Disability and quality of life in people with peripheral arterial disease and intermittent claudication

Material Information

Disability and quality of life in people with peripheral arterial disease and intermittent claudication
Scherer, Susan A
Publication Date:
Physical Description:
173 leaves : ; 28 cm

Thesis/Dissertation Information

Doctorate ( Doctor of Philosophy)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
Department of Health and Behavioral Sciences, CU Denver
Degree Disciplines:
Health and Behavioral Sciences
Committee Chair:
Regensteiner, Judith G.
Committee Co-Chair:
Hiatt, William R.
Committee Members:
Tracer, David
Zamudio, Stacy


Subjects / Keywords:
Peripheral vascular diseases ( lcsh )
Quality of life -- Evaluation ( lcsh )
Disability evaluation ( lcsh )
Walking -- Evaluation ( lcsh )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


Includes bibliographical references (leaves 152-173).
General Note:
Department of Health and Behavioral Sciences
Statement of Responsibility:
by Susan A. Scherer.

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:
54523267 ( OCLC )
LD1190.L566 2003d S33 ( lcc )

Full Text
Susan A. Scherer
B.S. University of Maryland, 1979
M.A. University of Phoenix, 1992
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree
Doctor of Philosophy
Health and Behavioral Sciences

This thesis for the Doctor of Philosophy
degree by
Susan A. Scherer
has been approved
JLtu as. 2003
iacy 2nmudio

Scherer, Susan A (Ph.D., Health and Behavioral Science)
Disability and Quality of Life in People with Peripheral Arterial Disease and
Intermittent Claudication
Thesis directed by Professor Judith G. Regensteiner
Peripheral arterial disease (PAD) is a common manifestation of
atherosclerosis involving narrowing of the lower extremity arteries and adversely
affecting health, walking and quality of life. Symptomatic PAD (PAD-IC) is
characterized by intermittent claudication (IC), a cramping leg pain that occurs
with walking, causing the individual to stop walking until the pain subsides.
The aims of this study were to describe the differences in walking speed,
walking distance and gait parameters between subjects with PAD-IC and controls,
and to describe the factors that explain physical function score (PFS) and quality
of life (PADQOL) in subjects with PAD-IC.
This cross-sectional study included 25 subjects with PAD-IC and 26 controls
similar in age, gender and physical activity status. Subjects participated in one
session where gait was measured and questionnaires administered. Differences in
means were analyzed using student t-tests. Univariate and multivariate linear
regression methods were used to identify the factors that explained PFS and
PADQOL in subjects with PAD-IC.

Groups were similar in demographic and psychosocial variables (p> .05).
There were no significant differences between groups in either usual or maximal
walking speed or any gait variable (step length, cadence, stride width, toe out
angle) at either usual or maximal walking speed (p>.05). Six-minute walk
distances were significantly less in the PAD-IC group than the control group
(p <.001). PAD group was the most significant predictor of PFS; this relationship
was not modified by any independent variable. Social support score was the most
significant contributor to PADQOL in subjects with PAD-IC (r = .623, p=. 001).
In PAD-IC, the level of disability appears to be a direct result of PAD and IC;
walking measurements are insignificant modifiers of the level of disability. Social
support levels contribute more to quality of life than either physical function or
personal beliefs. A treatment approach that includes exercise training and social
interaction is recommended to improve both physical function and quality of life.
This abstract accurately represents the content of the candidates thesis. I
recommend its publication.

I dedicate this thesis to two colleagues who are unable to see this project to
completion, but who were instrumental in encouraging me to pursue this doctoral
degree. This is dedicated to Polly Cerasoli and Jody Delehanty.

I could not have accomplished this endeavor without the help of many individuals.
Many thanks to my friends and family who offered encouragement, especially
Pam. I am grateful to my advisor, Judy Regensteiner for her comments, patience
and persistence over the course of the dissertation. My thanks to the others on my
committee who offered advice. I also appreciate the support from my fellow
faculty at the University of Colorado Physical Therapy Program.
This research was supported by the Adult General Clinical Research Center
(GCRC), University of Colorado Hospital, Grant number M01 RR00051.
Research support was also provided by the Vascular Research Program and the
ICR Committee of the Department of Rehabilitation Medicine (Doppler Unit).
Colorado Multiple Institutional Review Board (COMIRB)
Human Subjects Approval (02-104)

1. Introduction......................................................1
2. Review of Relevant Literature.....................................4
2.1 Overview..........................................................4
2.2 Peripheral Arterial Disease.......................................5
2.2.1 Pathophysiology...................................................5
2.2.2 Prevalence of PAD................................................10
2.2.3 Risk Factors.....................................................10
2.2.4 Prognosis........................................................11
2.2.5 Assessment and Treatment of Peripheral Arterial Disease and
Intermittent Claudication.......................................12
2.3 Conceptual Frameworks and Models for Disability
Associated with Chronic Disease ................................16
2.3.1 Sociologic and Rehabilitation Models of Disability...............16
2.3.2 Medical Models of Disability.....................................18
2.3.3 A Biopsychosocial Model of Disability/Ability....................21
2.3.4 Expanded Disablement Models......................................22

2.4 Factors Contributing to Physical Disability .................. 23
2.4.1 Disablement Model for Individuals with PAD.....................28
2.5 Role of Walking and Gait Parameters in Physical Function.......30
2.5.1 Walking, Gait and Physical Disability in People with PAD-IC....36
2.6 Conceptual Frameworks of Quality of Life.......................42
2.6.1 Health-related Quality of Life in PAD..........................48
2.7 Summary and Purpose........................................... 50
2.8 Preliminary Results............................................58
3.0 Research Design and Methodology................................60
3.1 Specific Aims and Hypotheses...................................60
3.2 Research Design Overview.......................................62
3.3 Subjects.......................................................63
3.3.1 S election of Study Population.................................63
3.3.2 Inclusion/Exclusion Criteria...................................64
3.4 Procedures ....................................................65
3.5 Methods of Assessment..........................................66
3.5.1 Demographic Characteristics....................................67
3.5.2 Measures of Walking Function and Gait Abnormalities............69 Walking Speed.................................................69 Gait Parameters- Stride length, Width, Cadence................72
3.5.3 Questionnaires and Survey Instruments..........................75

3.6 Statistical Analysis .............................................83
3.6.1 Sample Size Estimation............................................83
3.6.2 Analysis for Specific Aim 1.......................................84
3.6.3 Analysis for Specific Aim 2.......................................85
3.6.4 Analysis for Specific Aim 3.......................................86
4.0 Results...........................................................90
4.1 Demographics and Physical Function Characteristics................90
4.2 Differences in Gait Parameters between Subjects
with PAD and Controls.............................................91
4.3 Factors Which Contribute to Physical Function Score...............96
4.4 Contributors to Health-Related Quality of Life in Subjects
with PAD.........................................................103
5. Discussion.......................................................117
5.1 Overview of Significant Findings.................................117
5.2 Demographic and Clinical Characteristics.........................119
5.3 Walking Speed....................................................121
5.4 Gait Parameters .................................................127
5.4.1 Gait Parameters at Maximal Walking Speed.........................134
5.5 Walking Distance- Six Minute Walk................................135
5.6 Explaining Physical Function.....................................137
5.7 Quality of Life..................................................140
5.8 Limitations of Study.............................................143

5.9 Implications for Disablement Framework
5.10 Recommendations for Professional Practice
and Future Research.........................................147
A. Physical Function Score Questions............................150
B. Mastery Questions............................................151

2.1 Comparison of Disablement Frameworks.........................52
2.2. Expanded Model of Disablement................................53
2.3 Disablement Process in PAD...................................54
2.4 Phases of Gait...............................................55
2.5 Model of Contributors to Health-Related Quality
of Life in PAD..............................................56
3.1 Placement of Shoe markers for Gait Analysis..................87
3.2 Measurement of Step Length, Stride Length and Stride Width...88
3.3 Measurement of Toe Out Angle.............;..................89
4.1 Relationship of Usual Walk Speed and Stride Width ..........116

2.1 Reported Gait Abnormalities in PAD-IC............................57
4.1 Demographic Characteristics of Subjects with Peripheral
Arterial Disease (PAD) and Control Subjects (non-PAD)..........105
4.2 Physical Function Characteristics of Subjects with Peripheral
Arterial Disease (PAD) and Control Subjects (non-PAD)..........106
4.3 Gait Characteristics at Usual and Maximal Walking Speeds........107
4.4 Pearson Correlations of Gait Variables with Usual and Maximal
Walking Speed by Group..........................................108
4.5 Univariate Regressions for Predictors of Physical Function
(All Subjects)..................................................109
4.6 Multivariate Linear Regression Models for Predictors of Physical
Function Score in All Subjects.................................110
4.7 Univariate Regressions for Predictors of Physical Function
Score within Each Group ........................................Ill
4.8 Bivariate Regression (PAD group + a Single Confounding Variable)
for Physical Function Scores (All Subjects) ...................112
4.9 Multivariate Linear Regression Models for Predictors of
Physical Function Score in PAD-IC Group.........................113
4.10 Personal and Social Characteristics of Subjects with PAD........114
4.11 Multiple Linear Regression Model for Predictors of PADQOL......115

1. Introduction
Peripheral arterial disease (PAD) is a common manifestation of
atherosclerosis involving narrowing of the arterial circulation in the lower
extremities and adversely affecting health, walking and quality of life.The
prevalence of PAD has been estimated at 12% of the general population,2
a t
increasing to 20% in people over the age of 70. In people with PAD, less than
20% report symptoms indicative of low blood flow to the extremities.4
Symptomatic PAD is most characterized by intermittent claudication (IC), or
cramping leg pain that occurs with walking, causing the individual to stop walking
and rest until the pain subsides. Patients with PAD and IC (PAD-IC) have an
increased risk of mortality and morbidity related to the systemic atherosclerosis.5
Patients with PAD-IC report difficulties with walking quickly and for long
distances, resulting in limitations of ambulatory activities that can be significantly
disabling.1 Some authors suggest that people with PAD-IC have abnormal gait
parameters consisting of shorter steps and slower cadence.6 Patients also report
that PAD-IC has an adverse effect on health-related quality of life (HRQOL).7
The predominant research focus in PAD-IC has been on describing the
pathophysiology of peripheral atherosclerosis as well as characterizing the

resultant abnormalities in cell, nerve and muscle function. Some studies have
assessed the severity of walking limitations in people with PAD-IC. Recent
research suggests that PAD without symptoms is associated with poorer lower
extremity performance on walking tests performed in a clinical setting. This
dissertation evaluated the consequences of PAD-IC on gait, self-reported ability to
perform daily activities and HRQOL.
The International Classification of Impairment, Disability and Handicap
(ICIDH) is an appropriate model for describing the consequences of disease on an
individuals ability to participate in daily activities;9 however, this model has never
been used in subjects with PAD-IC. The ICIDH disease pathway is:
Pathology -Impairments> Functional limitation Disability.
The ICIDH model indicates that pathology leads to abnormalities in structure
or function of an organ system (impairment), inability of a person to perform basic
physical actions in a normal fashion as measured in a clinical setting (functional
limitation), and difficulty in performing activities normally required in daily life
(disability). For this dissertation, functional limitation measurements were walking
speed and distance (six-minute walk), gait parameters (cadence, step length and
width, toe out angle) and calf strength.10 Disability was defined as self-reported
difficulty with daily activities ranging from personal care to community-based
walking,11 and scored using the physical function subscale of the Medical

Outcomes Study short form (SF)-36.12 In this study, the disability measure was
labeled self-reported physical function score (PFS).
The negative effect of PAD-IC on HRQOL is presumably secondary to
severity of symptoms or difficulty in walking7; however, little is known about
whether an individuals personal characteristics or social supports also can predict
quality of life. This dissertation explored the relative contributions of personal
beliefs (Mastery) and social supports on self-reported HRQOL in subjects with
PAD-IC. A HRQOL instrument designed specifically for use in PAD (PADQOL)
was used as the outcome measure of quality of life.13
Therefore, the overall purpose of this study was to characterize the
consequences of PAD-IC on functional limitations (gait parameters), disability
(PFS), and quality of life (PADQOL).
The aims of this study were three-fold:
1) Describe the differences in walking speed, walking distance and gait
parameters between subjects with PAD-IC and a control group similar in
age, gender and physical activity status
2) Describe and evaluate the factors that explain PFS in subjects with
3) Explore the physical and psychosocial factors that explain PADQOL in
subjects with PAD-IC.

2. Review of Relevant Literature
2.1 Overview
Atherosclerosis is a major health problem in the elderly, and may present as
disease in the coronary, cerebral or peripheral vessels. While coronary artery and
cerebral vascular disease are commonly recognized as serious sequelae of
atherosclerosis, PAD is often overlooked, despite having the potential for similar
cardiovascular complications.14
This chapter reviews the pathophysiology of PAD, diagnosis and risk for the
condition; abnormalities related to the disease state and current treatment
approaches. A conceptual basis for explaining disability in the elderly is presented,
followed by a review of the current level of knowledge of physical limitations
observed in people with PAD-IC. Finally, the constructs and contributors to
health-related quality of life are discussed, including gaps in knowledge of
HRQOL in persons with PAD-IC.

2.2 Peripheral Arterial Disease (PAD)
2.2.1 Pathophysiology
PAD is a common manifestation of atherosclerosis, a systemic condition of
plaque deposition in the arteries leading to narrowing of large to medium-sized
arteries.15,16 Atherosclerosis may occur in the coronary, cerebral and peripheral
vessels, and is likely to be present in all three vascular beds. In a study of
prevalence of these forms of atherosclerosis in hospital based general practice,
84% of patients had coronary heart disease (CHD), 19% had stroke and 13% had
PAD. Forty-five percent had all three forms of atherosclerosis. In people with
PAD, CHD co-existed in 68% of these individuals and cerebrovascular disease
(CVD) in 42%.17
Atherosclerosis is characterized by damage to the endothelial lining of
arteries, leading to buildup of lipids, cholesterol, platelets and other compounds on
the interior surface of the artery. In peripheral arterial occlusive disease, the
arterial narrowing commonly occurs in the femoral/popliteal or aortoiliac arteries,
leading to decreased blood flow and diminished oxygen delivery to the muscles
supplied by the affected arteries. Artery diameter may decrease enough so that
peripheral pulses are diminished or absent. The clinical presentation of arterial
occlusion varies; patients may have no symptoms, symptoms may occur only
during exercise, when the demand for oxygen exceeds oxygen delivery, or there

may be ischemic pain at rest. The artery must be approximately 90% occluded to
produce a reduction in blood pressure and blood flow and consequently cause pain
at rest. In patients whose artery diameter is moderately decreased, there may be
pain only when the demand for oxygen exceeds the delivery ability of the artery.
Under exercise conditions with increased blood flow velocity, a 70% occlusion
1 8
leads to decreased blood pressure and flow.
The best noninvasive method to evaluate the presence of PAD is to determine
the ankle-brachial index (ABI). While the patient lies supine, systolic blood
pressures are measured in the brachial (arm), dorsalis pedis and posterior tibial
(leg) arteries, using Doppler ultrasonography. The ABI is calculated by dividing
the highest systolic blood pressure in each of the leg arteries by the highest arm
systolic pressure.19 The presence of PAD is defined as an ABI of less than or equal
to 0.90 m at least one leg. Other vascular assessment of peripheral flow can
assist in identifying the level of occlusion.22
The primary symptom of moderately severe PAD is termed intermittent
claudication (IC), a cramping pain in the legs that occurs with walking and
decreases with rest. It is characterized by being highly reproducible, occurring
with a consistent level of walking from one day to the next, disappearing after 2-3
minutes of rest and recurring at a consistent level of exercise.23 A number of

questionnaires can be used to assess the presence and extent of intermittent
A normal cardiovascular system will respond to lower extremity exercise by
increasing cardiac output and decreasing arterial resistance within the exercising
muscle, producing an increase in systolic blood pressure while maintaining normal
diastolic pressure, so that blood supply to the exercising muscle can meet the
oxygen demand. The normal vascular response to exercise is to increase systolic
blood pressures in both arm and leg vessels. Patients with PAD have altered lower
extremity blood flow and peripheral hemodynamics, particularly during exercise.
In patients with PAD, the ankle systolic pressures are unable to increase, and thus
blood flow is not adequate to meet the demand of the exercising muscle. As work
intensity increases, this supply-demand mismatch increases, leading to symptoms
of claudication pain. If the patient stops exercising, the supply-demand mismatch
improves, and the symptoms decrease. This abnormal exercise response may be
measured using an ABI measurement immediately after an exercise bout, and
determining the drop in ankle pressures. A 20% drop in ankle pressure after
exercise is considered an indication of PAD.
PAD appears to also be associated with decreased metabolic ability to produce
energy dining exercise. Oxygen delivery is impaired, as observed in faster tissue
deoxygenation in patients with PAD.25 The rate of ATP synthesis in the oxidative

pathway is decreased by about 42% in subjects with PAD compared to controls,
and ability to replenish energy stores of phosphocreatine is diminished as well.26,27
A limitation in ability to produce energy would lead to low aerobic capacity and
inability to sustain submaximal work for a period of time. Indeed, patients with
PAD have a low peak VO2, even when adjusted for differences in body weight.28
The secondary effects of this intermittent ischemia include changes in
mitochondrial content, muscle fiber, nerves and capillaries. Mitochondria are the
site of oxidative energy production and mitochondrial numbers can increase or
decrease to meet tissue demands. An increase in mitochondrial content is
associated with increases in aerobic capacity, while deconditioning leads to
decreased mitochondria.29 In PAD, this relationship is abnormal, as the low flow
state has been associated with either normal levels or an increase in mitochondria
content.30'32 Patients with PAD demonstrate accumulations of intermediate
components of oxidative metabolism, indicating an inefficient metabolic process.
One such intermediate compound is short chain acylcamitine, which is important
in freeing up acetyl co-enzyme A (Go-A) to be used in the Krebs cycle for energy
metabolism.34 Since acylcamitine and acyl Co-A are normally in equilibrium, an
accumulation of acylcamitine in the muscle indicates that acyl-CoA groups are not
being effectively oxidized for metabolism. The amount of muscle short chain

acylcamitine is inversely linked with exercise capacity, with an accumulation of
acylcamitine associated with decreased walking ability.35
Intermittent ischemia in peripheral muscles may lead to muscle fiber atrophy
and denervation. In the gastrocnemius muscle, which is the primary site of
claudication symptoms, there are conflicting reports as to whether percentages of
Type I and II fibers are normal or reduced. These structural changes may
produce decreased leg strength which could contribute to walking difficulty in
these individuals. Dorsi-flexor muscle strength has been demonstrated to be 20%
less in subjects with PAD compared to controls, and the dorsi-flexor weakness was
associated with slower time to rise from a chair.40 Plantar flexion strength was
decreased in subjects with PAD compared to controls, but the strength scores did
not correlate with walking ability.41 Deficits in muscle endurance have also been
Although it has been hypothesized that chronic ischemia could cause a muscle
to adapt by increasing capillary numbers, this has not been consistently
demonstrated in PAD 43 Despite the importance of blood flow in exercise capacity,
hemodynamic measures are poorly correlated with walking ability in subjects with

2.2.2 Prevalence of PAD
The prevalence of PAD has been estimated between 3% and 12% of the
general population,2445'47increasing to 20% in people over the age of 70. The
prevalence of PAD increases with age, ranging from approximately 6% in men
ages 55-60 and 9% in women of the same age range with an increased prevalence
of 52% in men over age 85 and almost 60% of women over age 85.24 Around age
60, men have a higher prevalence of PAD,248 although as age increases, men and
women have similar rates of PAD.24 Educational level, social class and marital
status do not appear to be significantly related to low ABI values, as demonstrated
in one study.48
There is an overlap, but not a total correlation between low ABI and
symptoms of intermittent claudication. In people with PAD, less than 20% report
symptoms indicative of low blood flow to the extremities.3446 When symptomatic,
IC appears to be associated with lower ABI values.48
2.2.3 Risk Factors
Risk factors for developing PAD are similar to those for other forms of
atherosclerosis. Hypertension, hyperlipidemia (particularly elevated triglycerides
and high density lipoprotein), elevated fibrinogen, current and former smoking,
body mass index and diabetes are all associated with increased risk of PAD.4647 Of
these, the two most significant risk factors for developing PAD-IC are smoking

and diabetes.45,49 Smoking is an independent risk factor for PAD and is a stronger
predictor for PAD than for CHD 50,51 Physical inactivity has also been implicated
as a risk factor. A retrospective study asked elderly men with PAD to indicate the
amount of physical activity performed between the ages of 35 and 45. Lower
levels of physical activity were associated with PAD in older men who had a
S? ^
smoking history.
2.2.4 Prognosis
Patients with intermittent claudication generally have a stable natural history
of symptoms. Many patients do not have worsening of symptoms, with
progression of symptoms in less than 25% of patients.54,55 If the arterial occlusion
is severe enough to impede blood flow at rest, the integrity of the tissues may be
compromised, leading to gangrene or amputation. Fortunately, less than 5% of
patients progress to amputation of the limb.5
Because persons with PAD (symptomatic or asymptomatic) commonly have
coronary or cerebral disease, they have an elevated risk of cardiovascular events
such as myocardial infarction and stroke as compared to patients without PAD.
For a patient with PAD, the relative risk (RR) of cardiovascular events has been
reported between 1.5 and 6.O.21,56,57 In addition, the relative risk of death from
cardiovascular events as well as all causes is higher in patients with PAD than
those without cardiovascular disease. The relative risk for all-cause mortality is

reported between 1.6 and 3.0 after adjusting for confounding factors.57 The relative
risk of death from cardiovascular disease is 4-5 times higher for those with PAD
versus those without.24 The annual rate for myocardial infarction and stroke deaths
is 5%.57
2.2.5 Assessment and Treatment of Peripheral Arterial Disease and
Intermittent Claudication
Treatment of the patient with PAD falls into two areas, treating systemic
disease and treating the symptoms of intermittent claudication. Since intermittent
claudication is a manifestation of atherosclerosis, recommendations for controlling
risk factors for future cardiovascular events are similar for all forms of
atherosclerosis. Blood pressure and cholesterol management are important in
decreasing risk of future cardiovascular events.58 Antiplatelet drugs reduce the risk
of vascular complications.59 People with PAD-IC should stop smoking, since
stopping smoking is associated with improvements in blood flow by ABI and in
one study, improved pain-free walking distance, but in general, does not help
decrease the level of claudication pain.60
In order to effectively treat the symptoms of IC, accurate measurement of
walking ability is necessary. Treadmill protocols provide two objective measures
of walking performance; pain-free walking time and maximal walking time. Pain
free walking time indicates the time to onset of claudication pain while maximal
walking time indicates the time to walk until claudication pain is severe and forces

the individual to stop walking. The variability in time to onset of claudication pain
has been reported to be about 25% from test to test.61 A drop of ankle systolic
pressure by more than 20% after exercise testing is considered abnormal.62,63 The
most frequently used community based walking measures are the six-minute walk
and walking speed at a comfortable and maximal pace over a specific distance. In
the six-minute walk, the patient is asked to cover as much distance over a
measured distance as is comfortable in six minutes.64 The six-minute walk test is
reliable for patients with PAD-IC, with an association between disease severity
(low ABI)65 and pain-free and maximal walking distance.66 Walking speed is
measured by asking the patient to walk at a comfortable pace over a specified
distance. Low ABI values are associated with slower walking at both usual and
maximal speeds.65
The Walking Impairment Questionnaire (WIQ)1 is the most commonly used
PAD-specific tool to assess self-reported walking ability. The WIQ characterizes
the patient's self reported difficulty in walking a specified distance, speed and
stairs and provides a score for each. The WIQ correlates with the gold standard of
treadmill walking performance and is sensitive to changes in performance after
exercise training.1
Treatment for the symptoms of IC is focused on relieving symptoms and
improving functional status and walking ability. The three main treatment options

for improving symptoms and improving functional status are drug therapy,
revascularization and exercise training.
Exercise training programs using intermittent treadmill walking in a
supervised setting universally improve pain-free and maximal walk time in
patients with PAD-IC.67'69 The most effective exercise interventions included
supervised walking to moderate claudication pain. An exercise program of 60
minutes, 3 times a week and for a duration of at least 12 weeks provided the most
benefit in both pain free and maximal walking time, with improvements up to
200%.68-71 Exercise training was superior to usual amounts of activity and to
strength training.41 Supervised walking exercise is more beneficial in increasing
walking distance than home-based therapy performed independently.72,73
The proposed mechanisms for improvement in walking ability include:
improvements in blood flow, metabolic adaptation (increase in oxidative enzyme
activity) in leg muscles or improvements in muscle strength.74 Blood flow
measurements correlate poorly with treadmill walking, either before or after
training.74,75 Measures of cellular metabolism improved in the exercise group in a
few studies.76,77 Muscle strength training was less effective in producing changes
in walking ability than walking exercise.41 Walking programs may improve
walking ability by other mechanisms such as improving cardiovascular fitness,
improving joint range of motion or increasing physical activity, although these

hypotheses have not been tested. Several authors have suggested that
improvements in walking ability after exercise training with walking may be due
to an improved gait economy or less oxygen consumed for a particular amount of
work.78,79 If gait economy is important, then baseline gait analyses are indicated as
an area of relevant study.
Medications approved in the United States for treating claudication symptoms
are limited to cilostazol (Pletal) and pentoxyfylline (Trental). Cilostazol, a
phosphodiesterase III inhibitor, inhibits platelet aggregation and lowers
triglycerides. Treatment with cilostazol significantly increased maximum walking
distance, improved walking ability as demonstrated by the walking impairment
questionnaire, and increased health related quality of life, as compared to treatment
with placebo. Pentoxifylline apparently works by decreasing platelet
aggregation, though its effect on walking is very modest.81
Patients with PAD unresponsive to conservative medical treatment or who
have intermittent claudication that severely limits daily walking may be considered
for limb revascularization. Since percutaneous angioplasty (PTA) and surgical
bypass carry some risk, conservative treatment is considered first. Outcomes of
surgical intervention include significant improvements in ABI, WIQ scores and
pain-free and maximal walking time.82 When compared to medical therapy (aspirin
and advice on stopping smoking,) PTA gave better results in terms of improving

treadmill walking distance and QOL.83 When PTA was compared to exercise in
one study, exercise therapy was better than PTA for increasing maximal walking
time and improving symptoms of IC.84 A comparison of surgery (bypass), surgery
plus exercise and exercise alone found surgery better than exercise in improving
treadmill walking distance, but the largest increases in pain-free walking occurred
oc t t
with surgery and exercise together. Although effective in specific cases, surgery
is more costly than exercise and carries additional risks for the patient.
2.3 Conceptual Frameworks and Models for Disability
Associated with Chronic Disease.
Disease and injury are significant causes of disability. Physical disability and
loss of independence contribute to a large percentage of health care expenditures.
The purpose of this section is to provide definitions and a conceptual framework
for understanding disability, describe the determinants of disability and indicate
the current knowledge of PAD-IC and disability. In describing the determinants of
disability, the focus is on how walking ability and gait parameters contribute to
physical disability.
2.3.1 Sociologic and Rehabilitation Models of Disability
Disability can be examined from either a biologic or social perspective. From
a biologic perspective, a person would be considered disabled due to disease or
injury that has altered a persons ability to perform critical daily tasks. However,
the inability to perform daily activities may lead to social disadvantages. For

example, a person in a wheelchair may be able to perform daily tasks adequately,
but finds that architectural barriers such as stairs impede his/her ability to locate
employment or adequate housing. The most common approach to disability
research is to combine these perspectives into a biopsychosocial model of
disability. This dual perspective is useful for medical practitioners, specifically
rehabilitation specialists, as well as sociologists. Rehabilitation therapists seek to
improve a persons functioning by improving the underlying cause of disability or
developing ways for individuals to compensate for their limitations. At issue for
sociologists is how to integrate disabled people into society and reduce the
disadvantages incurred from disability by changing social norms or policy. In this
paper, the conceptual issues of disability will be discussed from the rehabilitation
The sociologist Saad Nagi was one of the first to describe the conceptual
framework that has become the mainstay of disability research. Nagi defined a
main disease pathway consisting of 5 distinct levels of inter-related, but separate
entities that contribute to disability. Pathology is considered the disease or
condition that leads to disability. Impairments are abnormalities in structure or
function of body systems that are associated with pathology or remain after
pathology has resolved. Functional limitations are limits in abilities to perform
tasks and activities usually required for self-care or by social roles. The limitation

depends on the type of impairment as well as the type of activity. For example,
losing a finger may be more limiting for a pianist than for a farmer. Disability is
the pattern of behavior that evolves with long-term impairments and functional
2.3.2 Medical Models of Disability'
Nagi's original conceptual framework has been adopted by others and has
come to be known as the disablement framework. This disablement framework
was then used by the World Health Organization (WHO) and modified by both the
WHO and the Committee on a US National Agenda for the Prevention of
Disabilities. In the 1970's the WHO developed an instrument that could be used
worldwide to classify the consequences of disease. Underlying the development
of this model was a strong belief in the social model of disability. The original
WHO model was entitled the International Classification of Impairments,
Disability and Handicaps (ICEDH).89 The original definitions of the various levels
relied on Nagi's definitions.
Impairments were defined as any loss or abnormality of psychological,
physiological or anatomical structure or function". Impairments lead to disability,
that is, any restriction or lack (resulting from an impairment) of ability to
perform an activity in the manner or within the range considered normal for a
human being." Handicap is the term used to describe a disadvantage for a given

individual, resulting from an impairment or a disability, that limits or prevents the
fulfillment of a role that is normal (depending on age, sex and social and cultural
factors) for that individual.89
The ICEDH model was designed to examine both disease and social factors
contributing to disability. In actuality, the model implies a linear process. The
model suggests that people are disadvantaged (handicapped) as a result of their
impairments and disabilities, not by the lack the opportunities that exist for
disabled individuals. The original model does not measure the effect of the
physical or political environment on disability, but it is a reasonable model for
looking at effects of disease on an individual.
The National Center for Rehabilitation Research (NCMRR) modified Nagi's
model by clarifying definitions and adding a 5th category titled societal limitation.
This category of societal limitation is similar to the ICIDH definition of
Handicap.90 Figure 2.1 compares these three models.
One area of confusion is that although the models represent similar
descriptions of the impact of disease, they use different terminology to describe the
same concepts. The first difference is at the level of task performance. All models
indicate that the ability to perform daily tasks in a manner considered normal is an
important concept in the development of disability. All agree that these daily tasks
include body movements such as bending, stooping, and reaching, which are used

in daily activities such as getting dressed and providing meals. These tasks are
frequently classified into basic activities of daily living (ADL) and instrumental
activities of daily living (IADL) that involve cooking, cleaning, and grocery
shopping. The difference is that Nagi and NCMRR label the ability to perform
these tasks a functional limitation, while the ICIDH calls this a disability. When
reading the literature, it is frequently difficult to distinguish between measurement
of functional limitation and disability. For this reason, some authors use the term
functional status decline to indicate abnormalities in upper and lower extremity
function and difficulty in performing daily activities.91 In this paper, the term
physical disability is used to indicate difficulty in performing daily activities
including self-care and community based activities such as grocery shopping.
If disability is defined as difficulty in performing activities of daily living
(ADL), then both physical and mental or cognitive abilities contribute to
disability.92,93 Although mental function is an important component of daily
activity performance, this paper focuses on the physical abilities that are necessary
for ADLs.
If an individuals inability to perform these ADL tasks affects their social
roles, such as interfering with parenting or vocation, they are considered disabled.
Both Nagi and NCMRR label this disability, while the ICIDH model calls this a
handicap, and expands the definition by including a social disadvantage that

occurs when social roles are not fulfilled. The NCMRR model includes an
additional label of societal limitation, which is supposed to represent the social
barriers that exist for individuals with a disability.
All three models are similar in their representation of the process of disability;
however, the models are not totally consistent with the underlying philosophy of
disablement. The conceptual perspective underlying disablement is that disability
is an interaction of individual characteristics and the social environment, in a
process that is multi-dimensional. These three models portray disability primarily
as a linear process stemming from disease or impairments within an individual and
do not adequately explain the role of society or environment in causing disability.
These differences may be more than semantic. Research into predictors and
outcomes of disability relies heavily on the disablement model as an underlying
conceptual framework. Rehabilitation research is confounded by the inconsistent
use of terms relating to physical function and performance. Measuring handicap is
difficult because few instruments are effective at assessing the interaction of the
individual's disability with the requirements of society.
2.3.3 A Biopsychosocial Model of Disability/Ability
Recognizing that the original ICIDH model implied that people are
disadvantaged because of their limitations alone, both the WHO and the
Committee on a US National Agenda for the Prevention of Disabilities engaged in

a process of revision of the disablement framework. The new version of the ICIDH
titled the International Classification of Functioning (ICF)94 is considered a true
biopsychosocial model. Disablement is now understood as an identifiable
variation of human function.95 Wording has been changed, and the model
attempts to synthesize both medical and social approaches to disablement. The
three components of the model are now called impairments, activity limitations
and participation. Although this model represents a positive move toward a more
accurate model of disability, there is still disagreement on the definitions and
measurement of the activity limitation and participation constructs. Due to the lack
of operational definitions, this model will not be used.
2.3.4 Expanded Disablement Models
Several authors have pioneered the use of an expanded disablement model,
which addresses other factors than may contribute to disability.96 The expanded
model attempts to describe the multi-dimensional nature of disability by
explaining how secondary conditions, risk factors for disability and other factors
internal and external to the individual may modify the main disease pathway.
Figure 2.2 describes these relationships.
Secondary conditions are those that occur as a result of the primary pathology
or condition. Depression is a common secondary condition that occurs as a result
of the chronic illness, but may increase the level of disability. Factors that

predispose an individual to disability could be either within the individual, or
external to the individual. Both types of risk factors most likely influence disability
through their direct effect on pathology (such as sedentary lifestyle associated with
arterial disease) or impairments (physical inactivity leading to leg muscle
weakness). The factors considered impossible to modify, such as age, gender and
socioeconomic status, are defined as internal variables. External modifiers are
defined as those factors such as co-morbid conditions which are not directly
related to the disease process itself but which may influence the level of disability.
In summary, disablement models attempt to demonstrate the effects of
pathology or disease on an individuals ability to function in his/her own
environment and within society. These models are similar in constructs, but vary
in terminology. The most frequently used models suggest a main disease pathway
which consist of a series of related events that describe the effect of pathology,
which contributes to impairments, functional limitations and ultimately disability.
This model of disability was applied to people with PAD-IC.
2.4 Factors Contributing to Physical Disability
The main domains found to contribute to disability include depression,
alcohol use, cognitive changes, comorbidities, lower extremity function, physical
activity, aerobic capacity, body mass index and low frequency of social

contact.9197 Disease and aging are primary contributors to the cycle of disability,
while physical activity and fitness are potent modifiers in the disablement
Chronic diseases with the highest rates of disability include cardiovascular
disease, rheumatoid arthritis, stroke, spinal cord injury and diabetes. While not
specifically listed, PAD-IC is presumed to lead to significant disability. Having
one chronic disease is a significant predictor of functional decline, and the risk
increases significantly as the number of chronic conditions increases. The factors
of socioeconomic status, education, marital status and smoking had less influence
on disability than lower extremity impairment, health, physical activity and
Lower extremity performance is another significant contributor to disability.
Tests of lower extremity performance, such as ability to rise from a chair and
climb stairs, accurately predict disability in community dwelling populations.100
For example, in one large study, the relative risks of mobility-related disability
were 4.8 times higher for those with the lowest performance scores compared to
highest scores.100 Decreased muscle strength is associated both with difficulty in
functional tasks, and the presence of disability.101,102 These relationships also
appear to apply to those with PAD. McDermott used data from the Womens
Health and Aging Study to support that PAD diagnosed by ABI (symptomatic or

non-symptomatiq) is associated with poor lower extremity function, less walking
i m
per week and increased disability.
Age is a factor in the development of disability in all of the longitudinal
studies. The greater the age, the greater the prevalence of disability.104 This may be
explained in that the longer one lives, the greater the likelihood of developing
disease. In addition, age -related disability is influenced by physical activity and
physical function, with more severe disability occurring in the elderly who are less
physically active and more limited in physical function.
The effects of gender on disability are complex and mediated by a number of
other influences including strength, physical activity and social role requirements.
Older women typically spend more time in a disabled state because they live
longer than men. Women may have different daily activity patterns than men due
to differences in roles. For example, older women may perform more cooking and
cleaning activities than men, while men may be more active in outdoor
maintenance activities. Gender issues are influenced by socioeconomic status,
education and marital status.104 In addition; the gait and physical function changes
that occur with aging vary between genders and influence the gender effects of
Several important confounding factors that directly contribute to the
development of disability include body mass, physical activity and aerobic

capacity. Body Mass Index (BMI) influences mortality, muscle strength and
physical activity, but is not often considered in studies of disability. Increased
body mass appears to be associated with an increased risk of difficulty in daily
mobility activities, although the mechanism is not clear.105 In patients with
claudication, there is some evidence to suggest that additional weight leads to a
slower walking speed.106 Conversely, in the elderly, weight loss and a low body
mass index can be an indication of inadequate nutrition and frailty, with weight
loss associated with an increased mortality risk.107 If the low body mass is
associated with loss of muscle mass, then muscle strength may be low and thus
contribute to difficulty in performing activities of daily living. A recent study of
aerobic capacity in patients with PAD-IC indicates an association with low body
weight, low lean body mass, and low aerobic capacity.108 Taken together, the
available information about the relationship of body mass to physical function is
unclear, although excess weight and loss of lean body mass in low body weight
both appear to adversely affect some components of physical disability, including
walking speed and activities of daily living. The mechanism for such effects is
unclear, though likely the body weight interacts with other co-morbid conditions in
a complex fashion.
Physical activity is inversely related to disability. In the elderly, those with
higher levels of regular physical activity had less disability.109,110 Lifestyles that

involve physical exertion help maintain functional status and protect against
functional decline. The mechanism by which physical activity mediates disability
may be through its effects on muscle strength or overall health status. Greater
muscle strength is associated with greater physical activity, which is associated
with less functional decline. Patients in better health are more physically active
and less disabled than others who report worse health.1,0 Physical activity also
appears to play an important role in lower extremity functioning in people with
PAD. Lower levels of physical activity are associated with poorer lower extremity
function measured by a performance score which includes muscle strength
measures and walking speed.111
Despite the link between physical activity and disability, there is
inconsistency in how habitual physical activity is measured, making it difficult to
interpret the impact of physical activity. Some studies define physical activity as
regular exercise, while others define physically active as walking a mile at least
once a week.109 Other studies ask subjects to estimate the amount of activity
performed in the previous week, using questions such as number of blocks walked
and flights of stairs climbed or amount of heavy housework performed.99 From an
energy consumption standpoint, these activities are not equivalent, and therefore
make it difficult to compare activity levels in people between studies. In spite of

the variance in definitions, the concept of physical activity is important in any
study of disability.
Physical fitness is a result of habitual physical activity and thus is relevant to
disability. A major component of physical fitness is aerobic capacity. Aerobic
capacity is highly associated with functional limitations, but is not always
controlled for in longitudinal studies. Although a true measure of physiologic
aerobic capacity (VO2 max) is preferred, subjects with chronic disease may stop
exercising due to fatigue or other symptoms before they reach the absolute
maximum work. In this case, the highest level of oxygen consumption achieved is
defined as peak VO2 (VO2 peak)- Morey demonstrated that the cut off that
discriminated between high physical functioning and low physical functioning
occurred at a VO2 peak of 18.3 ml/kg/min, a value consistent with the oxygen
requirements of daily activities.97
2.4.1 Disablement Model for Individuals with PAD
The ICIDH model of disablement has been used to illustrate the consequences
of pathologies such as rheumatoid arthritis,'12 however, this pathway has never
been explicitly described for PAD-IC. Figure 2.3 describes a possible disablement
model of the impairments, functional limitations and disability as a result of PAD-
IC. PAD-IC is characterized by decreased blood flow in the peripheral vessels and
associated with altered hemodynamics, oxygen supply and extraction, and muscle

metabolism, as described earlier in this dissertation. Additional impairments of
low peak VCh, muscle strength and endurance and evidence of denervation are
observed. The primary limitation expressed by patients with PAD-IC is a
difficulty with walking quickly and for long distances due to the ischemic pain of
IC. Since walking is an essential component of daily activities, a separate section
of this dissertation is devoted to discussing the link between PAD-IC and gait
The confounding factors that contribute to the development of disability in
PAD-IC are age, gender and physical activity. Severe intermittent claudication is
associated with lower levels of physical activity, such that patients are considered
sedentary.113'115 Although patients as young as 45 may show symptoms of PAD,
the majority of patients are older than 60 when symptoms of the disease become
limiting. Thus, factors relating both to age and disease may increase the level of
disability. There may be a gender and age interaction, since before the age of 65,
PAD is more prevalent in men, while the proportion of men and women over age
65 with PAD is approximately equal.
While there is evidence that demonstrates that patients with PAD-IC have
impairments, functional limitations and disability, it is unclear whether PAD-IC or
the other factors are the primary predictors of disability. Therefore, the purpose of

this study was to characterize the consequences of PAD-IC on functional
limitations (gait parameters), disability (PFS) and quality of life (PADQOL).
2.5 Role of Walking and Gait Parameters in Physical
The purpose of locomotion is to propel the body from one location to another,
while maintaining balance and adapting to various environmental conditions. In
addition, locomotion must be energy efficient. In adult humans, the primary mode
of locomotion is upright, bipedal walking. Because walking is so important to
daily function, individuals substitute or compensate as necessary to maintain the
ability to walk. These adaptations may be inefficient or inadequate for maintaining
normal activities, and thus cause a disability.
Daily activities that involve walking both inside and outside the home require
the ability to vary ones walking speed to the environment.116 For example, an
individual may be able to walk at their own pace on the sidewalk, but would have
to increase his/her walking speed to cross a street in the time allotted by a crossing
light. Likewise, community activities such as grocery shopping involve walking
(or some other form of locomotion) over relatively long distances. Logically,
individuals whose walking speed or distance cannot meet these environmental
requirements would have difficulty with daily activities involving walking, and
thus have some level of disability.

Walking speed is one of the most useful measures of lower extremity
performance. Measurement of walking speed is easy to measure in the clinic and
the results are sensitive enough to detect small but clinically significant changes in
performance. Because walking speed is so crucial to daily activities measurement,
walking speed is as accurate as physical function tests in predicting
disability.117,118 Measurement of both the comfortable and maximal walking
speeds is recommended to understand the functional limitations of the individual.
The importance of walking speed in the development of disability is evident, but
the mechanics of walking also influence locomotion. The study of the mechanics
of walking is termed gait analysis. Each individual has a usual or customary
walking speed that represents an optimal functional balance of body mass, joint
mobility, muscle strength, length of the limb segments, neural control and
The elderly are at greater risk for walking related disability. Both usual
walking speed and maximal walking speed decline with age, between 0.1 to 0.7%
per year. Women of all ages walk slower than men, although the gender
differences are insignificant when walking speeds are normalized to height or leg
length.124 The elderly have average walking speeds of 1.1 m/sec, a speed which is
at or below the environment requirements for walking in the community.120,125,126
When asked to increase their walking speed by walking as fast as they can, older

adults without pathology can increase speed from 21-56% above a comfortable
pace.124 Normal variability in free speed walking velocity is between 4 and 7%.127
Gait is the term used to describe specific components of walking. Two
systems to describe gait parameters include the gait cycle and time/distance
qualities of stride. The gait cycle may be divided into phases, a strategy which
allows some gait parameters to be compared between individuals. The two main
phases are stance and swing. Stance time refers to the amount of time a foot is on
the ground, whereas swing applies to the time the foot is in the air, advancing
toward the next step. Stance times are described as single (one foot in contact with
the ground) and double (both feet on the ground at the same time). Normally,
stance is 60% of the gait cycle, and swing is 40%. A majority of stance time is
spent in single limb support, where only one leg is in contact with the ground.128
(Figure 2.4)
The time and distance gait parameters include step and stride length, stride
width, toe out angle, cadence and speed. Step length is the distance between the
heel of one foot to the heel of the other during double limb support. Stride length
represents the distance between two successive foot strikes of the same foot, e.g.
right to right foot. Cadence is the number of steps taken per unit time. Walking
speed is mathematically determined by length and frequency of the stride, for an
equation of velocity = stride length x cadence (steps/minute). There are some

typical gender differences in gait; men have a stride length averaging 14% greater
than women, while women have a slightly higher cadence than men to compensate
for their shorter stride length.128 As age increases, stride length decreases.
Stride width describes the distance between feet during periods of double
support, and is sometimes called base of support. A wider base of support is seen
in patients with balance difficulties,129 although base of support does not
adequately identify subjects who are unsteady from those without balance
deficits.130 While one study found that decreased stride width correlated with
falling in elderly individuals,131 other studies indicate that stride width increases in
older people with a history of falling.132 Toe out angle indicates the degree the foot
angles out from the line of progression. Toe out angle is influenced by the degree
of femoral or tibial rotation.
Lower extremity joint range of motion contributes to walking mechanics and
efficiency. The main determinants of normal gait mechanics are ability to
extend the hip, straighten the knee and dorsi-flex the ankle. These factors remain
the same regardless of whether the subject is walking fast or slow.134
Lower extremity muscle strength is a strong predictor of walking speed.
Walking is a series of actions by which the body weight falls forward and is caught
by the forward leg. Although forward movement of the body is generated by the
forward fall of the body, muscle action also serves to bring the body forward past

vertical. Muscles act to generate forward propulsion and constrain excessive
movement that would result in falling. Hip extensor, quadriceps and plantar flexor
muscles have been identified as the most important muscles in determining gait
speed. The one study that accounted for physical activity and age m
elderly, sedentary men found that hip extension torque was the only significant
predictor of usual walking speed, stride length and cadence.137 The relationship
between leg strength and walking speed is described as non-linear. In subjects with
greater leg strength, leg strength and gait speed are not associated, but there is a
relationship between leg strength and walking speed in subjects with muscle
t in
weakness. The authors proposed a threshold of muscle strength below which leg
strength is a critical contributor to gait speed. In that study, body weight and age
also explained gait speed, but gender and height did not. Physical activity
maintains muscle strength and therefore influences gait speed.125,135
The patterns of factors that contribute to walking speed differed slightly
between men and women.121 In men, walking speed was positively related to calf
strength, height, weight, and physical activity. Walking speed was negatively
related to age. In women, walking speed was also positively related to leg
strength, and physical activity, but negatively related to weight and leg pain.121 In
men, 44% of the variance in walking speed was accounted for by height, calf
strength and health problems. In women, 42% of the variance was explained by

height, calf strength, physical activity and leg pain. While height, weight and leg
pain are factors which are not easily modifiable, the main modifiable impairments
that contribute to usual walking speed were muscle strength, range of motion and
physical activity.
Engaging in a regular program of physical activity emphasizing walking may
have a positive influence on walking speed through effects on range of motion or
aerobic capacity. Walking is the primary activity that promotes hip extension, and
thus maintains joint range of motion.138 Lower levels of habitual physical activity
were associated with poorer physical functioning,139 however, few studies have
examined the relationship of aerobic capacity and walking speed.140 Energy
efficient gait occurs if the usual walking speed is less than 50% of the aerobic
capacity. In patients whose aerobic capacity is very low, the demands of normal
gait speed may be at a high percent of capacity, causing fatigue or a decrease in
walking speed.
Gait changes are seen in various pathologies. In the elderly, the decline in
walking speed is attributed primarily to decrease in stride length.127 Cadence may
increase as a result of decreased stride length,141 although this is not a universal
phenomenon. Other gait changes seen in the elderly are increase in double-limb
support time, decreased muscle power and a reduction in dynamic balance.125,133
While these changes may be construed as pathological, it is also possible that these

changes indicate adaptation towards a more stable gait pattern.141 Since patients
with PAD tend to be elderly, they may have gait changes that are similar to those
seen in elderly individuals, or they may have abnormalities in gait attributable to
the pain of intermittent claudication. An investigation into gait parameters in
people with PAD may provide insight into whether gait changes exist and if so,
whether these changes are similar to those seen in the elderly.
2.5.1 Walking, Gait and Physical Disability in People
with PAD-IC
Very little information exists about gait characteristics in people with PAD-
IC. Overall, previous reports indicate that people with PAD-IC tend to walk
slower and have a shorter stride length than healthy controls. Whether these gait
parameters are associated with disability is unclear. Although physical activity and
low aerobic capacity can influence walking speed, these factors are not
consistently reported. In addition, little is known about the impairments that may
contribute to walking speed, such as muscle strength or joint range of motion.
Slow walking speed appears to be the primary abnormality in people with
PAD-IC. The reported customary walking speed for people with PAD-IC ranges
from 0.94 m/sec 1.08 m/sec, all of which are at the limits needed to function in the
community. Walking speeds are also about 15% slower in people with PAD-IC
than controls.6,142,143

People with PAD-IC have changes in spatial and temporal gait parameters that
are greater than expected for age alone. These changes are most notable in stride
length and cadence. Stride length is significantly shorter by 14-22% in people with
PAD-IC as compared to controls in several studies.6,142 These walking studies
were performed over short distances during which subjects did not report
claudication pain. In contrast, McCully demonstrated no significant difference in
step length for subjects with PAD-IC vs. controls when walking without pain, but
found shorter step lengths for PAD-IC subjects with claudication pain.144 Gardner
reports similar findings, with the gait changes increasing as claudication pain
appears.142 Gait parameters in people with PAD are shown in Table 2.1. These
studies did not uniformly control for differences in physical activity, fitness or
body weight, all of which may impact walking speed and perhaps gait parameters.
If stride length decreases, cadence may increase to maintain walking speed, if
cadence decreases as well, then overall speed is slower. Two studies demonstrated
that people with PAD-IC walking a short distance had decreased cadence by about
3% 6,142 while another study indicated no significant difference in cadence.145
One other study of gait in animals with PAD did not report cadence.146 These
studies also did not control for differences in physical activity, fitness or body
weight. These reported studies examined the gait parameters of patients with PAD
while walking for short distances such as six meters. These distances do not

always produce claudication pain in patients with PAD-IC. Although significant
pain did not exist, there appeared to be some changes in gait speed as well as step
length and cadence that may contribute to poor physical functioning, although this
link has not been clearly demonstrated. Stride width and toe out angles have not
been described in subjects with PAD-IC compared to controls.
If subtle changes in gait are apparent in walking short distances, then the
abnormalities in gait may be magnified when patients have to walk longer
distances or with claudication pain. Walking longer distances is problematic for
patients with PAD-IC as evidenced by low scores on the WIQ.1 However, when
McDermott examined the changes in step length and cadence from the beginning
to the end of a six-minute walk, she demonstrated no significant difference.143 Step
length and cadences were not significantly different in people with PAD vs.
controls either at the beginning or the end of the six-minute walk, although she
suggests that step length may be more important than cadence in determining the
six minute walking distance in patients with severe disease (as measured by ABI).
Even if there are changes in gait, the contribution of step length and cadence to
self-reported difficulty with walking and activities of daily living are unknown.
In the elderly, plantar flexion strength and hip extensor torque have been
proposed as the most important contributors to walking speed. Hip extensor torque
was measured during kinematic gait analysis, a measure that is not easily obtained

in a clinical setting. There are no reports of hip extensor torque in people with
PAD-IC. In people who had more severe disease in one leg, plantar flexion
strength was found to be decreased in the involved leg as compared to the
uninvolved limb.40'42 Knee extensor and hip abductor strength of patients with
PAD-IC was less than controls.147 Muscle weakness is not consistently associated
with abnormalities in gait. For example, the plantar-flexor muscles (gastrocnemius
and soleus) control motion at the ankle and knee during the stance phase.
However, the plantar flexion strength needed to permit heel rise and control
motion is between 25 and 60% of maximal power, with a requirement of 75%
strength at only one point in the gait cycle.119 This implies that a significant
amount of muscle weakness may exist before observable changes in gait are
evident. Although muscle strength is decreased in PAD and may have an impact
on walking distance, there is no clearly established relationship between muscle
strength and walking speed in patients with PAD-IC.
People with PAD-IC have identifiable abnormalities in areas that have been
shown to contribute to disability; chronic disease, decreased lower extremity
function evidenced primarily by slower walking speed than controls, and
decreased physical activity levels. Only three studies have investigated the
relationship of low ABI with physical function in people with PAD. Using data
from the Multicenter Study of Osteoporotic Fractures (SOF), Vogt et al examined

the contributors to disability in women with PAD measured by ABI of 0.90 or less.
One percent of the group had symptoms of intermittent claudication. Disability
was defined as difficulty in carrying out IADLs such as shopping, preparing meals,
housework, 10 stairs and walking 2-3 blocks on level ground. Even in women
with PAD without symptoms of claudication, these subjects walked slower,
walked fewer blocks per day and were less likely to exercise than women without
disease. Decreases in muscle strength and balance were associated with the
presence of disease. In addition, the presence of PAD with or without symptoms
was an independent predictor of decreased functional status.147
McDermott used data from the Womens Health and Aging Study to support
the theory that low ABI ("symptomatic or non-symptomaticj was associated with
poor lower extremity function, less walking per week and increased disability.103
McDermott controlled for age, gender, smoking and comorbidities. From these
investigations of women subjects with PAD, it appears that women with PAD,
whether or not they have symptoms, do have more physical disability than those
without PAD.
One additional multi-center study of men and women with PAD-IC
demonstrated an apparent association between low ABI and low physical function.
An ABI of less than 0.50 was associated with walking shorter distance on a six

minute walk, slower walking over four meters, less physical activity and more
difficulty maintaining standing balance.
In summary, patients with PAD-IC have declines in lower extremity
performance, slower walking speed, decreased habitual physical activity and
decreased muscle strength. These factors are associated with physical disability.
The presence of PAD without symptoms also appears to be associated with
physical disability. Due to the confusion regarding terminology of physical
function and disability, in this dissertation, disability is defined as self-reported
difficulty with daily activities, labeled physical function score (PFS).
Gait characteristics appear to be different in subjects with PAD-IC versus
controls; however, the level of physical activity has not been controlled for in
previous studies. The gait characteristics of stride width and toe out angle have not
been investigated or compared to a control group similar in age, gender and
physical activity. In subjects with PAD-IC, it is not known whether walking speed
predicts the level of physical disability. Therefore, the purposes of this study were
to describe the differences in gait parameters between subjects with PAD-IC and a
control group similar in age, gender and physical activity status, and to identify the
factors that explain PFS in subjects with PAD-IC.

2.6 Conceptual Frameworks of Quality of Life
As people live longer with disability and chronic disease, quality, as opposed
to quantity of life, becomes a relevant question. Quality of life may be examined
in global terms or from the perspective that health is the primary contributor to
quality of life. The term 'quality of life (QOL) appears to have originated in the
United States in the post-war period and is used to describe the influence of
material goods on people's lives.149 The general term QOL reflected broad social
issues such as access to resources, income, housing or education, and is not
specific to health related questions. In the health professions, as advances in
medicine led to decreases in mortality, the focus of treatment shifted towards
improving quality of life. The rehabilitation profession, as an example, has always
focused on how to improve quality of life for the disabled individual. The
definition of quality of life is critical to both conceptual and methodological
procedures, but the lack of a consistent definition indicates no theoretical
consensus. This section discusses the conceptual frameworks for global, health-
related and disease-specific quality of life frameworks. Next, the domains of
health-related quality of life (HRQOL) and measurement issues in QOL research
are presented. Then, the current level of knowledge of HRQOL in people with
PAD-IC is reviewed.

On a global level, quality of life is a social construct, dependent on the
persons perspective. It is whatever the person says it is. Other terms that are used
to relay the same concept are life satisfaction and well-being. There are some
instruments that use a single score to measure global quality of life; however
quality of life consists of many different domains. Some authors have
conceptualized quality of life as an extension of the disablement model. The
factors that contribute to quality of life include aspects of health (or disease),
physical abilities, as well as an individuals personal beliefs and their social
supports.150 These factors are shown in Figure 2.5.
Some experts argue that global measures are too abstract to be used in a
medical context and that the conceptual framework for evaluating QOL should be
within the context of health and patient care.151,152 Others support this view by
proposing that health is the most important contributor to QOL.153 These views
have led to an increase in the number of studies which focus on health related
quality of life as an outcome of medical intervention. Specific measures of
HRQOL do not correlate well with other measures of life satisfaction, because
these instruments measure different constructs.
In the HRQOL field, sometimes the term health status is used interchangeably
with quality of life, but these terms are not equivalent. Health status implies a
measure of the progression or absence of a disease using some physical or clinical

measure, and is not a measurement of mental or social health or well-being.
Health- related quality of life studies often rely on measures of health status as a
substitute for QOL, although there may be other factors other than health that
contribute equally to QOL. HRQOL instruments were not all developed from a
patient perspective. Some instruments were developed by asking groups of patients
to identify factors important to their quality of life, while other instruments are the
result of medical personnel deciding what was important to QOL and proposing
those questions to patients. If quality of life is what the patient says it is, then it is
more appropriate to use QOL measures that are based on qualitative analysis of
patient interviews. The choice of instrument should be consistent with the type of
intervention provided and focused on factors pertinent to quality of life.
The quality of life literature generally agrees that health is "a
multidimensional concept encompassing, as a minimum, domains of physical,
psychological, and social function and well-being".151 These domains are
commonly evaluated in health-related quality of life instruments. Physical
functioning, for example, involves the ability to complete daily activities, move
around in one's environment, and would include physical abilities such as strength,
ambulation, or other measures of physical ability.154,155 Psychological function
might include emotional functioning, personal development and fulfillment, self-
esteem, or life satisfaction.149,154 Social well being might include the ability fulfill

one's social role or to be involved in social or civic activities. Most HRQOL
instruments measure the impact of disease on these domains. Some common
instruments that look at multiple domains include the Quality of Well-Being scale
(QWB) and the Medical Outcomes Study short-forms (SF- 36 or SF-20).154,156
A four-year study of health care delivery systems produced the Medical
Outcomes Study (MOS) instrument.156 This questionnaire has gained widespread
use in recent years. Several versions were developed; the most frequently used is
the short-form of 36 items (SF-36). The conceptual framework for this tool
explains quality of life from two main dimensions, physical and mental health.
These dimensions were assessed from five perspectives; general health
perception/satisfaction, clinical status, physical functioning, mental functioning,
and social functioning. Subjects are asked to respond to questions about whether
their physical or emotional health has influenced their daily activities. Widely
used, the SF-36 is considered reliable as well as sensitive to changes in
HRQOL.157 Although the measure is called a quality of life assessment, the
physical function subscale items closely resemble questions used to assess
disability. Examples of physical limitation questions include; Has your health
limited you in moderate activities, like moving a table or carrying groceries? The
physical function subscale may therefore be used as a proxy measure of

Some researchers suggest other related elements that should be considered in
quality of life. Somatic sensations such as pain and shortness of breath likely
influence an individual's quality of life.152 Economic factors are cited as
contributors to quality of life.160 Coping skills, expectations of health, or anxiety
might also influence one's quality of life.161
Early work on coping suggests that both psychological attributes and social
support are resources that mediate the effects of life stress on disability.
Psychological attributes such as mastery or self-efficacy can contribute to
successful coping in people in stressful situations or with chronic illness.162
Mastery, also known as perceived control, describes the extent to which one
perceives his/her life changes are under his/her control. Pearlin suggested a seven
question mastery scale with answers assigned a score based on agreement or
disagreement with each statement.163 An examples of a mastery question is:
There is little I can do to change many of the important things in my life. While
mastery is considered a belief in how much control one has over his/her life, self-
efficacy refers to a belief that one can successfully perform certain behaviors. For
this study, the appropriate concept is mastery, as there is no intervention against
which to measure self-efficacy.
Social support is also presumed to be an important resource in dealing with
the effects of stress. Many different perspectives have been utilized in studying the

role of social support on health, leading to many different definitions of social
support. A simple measurement of social support is to ask about the presence of a
spouse or significant other. This measure does not take into account the nature of
the relationship, but presumes that having a significant relationship is helpful.
Other measures of social relationships, such as the number of social contacts or
social groups, have been used as well.
Social support measures generally fall into three categories; social networks,
tangible support that has actually been received, or the degree of perceived
support.164 Actual support received may be important in decreasing the effects of
disability in those with significant limitations, such as a spinal cord injury, but
some research suggests that perceived support has the largest influence in
protecting health.165 With this perspective, the Interpersonal Support Evaluation
List is a 40-item questionnaire that measures the perceived availability of several
types of support. These include tangible, appraisal, self-esteem and belonging.166
Disease specific measures of quality of life may be useful because chronic
diseases have varying symptoms and prognoses. Disease specific instruments
should be more sensitive to changes in the disease process, especially as a result of
intervention. They are also especially good at identifying issues specific to an
illness. These instruments are not useful when the goal is to compare populations
with two different diseases, or when evaluating multi-disciplinary interventions.

2.6.1 Health-related Quality of Life in PAD
As indicated previously, patients with PAD-IC demonstrate or report declines
in physical function, associated with pain of intermittent claudication. One would
expect that these limitations would also influence quality of life. Quality of life
assessment in patients with PAD indicates that the domains of physical function
are affected more than the social functioning domains.
Several studies in patients with PAD-IC indicate that PAD-IC has a negative
impact on HRQOL.167 The studies using the SF-36 to measure HRQOL indicate
that patients with PAD report that their medical condition adversely impacts their
physical functioning. Studies indicate that social function is affected to a
lesser extent than is physical function.
The health-related quality of life measures used in the above studies do not
reflect all issues that are important to people with PAD-IC. Several researchers
used qualitative research methods to identify areas that were important to people
with PAC-IC but which are not captured by current HRQOL questionnaires. The
HRQOL domains important to people with PAD-IC were similar to other QOL
studies (disease, pain, physical, social and emotional functioning), but describe
more specific issues.13,174 Patients expressed frustration at delay in diagnosis and
knowing how to manage their disease. In addition to having significant pain,

patients were frustrated and profoundly limited in their ability to walk and do their
previous level of physical activities, especially of the leisure variety such as
dancing. People felt they were a burden to family and unable to work, and limited
in their ability or desire to participate in social activities that were enjoyable to
them. People also felt abnormal, unable to do activities they enjoyed and afraid of
future events. The multiple domains identified in these studies support the need to
examine the role of social and personal factors in addition to physical function
measures on quality of life.
From these studies, several approaches to measuring quality of life inpatients
with PAD-IC have been suggested. Hallin suggested that the SF-36 should be
supplemented with a life satisfaction questionnaire.171 Other PAD-specific QOL
questionnaires can also be used. The Intermittent Claudication Questionnaire
(ICQ),174 Vascular Quality of Life Questionnaire,175 and Sickness Impact Profile-
Intermittent Claudication (SIPic)176are examples of disease-specific
questionnaires that are similar in construct to the SF-36.
One other PAD-specific QOL questionnaire (PADQOL) was developed from
patient interviews, using qualitative methods and grounded theory to develop the
questionnaire items. This 60-item instrument asks subjects to agree or disagree
with a statement and also to rank the importance of each statement.13 This
questionnaire is unique in that it specifically addresses the impact of personal

beliefs on the disease process and the influence of the disease on social support
Studies of quality of life in patients with PAD-IC have used HRQOL
measures as an outcome of medical intervention, but no study has examined
whether personal coping skills and social supports system explain HRQOL in
these patients. The secondary purpose of this dissertation was to explore the
physical and psychosocial factors that explain PADQOL in subjects with PAD-IC.
2.7 Summary and Purpose
In summary, the primary symptom of PAD is intermittent claudication (IC), or
severe leg pain that occurs with walking and improves with rest. Patients with
PAD-IC are unable to walk even short distances without pain, and report
significant limitations in daily activities that can be defined as a disability. The
overall purpose of this dissertation was to characterize the functional limitations
(gait parameters), disability (PFS), and quality of life (PADQOL) in people with
PAD-IC. Since patients with PAD tend to be elderly, it is unclear whether the gait
changes are similar to those seen in elderly individuals, or whether gait
abnormalities are greater in subjects with intermittent claudication pain. The first
aim of the study was to describe the differences in walking speed, walking
distance, and gait parameters between subjects with PAD-IC and a control group
similar in age, gender and physical activity status. Secondly, in people with PAD-

IC, no strong link between measures of walking ability, abnormal gait and the
level of disability has been described. The second aim of this dissertation was to
identify the factors that explain PFS in subjects with PAD-IC. Finally, little is
known about the relative contributions of physical function, coping style and
social supports to the outcomes of quality of life in people with PAD-IC. This
dissertation explored the physical and psychosocial factors that explain PADQOL
in subjects with PAD-IC.

Figure 2.1 Comparison of Disablement Frameworks
Pathology Impairment Functional Limitation Disability Handicap/Societal Limitation
Definitions Problem in body system function Problem with the person performing activities Problem with person performing activities required by roles Social disadvantage occurring as a result of disability
Nagi Disease Loss of function of body system Limitation in ability to perform tasks and activities usually required by social roles or self-care Pattern of abnormal behavior that evolves with long-term impairments and functional limitations
ICIDH Disease Any loss or abnormality of psychological, physiological or anatomical structure or function Any restriction of ability to perform an activity in the manner or range considered normal for a human being Handicap: Disadvantage due to impairment or disability that limits fulfillment of normal role (for age, sex, culture)
NCMRR Abnormal Loss of structure or Restriction in ability to Limitation in Societal limitation:
physiological processes function perform task in normal manner performing tasks within social contexts Limited fulfillment of roles due to social barriers
Examples Decreased ROM, decreased strength, sensory loss, balance deficit Difficulty crouching, stooping, walking stairs, lifting 10 pounds, chair rise, gait difficulty Difficulty in any ADL in the persons usual environment Inability to participate in work or recreation, parenting
ROM = Range of Motion, ADL = Activity of Daily Living

Figure 2.2 Expanded Model of Disablement
Risk Factors
Factors External to Individual
Social Supports
Physical and Social
Lifestyle behaviors
Psychological attributes

Figure 2.3 Disablement Process in PAD
Age, gender, occupation, education

Figure 2.4 Phases of Gait
Right Left Left Right Right
Heel Toe Heel Toe Heel
Contact Off Contact Off Contact
Surface Surface
_________i_______i________i i_____________i_______i_______i--------1
0* 50% 100%
.......'"!......................... .............r " ...................i
Doutta I S OdUMt I S
Support j ; Support j j

Figure 2.5 Model of contributors to Health-Related Quality of Life in PAD

Table 2.1 Reported Gait Abnormalities in PAD-IC
Gait Parameter PAD Controls Author
Usual walking speed (m/sec) .90 - McDermott, Greenland, 2001
.99 1.16 Gardner, 2001
.98 1.08 McDermott, 1998
1.02 1.09 McCully, 1999
1.01 1.4 Scherer, 1998
1.18 1.26 McDermott, Ohlmiller, 2001
Stride length (cm) 119 135.8 Gardner, 2001
120 125 McCully, 1999
113.8 146.8 Scherer, 1998
Cadence (steps/min) 99.9 103.3 Gardner, 2001
107.2 116.7 Scherer, 1998
Maximal walking speed (m/sec) 1.45 1.86 Gardner, 2001
1.34 1.47 McDermott, 1998
Stride length at maximal speed (cm) 141.6 165 Gardner, 2001
Cadence at maximal speed (steps/min) 123 134.7 Gardner, 2001
Six-minute walk distance (ft) 1164 _ Montgomery, 1998
1234 McDermott, 1998
1279 Gardner. 2001

2.8 Preliminary Results
A preliminary study of gait characteristics of patients with claudication
provided background information pertinent to this research study.6 The purpose of
the preliminary study was to compare the differences in step length, cadence, and
walking speed during free speed walking between men with claudication and a
control group of similar age men.
Nineteen male subjects with peripheral arterial disease (PAD) and eleven
healthy controls were studied. Subjects with PAD had resting ABIs of < .95, with
exercise ABI decreased by 20%. Subjects were excluded if they had ischemic rest
pain, or if claudication did not limit daily activities. Subjects with clinically
significant cardiac, pulmonary or musculoskeletal conditions limiting walking
were also excluded.
Subjects performed five trials of walking on a 20-meter course in a tiled
hallway. Subjects were asked to walk at their usual speed over this course.
Walking began several feet prior to the start of the measured course, and continued
after the finish line. Two investigators measured walking parameters, one
positioned at the start line and indicating the start time. The other researcher was
positioned at the finish line, and was responsible for counting the number of steps
over the 20-meter course and for timing the walk. Walking speed was converted to

meters/minute; cadence was calculated by dividing the number of steps by the time
to walk 20 meters. Step length was calculated by dividing the number of steps by
the distance walked.
There was no difference between groups in regard to age and height (p > .05).
Subjects in the PAD group walked significantly slower than subjects in the control
group (p < .001). Step length and cadence were also significantly lower in the
PAD group compared to controls (p < .001). When step length was normalized to
height, step length was still significantly shorter in the PAD group (p .001).
Subjects with PAD appear to have decreased walking speed, cadence and step
length as compared to controls, regardless of pain symptoms and disease severity.
Future studies should include a measure of calf strength and a leg length
measurement to provide additional information.

3.0 Research Design and Methodology
This chapter describes the design and methods of the research project. These
methods include subject selection, instrumentation and procedures used in data
collection. The last part of this chapter includes description of data analysis
techniques. This is a cross-sectional study comparing two groups of subjects, one
with PAD-IC and a control group selected to be similar in age, gender and activity
level to the PAD-IC group. Characteristics of walking speed and distance, gait
characteristics and muscle strength were compared between groups. These
variables were used to describe the relative contributions of physical, personal and
social factors to the primary outcomes of self-reported physical disability (physical
function score) and health-related quality of life.
3.1 Specific Aims and Hypotheses
The purposes of this study were to describe the differences in walking speeds
and gait parameters between patients with PAD-IC and elderly controls, and to
describe the factors that predict physical function score and health-related quality
of life in the PAD-IC group.
Research Question 1: How do gait parameters at usual and maximal speeds
differ in people with and without PAD-IC?

Specific Aim 1: Describe and evaluate the differences in walking speed, distance
and gait parameters between subjects with PAD-IC and a control group similar in
age, gender and physical activity status.
Hypothesis 1: Subjects with PAD-IC will walk significantly slower than the
control group at usual and maximal walking speeds.
Hypothesis la: Subjects with PAD-IC will walk a significantly shorter distance in
a six-minute walk distance compared to the control group.
Hypotheses lb: Subjects with PAD-IC will have a significantly shorter stride
length at usual walking speed compared to a control group.
Research Question 2: How much self-reported physical function is explained by
walking speed in the PAD-IC group?
Specific Aim 2: Describe and evaluate the factors that predict physical function
score in subjects with PAD-IC.
Hypothesis 2: Self-selected, usual walking speed will significantly predict
physical function score in subjects with PAD-IC.
Research Question 3: What are the relative contributions of physical function,
personal characteristics (Mastery/perceived control) and social supports to quality
of life in those with PAD-IC?
Specific Aim 3: Explore the physical and psychosocial factors that explain
PADQOL in subjects with PAD-IC.

Hypothesis 3: Personal characteristic of mastery will be the most significant
predictor of PADQOL in individuals with PAD-IC.
3.2 Research Design Overview
This was a cross sectional study of case-control design intended to describe the
differences in walking speed, distance and gait parameters between subjects with
PAD-IC and a control group, and to describe the factors that explain physical
disability and health-related quality of life in patients with PAD-IC. Physical
disability was defined as self-reported limitations in performing normal daily
activities and scored using the physical function score (PFS) of the Medical
Outcomes study Short-Form 36 (SF-36).177 Health-related quality of life was
defined by the score on the Peripheral Arterial Disease Quality of Life instrument
(PADQOL), a self-report PAD-specific quality of life instrument.
Two groups of subjects, one with clinical PAD-IC and one group of control
subjects similar in age, gender and physical activity status were studied. Once
enrolled in the study, subjects participated in one session where gait was measured
and questionnaires administered. The study was approved by the Combined
Institutional Review Board (COMIRB) of the University of Colorado Health
Sciences Center and all patients gave consent to be studied.
Data analysis included both univariate and multivariate techniques. Analysis
included tests for differences between groups in walking speed and gait

characteristics. Univariate and multivariate regression analysis was used to
identify the explanatory variables for the outcomes of PFS and PADQOL in the
PAD-IC group.
3.3 Subjects
3.3.1 Selection of Study Population
The sample size estimated for this study was fifty adult men and women
subjects over the age of 50,25 subjects with PAD-IC and 25 control subjects. One
hundred thirty four individuals were contacted about participation in the study. Of
these, 47 could not be reached or did not return phone calls. Fourteen subjects
declined participation in the study, and another 19 were ineligible based on age,
difficulty walking or other factors. Fifty- five adult men and women subjects over
the age of 50 were screened for this study. Four subjects were deemed ineligible
at the screening visit; three subjects with suspected PAD had ABI values above
0.90 and were considered ineligible (too active) for the control group, one subject
was unable to adequately participate. Therefore, 25 subjects with PAD-IC and 26
control subjects were enrolled in the study. Control group subjects were selected to
be similar to the PAD-IC group in age, gender and physical activity. Individuals of
all ethnic groups were eligible.

3.3.2 Inclusion and Exclusion Criteria
Patients included in the PAD-IC group were eligible for the study if they met
the following conditions:
Reported symptoms of intermittent claudication which limited walking
Had an ankle-brachial index of < 0.90 in at least one leg without ischemic rest
Did not use assistive devices for ambulation
Were able to participate in the study and to answer the questionnaires.
Subjects in either group were excluded if they reported: Individuals in either
group were excluded if they reported any of the following conditions that
would affect their ability to walk:
Known arthritis or lower extremity trauma that affected walking ability
Walking ability was limited by cardiac or pulmonary symptoms such as chest
pain or shortness of breath
Diabetes associated with reported sensory changes in the lower extremity
Use of assistive devices for ambulation (Inability to walk independently)
Cognitive deficits that limited understanding of the questionnaires
Subjects with PAD-IC were recruited from the Vascular Research Laboratory
at University of Colorado Hospital and direct referrals. Control subjects were
recruited by general announcement through the University of Colorado Health

Sciences Center (UCHSC), from community centers and through direct referrals.
Initial contact with the patient included screening questions about activity level,
the use of walking devices and any lower extremity condition that would affect
walking pattern. Those who met the inclusion criteria were scheduled for a
screening visit with the principal investigator at the University of Colorado Health
Sciences Center.
3.4 Procedures
At the screening visit, subjects were given a written copy of the consent form
to read and sign, as the study was explained to them. The patient completed
subject information forms which included demographic information as well as
health history information and current health status, including co-morbid
conditions. Subjects were assisted with the questionnaires if requested. An ABI
was measured on all subjects to determine the degree of disease severity in
subjects with PAD-IC and to ascertain that control subjects did not have PAD.
Sensory testing with a 10 g monofilament was performed to the plantar surface of
the foot, a subject would be excluded if they had absent sensation indicative of
neuropathy. Height, weight, leg length and leg strength were measured prior to all
walking tests.
Habitual physical activity was assessed in all subjects by the principal
investigator, using the Low-Level Physical Activity Recall (LOPAR)

instrument.1781 Subjects completed the SF-36 questionnaire before the walking
tests. For the subjects with PAD-IC, the principal investigator administered the
walking impairment questionnaire (WIQ).
All study subjects completed the walking tests in the same order. Each subject
completed 2 trials of walking at comfortable walking speed and 2 trials of maximal
walking speed. All subjects walked with shoes on a tiled walkway. Usual walking
speed tests were administered first, followed by the maximal walking trials. All
subjects were allowed to rest in between each walking test so that no leg pain was
present at the beginning of each walking trial, including the 6-minute walk. These
tests were followed by the six-minute walking test performed by walking back and
forth on a 70-foot walkway. Subjects completed the remaining PAD questionnaires
after the walking tests. Subjects were assisted with the questions if they requested
help, in that the investigator read the questions out loud and recorded the subjects
3.5 Methods of Assessment
This section describes the purpose and appropriateness of each assessment or
instrument used in this study. In addition, measurement characteristics and
administration and scoring issues of each instrument are discussed. Demographic
characteristics are presented first, followed by the gait measurements and
concluding with discussion of the questionnaires used in this study.

3.5.1 Demographic Characteristics
Presence of disease (Co-morbidity): Subjects were queried about health
conditions to identify the presence of chronic co-morbid conditions suspected of
influencing walking ability or PFS. These conditions included heart and
pulmonary disease, cancer, diabetes, lower extremity arthritis, stroke or
depression. Heart disease was counted as a co-morbidity if the subject reported
history of myocardial infarction, bypass or angioplasty for coronary disease.
Subjects currently taking medication for hypertension were also counted as
positive in the heart disease category for co-morbidity. In addition, subjects were
asked to list the medications they were currently taking. Co-morbidity score was
determined by counting the number of current conditions of those listed.
Anthropometric measurements: Walking speed and other walking related
items are influenced by anthropometric characteristics. These include height,
weight and leg length. Height was measured using a standard wall-mounted
device. Weight was measured on a digital scale. Leg length was measured by the
principal investigator using a tape measure. Leg length was measured as the
distance from the anterior superior iliac spine of the pelvis to the medial malleolus
of the same leg with the patient lying supine on an examination table. ICC of this
tape measure method of measuring leg length has been reported as .79.179

Ankle/Brachial Index (ABI): The ankle-brachial index, or ratio of systolic blood
pressure between ankle and arm is the standard screening measurement for
1 QA
identifying the presence and severity of peripheral arterial disease. Intra-
observer reliability for ABI has been reported between .85 .99.181 In this study,
systolic blood pressure measurements in the arms and legs were obtained with the
patient lying supine using a Doppler flow-detector instrument (Nicolet
Instruments, Elite model number 100) with 8 mHz probe and standard blood
pressure cuffs. Subjects were lying flat for at least 10 minutes prior to blood
pressure measurement. In the legs, blood pressure was measured at both the
dorsalis pedis and posterior tibial arteries. The ratio of ankle to brachial systolic
pressure for each side was determined by dividing the highest ankle pressure by
the single highest brachial pressure for each side. ABI was reported for each leg.
Subjects with ABI lower than 0.90 in either one or both legs were included in the
PAD group.
Calf strength: Calf strength was assessed by having the subject perform 10
repetitions of an exercise consisting of lifting the heels and standing on the ball of
the foot. Subjects were allowed to use minimal upper extremity support to prevent
loss of balance. The number of times the subject completed the heel rise activity
without bending the knee was recorded as calf strength.182

3.5.2 Measures of Walking Function and Gait
Measures of walking function traditionally include self-selected walking
speed over short distances, or distance walked in a specified period of time.
Spatial and temporal factors of gait, including cadence, step and stride length,
width and degree of toe out angle are used to quantify gait abnormalities. This
section describes measurements of walking function and gait abnormalities. Walking Speed
Walking speed is typically measured over a relatively short distance. Subjects
may be asked to walk at both his/her own comfortable (usual) speed and at
maximal speed. The ability to increase ones speed indicates an ability to adjust to
environmental demands such as crossing a street in a specified time. Both
comfortable and maximal walking speeds are considered to be useful measures of
an individuals ability to function in his/her own environment.
Various methods for assessing walking speed have been described; all relying
on a timed measurement of how fast an individual can walk a specified distance.
Walking speed measurements demonstrate high intrarater reliability (ICC >
0.90)124 and test-retest reliability (r> .89).183
Usual Walking Speed: In this study, each subject walked at a self-selected
comfortable (usual) speed over a distance of 6 meters. Time to complete the
distance was recorded by stopwatch. Subjects were timed with a stopwatch, which

was started at the first complete step over the start line and stopped at the finish
line. Each subject completed two trials at usual speed. Subjects were allowed to
rest between trials so that no leg pain was reported prior to walking.
Maximal Walking Speed: Subjects were instructed to walk a distance of 6
meters as fast as possible without feeling unsafe and without running. Time to
complete the distance was recorded by stopwatch. Subjects were timed with a
stopwatch, which was started at the first complete step over the start line and
stopped at the finish line. Subjects completed two trials of maximal walking speed.
Subjects were allowed to rest between trials so that no leg pain was reported prior
to walking.
Six-Minute Walk: Timed walking tests have been useful measures of exercise
endurance, specifically in people with cardiac or pulmonary diseases.184 While
two, six and 12-minute tests have been used, the six-minute walk test is most
frequently utilized. These tests measure the maximum distance a subject can walk
in the six minutes. Although they are considered sub-maximal tests, one study
indicated that in healthy elderly subjects, the intensity of the six-minute walk test
corresponded to almost 80% of maximum aerobic capacity.185 In addition, this
walk test can be used to predict aerobic capacity in some populations.186
Montgomery demonstrated that the six-minute walk is feasible for use in subjects
with PAD, and further demonstrated that as there was no difference between two

trials of six-minute walk (test-retest reliability r = .94), one measure of six-minute
walk is sufficient.66
In this study, subjects were told to walk comfortably, but to cover as much
ground as possible as they walked a 70-foot course continuously for 6 minutes.
This course was an indoor tiled hallway that required the subject to turn around at
the end of each 70-foot distance. Distance in feet covered in the 6 minutes was the
measurement used for analysis. The tester walked the first lap with the subject to
instruct him/her where to turn around, then stood to the side of the hallway and
counted the laps walked. Standardized instructions were given to each subject
indicating he/she should cover as much ground as possible while walking
comfortably but without running. Subjects were given standard statements
including: you are doing great at one minute and told when they were halfway
done (three minutes) and at four and five minutes and again at the end of the test.
Subjects were allowed to rest or slow down if needed, although the time continued
to accrue. Subjects indicated the onset of claudication pain to the tester.
The six-minute walk appears to have good test-retest reliability in patients with
chronic disease, with intraclass correlation coefficients (ICC) values greater than
0.90 in subjects with heart disease.187 In patients with PAD, the six-minute walk
test has ICC correlation coefficients considered good (0.94).66 Values for the six-

minute walk are stable over time, as demonstrated by one-week test retest
reliability of 0.95188 but also sensitive to change.187
3.S.2.2 Gait Parameters- Stride Length, Width and
Although measures of gait parameters may be simple or complex, simple gait
parameters offer insight into the mechanics of walking, as well as being
inexpensive and clinically feasible. The primary spatial gait parameters include
measures of the length of the step, the horizontal distance between each limb and
the number of steps taken per unit distance. These are defined here. Step length is
defined as the distance between steps of one foot to the opposite foot. Stride length
is defined as the distance between consecutive heel strikes of the same foot (i.e.
left heel to left heel). Stride width is defined as the perpendicular distance
separating the midpoint of consecutive heel contacts. Number of steps per unit
distance is defined as cadence.
Boenig described a simple gait analysis involving the placement of moleskin
markers on subjects shoes before walking across a measured paper walkway. This
method was found to have a test-retest reliability of r > .925 for step and stride
length. This technique also had reliability values of r > .69 for toe out angle and
step width. In the present study, gait data was collected using the technique
described by Boenig.

Subjects wore shoes with heels not higher than 1 inch during the walking
tests, which were conducted on a tiled walkway. Subjects had a square moleskin
patch placed on the sole of the shoe at the midpoint of the heel, with a triangle
moleskin patch at the midpoint of the toe. (Figure 3.1) Red ink was applied to the
moleskin for trial one and black ink for trial two. Subjects walked down a 6-meter
length of paper taped to the floor, with the ink markers leaving a tracing of the
number of steps, step length, width, and toe angle. Walking speed was calculated
as time required to cover the distance assigned in meters per second (m/sec).
Cadence was defined as the number of steps taken per unit time (converted from
steps/second to steps/minute).
The midpoint of the heel square was used as a reference point for
measurement. A line was drawn through the mid-point of the heel square
perpendicular to the edge of the paper. The edge of the paper was considered the
line of forward progression. The principal investigator performed all gait
parameter measurements using standard tape measures and protractors.
Measurements performed on stride length and width measures on two separate
days indicated that intra-rater reliability for the principal investigator was excellent
(r= .99). See Figure 3.2 for a representation of the stride measures.
Step length was defined as the linear distance from the heel-strike of one foot
to the heel-strike on the next successive step of the opposite foot. The distance was

measured from the mid point of the heel square to the next consecutive step. The
distance from the start line to the first step was not measured. Of these measured
steps, steps 3-6 (the middle steps) were averaged, so that the influence of
acceleration and deceleration on step length could be minimized. The average step
length of the two trials was used in analysis. Stride length was calculated by
adding two consecutive step length measures together. The stride lengths were
averaged over the two trials. Intra-rater reliability of these gait measures was
excellent, with test-retest reliability on two separate occasions demonstrated as r =
Step width was defined as the transverse linear distance between midpoints
of heel squares on two successive steps, measured from the midpoint of the heel
square. The distance from the heel square to the line of progression was measured
for each step. Step width was calculated using the difference between the distances
for two consecutive steps.
Toe out angle indicates the amount of toe-in or toe out on each foot. A long axis
was drawn between the apex of the triangle and the midpoint of the heel square. A
line was drawn perpendicular to the line of progression (from edge of paper), but
intersecting with the midpoint of the heel square. A protractor was used to
determine degrees of toe in or toe-out from the line of progression. (Figure 3.3) If
the toe pointed towards the outside of the paper, the degree of toe out angle was

labeled positive, when the toe pointed towards the midline of the paper the toe out
angle was listed in negative degrees.
3.5.3 Questionnaires and Survey Instruments
Low-Level Physical Activity Recall: Habitual physical activity is one
contributor to walking speed in that it maintains muscle strength, flexibility and
aerobic capacity. Physical activity can be measured by self-report or with devices
such as accelerometers.190 Since some measures of energy consumption are
expensive, a simple, reliable questionnaire assessing physical activity is clinically
useful. A physical activity survey instrument was used to identify the level of
physical activity in each subject, thus ensuring similar levels of physical activity
between groups.
For this study, the Low-Level Physical Activity Recall (LO-PAR) was used.115
This questionnaire was based on the Stanford Five -City Project physical activity
recall questionnaire, adapted for use with sedentary subjects and validated for
subjects with claudication.115191 This questionnaire prompted subjects to recall the
amount of physical activity they had performed over the last week, by
remembering activities that required heavy, moderate and light and very light
work. Administered by interview, subjects were provided with examples of
activities that fell into each of these work categories.

In this study, the primary investigator asked the subject to recall the amount of
time spent in various activities over the last week. These activities were divided
into work, house/yard chores and recreation/leisure. First, subjects were asked to
recall the number of hours spent sleeping, on average, per night over the last week.
Subjects were then asked to recall the number of hours per week spent in heavy,
moderate, light or very light activities over the last week, using a chart as prompts
for the type of activity. Since it is easier to recall time spent in moderate or heavy
activities, and difficult to recall low-level activities, some hours are not accounted
for during initial questioning.192 In order to standardize the number of hours/week,
the difference between the hours reported and the total number of hours in a week
(168 hours) was added to the very light activity category, to ensure that all hours in
the week are accounted for. The amount of energy expended for each type of
activity was categorized, and expressed as metabolic equivalents (MET). One
MET is equivalent to the relative amount of oxygen consumed at rest, or 3.5
ml/kg/min. Activities are categorized as very light (0.9-2.0 METS), light (2.1-3.0
METS), Moderate (3.1-5.0 METs) and Heavy (5.1-7.0 METs). The number of
clock hours spent in each level of activity is multiplied by the mean MET level, for
example, the clock hours for heavy activity is multiplied by 6.05 METs, moderate
by 4.05 METs, light by 2.55 METs and very light by 1.45 METs.115

Results from self-reported physical activity have been compared with
measurement of activity from accelerometers. Correlation between accelerometer
and 7-day physical activity recall scores was reported as 0.602 using Spearmans
Rho coefficient in subjects with peripheral arterial disease.71
Walking Impairment Questionnaire: The Walking Impairment
Questionnaire (WIQ) is a PAD-IC specific instrument designed to characterize the
patients limitations in walking in his or her own environment.1 The answers
represent the subjects report of their ability to walk various distances and variable
speeds. Administered by an interviewer, the questions characterize the subjects
self reported degree of difficulty in walking various distances, speeds and stairs.
The six distance questions are scored by multiplying the degree of difficulty by the
distance that can be walked (1/2 block to 5 blocks). The speed score is similarly
derived from four questions rating degree of difficulty walking one block slowly,
at an average speed, quickly or running. The stair score is derived by multiplying
degree of difficulty with 1,2 and 3 flights of stairs. Scores for distance, speed and
stairs are obtained by dividing the raw scores by the maximum possible scores and
multiplying by 100.
The WIQ has been validated against walking time on treadmill tests, currently
considered the gold standard of measures of intermittent claudication symptoms.
Distance scores s correlated with peak treadmill time (r = 0.55-0.67)1 In regard to

sensitivity, the WIQ scores remain stable in control subjects, but change with
effective intervention in treatment groups. In addition, the distance WIQ scores
correlate with six-minute walk distances
(r = .557) and the speed scores correlate with usual walking speed (r =.528) and
fast walking speeds (r=.56 all with p < .05).193 In the current study, the WIQ was
used for baseline assessment in the PAD-IC group only since it is not sensitive in
the absence of PAD.
Physical Function Score: Physical disability is frequently defined as
limitations in performing normal daily activities. Activities of daily living include
those tasks normally considered part of a daily routine, ranging from self-care
activities such as washing or bathing to light housework and community activity
such as grocery shopping. In large studies, physical disability is often
dichotomized into two categories 1) needing assistance in activities of daily living
(ADL) or 2) independent or not needing assistance in activities of daily living. For
this study, a measure of physical disability that was more specific and scaled in
continuous fashion was preferred. The physical function scale of the Medical
Outcomes Study Short Form 36 (SF-36) was identified as an appropriate
measure.177 The 36 questions are broadly grouped into two categories of physical
function and mental function and further categorized into 8 domains of function,
listed below.

General health
Physical Functioning
Role limitations due to physical health
Role limitations due to emotional problems
Social functioning
Energy/ Vitality
Emotional Well-being
Of the 36 questions, eight questions are included in the physical function
domain. The physical function subscale includes questions about how limited
subjects are in ability to perform daily activities including bathing, light
housework and walking short and long distances. These questions are consistent
with a definition of physical disability as defined above. The physical component
score (PCS) of the SF-36 includes all questions from the physical domain, such as
whether health or pain limited the subjects ability to complete other activities.
Although the PFS and the PCS are similar, the PFS was considered to be more
closely match the disability definition of the disablement model, and thus is chosen
as the outcome measure for the second specific aim. Scoring of the SF-36 is done
according to standards protocols by the Medical Outcomes Study described later in
this chapter.

The MOS SF-36 measure has been used extensively in the United States and
abroad. The SF-36, as it is typically called, demonstrates excellent reliability
(Cronbachs alpha > 0.85 and reliability coefficients > 0.75) for all domains except
social functioning.194 In addition, this instrument is able to detect differences in
groups of adults in terms of health status, both mental and physical.195 The scores
are stable over time, and have good criterion validity.196 Because this instrument
has been used so extensively, there are established normative values, which allow
comparison of scores between groups with different medical conditions. The MOS
SF-36 has been administered in interview, phone and written formats.
The SF-36 was administered in a written format for this study. Scoring was
based on an algorithm developed by the RAND Corporation.197 The scores were
recorded as raw scores and transformed scores. Transformed scores equal the
actual raw score minus the lowest possible raw score divided by the possible raw
score range and multiplied by 100. Physical and mental component scores (PCS
and MCS) are scored using T-scores transformed for norm-based methods.
Normative scores for the US population are based on a mean PCS and MCS score
of 50, with a standard deviation of 10 points.197
PAD-Specific Quality of Life Measure: Although a generic health-related
quality of life measure such as the SF-36 can be useful when comparing outcomes
between different populations, these measures are unable to distinguish the

importance of the physical, personal and social factors in determining quality of
life in people with PAD. A HRQOL questionnaire specific to the concerns of
patients with PAD is useful for this research and may be used to identify new
therapeutic areas that may positively impact HRQOL.
The PADQOL questionnaire was developed using the qualitative research
methods of interviews and grounded theory.13 The results of these interviews
indicate that people with PAD have significant limitations in psychological, social,
and emotional functioning that has been inadequately identified in currently
existing instruments. Studies as to the reliability or validity of this new instrument
are in process. This instrument was chosen because it represents the most
thorough evaluation of HRQOL for the PAD patient at this time. The questionnaire
can be administered in written or interview format. This 60- item questionnaire is
in a Likert scale format, with two types of questions: participants are asked to rate
their agreement or disagreement with each statement and then to indicate the
importance of this item. Scoring is performed by multiplying the level of
agreement or disagreement by the importance ranking. Actual scores are
converted to a summary score, with higher scores indicating greater quality of life.
Perceived Control over Life Events: Psychological attributes can contribute
to successful coping in people in stressful situations or with chronic illness.
Mastery, also known as perceived control, describes the extent to which one

perceives his/her life changes are under his control, as opposed to having no
control over life situations. Pearlin developed a seven question Mastery scale with
answers assigned a score based on agreement or disagreement with each
statement. 163 Scores range from 7 to 35. Questions include items such as; I have
little control over the things that happen to me. This survey was administered in
written format and scored by the principal investigator. Internal reliability for
mastery has been reported as 0.73.158
Social Supports: The Interpersonal Support Evaluation List (ISEL) is
designed to measure the perceived availability of four independent sources of
social supports.166 The types of support include tangible, appraisal, self-esteem and
belonging support. Tangible support indicates availability of material assistance,
while appraisal indicates the perceived ability of someone with whom to discuss
ones problems. Self-esteem indicates whether ones perceived assessment of
him/her self is positive as compared with other acquaintances and the belonging
scale assesses whether the individual perceives he/she has availability of people
with whom to do social activities. This forty-item questionnaire asks subjects to
respond to each statement as true or false (and mostly true or mostly false). The
scoring algorithm assigns scores to each question. A summary score can be
reported, as can sub-scale scores in the 4 areas. Test-retest reliability for the ISEL
was reported as r = .993.198 Cronbachs internal reliability scores range between

.60 and .77 for the subscales and total score, indicating that the ISEL is a reliable
measure of social support and that the subscales do represent different domains of
social support.
3.6 Statistical Analysis
The purposes of this study were to describe the differences in walking speeds
and gait parameters between patients with PAD-IC and elderly controls, and to
describe the factors that predict physical function score and health-related quality
of life in the PAD-IC group. All analyses were performed with SPSS statistical
package (Version 11.5) with statistical significance set at p< 0.05. In the
following section, statistical analysis is provided for each specific aim.
3.6.1 Sample Size Estimation
The sample size was estimated for both hypotheses 1 and 2. For the
differences in walking speed between PAD-IC groups and controls, based on
previous work, 22 subjects were needed in each group for a power of 90%. For the
second hypothesis that self-selected, usual walking speed will significantly predict
physical function score in subjects with PAD-IC, no studies were found that used
PFS in this manner. Sample size estimation therefore, was based on research that
reported walking speed accounted for an for r2 of .14 for the outcome of disability
defined as being dependent in activities of daily living,102 It was estimated that
PAD identified by ABI <0.90 probably accounted for an additional variance of

10% (.10) for the outcome of physical disability. Using PASS 6.0, with variability
of .25 accounted for, and an alpha of 0.05, 50 subjects total provides a power of
90%. This allowed for 25 subjects in each group. Twenty-five subjects in each
group for a total of 50 subjects should demonstrate adequate power.
Hypothesis three was considered exploratory. Therefore, no sample size
determination was made for the hypothesis that mastery would predict PADQOL.
3.6.2 Analysis for Specific Aim 1
Describe and evaluate the differences in walking speed, distance and gait
parameters between subjects with PAD-IC and a control group similar in age,
gender and physical activity status.
Hypothesis 1: Subjects with PAD-IC will walk significantly slower than the
control group at usual and maximal walking speeds.
Hypothesis la: Subjects with PAD-IC will walk a significantly shorter distance in
a six-minute walk compared to the control group.
Hypotheses lb: Subjects with PAD-IC will have a significantly shorter stride
length at usual walking speed compared to the control group.
Statistical Analysis: Differences in means for each group regarding demographic
and physical function characteristics were examined using independent sample t-
tests for continuous variables and chi-square for categorical variables. Independent
sample t-tests were used for comparing means of walking speed, six-minute

walking distance and all gait variables between groups. Pearson Correlation
coefficients were used to evaluate the relationship between usual walking speed
and the gait parameters at usual speed (stride length, width, cadence, and toe
angle) and maximal walking speed and gait variables at maximal speed.
3.6.3 Analysis for Specific Aim 2
Describe and evaluate the factors that predict physical function score in subjects
with PAD-IC.
Hypothesis 2: Self-selected, usual walking speed will significantly predict
physical function score in subjects with PAD-IC.
Statistical Analysis: Simple (univariate) regression analyses were performed to
determine whether significant relationships existed between physical function
score and walking measures, gait parameters and relevant measures of health
status. These univariate relationships were examined in the entire sample, as well
as within each group (PAD-IC and control). Variables that were statistically
significant from the univariate analyses, as well as other clinically relevant
variables were used to explore multivariate models to explain PFS in subjects with
PAD-IC. To evaluate the significance of PAD-IC on PFS, a series of bivariate
regression analyses was performed. Each bivariate analysis included PAD-IC
group as one variable, and a single confounding independent variable as the other
variable. The following regression equation was used: Y = Bo + Bi(PAD group

assignment) + B2(independent variable), where Y is physical function score, Bo is
the intercept and Bi and B2 the coefficients for PAD group and the independent
variables, respectively. The coefficient, B, indicates the change in physical
function score for each unit of the independent variables, and thus is more
clinically useful than the R or r2 values. In addition, the B coefficient provides an
unbiased estimate of the effects of interest and therefore can be used to generalize
3.6.4 Analysis for Specific Aim 3
Explore the physical and psychosocial factors that explain PADQOL in
subjects with PAD-IC.
Hypothesis 3: Personal characteristic of mastery will be the most significant
predictor of PADQOL in individuals with PAD-IC.
Statistical Analysis: A multivariate linear regression was used to examine the
effect of physical, personal and social factors for the outcome of PADQOL.
Several models were fitted, using either self-report of physical function or actual
measures of physical function (walking speed and six-minute walk).


Figure 3.2 Measurements of Step Length, Stride Length and Stride Width