESI+PT) When asked to describe the main thing that helped h er get better, one woman stated how her positive attitude made all the difference: Do you want to know what the number one is? My dad used to say, she is a tough old bird. You know, you have to go with the flow, you h ave to deal with what comes and I think God has been very good to me. If I have pain I have pain. I take what I have to take to relieve it as much as possible. I donÂ’t commit m yself to anything that I know is going to be stressful or strenuous because I just be aski ng for it. So now, I live a very sedate life. And stuff like that. But thatÂ’s ok. IÂ’ve had a good life and I am not complaining, but I think itÂ’s my attitude, and I am not being bragga docios or anything like that, I have a very positive attitude. I have two beautiful grandd aughters. I have my sons and I try to be an example for them. (87-year-old female, ESI+PT)

PAGE 128

113 Quantitative and Qualitative: Combined Results Patients with LSS reported a variety of types of co ping behaviors that mediated both the severity and reduction of pain and disability inclu ding the a priori sources, self-efficacy, social support, perceived physical health and perceived me ntal health. Many related factors emerged from the patient interviews, including physical lim itations, lack of confidence in tasks and activities, feeling a lack of control, reduced soci al participation, feelings of vulnerability, poor mental state, frustration at needing support, and f inancial limitations. The main factors that increased the severity of pain and disability at en try to treatment are illustrated in Figure 4.5. In addition, many factors identified by patients contr ibuted to the reduction of pain and disability in the post-treatment period, including having greater confidence, knowing limitations, learning how to manage, having a push through attitude, having s upport, and having a positive outlook. The main coping behaviors found to mediate the reductio n in pain and disability post-treatment are illustrated in Figure 4.6.

PAGE 129

114 n n nnrn Bnrn "r ,r& Bnr r "r ,r&r n , rr &n n rn )rrn (r r r rr &n n rn ) ,n +rr n ,n )rrn (r n n n nnnnrn !nrrrn )n *rr 7nnnr nn 7nnnrnn 7nnrnn

PAGE 130

115 n nnrn !rr n "r "r&r ,r r ?rr rr nr # ?rr rr n & r Cnr rrrrn A ,n ,r r n# nn$nn nnrn % !nrrn nnnr nnr 7nnnr rnn 7nnnrnn

PAGE 131

116CHAPTER V DISCUSSION This study aimed to identify predictors of perceive d pain and disability in patients with LSS both upon entry to treatment and post-treatment 6 to 12 months after entry into the study. In addition, this study aimed to describe sources of s tress and respective coping as described by patients in their own words. Particular emphasis w as placed on self-efficacy, social support, and general health perceptions as they contribute to pa in and disability, using a theoretical framework from the Stress and Coping Model and the Stress-Dia thesis Model.107, 112, 125, 141, 248 This research sought to fill several important gaps in the LBP literature, particularly the lack of research investigating the contribution of psychosocial factors such as self-efficacy and social support in identifying severity for and redu ction of pain and disability in patients with LSS. In addition, this research has added greater unders tanding of the extent to which perceptions of physical and mental health contribute to the severi ty and outcomes of pain and disability. Finally, this research used a novel mixed methods design to uncover the patientÂ’s lived experience with LSS. The overall outcome of this study was to info rm translational research, clinical interventions, and policy recommendations, which ma y ultimately serve to identify treatment needs and provide additional resources for reducing pain and disability and improving the quality of life for patients with LSS. Key Findings: Quantitative Summary Multiple key findings emerged from this mixed metho ds study. The quantitative data illustrate that patients with LSS enter conservativ e treatment with high levels of pain and disability, indicating the importance of reducing t hese stressors of pain and disability in this population as an important clinical priority. Upon entry to treatment, 75% had experienced pain in the moderate or high category (4-10/10) with 44% in the high pain category (7-10/10) and all

PAGE 132

117 of the patients had moderate or greater levels of d isability with almost 40% at the severe or crippling level of disability. The data also illustrated that multiple factors int eracted positively during the treatment and post-treatment period to reduce pain and disabi lity. Post-treatment, the mean disability score was at the low end of moderate range (29.3) compare d to borderline high range (40.3) at entry to treatment. A quarter of the patients were in the lo w disability range (25%) post-treatment compared to none upon entry to treatment, but simil ar numbers at 25% in the severe range and 10% in the crippled range of disability for both pr e and post-treatment. Similar results were seen for pain as the mean pain post-treatment was at the low end of moderate range (4.14/10) compared to 5.27/10 upon entry to treatment. Post-t reatment, 60% of patients had either no pain or were in the low pain category (1-3/10) compared to only 27% at entry to treatment, while 10% were still in the high pain category up to 1 year p ost-treatment. Next, a profile emerged from this sample indicati ng patients experiencing greater disability upon entry to treatment had worse percei ved physical health, worse perceived mental health, and lower social support. As hypothesized, with three quarters (75%) of patients with LSS reporting elevated levels of disability, patients w ho perceived they had worse physical and mental health experienced an even more disproportionate bu rden of disability. The independent predictors for greater levels of disability upon en try to treatment emerged, with worse perceived physical health independently explaining 35.2% of t he variance in disability, worse perceived mental health independently explaining 17.2% of the variance in disability, and lower social support independently explaining 10% of the varianc e in disability. With all 4 variables of selfefficacy, social support, perceived physical health and perceived mental health in the model, 48.8% the variance in disability was explained by a perception of worse physical health and worse mental health predicting greater levels of di sability upon entry to treatment. With almost half the variance in baseline disability explained by self-perceived health status, this is an import contribution to the LSS literature.

PAGE 133

118 The only significant predictors of pain at entry to treatment were functional self-efficacy and total self-efficacy. Reduced functional self-ef ficacy was the only significant independent predictor of higher pain, predicting 10% of the var iance in pain upon entry to treatment. Significant but underpowered multivariate relations hips emerged with the first model including significant contribution from reduced functional se lf-efficacy explaining 15.7% of the variance in pain. The second model had significant contribution from reduced total self-efficacy explaining a total of 17.4% of the variance in pain. Both models identifying higher pain at entry to treatment are only exploratory at this stage and need to be i nterpreted with caution due to the risk for error. Because these findings are preliminary in nature, a dditional follow-up analysis is warranted to examine these and additional factors related to sev erity of pain in the larger RCT data as well as in future studies. Lastly, a profile emerged identifying predictors as sociated with outcomes of disability post-treatment. Specifically, having a better perce ption of oneÂ’s physical health independently explained 76.5% of the variance in disability which is quite an important contribution alone. In addition, total self-efficacy was a second importan t predictor independently explaining 42.3% of the variance in disability post-treatment. The grou p model resulted in better perceived physical health as the only significantly contributor to pre dicting less disability; accounting for 73% of the total variance in disability post-treatment. Predicting a reduction in pain post-treatment resul ted in 2 significant models. Only 1 model was adequately powered and explained approxim ately one third of the total variance in pain. This model only had significant contribution from better perceived physical health explaining 31.7% of the variance in pain post-treat ment. A second significant but underpowered model emerged, with a better perceived physical hea lth again explaining 24.7% of the variance in pain post-treatment. Given the small sample size an d preliminary nature of the quantitative results, the qualitative results were used to trian gulate the quantitative findings and contribute new insights associated with stressors and coping i n patients with LSS.

PAGE 134

119 Key Findings: Combined Summary The qualitative component of this study used patien t interview in order to better understand the constructs of self-efficacy, social support, perceived physical health, and perceived mental health as they relate to pain and disability in the eyes of patients with LSS. Emergent themes not tested by the quantitative surv eys were also uncovered. Interview transcripts were analyzed for themes that represent patient beliefs and coping behaviors. These expressions by the patients supported, clarified, a nd added to the survey results regarding the stressors and coping behaviors of patients with LSS Similar to what was found in the quantitative data, pain was a key focus by many patients when describing why they sought treatment. In expre ssing pain and issues related to pain, patients also described components of disability such as pro blems with body structure or function, issues with personal factors, influences of environmental factors, and limitations in activity and participation that reflected a more complete concep t of disability supported by ICF framework. Although the ICF framework uses a broad lens in whi ch to interpret disability, it was not a helpful model to use to narrow down factors related to pain and disability in LSS. The StressDiathesis Model resulted in a much more accurate fr amework in understanding pain and associated disability in LSS. Using this model, pat ients described overlapping factors that related to the stressors of pain and disability and confirm ed the importance of the appraisal and coping process. Because degenerative LSS is a chronic and re-occurring LBP condition, pain is complex and not isolated from issues of disability in these individuals and supports the use of the StressDiathesis Model in LSS. In examining behaviors and experiences of the patie nts, multiple messages regarding the influence of self-efficacy, perceived physical heal th, perceived mental health as well as conflicting meanings for social support were found that confirm the importance of these constructs in management of LSS. Patients in this s tudy were similar to most patients with LSS who are older and generally have more co-morbiditie s found in later stages in life. As a result

PAGE 135

120 patients with LSS have many limitations in discreti onary, committed, and obligatory activities that are tied to multiple factors. The clinical out comes of pain and disability are used exclusively as markers for severity and outcomes in the literat ure, yet little work has gone into understanding what factors impact these outcomes other than treat ment. Therefore goal for the mixed methods design was to help clinicians better understand how physical, mental, and social factors impact the severity and clinical outcomes of pain and disa bility in patients with LSS. Entry to Treatment: Increased Pain and Disability The patientsÂ’ experiences with LSS at entry to trea tment shed light on a variety of personal and environmental factors that relate to p erceived pain and disability in the LSS population. Based on the available data collected f rom this cohort, three predictors (perceived physical health, perceived mental health, social su pport) contribute to the variance of disability and one predictor (self-efficacy) contributes to th e variance of pain at entry to treatment. These personal and environmental factors described by pat ients are supported by the ICF and the StressDiathesis Model as factors that help to define the stressors of pain and disability and describe coping in patients with LSS. Additionally, eight th emes were identified by patients as reasons for increased perceptions of pain and disability upon e ntry to treatment. These themes included physical limitations lacking confidence in tasks and activities feeling a lack of control reduced social participation feelings of vulnerability poor mental health frustration about needing support, and financial limitations In line with the Stress and Coping Model, these t hemes confirmed that patients who had negative appraisals and fewer coping resources all experienced higher levels of pain and disability. Perceived physical health The largest statistical contributor to the variance in disability at entry to treatment was perceived physical health The quantitative results illustrated a clear inverse relationship between a worse perception of physical health and increased disability at entry to treatment. The interviews confirmed that p atients had general health worries as well as

PAGE 136

121 concerns across all activities due to their LSS con dition. Patients expressed general concerns about diminished physical health that revolved arou nd their inability to keep up with prior levels of activity and/or levels of activity of their peer s. The perception of declining physical health may be particularly noticeable in this population of pa tients with LBP because of the high likelihood of concurrent age-related declines and comorbidity factors as compared with other types of LBP.252 However, aging and illness related perceptions of p hysical decline in older adults do not always result in the perception of poor physical he alth. Tobin and Lieberman253 and Rosow254 found that many older adults that are physically il l and moderately incapacitated still appraise their health status and life circumstances positive ly. These concepts support the Stress and Coping Model107 and the study findings that specific patients who held beliefs of declining physical health contributed to negative appraisal o f the stressors of pain and disability and resulted in poor adaptational outcomes. Patients with LSS described concerns with general p hysical function, bodily pain, problems with physical roles across a spectrum of a ctivities, and decreased general physical health. These areas of concern are consistent with the areas measured by the PCS subscale of the SF-36. The negative perceptions of physical health reported in interviews were expressed in terms of losses related to participation in daily a ctivities, recreational activities or social activi ties, as well as the perceived inability to fulfill certa in roles (such as spouse, friend, or parent) becaus e of fatigue and general physical loss. In addition, many patients revealed that the role of physical health may have a strong link to social functioning which seemed to permeate their perception of physical health status. This result is consistent w ith literature on other chronic health conditions.10, 255, 256 At entry to treatment when LBP was at the greatest acuity, many of the patients also noted altered perceptions in physical health that c ombined with their ability to perform specific tasks. It is unclear at this time if the inability to do specific tasks contributed to eroding their

PAGE 137

122 confidence in their physical health or if their per ceived declining physical health status due to LSS contributed to the lack of confidence in task p erformance. It is likely that both played a role in augmenting and confirming fears of poor physical health due to the chronic condition of LSS. Even in this condition-specific population of older adults, there were differences in perceived physical health that were revealed both i n the PCS scores and in the patient interviews. Those who had a worse perception of their health al so communicated greater limitations in their lives. For example, one man described his inability to do most anything since most tasks required walking: “Yeah, I don’t do too much. A lot of the a ctivities I like involve lots of walking.” Or another man who had difficulty getting to the bathr oom every day equated his inability to do this specific task of walking to the bathroom with his i nability to do general physical activity: “So if going to the bathroom is so bad I figured I couldn’ t do much of anything.” The uncertainty and absence of control over both the performance and th e outcome of engaging in physical activities seemed to erode patients’ confidence and therefore actual performance of a general group of activities. An example of this uncertainty is illus trated as one woman stated: “Well it is a little bi t hard dealing with it. It seems like it comes and go es, one day I can feel pretty good, and the next day I can’t do anything.” All of these sentiments d escribing a reduced confidence in physical health supports prior research that self-rated heal th is an important factor to consider and has direct links to disability.107 Unexpectedly, perceived physical health was not a s ignificant predictor of pain at entry to treatment. In contrast to existing literature showi ng that perceived physical health is a powerful predictor of pain, this study did not find such a s trong relationship in this cohort.264, 265 The patients’ experiences did not shed any additional l ight on this relationship between general physical health perceptions and pain except indirec tly during interviews as a contributor to physical limitations. This is surprising as many of the biomedical theories for pain involve a large focus on physical components of pain.125, 226, 266 It may be that the experience of pain in this population with LSS is different than other types o f LBP due to older age and chronicity which

PAGE 138

123 has been shown in previous literature to affect one Â’s perception of pain.10, 116, 226 It also may be because perceived physical health as measured by th e PCS and as describe by the patientsÂ’ was a broader construct than just bodily pain and represe nted a component of illness behavior. Previous research has shown that somatic functioning has a l arge role to play in illness behaviors but it is the sum of the physical, psychological and social c ontributions to illness behaviors that result in health perceptions and outcomes.122, 267 These findings would support the Stress-Diathesis Model where pain is in the pathway towards disability but perceptions of physical health contribute after appraisals of pain and interact much closer to anxi ety and disability constructs. These findings warrant further investigation and suggest the need for a closer look into the experience of pain in older adults with the chronic condition of LSS. It is important to recognize that measures of selfrated physical health in the SF-36 PCS overlap with disability measurement using the ODI a s both represent domains of physical pain and functioning. The ODI is a condition-specific sc ale which measures low back related pain as it impacts function whereas the PCS, the physical heal th subscale of the SF-36, measures physical function but also encompasses measurement of genera l physical health. In addition to specific physical function and bodily pain, the PCS measures the physical roles of an individual and general health belief regarding oneÂ’s physical stat us which are not represented on the ODI. The strong association between the SF-36 and the OD I found in LSS at entry to treatment may in part be due to the similar domains of bodily pain and physical function in these two measurement tools. At this time, the nature of this relationship specifically in this cohort of patients with LSS cannot be determined due to the s mall sample size. However using the ICF framework to define disability, body structure and function represent only one area of disability. Disability according to this framework also include s personal and environmental influences. Similar to what was found in the data, additional c ontributions from perceived mental health and social support and qualitative data supporting self -efficacy were also associated with disability in

PAGE 139

124 patients with LSS at entry to treatment. These fact ors combined to represent a larger picture of disability in LSS as described by patients at entry to treatment. Perceived mental health The perception of poor mental health was another fa ctor associated with greater disability at entry to trea tment supported by both the quantitative and qualitative data. This studyÂ’s findings corroborate the results of several studies that found that patients with LSS have mental health issues pre-tre atment.76, 86, 88 An important contribution of this study is the finding that a having a worse per ception of mental health significantly contributed to determining elevated levels of disab ility upon entry to treatment. The qualitative findings supported this as well, suggesting that po or mental health and feelings of vulnerability contribute to concerns of patients with LSS and ser ve to increase their levels of stress and burden of disease. Patient interviews helped to clarify the role of me ntal health status as a contributor to the experience of disability at entry into treatment. T he interviews revealed that in 5 of 9 patients with higher pain and higher disability scores, a ge neral poor mental state and depression contributed to their condition severity. This poor mental state ranged from a depressed mood to thoughts of suicide by 2 patients during a period o f intense LBP before seeking treatment. These expressions of poor mental health illustrate mental health issues in the lives of patients with LSS and the relationship of mental health to pain and d isability in this population. Another theme related to general mental health, vul nerability, emerged from the qualitative data. Previous literature has not ident ified feelings of vulnerability as contributing to increased pain and disability in LSS. However suppo rted by the Stress-Diathesis Model, feelings of vulnerability can modify the appraisal of pain a nd magnify the trajectory towards disability. This is consistent with patientsÂ’ narratives descri bing the effects of pain but focusing on their vulnerability resulting from functional limitations .125 Patients also described a lack of control which con tributed to mental health beliefs as it not only identified physical concerns but also reve aled a high level of anxiety patients had about

PAGE 140

125 daily living. These anxieties about control over ev ents in their daily lives resulted in some degree of increased vigilance regarding pain and activity performance. These patientÂ’s descriptions regarding control, anxiety, and vulnerability as co mponents of mental health are supported by the Stress and Coping Model as an essential theme linki ng emotional states to appraisals of harm, threat, and challenge.107 Overall, the emotional states described by patient s with LSS support a link between perceptions of worse mental health and the negative appraisal of pain and disability. It is not clear to what extent emotional states and mental health contribute to experiences of pain and disability, versus how much the experie nce of pain and disability contribute to emotional states and perceptions of mental health. In this study, patients with expressed poor mental health status on the MCS also had a higher b urden of disability. The interviews reinforced the quantitative findings that perceived mental hea lth factors contributed to disability in LSS and supported the Stress-Diathesis Model. Future studie s should explore these relationships using additional tools to compare measures of perceived h ealth and emotional states with outcomes of pain and disability. This is one of the first studi es to explore this issue in LSS and it confirms tha t a perception of worse mental health is a concern fo r patients with LSS and contributes to their severity of disability. Social support A small but significant inverse relationship was found between decreased amounts of social support contributing to greater disability in bivariate regression analysis. Supporting the Stress and Coping Model,107 a reduction of social support equates to the reduction of a coping resource. Under this model, h aving different sources of support should protect an individual from the adverse effects of s tress. However these results were not supported by the multiple regression analysis as social suppo rt was not significantly associated with pain or disability at entry to treatment. These inconsisten t results for social support emphasize the need for exploring this factor in the context of the pat ientÂ’s perspective. One key theme from the patient interviews was the e xpressed frustration of patients at having to rely on others. These strong feelings of frustration about relying on othersÂ’ for support

PAGE 141

126 may be interacting with patient’s perceptions about social support resources and the role of these resources. As stated previously, the Stress and Cop ing Model describes having support available as a positive coping resource that can reduce stres sors. However, in this population, having to rely on others for support itself seemed to be a stresso r. The frustrations expressed by patients for needing support and even more by patients who had t o use support illustrates the possible mixed messages that revolved around social support, parti cularly when most vulnerable at entry to treatment. In the literature, Doeglas et al268 demonstrated that satisfaction with supportive transactions was more important for patients’ wellbeing than the existence or frequency of supportive transactions. Therefore, patients with L SS may have reduced satisfaction with support as a coping resource and as a result, having social support was less helpful in ameliorating their pain and disability. Self-efficacy Surprisingly, the quantitative data revealed that s elf-efficacy was not a predictor of disability and only contributed to exp eriences of pain upon entry to treatment. Since self-efficacy reflects one’s confidence in physical tasks, one might expect self-efficacy to be related to disability when physical function has de clined. However in this cohort, issues with selfefficacy at entry to treatment were only reflected by an association to pain. In line with the Stress and Coping Model, those with lower appraised total and functional self-efficacy as measured by the LoBACS survey experienced a disproportionate bu rden of pain compared to those with higher total self-efficacy in bivariate and multivariate a nalysis, although these relationships were underpowered in the current sample. The qualitative findings confirmed the role of self-efficacy contributing to the severity of pain but also suppo rt a larger role for self-efficacy in contributing to disability at entry to treatment. When patients spoke about their beliefs regarding o rganizing and executing necessary actions for life activities, they would most readil y use words like “confidence” for tasks and discuss what tasks they thought they could or could not perform. In the patient’s words, the a priori identified construct of self-efficacy was a theme that emerged most strongly as lacking

PAGE 142

127 confidence in tasks and activities Specific tasks such as standing doing dishes, sta ir climbing, walking to the bathroom or to the mail box all were examples used by patients of tasks for which they expressed lower confidence upon entering treat ment. In addition, the theme of lacking control particularly over specific physical tasks was pervasive across interviews. This theme reflected t he patients apprehensions for participating in tasks before treatment, especially low back related tasks such as ADLs that included standing, walking, lifting, bending, and stair climbing. Acco rding to Bandura,153 self-efficacy beliefs reflect confidence for performing tasks but only while regu lating one's own motivation, thought processes, affective states and actions, or changin g environmental conditions. When patients expressed a feeling of a lack of control over tasks it reflected their motivation and thought processes for executing specific actions. Patients described their lack of control as being unsure of their bodiesÂ’ performance or whether they would have pain during the performance of a task. This uncertainty helped to erode patientsÂ’ confiden ce for performing low back related tasks. Overall, those patients with feelings of a lack of control as well as a reduced self-confidence represented the construct of reduced self-efficacy and subsequently reported higher amounts of pain with activity supporting the quantitative resu lts. In addition, the qualitative findings emphasize tha t reduced self-efficacy is possibly more pervasive upon entry to treatment than found in the survey data. It is possible that expressions of reduced self-efficacy for specific tasks were overs hadowed upon entry to treatment by stronger concerns for an individualÂ’s overall physical and m ental health as seen in the quantitative data. Self-reported health may be more reflective of oneÂ’ s perception of disability during acute episodes of a chronic condition such as LSS, and co ncepts of self-efficacy may be more narrowly defined during this time to pain provocation during specific task performance. More research is warranted to explore these relationships in patient s with LSS.

PAGE 143

128 Entry to Treatment: Emergent Theme Financial limitations The role of financial limitations was an unexpec ted factor reported by patients that contributed to the experi ence of greater pain and greater disability that was not tested in the quantitative data. Those pat ients who had previously been unable to receive timely medical care or had received reduced care du e to financial limitations expressed greater levels of perceived pain and disability. In these i nstances, financial limitations represented environmental constraints to coping resources. Many patients complained of limited financial resources for ADLs that they could no longer do on their own such as home maintenance, yard maintenance, and transportation services. These lim ited financial resources were particularly present in those individuals who were retired and o n fixed incomes. Those few who still worked had fewer financial limitations but their problems seemed to revolve around trying to keep their job. In several cases, LBP became problematic to th eir work abilities and led to a loss of work time and even the loss of their job. Post-Treatment: Improving Outcomes of Pain and Disa bility Patients with LSS described a variety of sources co ntributing to a reduction in the stressors of pain and disability post-treatment. Th ese related to the a priori identified constructs of perceived physical health and total self-efficac y for low back related tasks. Additional themes from the patient interviews help explain some of th e additional variance in pain and disability post-treatment unexplained by the quantitative mode ls. Post-treatment, patients talked about their coping behaviors through the themes of knowing limitations learning how to manage having a push through attitude having a positive outlook and having social support In line with the Stress and Coping Model, patients who described mor e coping resources and improved coping behaviors post-treatment compared to pre-treatment reported better outcomes of pain and/or disability and subsequently reduced the effect of s tressors from this chronic condition. Perceived physical health The most significant factor contributing to the va riance in

PAGE 144

129 both pain and disability post-treatment was perceiv ed physical health. Better perceived physical health on the SF-36 PCS was both a bivariate and mu ltivariate predictor explaining approximately three quarters of the variance in dis ability outcomes. In addition, the perception of better physical health significantly contributed to the group model explaining 40% of the variance in pain post-treatment. The qualitative data supported the findings that a perception of better physical health was related to improvements in pain and disability post -treatment. During interviews, many patients expressed a greater confidence in their overall phy sical abilities and physical stamina. Improved perceptions of general physical health seemed to gr eatly improve patients’ motivation for coping efforts leading to a reduction the severity of thei r experience of pain and disability. The literature has shown that holding positive beliefs about one’s health increases resilience and the ability to cope with stressors.269 Positive self-rating of health has also been shown to predict active coping strategies in health challenges 4 years later.210 Patients’ general physical health beliefs were bols tered when specific tasks were performed or thought to be attainable. Specific tas ks become a proxy for a general class of activities that the patients’ wished to perform. Of ten these were activities patients used to engage in before their back got worse. Returning to some o f their “normal” functions post-treatment seemed to greatly improve patients’ perception of t heir physical health. However, in the interviews it was unclear whether beliefs for impro ving physical health led to increased confidence and control over specific tasks or if sp ecific task performance led to better perceptions regarding patients’ physical health status. It woul d be warranted as evidenced from these preliminary findings, to explore the relationship b etween self-efficacy beliefs and perceived physical health in patients with LSS in future stud ies. What is clear from the strong associations between the PCS scores and pain and disability scores in triangulation with patient statements is that improved perceptions of physical health was important for improving outcome expectations. This confirms that a positive perception of

PAGE 145

130 physical health contributes to improved coping beha viors in LSS. These themes for improved physical health were repeated often by patients who expressed a greater pain relief and a better overall ability to manage their low back condition. The interviews illustrated how perceived control over oneÂ’s health situation resulted in bet ter outcomes of pain and disability. Supported by previous literature in aging,270, 271 the findings in this cohort of patient with LSS su pport the importance of self-rated physical health in relatio nship to better long-term outcomes of pain and disability. As discussed above at entry to treatment, it is pos sible that the SF-36 PCS and the ODI overlap in the constructs each measure related to p hysical function and bodily pain. This is one potential reason why there is such a strong relatio nship between these two measures. In addition, LSS may represent a population where the focus of d isability is largely characterized by physical function over the long-term. Other studies have fou nd differences among groupings of conditions such as rheumatic conditions where different domain s of the SF-36 PCS represents some rheumatic conditions better than others.272 The usefulness of general measures such as the SF36 and itÂ’s subscales, is in their ability to allow co mparisons among patients with the same condition as well as between patients with different conditio ns. However, condition-specific measures such as the ODI has been reported to be more responsive to change in the condition under study, such as when comparing outcomes among individuals with L SS. Previous literature in LBP has found general health measures to be both more responsive,273 and less responsive274, 275 than condition specific measures such as the ODI a nd the Roland Morris Disability Index. The ODI had been sh own to be most sensitive to change compared to other measures in patients with LBP who have low function,10, 273 such as what is found in LSS. Other studies have shown similar resp onsiveness between generic health measures and condition specific measures in different types of health conditions.276, 277 However, the SF-36 has been found less responsive when representing co ndition-specific problems.274, 277 Specifically in LSS, Stucki et al275 found that a condition-specific measure was determ ined to be more

PAGE 146

131 responsive in measuring outcomes of physical functi on and symptom severity in LSS than general health measures, although neither the ODI o r the SF-36 were specifically tested making comparisons difficult. Typically it is recommended that both condition-spe cific and general measures be used clinically and for research. Utilization of both ty pes of surveys, however, increases respondent burden, creates redundancy, and adds to data collec tion and analysis burdens. If a general measure such as the SF-36 PCS was as responsive as a condition-specific measure such as the ODI for LSS, the benefits of the general measure co uld be maintained without risk of losing information. General health measures like the SF-36 include broader questions related to health related quality of life and therefore potentially r epresent more than what is routinely measured to determine clinically relevant change for a specific condition. Literature in other health conditions have found generic measures such as the SF-36 are i mportant to describe overall health and compare against other health conditions but conditi on-specific disability measures are important to compare outcomes within specific conditions such as LSS.278 At this time, the nature of the relationship between perceived physical health and disability in LSS using the SF-36 and ODI cannot be answered accept to acknowledge the strong association post-treatment. Future research needs to explore the relationships between these me asures of perceived physical health and the outcomes of disability as well as the emphasis on p hysical function in LSS. Self-efficacy The quantitative results demonstrated a signific ant relationship between higher self-efficacy and reduced disability post-tr eatment in bivariate analysis but not multivariate models. The qualitative results suppor ted the bivariate results and added further nuances regarding the role of self-efficacy as a co ping behavior in patients with LSS. Three themes helped explain patientÂ’s increased confidenc e for specific low back related tasks and contributed to the importance of increased total se lf-efficacy in the cohort of patients with LSS: greater confidence knowing limitations and learning how to manage

PAGE 147

132 Overall, patients expressed a greater confidence po st-treatment in performing specific tasks. This confidence originated from treatment ex periences as well as alternative health and self-motivated experiences in which they were able to do specific physical and social activities successfully. In addition, patients frequently stat ed that with treatment, information, and trial and error they developed a self-system of knowing what and when they reached or came close to their physical, mental, and fatigue related limitations. Identifying what helps improve patient confidence, a key aspect of self-efficacy, may be i mportant to investigate in future research in the reduction of disability in LSS and the proportion o f variance not explained by the statistical model. Social support At post-treatment, there were no significant quanti tative relationships between social support and outcomes of perceived pa in and disability. However, in the qualitative data, social support emerged as a theme related to improvements post-treatment. Yet even in the qualitative data, only some of the patients who had reduced pain and disability described social support as helpful in improving their condition. Ot her patients described social support as not helpful in improving their outcomes. These conflict ing results underscore the need to explore and understand the role of social support as a coping r esource as described by patients with LSS. Overall, the interviews illustrated a complex assoc iation between social support and the outcomes of pain and disability post-treatment. One theme emerged as several individuals with improved pain and disability outcomes did express social su pport as being helpful to their recovery. These patients found social support helpf ul particularly when it related to tangible support such as treatment from a health care profes sional and help around the house from family and friends. Others found helpful informational sup port from health care providers. Less frequently mentioned was emotional support which wa s found to be helpful for a few patients from sources such as health care providers, family, or friends to keep them motivated for recovery. These views of social support expressed b y the patients with LSS are similar to the positive effects of support as a stress buffer that has been described in the literature.107, 280, 281

PAGE 148

133 However, some patientsÂ’ descriptions of social supp ort as well as their evaluations of support depicted a different role for social suppor t. For these individuals, support was something they tried to avoid. Inherent in the definition of social support is the supportiveness of the relationship(s).279 These patients expressed embarrassment or reluctan ce about supportive functions. For those individuals who found support carried negative connotations, social support indicated weakness or helplessness. Asking and rece iving support was expressed as being a burden for their family and friends. Some stated th at even though they might have support sources, they would ask for and use support only if absolutely necessary. This response to support was due to pride, the need for independence, and ev en due to anxiety about family or friend relationships. Still others found health care profe ssionalsÂ’ support ineffective or disingenuous. For these individuals, just getting treatment did n ot indicate a role for support nor did they see the supportive nature of health care assistance. In addition, a number of patients with diminished p ain and reduced disability posttreatment stated they did not need or want to ask f or support. These individuals described themselves as having a push through attitude For these individuals, the reluctance for emotion al support and the desire not to be a burden on others seemed to overshadow the positive role social support could play as a coping resource. It is poss ible that these conflicting views of support and self-reliance illustrate why the underlying theorie s for social support as a positive coping resource were not supported in this research. These conflict s have been recognized in the literature as negative effects of social support.279, 282 In addition, some of the patients who had the most severe ratings of pain and disability post-treatment were the ones who found social suppo rt a positive coping resource. It is possible that those with higher disability needed to draw on relationships more frequently and therefore rated their support higher than those who possibly had the support available but did not need to use support for assistance in times of need. Accord ing to Lazarus and Folkman,107 social support is a resource that must be cultivated and used. The refore it is possible that an individual who has

PAGE 149

134 a chronic illness such as LSS who perceives he/she has a worse health status and greater need of support might cultivate and use his/her supportive resources more often regardless of the type of support. The qualitative findings confirm this expl anation as those who reported greater functional limitations and reduced ability to leave their house talked about how they created supportive relationships with family, friends, and neighbors to help provide them with the support they needed to go to medical appointments, get groc eries, or just go on outings. All of the negative and positive effects of social support described by patients’ with LSS illustrate the differing perceptions of individuals while engaged in cultivating and using social supports. Although no consensus on support was reac hed in this study, the conflicting results are consistent with what has been uncovered in other st udies and underscores the need to identify individual perceptions of social support as a part of the treatment planning process and determine the availability and acceptance of social support a s a coping resource.281, 282 Post-Treatment: Emergent Themes Several emergent themes were uncovered in the inter views that expressed reasons for patients’ relative improvements in perceived pain a nd disability post-treatment. These themes included having a push through attitude and a positive outlook Both of these themes reflect belief systems generalized by patients. According to Lazar us and Folkman,127 beliefs served to modify appraisals of a stressful situation and enhance cop ing behaviors. These two themes expressed by patients with LSS were ways that these patients des cribed how and why they improved posttreatment. Push through attitude Particularly those individuals who had decreases in disability scores post-treatment reflected the theme of pushin g through. This is similar to the “just do it” or “grin and bear it” attitude recognized in our socie ty as toughing it out despite pain or other obstacles. Some, but not all, who had lower pain sc ores post-treatment also talked about having this attitude. This inconsistency may be in part du e to the fact that those who pushed through had

PAGE 150

135 increases in pain at times but kept going neverthel ess. For those who were able to push through, their words indicated that they had returned to mor e activities and had higher belief that they were able to do things. For these individuals, this ment ality seemed to bolster their overall confidence and perspective on returning to “normal” or some de gree of their previously held lifestyle. However, there were a few patients with a push thro ugh attitude that did not show improved outcomes of pain or disability. These pati ents commented on pushing through out of necessity either due to the lack of support or lack of resources, particularly financial, at their disposal. For these individuals, pushing through ir ritated their condition, eroded their confidence, and often they resigned to doing less as a manageme nt strategy. Positive outlook Finally an emergent theme of having a positive att itude or outlook on their situation was helpful for many with LSS. Having a positive outlook seemed to be a beneficial complement in improving pain and disabil ity. The sense of positivity has been found to be a component of motivation and helps to prevent n egative states such as depression and anxiety.251 Supported by the Stress and Coping Model and the I CF framework, positive beliefs are a personal factor that can influence positive a ppraisals of stressors and the availability of coping resources. In research by Chipperfield,283 health optimists who tended to overstate their health, even with a number of health conditions, we re significantly more likely to live longer indicating a bias for survival and strong coping be haviors. It is unclear if patients with LSS who expressed a positive outlook were health optimists, but it illustrates that these individuals did possess some sort of resilience which may have infl uenced their coping behaviors. Patients’ stories about their positive attitudes along with t he theoretical framework of the Stress and Coping Model suggests that maintaining a positive h ealth outlook may serve as a cognitive buffering mechanism to protect the self from negati ve changes that can occur with health conditions such as LSS.

PAGE 151

136CHAPTER VI CONCLUSIONS Using existing theoretical frameworks, this study explored appraisals and coping behaviors related to the stressors of pain and disa bility in a sub-set of patients with LSS. The mixed methodology in this study enabled results of the qualitative analysis to enhance the findings from the quantitative analysis. This uniqu e approach in studying perceptions of pain and disability in patients with LSS generated possible explanations not found in the statistical models. Guided by the Stress and Coping Model, this study confirms the hypotheses that perceived physical health has an inverse relationsh ip with the degree of disability both prior to and long-term after treatment. In addition, it conf irms that perceived mental health contributes to and has an inverse relationship with disability at entry to treatment and self-efficacy has an inverse relationship to disability post-treatment. Taken together, these results demonstrate that a perception of worse physical health and worse menta l health contributed to the overall burden for disablement in this population. Improvements in dis ability were seen when complemented by coping behaviors for improving physical health beli efs as well as confidence for specific low back related tasks. Little evidence supported the role for social sup port as a coping resource in LSS. Patients with LSS expressed a mixed need, availability, and satisfaction with social support. Patients who did express positive support described tangible and informational support in the form of advice and treatment from health care professionals. In ad dition, patients stated a preference for tangible support from family and friends for obligatory, com mitted, and discretionary activities. Several patients discussed the availability of emotional su pport from family or friends but only a few stated they sought emotional support for their LSS condition. Moreover, most patients expressed not wanting to ask for or receive support for multi ple reasons such as pride, being a burden, and a preference for pushing themselves internally. These inconsistencies in social support can be

PAGE 152

137 found in literature citing both the positive and ne gative effects of social support as a coping resource.282 Overall, the study results regarding social suppor t as a coping resource might be clarified by further qualitative investigation as w ell as a closer investigation in a larger sample of patients with LSS before any conclusions can be dra wn. The qualitative analysis identified several emergen t themes that complemented the a priori constructs of self-efficacy, social support, perce ived physical health and perceived mental health. Eight themes helped to explain why a patie nt perceived greater pain and disability: physical limitations lacking confidence in tasks and activities feeling a lack of control reduced social participation feelings of vulnerability poor mental health frustrations about needing support and financial limitations In addition, 6 themes described patientsÂ’ views o n what helped them reduce the burden of pain and disability posttreatment. These included having greater confidence knowing limitations learning how to manage having a push through attitude having support and having a positive outlook These emergent themes, conceptualized as a reflec tion of coping resources and behaviors at entry and post-tr eatment related to overall appraisals of the stressors of pain and disability in patients with L SS. PatientsÂ’ descriptions of these appraisals and coping in accordance with the Stress and Coping Mod el may help to explain some of the variance that remained unexplained in the data, offering fut ure directions for intervention and management in patients with LSS. Limitations This study had several limitations that must be a cknowledged. The largest limitation of this study was the small sample size. Although prop osed research was originally planned with larger samples recruited for an RCT, challenges in recruitment and retention resulted in having far fewer subjects than intended. Consequently, thi s study involved a cohort of 34 patients at entry to treatment and 20 patients at follow-up. The smaller sample size introduces a few importan t limitations. First, findings from

PAGE 153

138 small samples may not be statistically generalizabl e to the population, with greater chance of Type II errors (i.e., failing to detect an associat ion when there is one). In this study, an a priori sample size determination was calculated based off of information from previous studies. Inferential statistics were computed on the results : particularly correlation, bivariate regression, and multiple regression. Some quantitative results were underpowered based on the a priori sample size determination. Therefore, there was a r educed ability to draw strong conclusions regarding self-efficacy, social support, and self-r ated physical and mental health as they contribute to a patientÂ’s perception of pain and di sability in LSS. To reduce the possibility of a Type II error, this study employed a post hoc analy sis on all models to help identify the magnitude of the observable data. Post-hoc analysis has some advantages as it determines the power estimates based off of the actual sample size and results of this study. However, the danger in being confident of the findings is when in fact the findings represent error. As with any finding, the strength of the results would be impro ved by achieving the same results in a separate repeat study. In addition, repeating the same analy sis once the RCT is completed will determine if the results are substantiated in a larger data set. In addition, the small sample only allowed a limi ted number of relationships to be examined.242 Treatment was not one of the main variables of int erest in this study. However, it is possible that the behavioral and social factors of interest in this study act as moderators in the relationship between treatment and outcomes of pain and disability. Therefore, the interaction of treatment type needs to be investigated upon comple tion of the larger RCT data set. Other important variables in addition to treatment that w ere not tested in this study include the effects of age, gender, comorbidity, and socioeconomic status. Moreover, there may be other potentially confounding variables (e.g. self-esteem) that shoul d be included when examining the relationships between patient characteristics, copi ng behaviors and outcomes. Some of these limitations in sample size were moderated by the ri chness of the dataset, using both quantitative

PAGE 154

139 surveys, 34 pre-treatment and 20 post-treatment, co mbined with 20 open-ended patient interviews. Multiple statistical comparison procedures were c onducted on this sample in an effort to determine the effects of the independent variables of self-efficacy, social support, and self-rated health on the outcomes of pain and disability. Alth ough the Type I error rate was set at alpha level of .05, it is possible with multiple comparis ons to increase the chance of erroneously finding a statistically significant impact. This could infl ate the Type I error rate for the combined multiple tests. The procedures that control for multiplicity to reduced Type I error rate also result in reduced statistical power. Due to the reduced stati stical power already present with the small sample, correction for multiple testing was not con ducted. Analysis focused on either the bivariate model independently examining each indepe ndent variable or the multivariate group model including all 4 independent variables of inte rest. Upon completion of the larger RCT data set, these analyses will need to be compared and co rrection procedures for multiple testing can be performed to appropriately reduce the chance for Ty pe I error. Another important limitation is that all of the m easures were self-report. While the intent of using measures of pain and disability was to cap ture the patientsÂ’ perspective, comparing perceptions of pain and disability to actual physic al measure of pain and/or disability in the future may also improve the understanding of patient perce ptions in LSS. In addition, repeat assessments were performed at the 2 time points (en try and post-treatment). At this time, change scores were not analyzed to determine if there was a change in scores across the treatment and self-management period. Even though each time point was analyzed separately, there was the potential for learning effect by the patients on th e repeat assessments. In addition, the use of selfreport instruments creates certain limitations such as issues with patientsÂ’ memory being incorrect, incomplete memory, and misrepresentation s due to patients attempting to show themselves positively.107 Patients may also have difficult interpreting ques tions and may unintentionally answer items incorrectly. In order to reduce these errors, the addition of the

PAGE 155

140 qualitative interviews helped to triangulate the re sults and uncover patient beliefs, concerns, and their lived experience with LSS which enhanced the analyses in this study. There were also limitations based on the type of analysis used in this study. Previously validated quantitative measures for self-efficacy, social support and perceived physical and mental health status were used to capture appraisal of these factors pre-treatment. The survey items were totaled and each measurement tool was co nverted to a total index score. This allowed the data to be examined through bivariate and multi ple regression analysis and multicollinearity was not found to be present among the predictor var iables, except in 1 of the significant findings which was eliminated. Both bivariate and multiple r egression analyses were used since the factors examined have been minimally examined in other stud ies with patients with LSS. Bivariate analysis allowed for determining effect in isolatio n to identify the strength of the predictor. Next, multivariate regression was used as it is more clin ically appropriate since patients rarely present with single factors in isolation. However, a closer investigation of the significant variables and their interactions is warranted as there was possib le overlap between the measurement tools. A follow up study looking at individual items in the surveys would allow for a better understanding of the nature of the interactions uncovered in this study. Only individuals who were referred to or self-ref erred to 2 spine clinics in Denver, Colorado and Greenville, South Carolina entered the study. Therefore, a crisis of representation was present since the small sample was not randomly selected from the larger population. The representation of findings is limited to patients i n communities of similar geography, social, cultural and economic makeup. Since health insuranc e is not equally distributed locally or across the United States, all individuals with LSS did not have equal opportunity for participation. This sample reflects predominately Caucasians (73%) who were moderate to highly educated (42% graduated college) although only 18% worked full ti me and most were on a fixed income including social security benefits, military benefi ts, and a few with job related retirement income. This clearly does not represent all individuals wit h LSS, particularly those who lack access or

PAGE 156

141 ability to engage in health care services due to li mited financial resources or lack of knowledge of treatment options. However, having multiple sites d oes help to broaden the application of the study to capture a larger clinical representation o f patients with LSS. In addition, the sample was well-matched to the actual LSS population in terms of diagnostic symptoms, gender, and median age. Information gathered from qualitative interviews w as collected and analyzed by only one researcher who was also a clinician and could there fore be biased. Using multiple perspectives from a qualitative research team would make the ana lysis even richer. However, triangulating the survey instruments with the patient interviews help ed to confirm findings as well as uncover interpretations that would not be present in a sing le method design. In addition, the strength of adding the qualitative interviews allows the reader to make connections between the different elements of the study and their own experiences wit h negative health conditions. This allows for a more complex discourse by researchers, clinicians a nd policy makers regarding the many contributors to severity and health outcomes of pai n and disability in LSS. Finally, there were limitations based on the resear ch design. Although validated measures were used, it is difficult to understand the phenom ena observed from only a quantitative standpoint. The addition of qualitative data post-t reatment helped to decipher appraisal and coping factors as they relate to the stressors of p ain and disability. In addition, the qualitative da ta post-treatment allowed for the exploration of the m eanings of these concepts in the patientÂ’s own words. However, given the design of the study, any reports by the patients regarding pretreatment appraisals were based on memory of their experiences upon entry to treatment and therefore subject to recall bias. An optimal design for a follow-up study would capture patientsÂ’ meanings of these concepts upon entry to treatment as well as post-treatment. Future research is warranted to further examine the expressions of pat ients with LSS at entry to treatment and all components of the Stress and Coping Model including primary appraisal, secondary appraisal, reappraisal and processes of coping.

PAGE 157

142 Significance and Future Directions As life expectancy and the number of older adults in our communities continue to increase, so does the need for research into improv ed treatment and effective support for patients with lumbar spinal stenosis. LSS is a chronic and disabling condition that cannot be eliminated but must be managed in the lives of those who have the condition. The effectiveness of treatment options is still in question particularly due to th e limited contributing factors that have been identified up to this point related to the severity and outcomes of pain and disability in these individuals. This study found that measurement of perceived ph ysical and mental health upon entry to treatment may be an effective way to assess severit y for heightened disability in LSS. In addition, assessing perceived physical health and s elf-efficacy for low back related tasks posttreatment may be particularly useful in terms of un derstanding coping behaviors in LSS to direct resources and improve outcomes. Increased education physical therapy, counseling, and motivational support are all ways that these improv ed coping behaviors can be encouraged. To date, this is the first study that has had the specific goal of understanding the appraisal and coping processes of patients with LSS. In addit ion, no intervention study has measured the levels of self-efficacy, self-rated health, and soc ial support before and after treatment to determine the level of pain and disability that can be attributed to these factors alone. The empirical evidence demonstrating that perceived phy sical health and perceived mental health have been identified as significant predictors of p ain and disability at entry to treatment, with perceived physical health and self-efficacy contrib uting post-treatment can be used as a starting point for intervention development. In addition, th e role of social support needs to be further explored with interventions targeted to individuals during management of this chronic disease. Current recommendations for clinical practice guid elines for LBP include the use of active pain coping strategies that decrease fear an d catastrophizing,284 as well as psychosocial intervention.285, 286 At this time, these recommendations are for nonspe cific LBP and none exist

PAGE 158

143 for the chronic condition of LSS. The qualitative findings of this study can help to inform further qualitative investigation into psychosocial factors This can include focus groups and other exploratory research in preparation for the develop ment of an LSS intervention study. Additional qualitative research might use focus groups to conf irm, challenge, or enhance findings from this study and/or identify novel factors that are of mos t relevance to patients with LSS. Added interviews or focus groups could also help to deter mine management strategies for patients needing greater self-efficacy, improved perceptions of their health as well as identifying and altering appraisals of social support as a coping r esource. More research is needed to determine the degree to which the findings of the present study are applicable to the broader population of p atients with LSS, including different clinical sites, geographic locations and a wider variety of patients with differing socioeconomic status and insurance coverage. Also, future research should s tratify subjects by treatment type to determine if psychosocial factors interact as moderators in t he outcomes of pain and disability. Finally, additional research is needed to identify and deter mine effective ways to support patients with LSS using cognitive-behavioral and social strategie s throughout the management continuum. While substantial previous research into physical intervention has been completed in LSS, little is known about the long-term burden of pain and disability in this patient population including all factors that contribute to or amelior ate pain and disability in these individuals. Given limited resources, management programs that meet im portant psychosocial needs of patients with LSS effectively and efficiently can be developed at the entry and post-treatment periods. While more research is needed, the results of this study provide groundwork for identifying the psychosocial needs of patients with LSS through app raisals of the stressors of pain and disability and identification of supporting coping behaviors f or this chronic and disabling condition.

PAGE 159

144 REFERENCES 1. Patel A, Ogle A. Diagnosis and management of acute low back pain. Am Fam Physician. 2000;61(6):1779-1786. 2. Luo X, Pietrobon R, Sun S, Liu G, Hey L. Estimates and Patterns of Direct Health Care Expenditures Among Individuals With Back Pain in th e United States. Spine. 2004;29(1):79. 3. Stewart JR, Thorne RP. Lumbar spinal stenosis. Orthopaedic Nursing. 1983;2(3):35-39. 4. Beizberg A. Michael J. Dorsi, MD. Pain Medicine and Management: Just the Facts. 2004. 5. Ciol M, Deyo R, Howell E, Kreif S. An assessment o f surgery for spinal stenosis: Time trends, geographic variations, complications and re operations. J Am Geriatr Soc. 1996;44(285-290). 6. Whitman JM, Flynn TW, Fritz JM. Nonsurgical manage ment of patients with lumbar spinal stenosis: a literature review and a case ser ies of three patients managed with physical therapy. Phys Med Rehabil Clin N Am. Feb 2003;14(1):77-101, vi-vii. 7. Burton AK, Waddell G. Clinical guidelines in the m anagement of low back pain. Baillieres Clin Rheumatol. Feb 1998;12(1):17-35. 8. Stewart W, Ricci J, Chee E, Morganstein D, Lipton R. Lost Productive Time and Cost Due to Common Pain Conditions in the US Workforce Vol 290: Am Med Assoc; 2003:2443-2454. 9. Airaksinen O, Herno A, Saari T. Surgical treatment of lumbar spinal stenosis: patients' postoperative disability and working capacity. European Spine Journal. 1994;3(5):261264. 10. Fanuele J, Birkmeyer N, Abdu W, Tosteson T, Weinst ein J. The Impact of Spinal Problems on the Health Status of Patients: Have we underestimated the Effect? Spine. 2000;25(12):1509-1514. 11. Onel D, Sari H, Donmez C. Lumbar spinal stenosis: clinical/radiologic therapeutic evaluation in 145 patients. Conservative treatment or surgical intervention? Spine. 1993;18(2):291-298.

PAGE 160

145 12. Vogt MT, Cawthon PM, Kang JD, Donaldson WF, Cauley JA, Nevitt MC. Prevalence of symptoms of cervical and lumbar stenosis among part icipants in the Osteoporotic Fractures in Men Study. Spine. 2006;31(13):1445-1451. 13. Postacchini F. Surgical management of lumbar spina l stenosis. Spine. 1999;24(10):10431047. 14. Szpalski M, Gunzburg R, Pope M. Lumbar Segmental Instability : Lippincott Williams and Wilkins; 1999. 15. Jenis L, An H. Spine Update: Lumbar Foraminal Sten osis. Spine. 2000;25(3):389. 16. Woolsey R. Lumbar spinal stenosis. Semin Neurol. 1986;6(4):385-389. 17. Katz JN, Dalgas M, Stucki G, et al. Degenerative l umbar spinal stenosis. Diagnostic value of the history and physical examination. Arthritis & Rheumatism. 1995;38(9):12361241. 18. Katz JN, Dalgas M, Stucki G, Lipson SJ. Diagnosis of lumbar spinal stenosis. Rheumatic Diseases Clinics of North America. 1994;20(2):471-483. 19. Turner JA, Ersek M, Herron L, Deyo R. Surgery for lumbar spinal stenosis. Attempted meta-analysis of the literature. Spine. 1992;17(1):1-8. 20. Johnsson KE, Rosen I, Uden A. The natural course o f lumbar spinal stenosis. Acta Orthopaedica Scandinavica. Supplementum. 1993;251:67-68. 21. Arnoldi CC, Brodsky AE, Cauchoix J, et al. Lumbar spinal stenosis and nerve root entrapment syndromes. Definition and classification Clin Orthop Relat Res. Mar-Apr 1976(115):4-5. 22. Penning L. Functional pathology of lumbar spinal s tenosis. Clin Biomech. 1992;7:3-17. 23. Spivak JM. Degenerative lumbar spinal stenosis. Journal of Bone & Joint Surgery American Volume. 1998;80(7):1053-1066. 24. Takahashi K, Miyazaki T, Takino T, Matsui T, Tomit a K. Epidural pressure measurements. Relationship between epidural pressur e and posture in patients with lumbar spinal stenosis. Spine. 1995;20(6):650-653.

PAGE 161

146 25. Porter R. Pathophysiology of neurogenic claudicati on. In: Wiesel S, Weinstein J, Herkowitz H, eds. The Lumbar Spine Philadelphia: WB Saunders Co; 1996:717-723. 26. Katz JN, Lipson SJ, Brick GW, et al. Clinical corr elates of patient satisfaction after laminectomy for degenerative lumbar spinal stenosis Spine. 1995;20(10):1155-1160. 27. Stucki G, Liang MH, Lipson SJ, Fossel AH, Katz JN. Contribution of neuromuscular impairment to physical functional status in patient s with lumbar spinal stenosis. Journal of Rheumatology. 1994;21(7):1338-1343. 28. Johnsson KE. Lumbar spinal stenosis. A retrospecti ve study of 163 cases in southern Sweden. Acta Orthopaedica Scandinavica. 1995;66(5):403-405. 29. Johnsson K, Uden A, Rosen I. The effect of decompr ession on the natural course of spinal stenosis. A comparison of surgically treated and untreated patients. Spine. Jun 1991;16(6):615-619. 30. Nagler W, Hausen HS. Conservative management of lu mbar spinal stenosis. Identifying patients likely to do well without surgery. Postgraduate Medicine. 1998;103(4):69-71. 31. Johnsson KE, Rosen I, Uden A. The natural course o f lumbar spinal stenosis. Clinical Orthopaedics & Related Research. 1992(279):82-86. 32. Kirkaldy-Willis W, Bernard T. Managing Low Back Pain: Edited by William H. Kirkaldy-Willis, Thomas N. Bernard : Churchill Livingstone; 1999. 33. Dawson E, Bernbeck J. The surgical treatment of lo w back pain. 34. Porchet F, Vader J, Larequi-Lauber T, Costanza M, Burnand B, Dubois R. The assessment of appropriate indications for laminecto my. Journal of Bone & Joint Surgery, British Volume. 1999;81(2):234-239. 35. Tatarek N. Changing Views of Health: Spinal Stenosis and Its I mpact on Well-being Ohio State University; 1999. 36. Gunzburg R, Szpalski M. The conservative surgical treatment of lumbar spinal stenosis in the elderly. European Spine Journal. 2003;12:176-180. 37. Sculco T. The Lumbar Spine. Orthopaedic Care of the Geriatric Patient. 1985:174.

PAGE 162

147 38. Weinstein JN, Tosteson TD, Lurie JD, et al. Surgic al versus nonsurgical therapy for lumbar spinal stenosis. N Engl J Med. Feb 21 2008;358(8):794-810. 39. Fritz J, Delitto A, Welch W, Erhard R. Lumbar spin al stenosis: A review of current concepts in evaluation, management, and outcome mea surements. Archives of Physical Medicine & Rehabilitation. 1998;79(6):700-708. 40. Whitman J, Flynn T, Fritz J. Nonsurgical managemen t of patients with lumbar spinal stenosis: A literature review and case series of th ree patients managed with physical therapy. Physical Medicine & Rehabilitation Clinics of North America. 2003;14:77-101. 41. Pinsky JL, Jette AM, Branch LG, Kannel WB, Feinlei b M. The Framingham Disability Study: relationship of various coronary heart disea se manifestations to disability in older persons living in the community. American journal of public health. 1990;80(11):13631367. 42. Shakil M, Vaccaro A, Albert T, Klein G. Efficacy o f Conservative Treatment of Lumbar Spinal Stenosis, 1999. 43. Hirsch MS, Liebert RM. The physical and psychologi cal experience of pain: the effects of labeling and cold pressor temperature on three p ain measures in college women. Pain. Jul 1998;77(1):41-48. 44. Gibson J, Grant I, Waddell G. The Cochrane Revieow of Surgery for Lumbar Disc Prolapse and Degenerative Lumbar Spondylosis. Spine. 1999;24(17):1820-1832. 45. Burton AK, Waddell G, Tillotson KM, Summerton N. I nformation and advice to patients with back pain can have a positive effect. A random ized controlled trial of a novel educational booklet in primary care. Spine. Dec 1 1999;24(23):2484-2491. 46. ECRI Health Technology Assessment Group. Treatment of degenerative lumbar spinal stenosis. Rockville, MD: Agency for Healthcare Research and Quality; June 2001 2001. 47. Fritz JM, Delitto A, Welch WC, Erhard RE. Lumbar s pinal stenosis: a review of current concepts in evaluation, management, and outcome mea surements. Archives of Physical Medicine & Rehabilitation. 1998;79(6):700-708. 48. ECRI HealthTechnology Assessment Group. Treatment o f Degenerative Lumbar Spinal Stenosis. Evidence report/Technology Assessment No 32 (Prepared by ECRI under Contract No. 290-97-0020). AHRW Publication No. 01 -E048. Rockville (MD): Agency for Healthcare Research and Quality; June 2001.

PAGE 163

148 49. Waddell G, Feder G, Lewis M. Systematic reviews of bed rest and advice to stay active for acute low back pain. Br J Gen Pract. Oct 1997;47(423):647-652. 50. Hunter S, Fritz J, Brennan G. Variables Associated with Outcomes of Physical Therapy for Patients with Lumbar Spinal Stenosis. J Orthop Sports Phys Ther. January 2005;35(1):A14. 51. Whitman JM, Flynn TW, Childs JD, et al. A comparis on between two physical therapy treatment programs for patients with lumbar spinal stenosis: a randomized clinical trial. Spine. Oct 15 2006;31(22):2541-2549. 52. Waddell G, Newton M, Henderson I, Somerville D, Ma in CJ. A Fear-Avoidance Beliefs Questionnaire (FABQ) and the role of fear-avoidance beliefs in chronic low back pain and disability. Pain. Feb 1993;52(2):157-168. 53. Flynn T, Fritz J, Whitman J, et al. A clinical pre diction rule for classifying patients with low back pain who demonstrate short-term improvemen t with spinal manipulation. Spine. Dec 15 2002;27(24):2835-2843. 54. Buer N, Linton S. Fear-avoidance beliefs and catas trophizing: occurrence and risk factor in back pain and ADL in the general population. Pain. 2002;99(3):485-491. 55. Fritz J, George S. Identifying psychosocial variab les in patients with acute work-related low back pain: the importance of fear-avoidance bel iefs. Physical Therapy. 2002;82(10):973. 56. Brox I, Jens M, Sorensen R, et al. Randomized Clin ical Trial of Lumbar Instrumented Fusion and Cognitive Intervention and Exercises in Patients with Chronic Low Back Pain and Disc Degeneration. Spine. 2003;28(17):1913. 57. Sunderland A, Tinson D, Bradley E, Fletcher D, Lan gton Hewer R, Wade D. Enhanced physical therapy improves recovery of arm function after stroke. A randomised controlled trial. British Medical Journal. 1992;55(7):530-535. 58. Akuthota V, Lento P, Sowa G. Pathogenesis of lumba r spinal stenosis pain: why does an asymptomatic stenotic patient flare? Phys Med Rehabil Clin N Am. Feb 2003;14(1):1728, v. 59. Clinchot DM, Kaplan PE, Lamb JF. Lumbar spinal ste nosis in an elderly patient. Journals of Gerontology. Series A, Biological Scien ces & Medical Sciences. 1998;53(1):M72-75.

PAGE 164

149 60. Aalto TJ, Malmivaara A, Kovacs F, et al. Preoperat ive predictors for postoperative clinical outcome in lumbar spinal stenosis: systema tic review. Spine. Aug 15 2006;31(18):E648-663. 61. Haig AJ, Tong HC, Yamakawa KS, et al. Predictors o f pain and function in persons with spinal stenosis, low back pain, and no back pain. Spine. Dec 1 2006;31(25):2950-2957. 62. Kroenke CH, Rosner B, Chen WY, Kawachi I, Colditz GA, Holmes MD. Functional impact of breast cancer by age at diagnosis. Journal of Clinical Oncology. 2004;22(10):1849-1856. 63. Valdmanis V, Smith DW, Page MR. Productivity and e conomic burden associated with diabetes. American journal of public health. 2001;91(1):129. 64. Brooks PM. Impact of osteoarthritis on individuals and society: how much disability? Social consequences and health economic implication s. Current opinion in rheumatology. 2002;14(5):573-577. 65. Bair M, Robinson R, Katon W, Kroenke K. Depression and Pain Comorbidity A Literature Review. Archives of internal medicine. 2003;163(20):2433-2445. 66. Atlas S, Deyo R, Patrick D, Convery K, Keller R, S inger D. The Quebec Task Force classification for Spinal Disorders and the severit y, treatment, and outcomes of sciatica and lumbar spinal stenosis. Spine. 1996;21(24):2885-2892. 67. Amundsen T, Weber H, Nordal HJ, Magnaes B, Abdelno or M, Lilleas F. Lumbar spinal stenosis: conservative or surgical management?: A p rospective 10-year study. Spine. 2000;25(11):1424-1435; discussion 1435-1426. 68. Atlas SJ, Keller RB, Robson D, Deyo RA, Singer DE. Surgical and nonsurgical management of lumbar spinal stenosis: four-year out comes from the maine lumbar spine study. Spine. Mar 1 2000;25(5):556-562. 69. Drew B, Bhandari M, Kulkarni AV, Louw D, Reddy K, Dunlop B. Reliability in grading the severity of lumbar spinal stenosis. Journal of Spinal Disorders & Techniques. 2000;13(3):253-258. 70. Jensen M, Brant-Zawadzki M, Obuchowski N, Modic M, Malkasian D, Ross J. Magnetic resonance imaging of the lumbar spine in people wit hout back pain. New England Journal of Medicine. 1994;331:69-73.

PAGE 165

150 71. Fairbanks J, Couper J, Davies J, O'Brien J. The Os westry low back pain disability questionnaire. Physiotherapy. 1980;66(271-273). 72. Roland M, Fairbank J. The Roland-Morris Disability Questionnaire and the Oswestry Disability Questionnaire. Spine (Phila Pa 1976). Dec 15 2000;25(24):3115-3124. 73. Ware JE, Jr., Sherbourne CD. The MOS 36-item short -form health survey (SF-36). I. Conceptual framework and item selection. Med Care. Jun 1992;30(6):473-483. 74. Hunt SM, McEwen J, McKenna SP. Measuring health st atus: a new tool for clinicians and epidemiologists. J R Coll Gen Pract. Apr 1985;35(273):185-188. 75. Whitman J, Flynn T, Childs J, et al. A Comparison Between Two Physical Therapy Treatment Programs for Patients With Lumbar Spinal Stenosis: A Randomized Clinical Trial. Spine. 2006;31(22):2541. 76. Sinikallio S, Aalto T, Airaksinen O, et al. Depres sion and associated factors in patients with lumbar spinal stenosis. Disability & Rehabilitation. 2006;28(7):415-422. 77. Sinikallio S, Aalto T, Airaksinen O, Herno A, Krag er H, Viinamaki H. Depressive burden in the preoperative and early recovery phase predicts poorer surgery outcome among lumbar spinal stenosis patients: a one-year p rospective follow-up study. Spine. 2009;34(23):2573. 78. Sinikallio S, Aalto T, Koivumaa-Honkanen H, et al. Life dissatisfaction is associated with a poorer surgery outcome and depression among lumba r spinal stenosis patients: a 2-year prospective study. European Spine Journal. 2009;18(8):1187-1193. 79. Sullivan MJL, Feuerstein M, Gatchel R, Linton SJ, Pransky G. Integrating psychosocial and behavioral interventions to achieve optimal reh abilitation outcomes. Journal of occupational rehabilitation. 2005;15(4):475-489. 80. Steiner WA, Ryser L, Huber E, Uebelhart D, Aeschli mann A, Stucki G. Use of the ICF model as a clinical problem-solving tool in physica l therapy and rehabilitation medicine. Physical Therapy. 2002;82(11):1098-1107. 81. Marinelli RP, Orto AED. The psychological and social impact of disability : Springer Publishing Company; 1999.

PAGE 166

151 82. Chou R, Qaseem A, Snow V, et al. Diagnosis and tre atment of low back pain: a joint clinical practice guideline from the American Colle ge of Physicians and the American Pain Society. Annals of internal medicine. 2007;147(7):478-491. 83. Pincus T, Burton AK, Vogel S, Field AP. A systemat ic review of psychological factors as predictors of chronicity/disability in prospective cohorts of low back pain. Spine. 2002;27(5):E109-E120. 84. Fayad F, Lefevre-Colau MM, Poiraudeau S, et al. [C hronicity, recurrence, and return to work in low back pain: common prognostic factors]. Paper presented at: Annales de readaptation et de medecine physique: revue scienti fique de la Societe francaise de reeducation fonctionnelle de readaptation et de med ecine physique, 2004. 85. Sinikallio S, Aalto T, Airaksinen O, et al. Somati c comorbidity and younger age are associated with life dissatisfaction among patients with lumbar spinal stenosis before surgical treatment. European Spine Journal. 2007;16(7):857-864. 86. Wood DW, Haig AJ, Yamakawa KSJ. Fear of movement/( re) injury and activity avoidance in persons with neurogenic versus vascula r claudication. The Spine Journal. 87. Al-Obaidi S, Nelson R, Al-Awadhi S, Al-Shuwaie N. The role of anticipation and fear of pain in the persistence of avoidance behavior in pa tients with chronic low back pain. Spine. 2000;25(9):1126. 88. Sinikallio S, Lehto SM, Aalto T, Airaksinen O, Vii namaki H. Low sense of coherence interferes with lumbar spinal stenosis patients pos toperative recovery: A prospective oneyear follow-up study. Journal of health psychology. 16(5):783-793. 89. Katz J, Lipson S, Brick G, et al. Clinical correla tes of patient satisfaction after laminectomy for degenerative lumbar spinal stenosis Spine. 1995;20(10):1155-1160. 90. Katz J, Stucki G, Lipson S, Fossel A, Grobler L, W einstein J. Predictors of surgical outcome in degenerative lumbar spinal stenosis. Spine. 1999;24(21):2229-2233. 91. Koivumaa-Honkanen H-T. Life satisfaction as a health predictor : Kupio University Printing Office; 1998. 92. Yeo M, Sawyer S. Chronic illness and disability. BMJ. 2005;330(7493):721-723. 93. Batti MC, Jones CA, Schopflocher DP, Hu RW. Heal th-related quality of life and comorbidities associated with lumbar spinal stenosi s. The Spine Journal.

PAGE 167

152 94. Gunzburg R, Keller T, Szpalski M, Vandeputte K, Sp ratt K. Clinical and psychofunctional measures of conservative decompres sion surgery for lumbar spinal stenosis: a prospective cohort study. European Spine Journal. 2003;12(2):197-204. 95. Pahl MA, Brislin B, Boden S, et al. The impact of four common lumbar spine diagnoses upon overall health status. 2006. 96. Lemcke J, Kiatzer S, Kiatzer R, Meier U. PLIF and ALIF for the degenerative spondylolisthesis of the lumbar spine. Zeitschrift Far Orthopadie Und Ihre Grenzgebiete. 2007;145(1):48-54. 97. Sharma R, Lutchman L, Kabir C, Crawford RJ. Underc utting Laminectomy and Medial facetectomy for Lumbar Spinal Stenosis: Clinical an d Radiological results. The Internet Journal of Orthopedic Surgery. 2008;9(1). 98. Slosar PJ, Reynolds JB, Schofferman J, Goldthwaite N, White AH, Keaney D. Patient satisfaction after circumferential lumbar fusion. Spine. 2000;25(6):722. 99. Orlin JR, Klevmark B, Bjornsen L, Hermansen P, San chez M. Two screening tests for urinary voiding dysfunction used in 209 consecutive patients undergoing lumbar spine operations. Scandinavian journal of urology and nephrology. 2010;44(2):106-112. 100. Sekiguchi M, Wakita T, Otani K, et al. Development and validation of a symptom scale for lumbar spinal stenosis. Spine. 37(3):232. 101. Painter JE, Borba CPC, Hynes M, Mays D, Karen Glan z Ph D MPH. The use of theory in health behavior research from 2000 to 2005: a syste matic review. Annals of Behavioral Medicine. 2008;35(3):358-362. 102. Allegrante JP, Marks R. Self-efficacy in managemen t of osteoarthritis. Rheumatic Disease Clinics of North America. 2003;29(4):747-768. 103. Taal E, Rasker J, Seydel E, Wiegman O. Health stat us, adherence with health recommendations, self-efficacy and social support i n patients with rheumatoid arthritis. Patient Education and Counseling. 1993;20(2-3):63-76. 104. Andersen B. Biobehavioral Outcomes Following Psych ological Interventions for Cancer Patients. Journal of Consulting and Clinical Psychology. 2002;70(3):590-610. 105. Black EK, White CA. Fear of recurrence, sense of c oherence and posttraumatic stress disorder in haematological cancer survivors. Psycho Oncology. 2005;14(6):510-515.

PAGE 168

153 106. Blank TO, Bellizzi KM. A gerontologic perspective on cancer and aging. Cancer. Jun 1 2008;112(11 Suppl):2569-2576. 107. Lazarus RS, Folkman S. Stress, appraisal, and coping New York: Springer Pub. Co.; 1984. 108. Shanahan MJ, Macmillan R. Biography and the sociological imagination : contex ts and contingencies 1st ed. New York: W.W. Norton; 2008. 109. Elder Jr GH. Time, human agency, and social change : Perspectives on the life course. Soc Psychol Q. 1994;57(1):4-15. 110. Rowland JH. Developmental Stage and Adaptation: Ad ult Model In: Holland JC, Rowland JH, eds. Handbook of psychooncology : psychological care of the patient with cancer New York: Oxford University Press; 1989:xiv, 785 p. 111. Bandura A. Toward a unifying theory of behavioral change. Psychol Rev. 1977;84(2):191-215. 112. Bandura A, ed. Social Learning Theory Englewood: Prentice-Hall; 1977. 113. Schwarzer R, Leppin A. Social support and health: A theoretical and empirical overview. Journal of social and personal relationships. 1991;8(1):99-127. 114. Glanz K, Rimer BK, Viswanath K. Health behavior and health education: theory, research, and practice : John Wiley & Sons; 2008. 115. Turner JA, Clancy S, Vitaliano PP. Relationships o f stress, appraisal and coping, to chronic low back pain. Behaviour research and therapy. 1987;25(4):281-288. 116. Rosenstiel AK, Keefe FJ. The use of coping strateg ies in chronic low back pain patients: relationship to patient characteristics and current adjustment. Pain. 1983;17(1):33-44. 117. Turner JA, Clancy S. Strategies for coping with ch ronic low back pain: Relationship to pain and disability. Pain. 1986;24(3):355-364. 118. Jensen MP, Karoly P. Control beliefs, coping effor ts, and adjustment to chronic pain. Journal of Consulting and Clinical Psychology. 1991;59(3):431.

PAGE 169

154 119. Descartes R. The philosophical writings of Descartes. Vol 2: Cambridge University Press; 1985. 120. Turk DC, Flor H. Chtonic Pain: A Biobehaviotal Pet specuve. Psychosocial factors in pain: Critical perspectives. 1999;1:18. 121. Boden S. The Use of Radiographic Imaging Studies i n the Evaluation of Patients Who Have Degenerative Disorders of the Lumbar Spine. Journal of Bone & Joint Surgery [Am]. January 1996;78-A(1):114-125. 122. Mechanic D. The concept of illness behavior. Journal of Chronic Diseases. 1962;15:189. 123. Engel GL. The clinical application of the biopsych osocial model. Am J Psychiatry. 1980;137(5):535-544. 124. Waddell G. A new clinical model for the treatment of low back pain. Spine. 1987;12(7):632-644. 125. Asmundson GJG, Wright KD. Biopsychosocial approach es to pain. Pain: psychological perspectives. 2004:35-57. 126. Sharp TJ, Harvey AG. Chronic pain and posttraumati c stress disorder: mutual maintenance? Clinical Psychology Review; Clinical Psychology Rev iew. 2001. 127. Lazarus RS, Folkman S. Stress, appraisal, and coping : Springer Publishing Company; 1984. 128. Price DD, Harkins SW, Baker C. Sensory-affective r elationships among different types of clinical and experimental pain. Pain. 1987;28(3):297-307. 129. Jensen MP, Turner JA, Romano JM, Lawler BK. Relati onship of pain-specific beliefs to chronic pain adjustment. Pain. 1994;57(3):301-309. 130. Von Korff M, Balderson BHK, Saunders K, et al. A t rial of an activating intervention for chronic back pain in primary care and physical ther apy settings. Pain. 2005;113(3):323330. 131. Nagi S. Some conceptual issues in disability and r ehabilitation. Sociology and Rehabilitation. Washington, DC: American Sociologic al Association. 1965:100–113.

PAGE 170

155 132. Nagi S. Disability concepts revisited: implication s for prevention. Disability in America: Toward a National Agenda for Prevention. 1991:309-327. 133. Verbrugge LM, Jette AM. The disablement process. Social science & medicine. 1994;38(1):1-14. 134. Verbrugge LM, Gruber-Baldini AL, Fozard JL. Age di fferences and age changes in activities: Baltimore Longitudinal Study of Aging. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences. 1996;51(1):S30. 135. Verbrugge LM. Disability. Rheumatic diseases clinics of North America. 1990;16(3):741. 136. Oliver M, Sapey, B. Thomas, P. Social work with disabled people 4th ed; 2012. 137. Bichkenbach J, Chatterji S, Badley E, Ustun T. Mod els of disablement, universalism and the international clssification of impairments, dis abilities and handicaps. Soc Sci Med. 1999;48:1173-1187. 138. Oliver M. Defining impairment and disability: issu es at stake. Exploring the divide: Illness and disability. 1996:39-54. 139. Stucki G, Ewert T. How to assess the impact of art hritis on the individual patient: the WHO ICF. Ann Rheum Dis. May 2005;64(5):664-668. 140. Threats T. Access for persons with neurogenic comm unication disorders: Influences of personal and environmental factors of the ICF. Aphasiology. 2007;21(1):67-80. 141. World Health Organization. Classification AS, Term inology T. ICIDH-2: International Classification of Functioning, Disability and Healt h: Final Draft, Full Version: Classification, Assessment, Surveys and Terminology Team, World Health Organization; 2001. 142. Waddell G, McCulloch JA, Kummel E, Venner RM. Nono rganic physical signs in lowback pain. Spine. Mar-Apr 1980;5(2):117-125. 143. Waddell G, Pilowsky I, Bond MR. Clinical assessmen t and interpretation of abnormal illness behaviour in low back pain. Pain. Oct 1989;39(1):41-53. 144. Krismer M, Van Tulder M. Low back pain (non-specif ic). Best Practice & Research Clinical Rheumatology. 2007;21(1):77-91.

PAGE 171

156 145. Sigl T, Cieza A, Brockow T, Chatterji S, Kostanjse k N, Stucki G. Content comparison of low back pain-specific measures based on the Intern ational Classification of Functioning, Disability and Health (ICF). The Clinical journal of pain. 2006;22(2):147-153. 146. Cieza A, Stucki G, Weigl M, et al. ICF Core Sets f or low back pain. Journal of Rehabilitation Medicine. 2004;36(0):69-74. 147. Gatchel RJ, Polatin PB, Mayer TG. The dominant rol e of psychosocial risk factors in the development of chronic low back pain disability. Spine. 1995;20(24):2702-2709. 148. Waddell G. Occupational low-back pain, illness beh avior, and disability. Spine. Jun 1991;16(6):683-685. 149. Lutgendorf S, Costanzo E. Psychoneuroimmunology an d health psychology: An integrative model. Brain Behavior and Immunity. 2003;17(4):225-232. 150. Bandura A, ed. Social foundations of thought and action: social co gnitive theory Englewood: Prentice-Hall; 1986. 151. Miller N, Dollard J, eds. Social Learning and Imitation New Haven: Yale University Press; 1941. 152. Bandura A. Social cognitive theory: An agentive pe rspective. Annual Review of Psychology. 2001;52:1-26. 153. Bandura A, ed. Self-Efficacy: The Exercise of Control New York: W. H. Freeman; 1997. 154. Bandura A, O'Leary A, Barr-Taylor C, Guauthier J, Gossard D. Percieved self-efficacy and pain control: Opiod and nonopiod mechanisms. J Pers Soc Psychol. 1989;53:563571. 155. Pintrich P, Schunk D, eds. Motivation in education: Theory, research, and appl ications Englewood: Prentice-Hall; 1996. 156. Davis F, Yates B. Self-efficacy expectancies versu s outcome expectancies as determinants of performance deficits and depressive affect. Cognitive Therapy and Research. 1982;6(1):23-35.

PAGE 172

157 157. Keefe FJ, Lefebvre JC, Maixner W, Salley AN, Caldw ell DS. Self-efficacy for arthritis pain: Relationship to perception of thermal laborat ory pain stimuli. Arthritis & Rheumatism. 2005;10(3):177-184. 158. Weisenberg M, Raz T, Hener T. The influence of fil m-induced mood on pain perception. Pain. 1998;76(3):365-375. 159. Woby SR, Roach NK, Urmston M, Watson PJ. The relat ion between cognitive factors and levels of pain and disability in chronic low ba ck pain patients presenting for physiotherapy. European Journal of Pain. 2007;11(8):869-877. 160. Bastone EC, Kerns RD. Effects of self-efficacy and perceived social support on recoveryrelated behaviors after coronary artery bypass graf t surgery. Annals of Behavioral Medicine. 1995;17(4):324-330. 161. Allen JK, Becker DM, Swank RT. Factors related to functional status after coronary artery bypass surgery. Heart & lung: the journal of critical care. 1990;19(4):337-343. 162. Waldrop D, Lightsey Jr OR, Ethington CA, Woemmel C A, Coke AL. Self-efficacy, optimism, health competence, and recovery from orth opedic surgery. Journal of Counseling Psychology. 2001;48(2):233. 163. Dohnke B, Knauper B, MalleraeFahrnow W. Perceived self-efficacy gained from, and health effects of, a rehabilitation program after h ip joint replacement. Arthritis Care & Research. 2005;53(4):585-592. 164. Orbell S, Johnston M, Rowley D, Davey P, Espley A. Self-efficacy and goal importance in the prediction of physical disability in people following hospitalization: A prospective study. British journal of health psychology. 2001;6(1):25-40. 165. Scherer Y, Schmieder L. The effect of a pulmonary rehabilitation program on selfefficacy, perception of dyspnea, and physical endur ance. Heart & Lung: Journal of Acute & Critical Care. 1997;26(1):15. 166. Carlson J, Norman G, Feltz D, Franklin B, Johnson J, Locke S. Self-efficacy, Psychosocial Factors, and Exercise Behavior in Trad itional Versus Modified Cardiac Rehabilitation. Journal of Cardiopulmonary Rehabilitation and Preve ntion. 2001;21(6):363. 167. Jeng C, Braun L. The influence of self-efficacy on exercise intensity, compliance rate and cardiac rehabilitation outcomes among coronary arte ry disease patients. Prog Cardiovasc Nurs. 1997;12(1):13-24.

PAGE 173

158 168. Rejeski W, Ettinger Jr W, Martin K, Morgan T. Trea ting disability in knee osteoarthritis with exercise therapy: a central role for self-effi cacy and pain. Arthritis Care Res. 1998;11(2):94-101. 169. Tinetti ME, Mendes de Leon CF, Doucette JT, Baker DI. Fear of falling and fall-related efficacy in relationship to functioning among commu nity-living elders. J Gerontol. May 1994;49(3):M140-147. 170. Salbach NM, Mayo NE, Robichaud-Ekstrand S, Hanley JA, Richards CL, WoodDauphinee S. Balance self-efficacy and its relevanc e to physical function and perceived health status after stroke. Arch Phys Med Rehabil. Mar 2006;87(3):364-370. 171. Lackner JM, Carosella AM. The relative influence o f perceived pain control, anxiety, and functional self efficacy on spinal function among p atients with chronic low back pain. Spine. Nov 1 1999;24(21):2254-2260; discussion 2260-2251. 172. Strahl C, Kleinknecht RA, Dinnel DL. The role of p ain anxiety, coping, and pain selfefficacy in rheumatoid arthritis patient functionin g. Behaviour research and therapy. 2000;38(9):863-873. 173. Lackner J, Carosella A. The Relative Influence of Perceived Pain Control, Anxiety, and Functional Self Efficacy on Spinal Function Among P atients With Chronic Low Back Pain. Spine. 1999;24(21):2254. 174. Putnam R. Bowling Alone: The Collapse and Revival of American Community : Simon & Schuster; 2000. 175. Guidi L, Tricerri A, Frasca D, Vangeli M, Errani A Bartoloni C. Psychoneuroimmunology and Aging. Logo. 1998;44(5). 176. Berkman L. The relationship of social networks and social support to morbidity and mortality. Social support and health. 1985:241-262. 177. Cohen S, Gottlieb B, Underwood L. Social relations hips and health. Social support measurement and intervention: A guide for health an d social scientists. 2000:3–25. 178. Heaney C, Israel B. Social networks and social sup port. Health Behavior and Health Education: Theory, Research, and Practice. 2002:185-209. 179. Cohen S. Stress, social support, and disorder. The meaning and measurement of social support. 1992:109-124.

PAGE 174

159 180. Krause N, Borawski-Clark E. Clarifying the functio ns of social support in later life. Research on Aging. 1994;16(3):251-279. 181. Cohen S, Lichtenstein E. Perceived stress, quittin g smoking, and smoking relapse. Health Psychology. 1990;9(4):466-478. 182. Coriell M, Cohen S. Concordance in the Face of a S tressful Event: When Do Members of a Dyad Agree That One Person Supported the Other? Journal of personality and social psychology. 1995;69:289-289. 183. Masters K, Stillman A, Spielmans G. Specificity of support for back pain patients: Do patients care who provides what? Journal of Behavioral Medicine. 2007;30(1):11-20. 184. Andr-Petersson L, Engstrm G, Hedblad B, Janzon L Rosvall M. Social support at work and the risk of myocardial infarction and stro ke in women and men. Social Science & Medicine. 2007;64(4):830-841. 185. Amick T, Ockene J. The role of social support in t he modification of risk factors for cardiovascular disease. Social support and cardiovascular disease. 1994:259-278. 186. Glass T, Matchar D, Belyea M, Feussner J. Impact o f social support on outcome in first stroke. Stroke. 1993;24(1):64-70. 187. Evans R, Bishop D, Matlock A, Stranahan S, Smith G Halar E. Family interaction and treatment adherence after stroke. Arch Phys Med Rehabil. 1987;68(8):513-517. 188. Friedland J, McColl M. Social support and psychoso cial dysfunction after stroke: buffering effects in a community sample. Arch Phys Med Rehabil. 1987;68(8):475-480. 189. McLeroy K, DeVellis R, DeVellis B, Kaplan B, Toole J. Social Support and Physical Recovery in a Stroke Population. Journal of Social and Personal Relationships. 1984;1(4):395. 190. Morris P, Robinson R, Raphael B, Bishop D. The rel ationship between the perception of social support and post-stroke depression in hospit alized patients. Psychiatry. 1991;54(3):306-316. 191. Stephens M. Social Networks as Assets and Liabilit iesin Recovery From Stroke by GeriatricPatients.

PAGE 175

160 192. Angeleri F, Angeleri V, Foschi N, Giaquinto S, Nol fe G. The influence of depression, social activity, and family stress on functional ou tcome after stroke. Stroke. 1993;24(10):1478-1483. 193. King R. Quality of Life After Stroke. Stroke. 1996;27(9):1467-1472. 194. Handen B. The influence of social support factors on the well-being of the elderly. Handbook of clinical behavior therapy with the elde rly client. 1991:121-139. 195. Kishi Y, Robinson R, Kosier J. Suicidal Ideation A mong Patients With Acute LifeThreatening Physical Illness. Psychosomatics. 2001;42:382-390. 196. Rao R, Jackson S, Howard R. Depression in older pe ople with mild stroke, carotid stenosis and peripheral vascular disease: a compari son with healthy controls. International Journal of Geriatric Psychiatry. 2001;16(2):175-183. 197. Marottoli R. Predictors of mortality and instituti onalization after hip fracture: the New Haven EPESE cohort. Established Populations for Epi demiologic Studies of the Elderly Vol 84: Am Public Health Assoc; 1994:1807-1812. 198. Kawachi I, Colditz G, Ascherio A, et al. A prospec tive study of social networks in relation to total mortality and cardiovascular dise ase in men in the USA. Journal of Epidemiology & Community Health. 1996;50(3):245-251. 199. Lacasse Y, Rousseau L, Maltais F. Prevalence of De pressive Symptoms and Depression in Patients With Severe Oxygen-Dependent Chronic Ob structive Pulmonary Disease. Journal of Cardiopulmonary Rehabilitation and Preve ntion. 2001;21(2):80. 200. Littlefield C, Rodin G, Murray M, Craven J. Influe nce of functional impairment and social support on depressive symptoms in persons wi th diabetes. Health Psychol. 1990;9(6):737-749. 201. Yates B. The relationships among social support an d shortand long-term recovery outcomes in men with coronary heart disease. Research in nursing & health. 1995;18(3):193-203. 202. Gulick E. Emotional Distress and Activities of Dai ly Living Functioning in Persons With Multiple Sclerosis. Nursing Research. 2001;50(3):147. 203. Weinberger M, Terney W, Booher P, Hiner S. Social support, stress and functional status in patients with osteoarthritis. Social science & medicine(1982). 1990;30(4):503-508.

PAGE 176

161 204. Ware Jr JE, Sherbourne CD. The MOS 36-item short-f orm health survey (SF-36): I. Conceptual framework and item selection. Medical care. 1992:473-483. 205. Garrity TF, Somes GW, Marx MB. Factors influencing self-assessment of health. Social Science & Medicine. Part A: Medical Psychology & Me dical Sociology. 1978;12:77-81. 206. Rodin J, McAvay G. Determinants of change in perce ived health in a longitudinal study of older adults. Journal of Gerontology. 1992;47(6):P373-P384. 207. Stewart AL, Hays RD, Ware JE. The MOS short-form g eneral health survey: reliability and validity in a patient population. Medical care. 1988;26(7):724-735. 208. Kaplan GA, Seeman TE, Cohen RD, Knudsen LP, Guraln ik J. Mortality among the elderly in the Alameda County Study: behavioral and demographic risk factors. American journal of public health. 1987;77(3):307-312. 209. Gold M, Franks P, Erickson P. Assessing the health of the nation: the predictive validity of a preference-based measure and self-rated health Medical care. 1996;34(2):163-177. 210. Menec VH, Chipperfield JG, Perry RP. Self-percepti ons of health: A prospective analysis of mortality, control, and health. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences. 1999;54(2):P85-P93. 211. Bailis DS, Segall A, Chipperfield JG. Two views of self-rated general health status. Social science & medicine. 2003;56(2):203-217. 212. Ferraro KF, Farmer MM, Wybraniec JA. Health trajec tories: long-term dynamics among black and white adults. Journal of Health and Social behavior. 1997:38-54. 213. Lazarus RS. Patterns of adjustment. 1976. 214. Lazarus RS, Cohen JB, Folkman S, Kanner A, Schaefe r C. Psychological stress and adaptation: Some unresolved issues. Selye's guide to stress research. 1980;1:90-117. 215. Brooks WB, Jordan JS, Divine GW, Smith KS, Neelon FA. The impact of psychologic factors on measurement of functional status: assess ment of the sickness impact profile. Medical care. 1990:793-804.

PAGE 177

162 216. Hicks GE, Manal TJ. Psychometric properties of com monly used low back disability questionnaires: are they useful for older adults wi th low back pain? Pain Med. Jan 2009;10(1):85-94. 217. Portney LG. Foundations Of Clinical Research: Appl ications To Practice Author: Leslie Gross Portney, Mary P. Watkins, Pu. 2008. 218. Onwuegbuzie AJ, Leech NL. Validity and qualitative research: An oxymoron? Quality & Quantity. 2007;41(2):233-249. 219. Koc Z, Ozcakir S, Sivrioglu K, Gurbet A, Kucukoglu S. Effectiveness of physical therapy and epidural steroid injections in lumbar spinal st enosis. Spine (Phila Pa 1976). May 1 2009;34(10):985-989. 220. Bryman A. Mixed methods. Vol 4: Sage; 2006. 221. Onwuegbuzie AJ, Johnson RB. The validity issue in mixed research. Research in the Schools. 2006;13(1):48-63. 222. Creswell JW, Klassen AC, Plano Clark VL, Smith KC. Best practices for mixed methods research in the health sciences. Bethesda (Maryland): National Institutes of Health. 223. Morse JM. Principles of mixed methods and multimet hod research design. Handbook of mixed methods in social and behavioral research. 2003:189-208. 224. Teddlie C, Tashakkori A. Major issues and controve ries inthe use of mixed methods in the social and behvioral sciences. Handbook of mixed methods in social & behavioral research. 2003:3-50. 225. Whtiman JM HJ, Cleland JA, Wainner RS, Childs JD. Psychometric Properties of the Modified Oswestry Disability Index, Numeric Pain Ra ting Scale, Modified Swiss Spinal Stenosis Scale, and Patient Specific Functional Sca le in Patients with Lumbar Spinal Stenosis. Am J Phys Med In review. 226. Peters ML, Patijn J, Lame I. Pain assessment in yo unger and older pain patients: psychometric properties and patient preference of f ive commonly used measures of pain intensity. Pain Med. Oct-Nov 2007;8(7):601-610. 227. Gagliese L, Weizblit N, Ellis W, Chan VW. The meas urement of postoperative pain: a comparison of intensity scales in younger and older surgical patients. Pain. Oct 2005;117(3):412-420.

PAGE 178

163 228. Yamada K LR, Popovich JM, Beneck GJ, Selkowitz DM, Kulig K. The Low Back Activity Confidence Scale (LoBACS): Development, Te st-Retest Reliability, And Preliminary Validation. Journal of Orthopaedic & Sports Physical Therapy. 2007;37(1):A46. 229. Sherbourne CD, Stewart AL. The MOS social support survey. Soc Sci Med. 1991;32(6):705-714. 230. McHorney CA, Ware Jr JE, Raczek AE. The MOS 36-Ite m Short-Form Health Survey (SF-36): II. Psychometric and clinical tests of val idity in measuring physical and mental health constructs. Medical care. 1993:247-263. 231. Ware Jr JE, Kosinski M, Bayliss MS, McHorney CA, R ogers WH, Raczek A. Comparison of methods for the scoring and statistic al analysis of SF-36 health profile and summary measures: summary of results from the Medic al Outcomes Study. Medical care. 1995. 232. Garratt A, Schmidt L, Mackintosh A, Fitzpatrick R. Quality of life measurement: bibliographic study of patient assessed health outc ome measures. BMJ. 2002;324(7351):1417. 233. Grevitt M, Khazim R, Webb J, Mulholland R, Shepper d J. The short form-36 health survey questionnaire in spine surgery. J Bone Joint Surg Br. Jan 1997;79(1):48-52. 234. McHorney CA, Ware JE, Jr., Raczek AE. The MOS 36-I tem Short-Form Health Survey (SF-36): II. Psychometric and clinical tests of val idity in measuring physical and mental health constructs. Med Care. Mar 1993;31(3):247-263. 235. Taylor SJ, Taylor AE, Foy MA, Fogg AJ. Responsiven ess of common outcome measures for patients with low back pain. Spine. Sep 1 1999;24(17):1805-1812. 236. McHorney CA WJ, Raczek AE. The MOS 36-item short f orm health survey (SF-36): II. Psychometric and clinical tests of validity measuri ng physical and mental health constructs. Med Care. 1993;31:247-263. 237. Ware JE, Kosinski M, Keller S. SF-36 physical and mental health summary scales: a user's manual : Health Assessment Lab.; 1994. 238. Ware Jr JE. The status of health assessment 1994. Annual review of public health. 1995;16(1):327-354.

PAGE 179

164 239. Portney L, Watkins M. Foundations of Clinical Research: Applications to P ractice. 2nd ed. Upper Saddle River: Prentice Hall Health; 2000. 240. Mac Nally R. Regression and model-building in cons ervation biology, biogeography and ecology: the distinction between and reconciliation of predictive and explanatory models. Biodiversity and Conservation. 2000;9(5):655-671. 241. Budescu DV. Dominance analysis: A new approach to the problem of relative importance of predictors in multiple regression. Psychological Bulletin. 1993;114(3):542. 242. Leech NL, Barrett KC, Morgan GA. SPSS for intermediate statistics: Use and interpretation : Lawrence Erlbaum; 2004. 243. Cohen J, ed. Statistical power analysis for the behavioural scie nces. Hillsdale: Lawrence Erlbaum Associates; 1988. 244. Ryan GW, Bernard HR. Data management and analysis methods. Handbook of qualitative research. 2000;2:769-802. 245. Brett JA, Heimendinger J, Boender C, Morin C, Mars hall JA. Using ethnography to improve intervention design. American Journal of Health Promotion. 2002;16(6):331340. 246. Patton M, ed. Qualitative Research and Evaluation Methods. 3 ed: SAGE; 2002. 247. Lazarus R. Stress, Appraisal, and Coping. 1966. 248. World Health O. International Classification of Fu nctioning, Disability and Health. Geneva. World Health Organization. 2001:1-303. 249. Fritz J, Erhard R, Vignovic M. A nonsurgical treat ment approach for patients with lumbar spinal stenosis. Phys Ther. 1997;77(9):962-973. 250. Weinstein PR. Diagnosis and management of lumbar s pinal stenosis. Clinical Neurosurgery. 1983;30:677-697. 251. Cummins RA. Objective and subjective quality of li fe: An interactive model. Social Indicators Research. 2000;52(1):55-72.

PAGE 180

165 252. Balague F, Mannion AF, Pellise F, Cedraschi C. Non -specific low back pain. The Lancet. 2012;379(9814):482-491. 253. Tobin SS, Lieberman MA. Last home for the aged : Jossey-Bass San Francisco; 1976. 254. Rosow I. Social integration of the aged : Free Press; 1967. 255. Martin J, Meltzer H, Elliot D. The prevalence of disability among adults. OPCS Su rveys of Great Britain. London: HMSO; 1988. 1. 256. Hoffman KE, McCarthy EP, Recklitis CJ, Ng AK. Psyc hological distress in long-term survivors of adult-onset cancer: results from a nat ional survey. Archives of Internal Medicine. Jul 27 2009;169(14):1274-1281. 257. Berthelot JM, Bertrand-Vasseur A, Rodet D, Maugars Y, Prost A. Lumbar spinal stenosis: a review. Revue Du Rhumatisme, English Edition. 1997;64(5):315-325. 258. Bridwell K. Lumbar Spinal Stenosis. Diagnosis, Man agement, and Treatment. Clinics in Geriatric Medicine. 1994;10(4):677-701. 259. Conway J, Tomkins CC, Haig AJ. Walking assessment in people with lumbar spinal stenosis: capacity, performance, and self-report me asures. Spine J. Sep;11(9):816-823. 260. Lai S-M, Perera S, Duncan PW, Bode R. Physical and Social Functioning After Stroke Comparison of the Stroke Impact Scale and Short For m-36. Stroke. 2003;34(2):488-493. 261. Hanson Frost M, Suman VJ, Rummans TA, et al. Physi cal, psychological and social well being of women with breast cancer: the influence of disease phase. PsychoOncology. 2000;9(3):221-231. 262. Sullivan MB, Sullivan LG, Kral JG. Quality of life assessment in obesity: physical, psychological, and social function. Gastroenterology Clinics of North America. 1987;16(3):433-442. 263. Stuck AE, Walthert JM, Nikolaus T, Bla CJ, Hohma nn C, Beck JC. Risk factors for functional status decline in community-living elder ly people: a systematic literature review. Social science & medicine. 1999;48(4):445-469. 264. Waddell G. 1987 Volvo award in clinical sciences. A new clinical model for the treatment of low-back pain. Spine. Sep 1987;12(7):632-644.

PAGE 181

166 265. McCracken LM, Turk DC. Behavioral and cognitive-be havioral treatment for chronic pain: outcome, predictors of outcome, and treatment process. Spine. Nov 15 2002;27(22):2564-2573. 266. Craig AD, Bushnell MC, Zhang ET, Blomqvist A. A th alamic nucleus specific for pain and temperature sensation. Nature. 1994;372(6508):770-773. 267. Waddell G, Main CJ, Morris EW, Di Paola M, Gray IC Chronic low-back pain, psychologic distress, and illness behavior. Spine. Mar 1984;9(2):209-213. 268. Doeglas D, Suurmeijer T, Briancon S, et al. An int ernational study on measuring social support: interactions and satisfaction. Social science & medicine. 1996;43(9):1389-1397. 269. Lazarus R, Folkman S. Coping and adaptation. The handbook of behavioral medicine. 1984:282-325. 270. Menec VH, Chipperfield JG. Remaining active in lat er life the role of locus of control in seniors' leisure activity participation, health, an d life satisfaction. Journal of Aging and Health. 1997;9(1):105-125. 271. Chipperfield JG, Greenslade L. Perceived control a s a buffer in the use of health care services. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences. 1999;54(3):P146-P154. 272. Strand V, Crawford B, Singh J, Choy E, Smolen JS, Khanna D. Use of spydergrams to present and interpret SF-36 health-related quality of life data across rheumatic diseases. Annals of the rheumatic diseases. 2009;68(12):1800-1804. 273. Walsh TL, Hanscom B, Lurie JD, Weinstein JN. Is a condition-specific instrument for patients with low back pain/leg symptoms really nec essary?: the responsiveness of the Oswestry Disability Index, MODEMS, and the SF-36. Spine. 2003;28(6):607-615. 274. Garratt AM, Moffett JK, Farrin AJ. Responsiveness of generic and specific measures of health outcome in low back pain. Spine. 2001;26(1):71-77. 275. Stucki G, Liang M, Fossel A, Katz J. Relative resp onsiveness of condition-specific and generic health status measures in degenerative lumb ar spinal stenosis. Journal of Clinical Epidemiology. 1995;48(11):1369-1378. 276. Taylor SJ, Taylor AE, Foy MA, Fogg AJB. Responsive ness of common outcome measures for patients with low back pain. Spine. 1999;24(17):1805.

PAGE 182

167 277. Garratt AM, Ruta DA, Abdalla MI, Russell IT. Respo nsiveness of the SF-36 and a condition-specific measure of health for patients w ith varicose veins. Quality of Life Research. 1996;5(2):223-234. 278. Khanna D, Tsevat J. Health-related quality of life -an introduction-page 2. American Journal of Managed Care. 2007;13(Suppl 9):S218-S223. 279. Thoits PA. Conceptual, methodological, and theoret ical problems in studying social support as a buffer against life stress. Journal of Health and Social behavior. 1982:145159. 280. Murawski BJ, Penman D, Schmitt M. Social support i n health and illness: The concept and its measurement. Cancer nursing. 1978;1(5):365-372. 281. Thoits PA. Stress, coping, and social support proc esses: Where are we? What next? Journal of Health and Social behavior. 1995:53-79. 282. Suls J. Social support, interpersonal relations, a nd health: Benefits and liabilities. Social psychology of health and illness. 1982:255-277. 283. Chipperfield JG. Incongruence between Health Perce ptions and Health Problems Implications for Survival among Seniors. Journal of Aging and Health. 1993;5(4):475496. 284. Delitto A, George S, Van Dillen L, et al. Low Back Pain. Journal of Orthopaedic & Sports Physical Therapy. 2013/10/11 2012;42(4):A1-A57. 285. Hoffman BM, Papas RK, Chatkoff DK, Kerns RD. Metaanalysis of psychological interventions for chronic low back pain. Health psychology. 2007;26(1):1. 286. Ostelo RW, Van Tulder MW, Vlaeyen JW, Linton SJ, M orley SJ, Assendelft WJ. Behavioural treatment for chronic low-back pain. Cochrane Database Syst Rev. 2005;1(1). 287. Boden S, Davis D, Dina T, Patronas N, Wiesel S. Ab normal magnetic resonance scans of the lumbar spine in asymptomatic subjects: a prospe ctive investigation. Journal of Bone and Joint Surgery [Am]. 1990;72:403-408. 288. Botwin KP, Thomas S, Gruber RD, et al. Radiation e xposure of the spinal interventionalist performing fluoroscopically guide d lumbar transforaminal epidural steroid injections. Arch Phys Med Rehabil. May 2002;83(5):697-701.

PAGE 183

168 APPENDIX A MANUAL OF STANDARD OPERATING PROCEDURES SUBJECT RECRUITMENT PROCEDURES: Identify potential patients with low back pain (LBP ) and lower extremity symptoms. These will only be patients who are referred to the physician and diagnosed with lumbar spinal stenosis (LSS). Do not “cold-call” previous patients to part icipate in this study or invite individuals who may call about the possibility of participating. On ly take patients from physicians at the University Spine Clinic. The physician will describ e the study to the patient and invite them to participate. INFORMED CONSENT: Have subjects read and sign the Informed Consent do cument. All subjects must complete the Informed Consent document prior to any data collect ion procedures can begin. Any of the participating clinicians may consent a patient to p articipate; however, no other individuals may consent a patient into the study. Ensure the patien t initializes all pages of each of the two copies of the Informed Consent document (except for the la st page where their signature will be present). Both copies should then be signed by one of the par ticipating clinicians in the “Investigator’s Signature” block and by another clinic staff member in the “Witness Signature” block. Do not have a friend or family member of the subject sign in the “Witness Signature” block. Hand one copy of the Informed Consent document to the subjec t for them to keep and place the other copy of the Informed Consent document in the unmasked fo lder in a secure location that is separate from the masked study folder which contains the maj ority of the subject’s data collection records. Patients are considered enrolled into the study onc e they sign this statement. INCLUSION/EXCLUSION CRITERIA: The physician, resident, or physician’s assistant w ill screen subjects according to the inclusion/exclusion criteria. Inclusion Criteria: In order to be eligible for participation an indivi dual must satisfy all of the criteria listed below: 1. Lumbar spinal stenosis identified by MRI or CT scan and interpreted by a radiologist independent of the study. The criteria of Boden e t al287 will be used to define LSS on MRI: non-discogenic loss of signal in the epidural fat with compression of neural tissues. 2. Chief complaint of pain in the low back, buttock, a nd/or lower extremity(s). The patient must have LE symptoms consistent with neurogenic cl audication or radicular pain. 3. Patient-reported difficulty walking due to lower ex tremity pain and/or cramping. 4. Rates sitting as a better position with respect to symptom severity compared to standing or walking. 5. Consent of the patient to undergo education, epidur al steroid injection(s), and attend specified physical therapy sessions.

PAGE 184

169 6. Individuals with no language barrier, that are coop erative, have transportation to the Spine Center, and who sign an informed consent form 7. Age greater than or equal to 50 years. Exclusion Criteria: An individual meeting any one of the following crit eria will not be eligible for participation: 1. Patients with organic brain syndrome or dementia. 2. Severe vascular, pulmonary or coronary artery disea se which limits ambulation. 3. Recent myocardial infarction (within last 6 months) 4. Spondylolisthesis requiring surgical fusion (i.e., greater than 5mm of slippage). 5. Previous spinal surgery that included fusion of two or more vertebrae. 6. Severe osteoporosis as defined by multiple compress ion fractures or a fracture at the same level as the stenosis. 7. Metastatic cancer. 8. Excessive alcohol consumption or evidence of non-pr escribed or illegal drug use. 9. Other orthopedic conditions or physical impairments of unrelated nature which would limit ambulation or prevent the subject from fully participating in any aspect of the rehabilitation exercises. 10. Epidural steroid injection within the last 365 days 11. Vascular or other non-musculoskeletal condition oth er than LSS suspected to be the primary source of the patient’s symptoms. Be sure to keep track of subjects that you screen b ut who do not meet the eligibility criteria on the “Tally Sheet for Subject Ineligibility/Refusal” tra cking form provided. It is designed so all that is need to place a tick mark next to the pertinent inc lusion/exclusion criteria. Next the patients will undergo the baseline examination. The research pack et includes the following forms: NOTE: Once the baseline physical exam is completed by th e physician, the patient will meet with an office assistant who will assist with compl etion of the following. 1. 3 copies of the Informed Consent document 2. Subject ID-Name Link 3. Numeric Pain Rating Scale (NPRS) 4. Oswestry Disability Questionnaire (OSW) 5. Spinal Stenosis Scale (SSS) form 6. Medical Outcomes Study (MOS) SF-36 form 7. Beck Depression Index (BDI) form 8. MOS Social Support Survey 9. Low Back Confidence Scale (LoBACS) 10. Demographic Information form 11. Physical Examination form COMPLETION OF FORMS Be sure that subjects and clinicians complete every form and all items of every form. It is easy to skip over questions and components of the form, and some of these items may be key pieces of

PAGE 185

170 information necessary to properly analyze the data. I would recommend that you verify the completeness of every form as subjects hand them ba ck to you and after clinicians complete any forms. By doing this, if an item is missing, you ca n immediately have them complete the item, rather than having them recall the information seve ral days after the form was completed. Guide to the Baseline Physical Examination Performe d by Physician: Date of onset: The subject has already recorded this once on p. 1 of the Demographic Information form; however, please ask this question again to confirm the subject gives you the same answer. The date of onset should be with respe ct to the most recent episode for which they are currently presenting, not the date at which the y first ever had an episode of back pain. I. Historical Information 1. Mode of Onset : Three options are available, only one may be chosen : 1. Gradual Patient is unable to identify a discrete moment when LBP or buttock/leg pain began. 2. Sudden (Minimal/No Perturbation) – LBP or buttoc k/leg pain began at a discrete moment in time, but was not associated with any abnormal move ments or trauma, or was associated with an routine activity that involves very low stresses (e.g. picking up a light object from the floor) 3. Traumatic LBP or buttock/leg pain began at a d iscrete moment, and was associated with an event or activity involving moderate or high stress on the spine. Chose the most appropriate option describing the ci rcumstances surrounding the onset, or select other and explain. 2. Distribution of Symptoms: This information should be gathered from the pain d iagram and confirmed by asking the patient if the diagram accurately reflects the symptoms that h e or she experiences. Each of the five anatomical areas should be specified with the appro priate symptom distribution or marked as “No Symptoms”. The options for the type of symptoms, lo cation, and nature are mutually exclusive, thus only one option may be selected for each area. 1. Lumbar spine is defined as the area at or above the lumbosacral junction. Central symptoms occur at or very near the spinal column. Bilateral symptoms occur to both sides, but occur outwards from the spinal column. Right or left refe r to symptoms occurring out from the spinal column on one side only. 2. Buttock refers to the area below the lumbosacral junction and above the gluteal fold. Central symptoms occur along or very near the sacrum spinal processes. Bilateral symptoms occur to both sides, but occur outwards from the sacral spin ous processes. This would include the region of the PSIS and sacral sulcus. Right or left refer to symptoms occurring out from the sacral spinous processes, including the regions of the PSIS and sacral sulcus but on one side only.

PAGE 186

171 3. Groin refers to the inguinal region of the proxi mal anterior thigh. 4. Thigh is the area below the gluteal fold and abo ve the popliteal fold of the knee on the posterior aspect of the leg. 5. Lower leg/foot is the area below the popliteal f old of the knee. The nature of the symptoms in each of the anatomica l areas is described in one of three ways, only one option may be chosen for each anatomical r egion: Constant Always present with no variation in inte nsity Intermittent Present at times, completely absent at other times Variable Always present, but intensity varies 3. Ordering of Symptoms: The patient is asked to identify which posture (sit ting, standing, and walking) is the worst and best with regards to symptoms. More than one postur e may be selected for each category (best and worst); however, the same posture should not be checked for both categories. A posture may only be identified for one of the categories, with the other left blank (e.g. walking is clearly the worst, and sitting is identified as best). 4. Prior History of LBP or buttock/leg pain: a. The patient is asked about prior episodes of pai n that have caused him to reduce his functional activity level. If the answer is no, the subsequent questions are not answered. b. The number of prior episodes is established into one of the four categories listed. c. The frequency of prior episodes is established i nto one of the three categories listed. d. The location of symptoms of previous episodes is established. If any of the prior episodes involved leg pain, this option is marked. Both opti ons may be marked. e. The events leading up to prior episodes are dete rmined; more than one option may be selected if the patient has had multiple prior episodes. f. The patient is questioned regarding any treatmen ts attempted for previous episodes. If a treatment has been used, the response to the treatm ent is ascertained. II. Neurological Screening 1. Sensory Examination: Sensory examination is carried out with pin prick e xamination in the specified anatomic areas bilaterally, while the patient has his eyes closed. The patient is asked if the sharp sensation is of equal intensity on both sides (WNL), or if one side feels more dull than the other (Diminished), or if one side is unable to be felt (Absent). 2. Motor Examination:

PAGE 187

172 Manual muscle testing is performed bilaterally. Gra ding is either WNL (equal bilaterally) or diminished (less strength than the other side). Eac h movement is also assessed as painful or not painful during testing. All testing is performed wi th the patient seated: Hip flexion the hip is flexed to near end range a nd pressure is applied to the anterior thigh into hip extension. Knee extension The knee is placed in a position s lightly less than full extension. One hand stabilizes the patient’s thigh, while the other app lies pressure on the anterior tibia into knee flexion. Dorsiflexion The foot is placed in full dorsiflex ion with some inversion. One hand stabilizes the distal tibia, while the other hand applies pres sure on the dorsum of the foot into plantar flexion with some eversion. Plantarflexion – The foot is placed in full plantar flexion with some eversion. One hand stabilizes the distal tibia, while the other hand applies pres sure on the sole of the foot into dorsiflexion with some inversion. Hallux Extension With the shoes off, the great to e is placed in extension. One hand stabilizes the foot, while the other hand applies pressure on the dorsum of the distal phalanx of the great toe into flexion. 3. Deep Tendon Reflexes: Reflexes of the lower extremity are tested bilatera lly with the patient seated. The quadriceps reflex is tested by tapping the patellar tendon and observing for knee extension. The ankle (Achilles) reflex is tested by grasping the patient ’s foot and placing it into slight dorsiflexion. Th e Achilles tendon is tapped and the examiner observes and feels for ankle plantar flexion. Reflexes are graded as “WNL” when equal to the other side, “ Diminished” if a response is of reduced vigor as compared to the other side, or “Absent” if no response is elicited. 4. Tension Signs: 1. Straight Leg Raise is performed with the patient supine and the head relaxed on a pillow and both hips and knees extended resting on the table. The examiner grasps under the ankle on the side to be tested and passively lifts the leg. The leg should remain straight and in neutral hip rotation. The leg is lifted until the patient repor ts that pain is produced. A test is considered positive when raising the leg less than 450 reproduces pain in the lower extremity (below the knee). 2. Slump Test is performed with the patient in sitt ing. The patient begins the test sitting in an upright position. The examiner passively flexes the patient’s spine and neck. The examiner then passively extends the patient’s knee and dorsi flexes the patient’s ankle. The examiner notes any report of pain from the patient. The test is repeated on the contralateral lower extremity. The test will be coded as 1) no reprodu ction of familiar symptoms, 2) reproduction of the patient’s LBP, 3) reproduction of patient’s LE symptoms proximal to the thigh, 4) reproduction of patient’s LE symptoms distal to the thigh. 3. Femoral Nerve Stretch is performed with the pati ent prone or prone with a pillow under the abdomen if the patient can not tolerate flat prone lying. The examiner passively flexes the

PAGE 188

173 patient’s knee and notes any report of pain from th e patient. The test is considered positive if pain is reproduced in the patient’s anterior thigh. This is determined by evaluating the location and nature of the pain produced during the test mov ement. If the anterior thigh symptoms are reproduced, the test is judged to be positive. Howe ver, if pain is reproduced only in the back, the test is judged to be negative. III. Additional Physical Examination 1. Postural Observation: The posture is observed. Postural deformities shoul d be considered as postural adaptations to injury or pain, not normal variants of posture freq uently observed in both healthy and symptomatic individuals (e.g. reduced or accentuate d lumbar lordosis). Three options are possible. Both an acute kyphosis and a lateral shif t may be present in the same individual: 1. WNL – This should be the selection if and acute kyp hosis and a lateral shift are not present. In other words, if you are having to debate about the presence of an acute kyphosis or a lateral shift, mark it as WNL. 2. Acute Kyphosis A frontal plane deformity in which the patient adopts a flexed posture due to injury or pain. 3. Lateral Shift A sagittal plane deformity in which the shoulders are notably displaced to the left or right in the frontal plane with reference t o the pelvis. The direction of the lateral shift is determined by the direction of the shoulders relati ve to the pelvis. 2. Range of Motion Provocation: The patient will be taken through a range of motion for the lumbar spine, SI joint, and hip joint to determine pain provocation in both supine and st anding positions. A positive pain provocation is indicated when the patient reports p ain during the ROM testing procedure. The position of the joint range and the pain severity f rom 1-10 will be recorded for any pain provocation test. 3. Distal Pulses, Vital Signs: The distal pulses and vital signs of the subjects w ill be taken. Distal pulses will include pulses found on the dorsal portion of the foot (dorsalis p edis) and behind the ankle (posterior tibial). Pulses will be recorded as normal, abnormal, or abs ent. Vital signs will include resting heart rate and resting blood pressure. 4. Body Mass Index: The Body Mass Index (BMI) will be recorded by measu ring the weight (lbs) and height (inches) of the subjects. These values will be cal culated and compared to normative values based off of the 2007 American College of Sports Me dicine Guidelines. ASSIGNMENT OF SUBJECT ID

PAGE 189

174 Once the baseline physical examination, and the MRI has been read, and the inclusion/exclusion criteria has been verified, assign the patient a Su bject ID. Then, once the patient completes the Informed Consent document they will report to one o f the office staff at the Spine Center and will be given the questionnaires listed above. Once the questionnaires are completed they will be randomly assigned to either the ESI or ESI+PT group RANDOMIZATION Once the patient completes the self-report measures the office staff member will open the randomization envelope indicating the patientÂ’s tre atment group assignment that corresponds to the patientÂ’s unique identification number. A rando m number generator will be used to establish randomization lists prior to the initiation of the study. Individual randomization assignments will be concealed according to the following procedure: 1) The group assignment will be recorded on a label that is affixed to a 3.5 X 5 inch index car d; 2) This card will be folded in half such that th e label with the patientÂ’s group assignment will be o n the inside of the fold; 3) The folded index card will then be placed inside the envelope, and t he envelope will be sealed. Assign subject ids in a consecutive order as subjec ts are enrolled into the study. In other words, just take the next envelope on top of the stack. Do not skip envelopes for any reason. Upon opening the envelop, the office staff will instruct the patient to either 1) schedule an appointment with the physician for their first epidural injecti on (ESI group), or 2) schedule an appointment with the physician for an epidural injection and sc hedule an initial physical therapy visit (ESI+PT group). Patients in the ESI group will follow-up wi thin 10 days with the treating physician for ESI. Patients in the ESI+PT group will follow-up w ithin 10 days with the treating physician for ESI and schedule an appointment with a licensed phy sical therapist. Patients will be monitored to confirm they have made their follow up appointments by the study coordinator. Patients who do not schedule their appointments as instructed will be called by the study coordinator to assist the patient in progressing to the treatment stage. TREATMENT PROCEDURES Epidural Steroid Injection Group (ESI) Epidural Steroid Injection Treatments: Patients in the ESI Group will be treated with 2-3 epidural steroid injections, standardized educational support, and general care by the treati ng physician. Prior to the injection, the physician will verbally confirm that the patient do es not have any allergies and is not taking blood thinners. Standardized technique as describe d by Botwin et al288 will be used for all epidural steroid injections. Plain radiographs in t he anteroposterior and lateral views will be taken after all injections to document both the contrast pattern and needle placement. All patients will be monitored by pulse oximetry, blood pressure, and electrocardiogram before, during, and after the procedure. Patients will be transferred to the recovery unit for 40 min. All patients will be seen by the physician who performed the injection a nd by a registered nurse before discharge. Patients will be instructed to perform activity as tolerated throughout the time of the study. A maximum of two to three ESI will be given in total to each patient throughout the study according to a clinical Epidural Steroid Injections

PAGE 190

175 algorithm for ESI treatment located below. The num ber of injections received by each individual and any medications taken will be recorded and anal yzed to determine similarities between groups. The patient will receive maximum of 2-3 epi dural steroid injections during the RCT for lumbar spinal stenosis following the algorithm list ed below: Assessment Treatment 1. Make verbal confirmation Before starting treatment, make verbal confirmation that the patient 1) does not have any allergies, and 2) is not on any b lood thinning medications. 2. 1 st Injection Identify the level of greatest stenosis pathology a ccording to MRI or CT scan that correlates the most to the patientÂ’s s ymptoms. Use a transforaminal approach at the level identifi ed. Determine if bilateral sites (2 injections at both sides of the level identified) or unilateral site (1 injection on only one side of the level identified). If >90% of the patientÂ’s symptoms ar e one only one side, perform unilateral site injection. Otherwise, perf orm bilateral site injection. Utilize exactly 1.5ml of steroid at each site injec ted. 3. Re-assess 3-4 weeks for possible 2nd injection If patient reports >90% improvement in overall symp tom; discontinue injections. If patient reports 50-90% improvement in overall sy mptoms; repeat injection #1 procedure at the same level indicated. If patient reports <50% improvement in overall symp toms; change levels to above or below 1st injection determined by stenosis pathology indicated at other levels on MRI or CT sc an and patient symptoms. 4. Re-assess 3-4 weeks for possible 3rd injection If patient reports >90% improvement in back/buttock /leg pain; discontinue injections. If patient reports 50-90% improvement in back/butto ck/leg pain; repeat injection #1 procedure at the same level ind icated. If patient reports <50% improvement in back/buttock /leg pain; change levels to above or below 1st and 2nd injection determined by stenosis pathology indicated at other levels on MRI or CT sc an. 5. Re-assess 3-4 weeks after 3rd injection Determine the level of patient reported improvement s in back/buttock/leg pain at this time and document ove rall symptom improvements or changes. No further epidural steroid injections at this time

PAGE 191

176 ESI + Physical Therapy Group (ESI + PT) Epidural Steroid Injection Treatments (same as ESI group above) Physical Therapy Examination that Therapists will U se to Determine Progression of Interventions: I. Standing Examination 1. Single Movement Testing and Status Change with T runk Movements: Active range of motion is tested with the patient s tanding. Single movement testing is performed. The patient is first asked about his symptoms while standing, prior to any movement testing. The patient is instructed that these symptoms will serve as a baseline level, and it is the change in symptom location and/or intensity that sh ould be reported. The range of motion values are measured in the following manner: Single Movement Testing: Total Flexion The center of the inclinometer is c entered over the spinous process of T12. The inclinometer is zeroed. The patient is instructed t o bend forwards as far as possible without bending the knees. Left and Right Sidebending The spinous process of T12 is identified. The center of the inclinometer is placed just above this point parall el to the axis of the spinal column, and is zeroed. To measure right sidebending, the inclinome ter is placed on the left side and to measure left sidebending, the inclinometer is place d on the right side. The patient is instructed to bend to the right or left as far as possible wit h the fingertips reaching as far down the side of the thigh. Extension The center of the inclinometer is cente red over the T12 spinous process. The inclinometer is zeroed. The patient is instructed t o run their hands down along the back of the thighs while bending backwards. The patient is to bend backwards as far as possible without bending the knees. After this movement, have the p atient bend perform an easing movement (bend forward, sit, etc) until any symptoms elicite d ease. Sustained extension (up to 15 seconds) or repeated movements (up to 5) to assess production of symptoms (ask specifically if symptoms are above or below the gluteal folds) From the active movement testing, determine an “ast erisk” sign that will be utilized for reassessment. Quadrant testing with overpressure may also be used (extension with side bending and rotation to the same side). The treating therapist would mo st often opt to perform this test if other ROM and provocative testing has not reproduced the patient’s familiar symptoms. Status Change with Trunk Movements: With each movement performed, the examiner makes a judgment as to whether the patient’s symptoms are improved or worsened after each moveme nt. The examiner should also record whether the improvement or worsening in status is d ue to changes in pain (“My pain is worse.” or “Now my pain is less.”) or changes in the location of the symptoms (“My symptoms have moved up my leg closer to my back” or “My symptoms have m oved closer to my foot.”). Centralization

PAGE 192

177 of symptoms should indicate an improvement in statu s, and peripheralization of symptoms with movement should indicate a worsening in status. Onl y one term (“Improve” or “Worsen”) can be used with each movement tested. However, it is poss ible that with flexion, for example, the patient’s symptoms may centralize (i.e. improvement in status), but the patient may describe their pain as getting worse. If this is the case, alterat ions in paresthesias take precedence over alterations in pain, thus this patient would be jud ged to have improved with this movement, and the improvement would be marked as occurring becaus e of centralization. Improvement and worsening in status are defined below: 1. Worsen a. Symptoms present (or produced) increase in inten sity with the test movement. When the neutral position is resumed the intensity remains h igher than baseline for at least 30 seconds after completion of the movement. b. Or, a paresthesia is produced which was not pres ent prior to the movement, c. Or, a pain or paresthesia moves distally away fr om the spine 2. Improve a. Symptoms present are diminished or abolished dur ing the movement. When the neutral position is resumed, the symptoms remain decreased in intensity for at least 30 seconds. b. Or, a paresthesia present at rest is abolished. c. Or, a pain or paresthesia moves centrally toward s the spine. 3. Status Quo (ISQ) a. Test movement does not cause improving or worsen ing of symptoms II. Seated Examination 1. Motor: The following key muscles will be tested: Hip flexion (L1-L2) the hip is flexed to near end range and pressure is applied to the anterior thigh into hip extension. Knee extension (L2-L4) The knee is placed in a po sition slightly less than full extension. One hand stabilizes the airman’s thigh, the other a pplies pressure on the anterior tibia into knee flexion. Dorsiflexion (L4-L5) – The best way to test dorsifl exion is with heel walking. For a nonweightbearing test, the foot is placed in full dors iflexion with some inversion. One hand stabilizes the distal tibia, the other hand applies pressure on the dorsum of the foot into plantarflexion with some eversion. Great Toe Extension (L5) With the shoes off, the great toe is placed in extension. One hand stabilizes the foot, the other hand applies pressur e on the dorsum of the distal phalanx of the great toe into flexion.

PAGE 193

178 Ankle Eversion (S1-S2) The foot is placed in full eversion and plantarflexion. One hand stabilizes the distal tibia, the other applies pres sure on the lateral aspect of the foot into dorsiflexion and inversion. Ankle Plantarflexion (S1-S2) – The foot is placed in full plantarflexion. One hand stabilizes the distal tibia, the other applies pressure on the plantar aspect of the foot into plantarflexion. 2. Muscle Stretch Reflexes: The Achilles’ reflex is performed to test the integ rity of the S1-S2 nerve roots. The patient is seated and relaxed with the ankle supported at appr oximately neutral dorsiflexion. The Achilles tendon is struck with the reflex hammer and reflexi ve plantarflexion is felt with the supporting hand. The patellar tendon reflex is performed to te st the integrity of the L2-L4 nerve roots. The patient is seated with the knee flexed to approxima tely 900. The patellar tendon is struck with the reflex hammer and reflexive knee extension is obser ved. 3. Babinski Test: The Babinski responses will be tested bilaterally t o assist in ruling out upper motor neuron disease. The test is performed with the patient sit ting and relaxed. The examiner places a moderately sharp object (end of the handle of a ref lex hammer) at the lateral plantar surface of the foot near the heel and, with moderate pressure, str okes the foot on the lateral surface to the ball of the foot medially. Extension of the great toe and f anning of the other toes is considered a present Babinski Sign, which is considered abnormal in matu re adults. 4. Sensation: Evaluation for sensory loss is performed by lightly brushing the hand over key dermatomal regions. Any deficit noted should be tested further with the use of a pin to clearly map out the area of sensory deficit. Inguinal area (L1) Anterior mid-thigh (L2) Distal anterior thigh and medial knee (L3) Medial lower leg and foot (L4) Lateral lower leg and foot (L5) Posterior calf (S1) Medial calcaneus (S2) III. Supine Examination 1. Capsular mobility of the hip:

PAGE 194

179 The passive accessory mobility of the hip will be a ssessed by applying passive glides to end range into: Distraction Caudal glide Posterior glide 3. Modified Thomas Test: The patient is supine sitting on the edge of the ta ble with the pelvis level and square to the trunk. The patient flexes both hips (with the help of the examiner) and brings the thighs up towards the trunk. The pelvis is stabilized by placing a hand u nder the patient's lumbar spine. The patient holds one leg to his/her chest and lets the other l eg lower down until it is flat on the table. If the thigh does not reach horizontal, the psoas major or the rectus femoris may be tight. The knee is then extended. If the hip extends further, the tes t is positive for a tight rectus femoris, if it doe s not extend to neutral, then at least the psoas is t ight (and perhaps the rectus femoris is tight, too) 4. Hip Flexion Test: With the subject in the supine position, the primar y examiner passively flexes the hip to 90 and zeroes the inclinometer at the apex of the knee. Th e hip is then flexed until the opposite thigh begins to rise off of the table (be sure to adjust the patientÂ’s clothing so that tight or restrictive clothing does not interfere with the test). A measu rement is taken. The hip is then passively moved into end range flexion with gentle overpressu re. The effect on symptoms is recorded. 5. Quadrant test of the hip: With the subject in the supine position, the examin er passively flexes, adducts and internally rotates the hip such that the knee is guided toward s the contralateral shoulder while applying gentle overpressure and compression through the axi s of the femur. The amount of motion and the degree of patient discomfort is noted. 6. FABER Test: The FABER test is administered with the subject in supine, the heel of the lower extremity to be tested placed over the opposite knee. If the patie nt does not have enough mobility for this position, rest the ipsilateral foot on the table ju st medial to the contralateral knee) The hip joint is passively externally rotated and abducted by placin g pressure over the ipsilateral knee, while stabilizing the contralateral innominate. After bei ng zeroed against a wall, an inclinometer is placed on the medial tibia of the lower extremity t o be tested, just distal to the medial tibial condyle. The range of motion is measured at the po int of maximal passive resistance or at the point where the subject stops the test secondary to pain. 7. Transversus abdominis (TrA) testing:

PAGE 195

180 The therapist will palpate the muscle medial and in ferior to the ASIS and ask the subject to slowly draw their navel up and in towards the table Therapist should feel a tensioning under their fingers with retraction of the abdominal wall posteriorly without a “bulging” sensation which would indicate contraction of the internal ob lique. Patients will be instructed in TrA exercises as described below. IV. Sidelying Examination 1. Hip abduction strength: In sidelying, the therapist will passively lift the subject’s leg into end range abduction while maintaining the hip in a neutral sagittal plane pos ition. The patient will then be asked to hold the leg up against gravity. If the patient is able to hold the leg up, the therapist will assess the strength by attempting to push the leg down towards the table. A judgment will be made as to whether the hip abductors are normal or weak. V. Prone Examination 1. Spring Testing of the Thoracic and Lumbar Spine: Spring testing in prone (can use up to two pillows) is performed over the spinous processes of the thoracic and lumbar vertebrae and the sacrum. Sprin g testing is both a provocation test and a test of segmental mobility. The following options are av ailable for each level tested: Normal Mobility Passive mobility is judged to be normal. Hyper/Hypomobile Judgments based on the passive m obility of the tested segment relative to adjacent segments and the expectation of the exa miner. One of the two options may be selected. No Pain No painful symptoms are produced. Pain Judgment based on the provocation of pain. L ocal refers to pain produced directly under the examiner’s hand, whereas distant pain ref ers to provocation at an anatomical area not directly under the examiner’s hand. One of the two options may be chosen. Spring testing is performed by placing the hypothen ar eminence of the hand over the spinous process of the segment to be tested. With the elbow and wrist extended, the examiner applies a gentle but firm, anteriorly-directed pressure on th e spinous process. Interpretation of whether a segment is hypomobile should be based on the examin er’s anticipation of what normal mobility would feel like at that level and compared to the m obility detected in the segment above and below. In previous studies of similar subjects, it was unusual for examiners not to identify at least one hypomobile segment. 3. Measurement of hip internal rotation (IR) and ex ternal rotation (ER): The patient lies prone. The examiner places the opp osite leg of the leg to be measured in 30 of hip abduction to enable the tested hip to be freely moved into external rotation. The lower extremity of the side to be tested is kept in line with the body, and the knee on that side is flexed to 90 with the ankle in the neutral position, and the le g in the vertical position. The inclinometer is first zeroed on a vertical surface and then plac ed on the distal aspect of the fibula in line with the bone. The leg should be oriented in such a fash ion that the inclinometer reads zero degrees.

PAGE 196

181 Measurement of hip IR (hip rotated in a lateral dir ection [leg moved toward the edge of the plinth) and ER (hip rotated in a medial direction [ leg moved toward the middle of the table]) is recorded at the point in which the pelvis first beg ins to move. The measurement should be recorded bilaterally. The effect of overpressure a t end range will also be assessed. 4. Measurement of hip extension: The patient lies prone (if tolerated). Stabilize t he pelvis by using one hand to push the ischial tuberosity towards the table, and then passively ex tend the hip. Use up to 2 pillows under the abdomen if needed for comfort. This should be performed bilaterally and the clinic ian should make a judgment if it is tight or restricted. If the hip does not extend to >/= 10 de grees extension, the examiner should record this as restricted or limited hip extension. The effect of overpressure at end range will also be assessed. 5. Hip extensor strength: In prone, the therapist will passively lift the sub ject’s leg passively into end range hip extension while maintaining the hip in neutral in regards to abduction and adduction and the knee in full extension. The patient will then be asked to hold the leg up against gravity. If the patient is able to hold the leg up, the therapist will assess the s trength by attempting to push the leg down towards the table. A judgment will be made as to w hether the hip extensors are normal or weak. Note that the clinician should take precautions aga inst aggravating the patient’s LBP and LE symptoms while testing the patient as needed (have patient tighten the abdominal muscles, use a pillow under the abdomen, reduce the degree of hip extension for testing, etc). Physical Therapy Treatment: Patients randomly assigned to the physical therapy group will be treated in physical therapy for 810 weeks, with a frequency of 1-2 visits per week f or the first 6 weeks and 1 visit every two weeks for the following 4 weeks with no more than 1 0 visits total. Each patient will also receive additional instruction in an individualized home ex ercise program that will be updated during the treatments. The algorithm for physical therapy trea tments can be found below: The patient will receive interventions targeting lu mbo-pelvic impairments on day one. Other interventions indicated based on the algorithm belo w and will be provided no later than the end of the third visit. Prioritization is as follows: 1) lumbo-pelvic spine, 2) hips / hip flexors; 3) abdominal muscles / ‘stabilization’ exercise; 4) th oracic spine mobility. SUBJECT PAYMENT Subjects will be paid $50.00 for each of the five r e-evaluation packets of the study, for a total of $250.00 compensation. Additionally, subjects who pa rticipate in the qualitative interview with the PI will receive an additional $25 for completio n of the interview. Subjects will receive one check at the end of their participation in the stud y for the degree of the study completed. Remind

PAGE 197

182 subjects that payment usually takes from 6-8 weeks from the time the therapist faxes me the completed payment form. DATA MANAGEMENT Completed study folders will be packaged in a box a nd picked up by the PI, Amy Hammerich. Make sure the folders will not be able to excessive ly move inside the box, or it could become difficult to re-organize the data back to the corre ct subject.

PAGE 198

183 APPENDIX B IRB APPROVALS 1. COMIRB – University of Colorado Multidiciplinary In vestigator Review Board 2. Hawkins Foundation – Greenville Hospital System

PAGE 199

184

PAGE 200

185 November 9, 2012 Thomas Denninger, DPT, CSCS Attn: Allyson Sandago, MPH, ATC The Hawkins Foundation 200 Patewood Drive, Suite C-100 Greenville, SC 29615 RE: IRB File #Pro00013317 Study Title: Epidural Steroid Injection versus Epidural Steroid Injection and Manual Physical Therapy and Exercise in the Managem ent of Lumbar Spinal Stenosis: A Randomized Clinical Trial Items Submitted for IRB Review: Protocol and Consent Form Continuing Review Dear Dr. Denninger: On November 5, 2012, the Chairman of the Institutio nal Review Board/Committee-A (IRB) of the Greenville Hospital System reviewed the continu ation of your above-mentioned study. Expedited approval was given for one year. Your study will expire on November 4, 2013. It is the investigatorÂ’s responsibility to make sur e the proper reapproval information is submitted to t he IRB. This information must be submitted to the IRB in October 2013. The same requirements as previously outlined for yo u by the IRB remain in effect as long as the study is ongoing. Please refer to your initial app roval letter for these requirements. Thank you for your assistance in this matter. If you have any qu estions, please feel free to call the IRB Office at 455-4984. Sincerely, Christopher C. Wright, MD, Chairperson Institutional Review Board / Committee-A 900 West Faris Road Greenville, SC 29605

PAGE 201

186 APPENDIX C INFORMED CONSENT 1. University of Colorado 2. Hawkins Foundation

PAGE 202

187

PAGE 203

188

PAGE 204

189

PAGE 205

190

PAGE 206

191

PAGE 207

192

PAGE 208

193

PAGE 209

194

PAGE 210

195 Consent to Participate in a Research Study Epidural Steroid Injection vs. Epidural Steroid Inj ection, Manual Physical Therapy, and Exercise in the Management of Lumbar Spinal Stenosi s: A Randomized Clinical Trial Study to be Conducted at : Steadman Hawkins Clinic of the Carolinas 200 Patewood Drive, C100 Greenville, SC 29615 Proaxis Therapy 111 Doctors Drive Greenville, SC 29607 Sponsor Name: American Physical Therapy Association (Orthopedic S ection) and the Bob Doctor Research Program Principal Investigator: Thomas Denninger, PT, DPT (864) 797-7020 INTRODUCTION You are being asked to participate in a research st udy. The Institutional Review Board of the Greenville Hospital System has reviewed this study for the protection of the rights of human participants in research studies, in accordance wit h federal and state regulations. However, before you choose to be a research participant, it is important that you read the following information and ask as many questions as necessary to be sure that you understand what your participation will involve. Your signature on this consent form will acknowledge that you received all of the following information and expla nations verbally and have been given an opportunity to discuss your questions and concerns with the principal investigator or a coinvestigator. A description of this clinical trial will be availa ble on http://www.ClinicalTrials.gov as required by U.S. Law. This Web site will not include informa tion that can identify you. At most, the Web site will include a summary of the results. You can search this Web site at any time. PURPOSE You are being asked to participate in a research st udy because you have the type of lower back pain and/or leg pain that may come from lumbar spin al stenosis. Spinal stenosis is a type of spinal problem that is caused by a narrowing of the spaces around the nerves in your lower back. The purpose of this study is to determine the effec tiveness of various treatment techniques for spinal stenosis. This study is designed to provide information on factors which may help predict who will respond to non-surgical treatment for spin al stenosis and to compare different forms of testing and treatment for individuals with this con dition. There will be 80 patients asked to participate in t his study. The total amount of time you will be involved in this study is three years. PROCEDURES If you choose to participate in this study, you wil l be randomized (assigned to a treatment group by chance, like flipping a coin) to receive one of two options for your treatment after your injection. You will answer questionnaires regardin g your back and leg pain before you start treatment (injection plus, potentially, physical th erapy). The questionnaires should take 30-45

PAGE 211

196 minutes to complete. You will answer these same qu estionnaires again at 3 months, 6 months, 12 months, 24 months and 36 months after completion of the treatment. There will also be a qualitative interview at 12 months to determine any factors that contribute to both success and obstacles for success in a physical rehabilitation program, as well as any factors that may have limited your success due to any social or home envi ronment obstacles. The total number of appointments will be one for evaluation by a specia list and up to fourteen visits for testing and treatment by clinicians. If you are randomized to Group One, you will be ins tructed in exercises designed to improve the motion of your spine and hips. You will be asked t o continue your exercises at home as well as start an aerobic conditioning program, either walki ng or cycling. Also, you will receive a book of instructions on how to take care of your lower back and avoid motions that will irritate your lower back. In addition, you will receive 1-3 epid ural steroid injections from your physician that will help reduce your lower back pain and allow you to continue physical activity. If you are randomized to Group Two, you will be ins tructed in exercises designed to improve the motion of your spine and hips. Also, you will rece ive a book of instructions on how to take care of your lower back and avoid motions that will irri tate your lower back. You will receive 1-3 epidural steroid injections from your physician tha t will help reduce your lower back pain and allow you to continue physical activity. You will also undergo an individualized manual therapy and exercise program that will help improve motion in your spine and hips. The manual therapy consists of joint movement/motion using a variety o f techniques commonly used by physical therapists. In these treatments, the therapist wil l manually stretch various joints, to include your lower back and your hip joints. The therapist will also help you stretch muscles in your buttocks and legs. You will see the therapist 1-2 times per week for a maximum of ten weeks for treatment. In addition, you will walk on a treadmi ll or ride a bicycle 1-2 times per week. Both treatments are standard care procedures with k nown beneficial effects for patients with lumbar spinal stenosis. We are investigating which post-injection procedure provides the better outcome for the patient through a series of questio nnaires given at the indicated follow-up schedule. POSSIBLE RISKS Any treatment has possible side effects. The treat ments and procedures used in this study may cause all, some, or none of the side effects listed There is always the risk of very uncommon or previously unknown side effects happening. There m ay be risks or side effects which are unknown at this time. Side effects may be mild or very serious. You should always talk to the Principal Investigator about any side effects that you have while taking part in the study so he may properly monitor your health. It should first be recognized that participants eli gible for this study are those that have been identified as appropriate candidates for Epidural S teroid Injections (ESI), and referred to the coinvestigator for consideration for this procedure. Therefore, regardless of your participation in this study, the treating physician would recommend an ESI for you as is generally accepted current standard of care. Potential adverse effect s with the injection include pain at the injection site, facial flushing, headache, insomnia (inabilit y to sleep), and fever. The risks associated with physical therapy treatmen t are 1.an increase in pain intensity from completing the exercises and 2.mild muscle soreness after manual therapy technique s are performed. We will attempt to minimize these risks by having a licensed physical therapist specifically trained in the study procedures carry out all treatments. There is also a slight risk

PAGE 212

197 associated with the exercises you will be asked to perform at home, including muscle soreness in your lower back or legs. Some of the questions you will answer may be of a p ersonal nature, including questions about the impact of low back pain on various aspects of your life. You do not have to answer any questions you do not feel comfortable answering. EXCLUSIONS You cannot participate in this study if you have or have had the following: Less than 50 years of age Organic brain syndrome (decreased brain function du e to a medical disease) or dementia Severe vascular, pulmonary or coronary artery disea se which limits movement Recent heart attack (within last 6 months) Spondylolisthesis (degeneration or slippage of part of the spine) requiring surgical fusion Pervious spinal surgery that included fusion of two or more vertebrae Severe osteoporosis as defined by multiple compress ion fractures or a fracture at the same level as the stenosis (narrowing of area in th e spine) Metastatic cancer (cancer spread throughout various regions in the body) Excessive alcohol consumption or evidence of non-pr escribed or illegal drug use Other orthopedic conditions or physical impairments of unrelated nature which would limit ambulation (moving/walking) or prevent you fr om fully participating in any aspect of the rehabilitation exercises Epidural steroid injection within the last year Vascular or other non-musculoskeletal condition oth er that LSS suspected to be the primary source of your symptoms Pregnancy during the period of the study POSSIBLE BENEFITS It is not possible to know whether or not you may b enefit from participating in this study. Future lumbar spinal stenosis patients may benefit from th e results of this research. ALTERNATIVE (OTHER) TREATMENTS You can still receive evaluation and treatment for your condition if you do not participate in this study. Discuss any alternative treatments with you r regular doctor and/or the study doctor before you decide to participate in the study. Some alter native treatments for your condition include surgery, orthopedic support(s) tools to decrease pa in/inflammation (cold/hot pack, electrical stimulation, etc) and acupuncture. Your decision i s entirely up to you. If you decide not to participate in the study, you will not be penalized or lose any benefits and your decision will not affect your relationship with your doctor or hospit al COSTS TO YOU FOR PARTICIPATING IN THIS STUDY You and/or your insurance company will be billed fo r the cost of those procedures which are considered to be normal and usual medical practice for someone with your condition (including the injection). You will be responsible for all ot her charges, including your office visits and physical therapy. PAYMENT FOR PARTICIPATION To You: You will be paid $50 for each re-evaluation at 3, 6 12, 24, and 36 months, for a total of $250 if all follow-up questionnaires are completed. You will also be paid $25 for a 30-45 minute

PAGE 213

198 interview at 12 months. If you have to leave the s tudy early, or if we have to remove you from the study, you will be paid only for the re-evaluat ion packets or interviews you have completed. It is important to know that payments for participa tion in a study are taxable income. To Investigators: The investigators will not be paid for participatin g in the study. To Institution: The Greenville Hospital System and Proaxis Therapy are not being paid for this study. COMPENSATION FOR INJURY AS A RESULT OF STUDY PARTIC IPATION If you get hurt or sick because of treatment you ha ve received in this study, emergency medical treatment is available but will be provided at the usual charge. The study sponsor may or may not pay for this treatment. You will be responsible fo r any charges not paid for by the sponsor. No financial compensation (payment) will be availab le to you from the study sponsor, the Greenville Hospital System or the investigators. Y ou or your insurance company will be charged for continuing medical care and/or hospitalization. You understand that you have not given up any of your legal rights by signing this consent fo rm. VOLUNTARY PARTICIPATION Participation in this study is completely voluntary (your choice). You may refuse to participate or withdraw from the study at any time. If you refuse to participate or withdraw from the study, you will not be penalized or lose any benefits. Your d ecision will not affect your relationship with your doctor or hospital. NEW INFORMATION During this study, you will be told of any importan t new information that may affect your willingness to participate in this study. You will be informed of any new findings that might change your decision to be in this study. AUTHORIZATION TO USE AND DISCLOSE (RELEASE) MEDICAL INFORMATION As part of this research study, your study doctor a nd his/her research team will keep records of your participation in this study. These study reco rds may be kept on a computer and will include all information collected during the research study and any health information in your medical records that is related to the research study. You r study doctor and his/her research team will use and disclose (release) your health information as t hey conduct this study. To evaluate the results of the study and for compliance with federal and st ate law, your health information may be examined and copied by the Food and Drug Administra tion (FDA), other governmental regulatory agencies, the Institutional Review Board of the Greenville Hospital System, the study sponsor and the sponsorÂ’s authorized representative (s). This study may result in scientific presentations and publications, but steps will be t aken to make sure you are not identified. Under federal privacy laws, your study records cann ot be used or released for research purposes unless you agree. If you sign this consent form, y ou are agreeing to the use and release of your health information. If you do not agree to this us e, you will not be able to participate in this study. The right to use your health information fo r research purposes does not expire unless you withdraw your agreement. You have the right to wit hdraw your agreement at any time. You can do this by giving written notice to your study doct or. If you withdraw your agreement, you will not be allowed to continue participation in this re search study. However, the information that has already been collected will still be used and relea sed as described above. You have the right to review your health information that is created duri ng your participation in this study. After the study is completed, you may request this informatio n.

PAGE 214

199 Once your health information has been released, fed eral privacy laws may no longer protect it from further release and use. If you have any ques tions about the privacy of your health information please ask your study doctor. CONTACT FOR QUESTIONS For more information concerning this study and rese arch-related risks or injuries, or to give comments or express concerns or complaints, you may contact the principal investigator, Thomas Denninger, PT, DPT, at (864) 797-7020. You may als o contact a representative of the Institutional Review Board of the Greenville Hospit al System for information regarding your rights as a participant involved in a research stud y or to give comments or express concerns, complaints or offer input. You may obtain the name and number of this person by calling (864) 455-8997. A survey about your experience with this informed c onsent process is located at the following website: http://www.ghs.org/Research-and-Clinical-Trials Participation in the survey is completely anonymous and voluntary and will not affect your relationship with your doctor or the Greenville Hos pital System. If you would like to have a paper copy of this survey, please tell your study d octor. CONSENT TO PARTICIPATE My study doctor, __________________________________ __________, has explained the nature and purpose of this study to me. I have bee n given the time and place to read and review this consent form and I choose to participate in th is study. I have been given the opportunity to ask questions about this study and my questions hav e been answered to my satisfaction. I have been given a copy of my study doctorÂ’s Notice of Pr ivacy Practices. I agree that my health information may be used and disclosed (released) as described in this consent form. After I sign this consent form, I understand I will receive a co py of it for my own records. I do not give up any of my legal rights by signing this consent form _____________________________________________ Printed Name of Participant ________________________________ ___________ __________ Signature of Participant Date Time _______________________________ ___________ __ ________ Signature of Witness Date Time INVESTIGATOR STATEMENT I have carefully explained to the participant the n ature and purpose of this study. The participant signing this consent form has (1) been given the ti me and place to read and review this consent form; (2) been given an opportunity to ask question s regarding the nature, risks and benefits of participation in this research study; and (3) appea rs to understand the nature and purpose of the study and the demands required of participation. T he participant has signed this consent form prior to having any study-related procedures perfor med. _________________________________ __________ ____ ______ Signature of Investigator Date Time Principal Investigator: Thomas Denninger, PT, DPT (864) 797-7020 Co-Investigators: Thomas Jarecky, MD (864) 454-742 2

PAGE 215

200 APPENDIX D STUDY QUESTIONAIRES 1. Demographic survey 2. Modified Oswestry Disability Index 3. Numeric Pain Rating Scale 4. Lower Back Activity Confidence Scale 5. Medical Outcomes Survey Social Support Scale 6. Medical Outcomes Survey SF-36

PAGE 216

201 Demographic Information Thank you for completing this questionnaire. This q uestionnaire will help us to better understand your general health and any problems related to bon e and muscle conditions. Your responses will be held in the strictest confidence. Please answer every question. Some questions may look like the others, but each one is different. There is no right or wrong answer. If you are not sure how to answer a question, just give the best answer you ca n. Subject ID: ____________________________ TodayÂ’s D ate: _____/_____/_____ mm dd yy Date of Birth: _____/_____/____ Height:_________ Weight:_________ mm dd yy Gender: Female Male Do you expect you will be able to complete all treatment sessions over a 10-week period (i.e. not going on vacation, no extended business trips scheduled, etc.)? No Yes Race: American Indian Asian Pacific Islander Black or African American White or Caucasian Hispanic Other ________________

PAGE 217

1. Are you presently seeking treatment from any other specialists for your back pain? No Yes (If yes, please check all that apply below: 2. Prior to your coming to the physician, what treatme nt(s) have you had for this episode of your low back pain? (Please mark all that apply.) € None € Surgery (Date and type of surgery:___________________________________________ _________) € Physical/Occupational Therapy € Medication (Date and type of medication:________________________________________ ______) € Chiropractic € Massage Therapy € Splint, Brace, or Cast € Shoe Inserts € Other (Please specify:___________________________________________ ___________________) 1. If you had to spend the rest of your life with the low back symptoms you have right now, how would you feel about it? € Very dissatisfied € Somewhat dissatisfied € Neutral € Somewhat satisfied € Very satisfied € Acupuncturist € Chiropractor € Emergency Room € General Practitioner € Internist € Massage Therapist € Neurosurgeon € Osteopath € Pain Clinic € Physical Therapist € Rheumatologist € Work Hardening Clinic € Nurse Practitioner € Other:__________________________

PAGE 218

203 2. What results do you expect from your treatment? (Ch eck one response on each row.) Definitely yes Probably yes Not sure Probably not Definitely not Not applicable Complete relief from symptoms (pain, stiffness, swelling, numbness, weakness, instability)       Moderate relief from symptoms (pain, stiffness, swelling, numbness, weakness, instability)       To do more every day household or yard activities       To sleep more comfortably       To go back to my usual tasks       To exercise and do more recreational activities       To prevent future disability       3. During the past week, how bothersome have these sym ptoms been? (Check one response on each row that best describes your average symptoms over the past week.) Not at all bothersome Slightly bothersome Somewhat bothersome Moderately bothersome Very bothersome Extremely bothersome Low back and/or buttock pain       Leg pain       Numbness or tingling in leg and/or foot      

PAGE 219

204 1. What is your current marital status? Single Married Living with significant other Divorced/separated Widowed 2. What level of education have you completed? Less than high school Graduated from high school Some college Graduated from college Some post-graduate course work Completed post-graduate degree The following is a list of common health problems. In the first column please indicate if you currently or have ever had any of the problems in t he past. In the second column please indicate if you are currently receiving treatment for the pr oblem. In the last column please indicate if the problem limits any of your daily activities. Do you or have you had the problem? Do you currently receive treatment for this problem? Does this problem limit your daily activities? Osteoporosis Yes No Yes No Yes No Spinal Compression Fracture Yes No Yes No Yes No Heart Disease Yes No Yes No Yes No High Blood Pressure Yes No Yes No Yes No Low Blood Pressure Yes No Yes No Yes No Lung Disease Yes No Yes No Yes No Diabetes Yes No Yes No Yes No Weakness in leg and/or foot      

PAGE 220

205 Ulcer or Stomach Disease Yes No Yes No Yes No Kidney Disease Yes No Yes No Yes No Liver Disease Yes No Yes No Yes No Anemia or Other Blood Disease Yes No Yes No Yes No Cancer Yes No Yes No Yes No Seizures Yes No Yes No Yes No Fainting Yes No Yes No Yes No Dizziness or Vertigo Yes No Yes No Yes No Nerve Disease or Disorder Yes No Yes No Yes No Muscle Disease or Disorder Yes No Yes No Yes No Hearing Loss Yes No Yes No Yes No Eye Disease or Injury Yes No Yes No Yes No Osteoarthritis Yes No Yes No Yes No Rheumatoid arthritis Yes No Yes No Yes No Skin Disease Yes No Yes No Yes No Other Medical Problems (please specify) ____________________________________________________ Yes No Yes No Yes No 1. Are you currently taking any medications (over the counter and/or prescribed)? No Yes (If yes, please list the medications that you a re currently taking.)

PAGE 221

206 Name of Medicine Dose (Milligrams) How many pills? How many times per day? 2. During the past week how often have you taken pain medication, includi ng narcotics or over-the-counter medications for your low back pain ? Not at all Once a week Once every couple of days Once or twice a day Three or more times a day 3. Have you smoked at least 100 cigarettes in your lif etime? No Yes 4. If you smoked more than 100 cigarettes in your life time (Proceed to the lifestyle section if you have not smoked more than 100 cigarettes in you r lifetime): 5. On average during all of the years that you have sm oked, how many cigarettes did you usually smoke per day? 1 10 11 20 21 40 More than 40 6. Do you smoke cigarettes now? No Yes 7. Except for the times that you quit, how many years all together have you smoked cigarettes? 0 5 6 10

PAGE 222

207 11 20 More than 20 years Lifestyle 1. Which statement best describes the work you do? (If retired, answer based on everyday activities.) € Mostly sedentary € Sedentary, substantial amount of walking required € Moderately active; walking, some lifting, and carry ing € Demanding physical activity, heavy lifting, and car rying 2. Describe your employment status. Work regular duty full time Work regular duty part time Work light duty or modified position full time Work light duty or modified position part time Temporarily unable to work due to health status Permanently unable to work or retired due to health status Retired (not due to health status) Unemployed Homemaker (not working outside the home) Student (not currently working) 3. How much physical activity do you do in a typical w eek. Less than 3 days a week 3 to 4 days a week 5 days a week Greater than 5 days a week 4. Mark the type of physical activity you participate in: Very light level. Light activity. Moderate level activity. Vigorous level of activity. I do not participate in physical activity.

PAGE 223

208 Indicate by circling the comment next to the treatm ent that corresponds to your amount of agreement with the following statement. Substitute each treatment into the blank as you consider your response. I believe ________________ will significantly help to improve this episode of my back pain. Medication completely disagree somewhat disagree neutral somewhat agree completely agree Rest completely disagree somewhat disagree neutral somewhat agree completely agree Surgery completely disagree somewhat disagree neutral somewhat agree completely agree Epidural Steroid Injections completely disagree somewhat disagree neutral somewhat agree completely agree Massage completely disagree somewhat disagree neutral somewhat agree completely agree Joint Manipulation completely disagree somewhat disagree neutral somewhat agree completely agree Traction completely disagree somewhat disagree neutral somewhat agree completely agree Aerobic exercise (i.e. walking, stationary cycling, Stairmaster, etc.) completely disagree somewhat disagree neutral somewhat agree completely agree Range of motion exercises (i.e. stretching) completely disagree somewhat disagree neutral somewhat agree completely agree Strengthening exercises completely disagree somewhat disagree neutral somewhat agree completely agree

PAGE 224

84= MODIFIED OSWESTRY LOW BACK PAIN DISABILITY QUESTION NAIRE1 Section 1: To be completed by patient Age:_ ______ Date:_________ Gender: Male / Female Section 2: To be completed by patient This questionnaire has been designed to give your t herapist information as to how your back pain has a ffected your ability to manage in every day life. Please answer every question by placing a mark on the line that b est describes your condition today. We realize you may feel that two of the statements may describe your condition, but please mark only the line which most closely describes you r current condition. Pain Intensity _____The pain is mild and comes and goes. _____The pain is mild and does not vary much. _____The pain is moderate and comes and goes. _____The pain is moderate and does not vary much. _____The pain is severe and comes and goes. _____The pain is severe and does not vary much. Personal Care (Washing, Dressing, etc.) _____I do not have to change the way I wash and dr ess myself to avoid pain. _____I do not normally change the way I wash or dr ess myself even though it causes some pain. _____Washing and dressing increases my pain, but I can do it without changing my way of doing it. _____Washing and dressing increases my pain, and I find it necessary to change the way I do it. _____Because of my pain I am partially unable to w ash and dress without help. _____Because of my pain I am completely unable to wash or dress without help. Lifting _____I can lift heavy weights without increased pa in. _____I can lift heavy weights but it causes increa sed pain _____Pain prevents me from lifting heavy weights o ff of the floor, but I can manage if they are conve niently positioned (ex. on a table, etc. ). _____Pain prevents me from lifting heavy weights o ff of the floor, but I can manage light to medium w eights if they are conveniently positi oned. _____I can lift only very light weights. _____I cannot lift or carry anything at all. Walking _____I have no pain when walking. _____I have pain when walking, but I can still wal k my required normal distances. _____Pain prevents me from walking long distances. _____Pain prevents me from walking intermediate di stances. _____Pain prevents me from walking even short dist ances. _____Pain prevents me from walking at all. Sitting _____Sitting does not cause me any pain. _____I can only sit as long as I like providing th at I have my choice of seating surfaces. _____Pain prevents me from sitting for more than 1 hour. _____Pain prevents me from sitting for more than 1 /2 hour. _____Pain prevents me from sitting for more than 1 0 minutes. _____Pain prevents me from sitting at all.

PAGE 225

834 1 adapted from Hudson-Cook N, Tomes-Nicholson K, Bre en A. A revised oswestry disability questionnaire. In: Roland M, Jenner J, eds. Back Pa in: New Approaches to Rehabilitation and Education. New York: Manchester University Press; 1 989. p. 187-204. [Prepared May 1999] Section 2 (cont): To be completed by patient Standing _____I can stand as long as I want without increas ed pain. _____I can stand as long as I want but my pain inc reases with time. _____Pain prevents me from standing more than 1 ho ur. _____Pain prevents me from standing more than 1/2 hour. _____Pain prevents me from standing more than 10 m inutes. _____I avoid standing because it increases my pain right away. Sleeping _____I get no pain when I am in bed. _____I get pain in bed, but it does not prevent me from sleeping well. _____Because of my pain, my sleep is only 3/4 of m y normal amount. _____Because of my pain, my sleep is only 1/2 of m y normal amount. _____Because of my pain, my sleep is only 1/4 of m y normal amount. _____Pain prevents me from sleeping at all. Social Life _____My social life is normal and does not increas e my pain. _____My social life is normal, but it increases my level of pain. _____Pain prevents me from participating in more e nergetic activities (ex. sports, dancing, etc.) _____Pain prevents me from going out very often. _____Pain has restricted my social life to my home _____I have hardly any social life because of my p ain. Traveling _____I get no increased pain when traveling. _____I get some pain while traveling, but none of my usual forms of travel make it any worse. _____I get increased pain while traveling, but it does not cause me to seek alternative forms of trav el. _____I get increased pain while traveling which ca uses me to seek alternative forms of travel. _____My pain restricts all forms of travel except that which is done while I am lying down. _____My pain restricts all forms of travel. Employment/Homemaking _____My normal job/homemaking activities do not ca use pain. _____My normal job/homemaking activities increase my pain, but I can still perform all that is requir ed of me. _____I can perform most of my job/homemaking dutie s, but pain prevents me from performing more physically stressful activities (ex. l ifting, vacuuming) _____Pain prevents me from doing anything but ligh t duties. _____Pain prevents me from doing even light duties _____Pain prevents me from performing any job or h omemaking chores. Section 3: To be completed by physical therapist SCORE:___________ or ___________%

PAGE 226

833 Pain Diagram and Pain Rating Name:_______________________________ Date:_____/___ __/_____ Please use the diagram below to indicate the sympto ms you have experienced over the past 24 hours. Use the key to indicate the type of symptoms nrnrnrnn rnr !" rnrn#r$r nnnrr nnrnnrnr !" rnr n#r$r

PAGE 227

838 LOW BACK ACTIVITY CONFIDENCE SCALE The following items are designed to determine what types of activities you can do easily, which are more difficult, and which you cannot do success fully. Please indicate your level of confidence, at the present time, in doing the activ ity in question by circling the appropriate percentage. Select the response that most closely matches you r own, remembering that there are no right or wrong answers. For example, in question #1 if you have almost comp lete confidence that you could carry a box that weighs 25 lbs. from a car into your home, you might circle 90%. If, however, you had no confidence that you could carry a box that w eighs 25 lbs. from a car into your home, you would circle 0%. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% NO MODERATE COMPLETE CONFIDENCE CONFIDENCE CONFIDENCE I believe that I can Â… 1. carry a box that weighs 25 lbs. from a car into my home. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 2. move a heavy phone book from an overhead cabinet to a low shelf. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 3. push a sofa 10 feet into a new location across c arpet. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 4. sit for a 6-hour plane trip. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 5. climb 3 flights of stairs. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 6. walk 1 mile (10 city blocks) non-stop. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 7. stand in a slow-moving line for 3 hours. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% NO MODERATE COMPLETE CONFIDENCE CONFIDENCE CONFIDENCE

PAGE 228

839 I believe that I can Â… 8. do what I have to do to take care of my back. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 9. control my low back problem in such a way that I can do the things I enjoy. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 10. find strength within myself to deal with the f rustration of low back pain. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 11. continue to do my exercise program even when I have pain or discomfort. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 12. continue to do my exercise program even if I ha ve no current symptoms of my low back problem. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 13. exercise regularly even if I was bored by the program or activity. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 14. exercise when there is no one around to offer encouragement. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 15. exercise when I need to start up with the prog ram again after lapsing. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

PAGE 229

83; None of the time A little of the time Some of the time Most of the time All of the time MOS Social Support Survey Someone you can count on to listen to you when you need to talk 1 2 3 4 5 Someone to give you information to help you understand a situation 1 2 3 4 5 Someone to give you good advice about a crisis 1 2 3 4 5 Someone to confide in or talk to about yourself or your problems 1 2 3 4 5 Someone whose advice you really want 1 2 3 4 5 Someone to share your most private worries and fears with 1 2 3 4 5 Someone to turn to for suggestions about how to deal with a personal problem 1 2 3 4 5 Someone who understands your problems 1 2 3 4 5 Someone to help you if you were confined to bed 1 2 3 4 5 Someone to take you to the doctor if you needed it 1 2 3 4 5 Someone to prepare your meals if you were unable to do it yourself 1 2 3 4 5 Someone to help with daily chores if you were sick 1 2 3 4 5 Someone who shows you love and affection 1 2 3 4 5 Someone to love and make you feel wanted 1 2 3 4 5 People sometimes look to others for companionship, assistance, or other types of support. How often is each of the following kinds of support available to you if you need it? Circl e one number on each line.

PAGE 230

83D Someone who hugs you 1 2 3 4 5 Someone to have a good time with 1 2 3 4 5 Someone to get together with for relaxation 1 2 3 4 5 Someone to do something enjoyable with 1 2 3 4 5 Someone to do things with to help you get your mind off things 1 2 3 4 5

PAGE 231

83: The SF-36v2 Health Survey Instructions for Completing the Questionnaire This survey asks for your views about your health. This information will help keep track of how you feel and how well you are able to do your u sual activities. Please answer every question by marking the answer as indicated. If yo u are unsure about how to answer a question, please give the best answer you can. Your Health in General 1. In general, would you say your health is (check one ): Excellent Very good Good Fair Poor 2. Compared to one year ago, how would you rate your health in general now? (check one) Much better now than one year ago Somewhat better now than one year ago About the same as one year ago Somewhat worse now than one year ago Much worse now than one year ago

PAGE 232

83E 3. The following questions are about activities you mi ght do during a typical day. Does your HEALTH now limit you in these activities? If so, how much? (check one on each line) Yes, limited a lot Yes, limited a little No, not limited at all a) Vigorous activities such as running, lifting heavy objects, participating in strenuous sports b) Moderate activities such as moving a table, pushing a vacuum cleaner, bowling, or playing golf c) Lifting or carrying groceries d) Climbing several flights of stairs e) Climbing one flight of stairs f) Bending, kneeling, or stooping g) Walking more than a mile h) Walking several blocks Yes, limited a lot Yes, limited a little No, not limited at all i) Walking one block j) Bathing or dressing yourself

PAGE 233

832 4. During the PAST WEEK, how much of the time have you had any of the following problems with your work or other regular daily activities as a result of your physical health? (check one on each line) All of the time Most of the time Some of the time A little of the time None of the time a) Cut down on the amount of time you spent on work or other activities b) Accomplished less than you would like c) Were limited in the kind of work or other activities d) Had difficulty performing the work or other activities (for example, it took extra effort) 5. During the PAST WEEK, how much of the time have you had any of the following problems with your work or other regular daily activities as a result of any emotional problems (such as feeling depresse d or anxious)? (check one on each line) All of the time Most of the time Some of the time A little of the time None of the time a) Cut down on the amount of time you spent on work or other activities b) Accomplished less than you would like c) Did work or other activities less carefully than usual

PAGE 234

83= 6. During the PAST WEEK, to what extent has your physi cal health or emotional problems interfered with your normal soci al activities with family, friends, neighbors, or groups? (check one) Not at all Slightly Moderately Quite a bit Extremel y 7. How much bodily pain have you had during the PAST W EEK? (check one) None Very mild Mild Moderate Severe Very severe 8. During the PAST WEEK, how much did pain interfere w ith your normal work (including both work outside the home and hous ework)? (check one) Not at all A little bit Moderately Quite a bit Extr emely 9. These questions are about how you feel and how thin gs have been with you during the PAST WEEK. For each question, please give the one answer that comes closest to the way you have been feeling. How much of the time du ring the PAST WEEK... (check one answer on each line) All of the time Most of the time Some of the time A little of the time None of the time a) Did you feel full of life? b) Have you been a very nervous person? c) Have you felt so down in the dumps that nothing could cheer you up? d) Have you felt calm and peaceful? e) Did you have a lot of energy? f) Have you felt downhearted and blue? g) Did you feel worn out? h) Have you been a happy person? i) Did you feel tired?

PAGE 235

884 10. During the PAST WEEK, how much of the time has your physical health or emotional problems interfered with your social activities (li ke visiting friends, relatives, etc.)? (check one) All of the time Most of the time Some of the time A little of the time None of the time 11. How TRUE or FALSE is each of the following statemen ts for you? (check one answer for each line) Definitely true Mostly true Don't know Mostly false Definitely false a) I seem to get sick a little easier than other people b) I am as healthy as anybody I know c) I expect my health to get worse d) My health is excellent THANK YOU FOR COMPLETING THIS QUESTIONNAIRE!

PAGE 236

89E APPENDIX E INTERVIEW GUIDE The Patient Interview Questions I am going to ask you a series of open ended questi ons about your experiences with having lumbar spinal stenosis. I want to hear in your wor ds what your feelings are and experiences have been with this condition. Feel free to elabor ate or explain your thoughts as you feel necessary. At the end, if there is anything I have n’t asked you that you would like to share, I would like you to have the opportunity to share thi s information. Main disease problems Tell me in your own words what it is like to have L SS. Describe to me how your life is different now that you have LSS. Does having LSS impact your interactions? For exam ple: Your perceptions about yourself and what can do Your relationships Your ability to do daily activities Your ability to do social or recreational activitie s Rehabilitation Treatment Tell me about your experiences with Epidurals Tell me about your experiences with physical therap y Tell me about what factors made your experience in physical rehabilitation (ESI and PT) Supportive o Positive – helpful o Negative – helpful Challenges Describe any education or advice you were given. Describe any tasks you were given. Did you experienced problems with adhering to these and why? Describe how your pain was improved or got worse du ring rehabilitation o What factors do you think contributed to the change in your pain? Contextual Factors In General/Everyday: What do you think influenced your ability to manage LSS the most? Rehab: What do you think influenced your ability to partic ipating in rehabilitation (either ESI or PT) the most?

PAGE 237

892 Do you remember being confident or not confident to participate in rehabilitation before you started your treatment program? Confidence with Tasks: Describe any task you originally felt you were not confident to perform physically, mentally, or socially. Are there now tasks you feel you can now perform? Describe how this relates to your back problems Are there any tasks you feel you now cannot perform ? Describe how this relates to you back problems Support Sources What did your family or friends think of all of thi s? Did they support you? Was it helpful or not helpful? Who do you think influenced you the most during thi s time? Were any friends or family members, neighbors or an yone else sources of support sources during his stressful health experience? Did you feel you had: Emotional support? Informational support? Tangible support? Affectionate support? o Describe who provided this support. o Describe if the support was either helpful or unhel pful