Citation
Examination of the role of patient psychological and chronic illness comorbidites in behavioral health provider adherence to integrated primary care model metrics in five federally qualified health centers

Material Information

Title:
Examination of the role of patient psychological and chronic illness comorbidites in behavioral health provider adherence to integrated primary care model metrics in five federally qualified health centers
Creator:
Kinman, Carissa Rose
Publication Date:
Language:
English
Physical Description:
x, 76 leaves : ; 28 cm

Subjects

Subjects / Keywords:
Comorbidity ( lcsh )
Mental health services ( lcsh )
Primary care (Medicine) ( lcsh )
Medical centers ( lcsh )
Integrated delivery of health care ( lcsh )
Comorbidity ( fast )
Integrated delivery of health care ( fast )
Medical centers ( fast )
Mental health services ( fast )
Primary care (Medicine) ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 61-76).
General Note:
Department of Psychology
Statement of Responsibility:
by Carissa Rose Kinman.

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:
710982542 ( OCLC )
ocn710982542
Classification:
LD1193.L645 2010m K56 ( lcc )

Full Text
EXAMINATION OF THE ROLE OF PATIENT PSYCHOLOGICAL AND
CHRONIC ILLNESS COMORBIDITIES IN BEHAVIORAL HEALTH PROVIDER
ADHERENCE TO INTEGRATED PRIMARY CARE MODEL METRICS IN FIVE
FEDERALLY QUALIFIED HEALTH CENTERS
by
Carissa Rose Kinman
B.A., William Jewell College, 2005
A thesis submitted to the
University of Colorado Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts
Clinical Psychology
2010


This thesis for the Master of Arts
degree by
Carissa Rose Kinman
has been approved
by
David Albeck
0(/ 7 2e>/o
Date
borah Seymour


Kinman, Carissa Rose (M.A. Clinical Psychology)
Examination of the Role of Patient Psychological and Chronic Illness Comorbidities
in Behavioral Health Provider Adherence to Integrated Primary Care Model Metrics
in Five Federally Qualified Health Centers
Thesis directed by Assistant Professor Abbie O. Beacham
ABSTRACT
The prevalence of mental illness has created a significant burden for individuals and
the healthcare system. Unfortunately, very few people with psychological or
behavioral concerns actually receive appropriate mental health treatment. Moreover,
an increasing number of people suffer from chronic illnesses that are often comorbid
with psychological or behavior problems. For a variety of reasons, the primary care
system has become the de facto mental health care system and is not well equipped
to effectively handle all of the needs of many primary care patients. Efforts to
improve the identification and treatment of psychological and behavioral problems
have lead to the integration of primary care or health psychologists into primary care
settings. Current research has shown positive outcomes from integrated primary care
models of service delivery, however, the key components of integrated primary care
models are still being identified. Additionally, little is known about the effects of
psychological and physical comorbidities on integrated primary care. The purpose of
the present study is to examine the role of multiple psychological and chronic illness


comorbidities in Behavioral Health Provider adherence to integrated primary care
model metrics in federally qualified health centers. Results indicate that the number
of psychological comorbidities that a patient has is positively related to the length of a
session with a Behavioral Health Provider. Additionally, adherence to the integrated
primary care model metrics was predicted by the number of psychological
comorbidites and the number of years of clinical training the Behavioral Health
Provider had received. This study provides information to better understand how
patient comorbidities affect integrated primary care. This study also highlights the
need for Behavioral Health Providers working in primary care settings to have
specialty training in health psychology in order to effectively treat the wide range of
complexities that patients present with in primary care settings.
This abstract accurately represents the content of the candidates thesis. I recommend
its publication.
Signed
Abbie O. Beacham


DEDICATION
This thesis is dedicated to my loving husband for his unfaltering support, patience and
understanding and for sacrificing so much to help me complete this thesis. I simply
could not have done this without him. I would also like to dedicate this to my family
for teaching me the value of hard work, and for helping me achieve my goals. I
cannot thank them enough for all of their love, support, and sacrifice throughout my
life.


ACKNOWLEDGEMENT
I would like to especially thank my chair, Abbie Beacham, Ph.D. for her unwavering
support and guidance throughout this project. I would also like to thank my
committee members, David Albeck, Ph.D. and Deborah Seymour, Psy.D., for all of
their helpful feedback as well as the time they devoted to reviewing my work and
participating in my proposal and defense presentations. I also want to express my
gratitude to my fellow trainees, Andrew Herbst, B.A., Dana Brown, M.A., and Jessica
Payne-Murphy, B.A., for their assistance and encouragement through this entire
project.


TABLE OF CONTENTS
Tables.................................................................x
CHAPTER
1. INTRODUCTION........................................................1
Mental Illness in the United States.............................1
Chronic Illness in the United States............................3
Behavioral Factors Related to Chronic Illness.................4
Comorbid Mental Illnesses and Chronic Medical Illnesses.........5
Primary Care as the De Facto Mental Health System...............7
Models of Behavioral Health Service Delivery in
Primary Care....................................................9
Coordinated..................................................10
Co-Located Model.............................................11
Fully Integrated Model.......................................12
Primary Care Behavioral Health Consultation Model.........12
Application and Outcomes of Integrated
Primary Care.................................................14
Primary Care Patient Characteristics and Integrated
Primary Care.................................................18
Purpose and Hypotheses of the Present Study....................24
vii


2. METHOD
26
Participants......................................................26
Procedure.........................................................27
Measures..........................................................28
Psychological Diagnoses.........................................28
Medical Diagnoses...............................................28
Length of Patient Session.......................................28
Years of Clinical Training......................................29
Metrics of the Integrated Primary Care Model....................29
Data Analysis.....................................................29
3. RESULTS...............................................................32
Participant Sample................................................32
Medical Provider Characteristics..................................36
Behavioral Health Consultant Characteristics......................37
System Characteristics............................................41
Hypothesis One....................................................42
Hypothesis Two....................................................43
Hypothesis Three..................................................43
Hypothesis Four...................................................43
viii


Hypothesis Five..............................................45
Supplementary Analysis.......................................45
4. DISCUSSION...............................................48
Sample Characteristics.......................................48
Hypothesis One...............................................51
Hypothesis Two...............................................52
Hypothesis Three.............................................53
Hypothesis Four..............................................54
Hypothesis Five..............................................55
Limitations..................................................58
Future Directions............................................59
REFERENCES...............................................................61
ix


TABLES
Table
3.1 Participant characteristics by gender and overall sample......................33
3.2 Patient psychological diagnoses by gender and overall sample..................34
3.3 Patient medical diagnoses by gender and overall sample........................35
3.4 Clinic location of referrals by medical provider..............................36
3.5 Type of referral by medical provider..........................................37
3.6 Clinic location of referrals by BHC...........................................38
3.7 Type of patient referral by BHC...............................................39
3.8 Number of patient sessions by BHC.............................................40
3.9 Referral source by clinic location............................................41
3.10 Type of referral by clinic location...........................................42
3.11 Logistic regression analysis for variables predicting success/failure to
adhere to the model length of session parameter..............................44
3.12 Means, standard deviations, and intercorrelations for number of follow
up appointments and predictor variables......................................46
3.13 Multiple regression analysis for variables predicting number of patient
follow up visits with a BHC..................................................47
x


CHAPTER I
INTRODUCTION
Mental Illness in the United States
Nearly 30% of the adult population or an estimated 57.7 million adults in the
United States experiences a diagnosable mental illness within a given year (Kessler,
Chiu, Dernier, & Walters, 2005; World Health Organization [WHO], 2008).
However it is difficult to assert this statistic with confidence considering only about
30% of those with a mental illness actually receive some type of treatment, thus,
making it difficult to accurately estimate the true prevalence (Narrow, Rae, Robins, &
Regier, 2002; Young, Klap, Shoai, & Wells, 2008). The most frequently researched
mental illnesses include depression and anxiety with estimated prevalence rates of
6.7% and 18% respectively within the general adult population during a twelve month
period of time (Kessler et al., 2005).
In addition to those who experience a diagnosable mental illness or disorder, a
large majority of adults also experience significant psychological symptoms that,
while may not meet criteria for a DSM diagnosis, do cause a considerable amount of
distress (Strosahl, 2001). Mental illnesses and other psychosocial stressors, if left
untreated can significantly impact a persons overall functioning and quality of life
(Kroenke, Spitzer, Williams, Monahan, & Lowe, 2007; Simon, 2003). Specifically,
1


mental illnesses can contribute to non-adherence to prescribed medical treatments,
diminished immune functioning, and unhealthy behaviors (WHO, 2008). Untreated
mental illnesses can cause loss of productivity at work which can lead to significant
increases in days of work missed due to illness or disability, especially when
comorbid medical conditions are present (Olfson et al., 2000; Simon, 2003).
Persistent depression has been shown to be strongly associated with high rates of
suicidal ideation, family distress, functional impairments and lower levels of quality
of life especially in areas of social and emotional functioning (Olfson, et al., 2000;
Simon, 2003; Young, et al., 2008). Taken together, diagnosis, treatment and
management of mental illness, and psychological symptomatology presents a
significant burden for the United States health care system. Conservative estimates
suggest that the indirect cost of mental illness is $79 billion annually while more than
$99 billion is spent on direct treatments for mental disorders (U.S. Public Health
Service, 1999).
Unfortunately, most people with a mental illness will never have access to any
type of mental health service due to a myriad of barriers that limit access to these
services. Some estimates suggest that only 30% of those with a mental illness
actually receive necessary treatment and moreover, only 15%-20% of those treated
receive treatment that is minimally adequate or evidence based (Kessler et al., 2003;
Narrow, et al., 2002; Wang, Berglund, & Kessler, 2000; Wang et al., 2005; Young, et
2


al., 2008). In one national sample of 9,585 people in the United States, only 19% of
those with a diagnosed depressive or anxiety disorder received care from a mental
health specialist and less than 6% received both the appropriate medications and at
least four counseling sessions (Young, et al., 2008).
Chronic Illness in the United States
In 2005, there were approximately 133 million Americans living with at least
one chronic illness such as diabetes, stroke, lung disorders, or chronic pain (Ogden,
Carroll, McDowell, & Flegal, 2007) and many patients have more than one chronic
illness (Rothman & Wagner, 2003). These chronic illnesses and conditions can lead
to very expensive treatment and management over time and have been estimated to
account for as much as 78% of total health care costs (Anderson & Horvath, 2004;
Fisher et al., 2008; Hoffman, Rice, & Sung, 1996). As the number of diagnosed
chronic conditions per person increases, the healthcare cost per person likewise
increases significantly. Currently, it is estimated that the average cost for a patient
with one chronic condition is $1900 per year while the cost for a patient with five or
more chronic conditions is an estimated $11,500 (Anderson & Horvath, 2004).
A large portion of the increased costs for patients with chronic illnesses can be
attributed to the significant increase in healthcare utilization in this population.
Anderson and Horvath (2004) examined the burden of chronic illness by analyzing
3


data from 24,072 non-institutionalized Americans who participated in the Household
Component of the 1998 Medical Expenditure Panel Survey. Results from this survey
showed that those people with chronic illnesses utilized the most medical care. Those
who reported having more than one chronic illness utilized 80% of home health
services, 67% of prescriptions, 48% of physician visits, and 56% of inpatient stays
(Anderson & Horvath, 2004). Additionally, a persons utilization of medical services
increases dramatically as they are diagnosed with more chronic illnesses. Medicare
beneficiaries with five or more chronic conditions average 37 annual physician visits
to almost 14 different physicians compared to the average Medicare beneficiary who
sees six or seven physicians annually (Berenson & Horvath, 2003).
Behavioral Factors Related to Chronic Illness
According to the Unites States Department of Health and Human Services
(DHHS) roughly half of all deaths in the United States can be attributed to
preventable lifestyle behaviors (2000) with the top three factors being tobacco, poor
diet and physical activity, and alcohol consumption (Mokdad, Marks, Stroup, &
Gerberding, 2004). In fact, poor diet and lack of physical activity alone account for
approximately 400,000 deaths per year (Mokdad, et al., 2004). These behaviors are
directly and independently related to elevated risk for onset and progression of a
number of chronic illness conditions including diabetes, cardiovascular disease and
4


certain types of cancers (Fisher, et al., 2008; U.S. Department of Health and Human
Services, 2000).
Comorbid Mental Illnesses and Chronic
Medical Illnesses
Comorbid mental illnesses are often common in patients with chronic medical
conditions such as chronic pain, cancer, cardiovascular disease, diabetes, arthritis, and
asthma (Chapman, Perry, & Strine, 2005; Katon, 2003; Katon & Ciechanowski, 2002;
Katon, Lin, & Kroenke, 2007; WHO, 2008). Elevated levels of anxiety and
depression are particularly common in patients with medical disorders (Katon, 2003;
Katon & Ciechanowski, 2002; Katon, et al., 2007). Studies examining comorbid
depression and chronic illness have demonstrated prevalence rates around 40-50%
(Goldman, McCulloch, & Cuffel, 2003; Yates et al., 2007).
Research suggests a bidirectional relationship between the effects of
depression/anxiety and medical illnesses (Chapman, et al., 2005; Katon, 2003;
Rudisch & Nemeroff, 2003; Simon et al., 2007). Since depression and anxiety are
associated with adverse health-risk behaviors such as smoking, poor, diet, and lack of
physical activity (Katon, 2003; Rosal et al., 2001) this can lead to an increased risk of
developing a chronic illness such as coronary artery disease (Lett et al., 2004) and
diabetes (Katon, 2003). On the other hand, burden of chronic illnesses and increased
5


functional impairment related to the illness may lead to the development or
exacerbation of depression and anxiety (Katon, 2003). For example, having a heart
attack increases the likelihood that a person will develop depression (Strik, Lousberg,
Cheriex, & Honig, 2004). Increased symptoms of depression have also been found to
interfere with prescribed medical treatment through decreased self care and poor
treatment adherence (Hoffman, et al., 1996; Katon, 2003; Lett, et al., 2004) which can
lead to the exacerbation of medical symptoms and increased functional disability
(Katon, 2003; Katon, et al., 2007). For instance, patients with diabetes and comorbid
depression experience increased symptoms of diabetes with increasing depression
severity (Ciechanowski, Katon, Russo, & Hirsch, 2003), and often have poorer
physical functioning as well as lower adherence to dietary recommendations and
exercise (Ciechanowski, et al., 2003; Katon et al., 2010). Katon, Lin, and Kronenke
(2007) recently reviewed the current literature to examine the association between
depression or anxiety with medical symptom burden in patients with diabetes, heart
disease, pulmonary disease, and arthritis. In their review of 31 studies, they found
that overall, patients with these chronic illnesses and comorbid anxiety or depression
reported more medical symptoms while controlling for the severity of the disease
(Katon et al., 2007). Additionally, depressed patients in primary care settings with
chronic medical conditions have a 40%-75% increase in costs for health services
(Ciechanowski, Katon, & Russo, 2000; Simon, 2003).
6


Primary Care as the De Facto Mental
Health System
It has been estimated that 70-80% of patient visits in primary care are for
symptoms that stem from psychosocial issues (Gatchel & Oordt, 2003; Kroenke &
Mangelsdorff, 1989). A significant number of these symptoms can be explained by
mental illnesses. One study of patients in primary care revealed that 18.9% had major
depressive disorder, 14.8% had generalized anxiety disorder, 8.2% had panic disorder
and 7.9% had a substance use disorder and the majority of patients met diagnostic
criteria for two or more psychological disorders (Olfson, et al., 2000). Additionally,
many primary care visits are related to symptoms such as chest pain, fatigue,
dizziness and abdominal pain that have no known organic cause (Kroenke &
Mangelsdorff, 1989), which increases the difficulty to accurately diagnose patients
and may lead to unnecessary tests, treatments, and use of prescriptions (Katon et al.,
1990; O Donohue, Byrd, Cummings, & Henderson 2005).
Of those patients currently receiving mental health treatment, 40% to 50% get
this treatment only in general medical settings (Kessler et al., 2005; Uebelacker,
Wang, Berglund, & Kessler, 2006) and often only receive treatment in the form of
medication (Ledoux, Barnett, Garcini, & Baker, 2009; Wang, et al., 2000; Wang, et
al., 2005). Frequently, psychological and behavioral concerns are often undetected
by primary care providers and when they are recognized, few patients receive
7


adequate treatment. One study suggests that 83% of people with depression see a
medical provider during a depressive episode, but less than half receive a diagnosis
from their primary care provider and only 30% received appropriate mental health
treatment (Young, Klap, Sherbourne, & Wells, 2001). These patients with mental
illnesses have been shown to utilize large amounts of resources in primary care.
Patients diagnosed with depression demonstrated health care costs of approximately
$4246 annually compared to $2371 for those patients with no depression diagnosis
(Simon, VonKorff, & Barlow, 1995).
For a variety of reasons, including dwindling access to mental health care and
third party payor restrictions, most people who experience psychological symptoms,
seek and receive treatment solely from their primary care medical provider. In
addition to diagnosable mental illnesses, many, if not most, medical problems such as
chronic illness have psychological and behavioral components often only managed by
a patients primary care provider (Belar, 2008b; Gatchel & Oordt, 2003). Many have
come to regard primary care as the de facto mental health system (Regier et al.,
1993). This creates a system in which those needing behavioral health services are
seen only in general medical settings by primary care providers with limited time and
training (Blount et al., 2007; Levant, 2005; WHO, 2008).
Ultimately, despite its best efforts, the healthcare system is currently
providing effective care for only a small percentage of those who need it. Given that
8


roughly 80% of the population visits a Primary Care Provider (PCP) for at least one
visit per year, it follows that primary health care is a promising venue for patients to
receive treatment for psychological and behavioral problems (Robinson & Strosahl,
2009; Strosahl, 1998; Young, et al., 2008). Therefore, if behavioral health
professionals (e.g., psychologists) could be consistently integrated directly into
primary care clinics as a regular member of the health care team, far greater numbers
of patients in need of these services could be served (Levant, 2005).
Models of Behavioral Health Service
Delivery in Primary Care
Since the 1980s, mental health treatment has been carved out to managed care
organizations that treat mental illnesses completely separate from medical illnesses
(Gray, Brody, & Johnson, 2005). In this current model of service provision, patients
must seek and arrange services in a system that is physically and organizationally
separate from other health care providers. This arrangement has created a system that
is difficult for physicians to make an appropriate referral to mental health. In
addition, funding for these services has been reduced significantly (Strosahl, 2001,
2005; Strosahl & Quirk, 1994). The model the cornerstone of the community
mental health system is rapidly becoming regarded as one which is impractical and
ineffective (Robinson & Reiter, 2007; Strosahl, 2005). Strosahl asserts that this
9


model is not only at capacity, it is largely ineffective (Strosahl, 2001; Strosahl &
Quirk, 1994). Therefore, the need to identify and treat psychological symptoms
earlier is paramount and requires a different model of care (Strosahl & Quirk, 1994).
Efforts to improve the identification and treatment of psychological and
behavioral problems have lead to the integration of primary-care or health
psychologists into community health centers and various primary care practices
(Packard, 2007). Models of care that utilize mental and behavioral health
professionals in primary care settings are becoming more common. Currently, there
are several types of models, and labels of models, used to describe primary care
behavioral health services.
Coordinated
Coordinated care is one such model that refers to services provided separately
by a medical professional and a behavioral health professional in their respective
locations (Blount, 2003). A referral to one provider initiates a level of enhanced
communication between the healthcare providers. Information regarding a patients
treatment at one location is regularly communicated to the other provider in an effort
to improve the quality of care that the patient is receiving.
10


Co-Located. Model
The co-located model refers to services provided by behavioral health and
medical professionals from the same facility (Blount, 2003). Some depict co-located
services not as a specific model but simply a description of the physical location of
services (Miller, Mendenhall, & Malik, 2009). However, others describe this as a
separate model in which the healthcare providers operate independently in their own
offices and systems but may share the same waiting room and support staff (Gatchel
& Oordt, 2003). Typically a referral system exists in which a patient is referred by
their PCP to the behavioral health provider who operates in a traditional mental health
service delivery model that includes a comprehensive intake and 50 minute sessions
with patients (Gatchel & Oordt, 2003; Strosahl, 2005). However, communication
between providers should continue once a behavioral health provider has seen the
patient. By having the separate services in close proximity to one another, the
providers can consult more regularly regarding the treatment of patient. Through
more regular consultation with the behavioral health providers, medical providers can
become more familiar with types of treatment and services that behavioral health
professionals provide. The co-located model can help to improve the efficiency of
the referral process for patients needing behavioral health services as well as help the
medical providers improve their own skills in addressing behavioral health issues
with their patients (Blount, 2003).
11


Fully Integrated Model
The fully integrated model describes services in which Behavioral Health
Consultants (BHCs) work alongside Primary Care Providers (PCPs) in the same
clinic with most of the same patients (Blount, 2003). In this model, the BHC is
considered a regular part of the health care team (Robinson & Reiter, 2007) and the
two groups of professionals work together to develop care plans for patients that
incorporate both medical and behavioral components (Blount, 2003). This
collaboration stems from a biopsychosocial model of care (Engel,1977) and allows
for treatment of the complete individual in one location.
Primary Care Behavioral Health Consultation Model
The Primary Care Behavioral Health Consultation Model (PCBH), developed
through the work of Kirk Strosahl (Strosahl, 1996a, 1996b, 1998) is one example of a
fully integrated primary care model that is widely used. The mission of PCBH model
is to improve the overall health of the population (Strosahl & Robinson, 2008) partly
by providing patients easy access to behavioral health care. Across the United States
it has been implemented in over 100 Federally Qualified Health Centers and is largely
used in primary care clinics within the United States Air Force and Navy, the
Veterans Administration, and Kaiser Permanente (Robinson & Reiter, 2007; Strosahl
& Robinson, 2008).
12


In the PCBH model, the PCP refers a patient to the BHC through a few
different methods including a traditional referral in which the patient schedules an
appointment with a BHC at a later time or a warm hand-off. The warm hand off
occurs when a PCP identifies psychological or behavioral health concerns during an
appointment with a patient and asks the BHC for immediate assistance with the
patient (Robinson & Reiter, 2007). The BHC will usually conduct a brief functional
analysis with the patient focusing on the concerns that led to the referral then provide
the patient with treatment recommendations and brief interventions. The BHC may
see 10-15 patients per day but the interventions are typically performed in one to four
appointments and are only 15-30 minutes long, allowing the BHC to work within the
primary care model to provide effective, brief behavioral interventions (Robinson &
Reiter, 2007). The BHC will then follow up with the PCP to provide feedback and
discuss ongoing treatment recommendations for the patient (Bryan, Morrow, &
Appolonio, 2009; Gatchel & Oordt, 2003). The treatment plans developed in
conjunction with the patient and the PCP are typically focused on functional
improvements rather than on symptom reduction (Robinson & Reiter, 2007). The
BHCs goal is not to eliminate all of the patients symptoms, but rather improve the
patients functional status and consult with the PCP to identify goals that the PCP and
patient can continue to work on (Strosahl & Robinson, 2008). In this model, the
BHC acts as a consultant within the primary care setting, and the full responsibility of
13


the patients treatment remains that of the PCPs (Bryan, et al., 2009; Robinson &
Reiter, 2007).
Application and Outcomes of Integrated
Primary Care
One goal of the PCBH model is to assist with the mental health problems of
the patients in primary care. Research has demonstrated that increased integration of
services in primary care leads to improved mental health outcomes (Hedrick et al.,
2003; Hunkeler et al., 2006; Price, Beck, Nimmer, & Bensen, 2000; Roy-Byrne et al.,
2005). Partially integrated models have also been shown to reduce no-show rates for
mental/behavioral health services compared to usual care and as the level of
integration increases, the rate of no-shows decreases (Guck, Guck, Brack, & Frey,
2007). One randomized trial indicated that primary care patients who received brief
depression treatment in primary care had better clinical outcomes, a low patient drop-
out rate, improved medication adherence, and were more likely to follow relapse
prevention plans compared to a group of patients receiving usual primary care (Katon
et al., 1996).
Roy-Byrne and colleagues (2005) examined the effectiveness of brief
Cognitive Behavioral Therapy (CBT) plus medications compared to usual care in a
group of primary care patients with panic disorder. The intervention group received
14


up to six sessions of CBT with a masters or doctoral-level behavioral health specialist
over a three month time period plus up to six brief follow-up phone calls for the
remainder of the year (Roy-Byrne et al, 2005). More than 70% of the patients in this
sample had at least one comorbid mood or anxiety disorder and two-thirds had a
comorbid medical condition. The intervention group demonstrated a significantly
greater proportion of patients with zero panic attacks and minimal anticipatory
anxiety and phobic avoidance at 3, 6, 9, and 12 month follow-ups. The intervention
group also had significantly greater improvements in measures of disability and
mental health functioning compared to the usual care group (Roy-Byrne et al, 2005).
Other studies examining integrated primary care models have found similar
results. When compared to patients receiving treatment as usual in primary care
settings, patients who receive mental health treatment in an collaborative care model
report more days without depressive symptoms (Simon et al., 2001) as well as more
anxiety free days (Katon, Roy-Byrne, Russo, & Cowley, 2002). In a study using the
PCBH model, Bryan, Morrow, and Appolonio (2009), found that those patients with
up to three appointments with a BHC demonstrated improvements in subjective well-
being, global mental health, emotional distress, and life functioning. The results of
this study indicate that rapid improvements in mental health and life functioning can
occur in the brief behaviorally focused interventions that are the central to the PCBH
model (Bryan et al., 2009).
15


Psychological and behavioral interventions focused on lifestyle changes have
consistently been shown to be effective treatments in patients with chronic illnesses
(Foreyt & Goodrick, 1993; Kessler & Stafford, 2008a; Stetson, Carrico, Beacham,
Ziegler, & Mokshagundam, 2006). One pilot study examined the efficacy and
feasibility of a brief CBT intervention for adults with diabetes in real world
outpatient diabetes clinic (Stetson, et al., 2006). The intervention consisted of six
weekly sessions that focused on realistic goal setting and enhancing positive self-care
behaviors that impact diabetes control and cardiovascular risk such as a healthy diet,
stress/mood management, physical activity, and blood glucose testing (Stetson et al.,
2006). Post-intervention measures indicated that patients who kept a record of their
physical activity had greater behavioral changes in this area. Overall, patients were
able to set more specific behavioral goals after completing the intervention (Stetson et
al., 2006).
One diabetes treatment protocol called Take Charge! incorporates five
recommended areas for psychoeducational treatments for patients with depression
which include, educational information, coping skills and stress management, social
support, diet, and education (Callaghan, Gregg, Ortega & Berlin, 2005; Gregg,
Callaghan, Hayes, & Glenn-Lawson, 2007). Preliminary results from an evaluation
of this treatment protocol in a primary care clinic indicate that it may help to increase
psychological functioning, increase quality of life, increase adherence to medical
16


treatment, and decrease blood glucose levels in patients with type 2 diabetes
(Callaghan, et al., 2005).
A recent randomized controlled trial examined the impact of an Acceptance
and Commitment Therapy (ACT) intervention in a sample of patients with Type 2
diabetes in a community health center (Gregg, et al., 2007). In this study, patients
either received only education about diabetes management or education in addition to
mindfulness and acceptance training to learn skills to apply to difficult thoughts and
feelings they might have related to their diabetes. Gregg et al. found (2007) that both
groups showed overall improvements, but at the three-month follow-up, patients in
the ACT group had better diabetes self-care and had blood glucose levels in a normal
range.
When applied as proposed, the PCBH model is posited to be effective and
increase access to care. During a primary care visit, the patient is just as likely to see
a BHC as s/he is to see a medical professional (Robinson & Reiter, 2007). The BHC
is part of the primary care team and collaborates regularly with the PCP and is active
in the treatment planning for a patient (Robinson & Strosahl, 2009). This model
allows for the increased availability of the BHC, which helps reduce some of the
barriers to mental/behavioral treatment and increase access to adequate care for those
who need it (Robinson & Reiter, 2007). BHCs working in this model should have the
ability to work with the wide range of problems that patients present with in primary
17


care including mental health concerns such as depression or anxiety in conjunction
with medical problems such as chronic pain, hypertension, and obesity (Robinson &
Reiter, 2007).
Primary Care Patient Characteristics and
Integrated Primary Care
Models of service delivery in which medical providers work more closely
with behavioral health providers to treat a patient have shown positive outcomes in
primary care settings. A recent systematic review conducted by the Agency for
Healthcare Research and Quality (AHRQ) examining 33 trials of current models of
integrated care in the United States found that overall, integrated care does produce
positive outcomes (Butler et al., 2008). However, the key components for the most
effective integrated primary care treatment models are still being identified (Butler, et
al., 2008; Miller, et al., 2009) and few studies examining the effectiveness of the
PCBH integrated care model have been conducted (Bryan, et al., 2009; Robinson &
Strosahl, 2009). So far, much of the research examining collaborative/integrated
primary care models has focused on targeting specific diseases (Butler, et al., 2008).
Considering the wide range of mental/behavioral health and medical concerns that
patients have in primary care settings, it can often be difficult to translate this
research into real world medical settings.
18


Recently, the Four Quadrant Clinical Integration Model has been introduced
as a population-based planning tool to help determine the appropriate level of
integration based on the behavioral health and medical risk factors of the system
population (Mauer, 2006). According to this model, patients in Quadrant I have low
behavioral health needs and low physical needs and can be served in a primary care
setting. Patients in Quadrant II are characterized as having high behavioral health
needs but low to moderate physical health risk/complexity and can be served in
specialty mental health settings through collaboration with medical services (Mauer,
2006). Quadrant III describes patients who have a low to moderate behavioral health
risk/complexity but high physical needs. These patients are typically best served in
the general medical setting but still may need mental/behavioral health services
(Collins, Hewson, Munger, & Wade, 2010). Quadrant IV patients require the highest
level of integration of services as they typically have a high level of behavioral health
and physical needs (Mauer, 2006). The PCBH model is one example of a fully
integrated model that is appropriate to address the needs of Quadrant IV patients
(Collins et al., 2010). However, most of the current research examines integrated
care models that are only similar to the PCBH model and very few studies are
conducted in typical medical settings.
The recent systematic review conducted by the Agency for Healthcare
Research and Quality (AHRQ) examined 33 RCTs and high quality quasi-
19


experimental studies of collaborative/integrated care models that mostly addressed
depression (n = 26) and anxiety (n = 4) (Butler et al., 2008). According to the
authors, most of the studies demonstrated that integrated care significantly improves
patient treatment response and remission rates compared to usual primary care (Butler
et al., 2008). However, in this report, there did not seem to be a correlation between
increased levels of integration and the outcome measures. Additionally, none of the
included studies were conducted in typical medical settings and the majority of the
studies that addressed depression excluded patients that had comorbid psychological
and physical conditions. From this report, it is difficult to ascertain which
components of integrated care are essential to produce desired outcomes. It is also
unclear who will most likely benefit from integrated care since there is little research
that examines the efficacy of these models in a real world medical setting (Butler et
al., 2008).
Improving Mood: Promoting Access to Collaborative Treatment (IMPACT)
is one multisite randomized clinical trial of collaborative care that targets depression
in elderly patients that was reviewed in the AHRQ report (Hunkeler, et al., 2006).
This collaborative care model provided patients with brief behavior based
psychotherapy and pharmacotherapy delivered by a care team that included a
depression care manager, a consulting psychiatrist, the patients PCP and a liaison
primary care doctor. Compared to usual care, after 12 months of treatment, the
20


patients who received care in the collaborative care model, demonstrated significantly
better adherence to antidepressant treatment, decreased depressive symptoms,
remission of depression, improved quality of life, better physical functioning, self-
efficacy, and satisfaction with treatment at 18 and 24 month follow-ups. Even one
year after the IMPACT resources had been removed from the settings, intervention
patients still demonstrated a significant difference in depression scores (Hunkeler, et
al., 2006).
Cigrang, Dobmeyer, Becknell, Roa-Navarrete, and Yerian (2006) found that
patients having one to four appointments with a BHC located with-in a primary care
clinic demonstrate significant decreases in symptomatic distress. In this study,
patients who demonstrated more distress at the initial appointment required more
visits with a BHC to recover (Cigrang et al, 2006). This study examined the impact
of the integrated model on symptomatic distress, but did not investigate the impact on
patients functional capacity or the role of comorbidities in adherence to the model.
One goal of integrated primary care is to decrease the fragmentation between
primary care and mental/behavioral health providers by increasing the collaboration
between providers (Robinson & Reiter, 2007). Felker and colleagues (2004) reported
positive results from one collaborative care model very similar to the PCBH model in
a Veterans Affairs internal medicine primary care clinic. They created a
multidisciplinary primary care mental health team that was available in the primary
21


care setting to provide services to patients and to consult with the physicians for
urgent concerns regarding patients. The patients were seen by the primary care
mental health team for a variety of disorders for an average of 2.5 sessions. Referrals
to the primary care mental health team increased significantly while referrals to
specialty mental health were significantly reduced (Felker et al., 2004). These
findings are similar to those at the St. Louis Veterans Affairs Medical Center where
the PCBH model has been integrated (Robinson & Strosahl, 2009). This program has
seen a decrease of PCP referrals to specialty mental health by 48% and an increase in
access to mental/behavioral health services in primary care by 170% (Felker et al.,
2004). This provides further support that treating patients in integrated primary care
can increase access to care and improve collaboration among providers.
Brief Cognitive-behavioral treatments for insomnia delivered within the
PCBH model in a family medicine setting have also shown positive outcomes
(Goodie, Isler, Hunter, & Peterson, 2009). Patients who completed the study
intervention saw a BHC for three sessions that were 10-25 minutes in length over an
average of 36 days. Following treatment, the patients demonstrated a mean sleep
efficiency rating greater than 85% compared to 14% at baseline and significantly
fewer patients reported using medication to aid with sleep (Goodie et al., 2009). This
study is one of only a few that examines the effectiveness of an integrated primary
care model in a real world medical setting as opposed to a research setting. Patients
22


with comorbid psychological and medical diagnoses were included in this study and
still demonstrated improvements in the outcome measures (Goodie et al, 2009). This
study does not however, examine the effect of these comorbidites on patient
outcomes or on fidelity to the integrated model.
To date, data documenting the presenting symptoms and problems of primary
care patients is limited/scarce making it difficult to understand this population of
patients and impossible to develop effective integrated treatment methods (Levant,
2005). Patient populations with the greatest need and highest risk of chronic illness
and comorbid medical problems are often those who have the fewest resources.
These patient populations often do not have access to mental/behavioral health
professional and are often managed through Federally Qualified Health Centers
which have been active in integrating mental health into these primary care settings
(DeLeon, Kenkel, & Belar, 2007).
The current research on integrated care often fails to consider the additional
biopsychosocial factors and comorbid psychological and medical conditions that may
be contributing to the patients symptoms. Unfortunately, attention to comorbid
medical problems and key behavioral components in these populations may not be
well understood and hence, not adequately addressed. In order to fully understand
these populations we need to integrate biological, psychological, and socio-cultural
knowledge and gain a better understanding of the complexity of these patients (Peek,
23


2009) and how this affects the adherence to integrated primary care models. We must
understand all aspects of primary care patients to most effectively treat them
(DeLeon, et al., 2007; McDaniel, Belar, Schroeder, Hargrove, & Freeman, 2002) and
meet their needs with seamlessly integrated services (Peek, 2009).
Purpose and Hypotheses of the
Present Study
One goal of this line of research is to inform psychologists, primary care
physicians, and policy makers about characteristics of patients in primary care in
order to more effectively conceptualize and develop integrated care models of
treatment in primary care settings. Additionally, little is known about the effects of
psychological and physical comorbidities on integrated primary care (Butler, et al.,
2008). Therefore, the purpose of the present study is to examine the role of multiple
psychological and chronic illness comorbidities in Behavioral Health Provider
adherence to integrated primary care model metrics in federally qualified health
centers. To this end, the following hypotheses are presented:
24


Hypothesis 1. It is hypothesized that greater than or equal to 50% of patients
referred to a BHC will have at least one documented chronic medical illness.
Hypothesis 2. It is hypothesized that the number of comorbid psychological
diagnoses will be positively related to the length of the behavioral health session.
Hypothesis 3. It is hypothesized that the number of comorbid chronic illness
diagnoses will be positively related to the length of the behavioral health session.
Hypothesis 4. Success versus failure in adhering to the integrated model
length of session (< 30 minutes) parameter will be predicted by the number of
comorbid psychological diagnoses and number of chronic illness diagnoses when
controlling for the effects of the number of years of clinical training that the BHC has
received.
Hypothesis 5. Success versus failure in adhering to the integrated model
number of follow-up sessions (< 4) parameter will be predicted by the number of
comorbid psychological diagnoses and number of chronic illness diagnoses when
controlling for the effects of the number of years of clinical training that the BHC has
received.
25


CHAPTER II
METHOD
The proposed study is an analysis of archival data from a previous clinical
program evaluation at Spalding University in Louisville, KY. Since the original
study was of a clinical program and not a research project, approval from a human
subjects committee was not required. This study was therefore approved for
exemption by the Colorado Multiple Institutional Review Board (COMIRB Protocol
10-0727).
Participants
The archival data set is comprised of data recorded for 843 primary care
patient contacts over a 12-month period. Patients were referred by their Primary Care
Provider (PCP) to a Behavioral Health Consultant (BHC) in five urban, medically
underserved, economically disadvantaged primary health clinics in Louisville,
Kentucky during a twelve-month period of time. The age of the clinic patients ranged
from 5 to 82 years old, but for the purposes of this study, only the data for those 18
and over were analyzed. Therefore, there were 709 adult patients with whom the
BHCs had some type of initial contact. These contacts may have been either by
phone or face-to-face. The initial contacts with patients resulted in 468 face-to-face
26


sessions with a BHC. During an appointment with their PCP, the patients were
identified to have psychological or behavioral needs and were thus referred to the
BHCs. The BHCs included in this data set were six clinical psychology doctoral
students each with a different number of years of clinical training and experience.
Procedure
The patients were referred to the BHCs solely through the PCP (i.e., MD or
Nurse Practitioner) referral. Some were referred through a warm hand off which
occurred when the PCP asked a BHC for immediate assistance with a patient or
during a consultation between a BHC and the PCP. Other patients were referred
through a traditional referral method in which they scheduled an appointment with a
BHC for a later date. A referral was considered a follow up if the patient had already
seen the BHC for a previous appointment. If a BHC consulted with a PCP about the
care of a patient, this was considered a referral and will be referred to in this study as
a consultation with MD. Patients were referred for a variety of behavioral and
psychological problems. In this study, BHCs were instructed to target the presenting
behavioral or psychological concerns, but not the patients medical condition. The
BHCs met with the patients for an initial assessment or intervention and saw the
client for up to four appointments. Initial appointments/sessions were designed to
have a duration of 15-30 minutes in which the BHCs conducted a brief functional
27


analysis focusing on the referral problem. The BHCs provided the patients with the
appropriate behavioral intervention for the presenting problem. The length of time
the BHC spent directly with the patient was tracked in minutes, as well as the amount
of indirect time the BHC spent consulting with the PCP.
Measures
Psychological Diagnoses
Psychological working diagnoses for each patient were collected by the BHC
through review of the patients medical chart and from PCP and patient report. These
data points were collected at the time of the patient contact/visit.
Medical Diagnoses
For each patient, the BHC collected working diagnoses for medical conditions
by reviewing the patients medical chart and from PCP and patient report. These data
points were collected at the time of the patient contact/visit.
Length of Patient Session
The length of the patient session was tracked in minutes by each respective
BHC and subsequently converted to 15 min increments in accordance with the PCBH
model (Robinson & Reiter, 2007).
28


Years of Clinical Training
Years of clinical training was defined as the number of completed years of
clinical practicum training in a clinical psychology doctoral program.
Metrics of the Integrated Primary Care Model
Within the PCBH model, the BHC sees up to 10 to 15 patients per day in the
primary care setting for brief sessions lasting 15-30 minutes (Robinson & Reiter,
2007). The BHC may follow up with a patient but this is limited to one to four
sessions.
Data Analysis
The following analyses were conducted using SPSS for Windows, version
17.0. Participant demographics, as well as provider and system characteristics were
examined using descriptive statistics and frequencies. Hypothesis one was analyzed
using a frequency distribution. Hypothesis two was analyzed with a bivariate
correlation using the independent variable, total number of psychological diagnoses.
This variable was created by adding the total number of documented psychological
diagnoses for each patient that had a face-to-face session with a BHC. The dependent
variable was the length of the behavioral health session in minutes. Hypothesis three
was also analyzed using a bivariate correlation and the same dependent variable used
29


in hypothesis two. For the independent variable, total number of chronic illnesses, all
documented chronic illness diagnoses for each patient were added to create this
variable. The following diagnoses were considered chronic illnesses, pain, diabetes,
respiratory disorders, cardiac conditions, congestive heart failure, and hypertension.
To analyze hypothesis four, hierarchical logistic regression was used to
predict success/failure to adhere to the integrated primary care model parameter of the
session length being less than or equal to 30 minutes. The independent variable,
years of training, was measured by the total number of completed years of training
that the BHC had received. The dichotomous dependent variable was created by
coding the sessions that were less than or equal to 30 minutes with a 1 (adherence to
the model) and the sessions that were greater than 30 minutes as 0 (failure to adhere
to the model). The independent variables were the same as those used to test
hypotheses two and three.
Hypothesis five, prediction of success/failure to adhere to the model
parameter number of sessions with a BHC being less than or equal to four was also
analyzed with hierarchical logistic regression. The dichotomous dependent variable
was measured by using the total number of face-to-face sessions that a patient had
with a BHC. Those with less than or equal to four sessions were coded as 1
(adherence to the model) and those with greater than four session were coded with a 0
(failure to adhere to the model).
30


A hierarchical multiple regression was performed to further investigate
whether the number of patient sessions with a BHC would be predicted by the same
independent variables from the previous analysis. The continuous dependent
variable, total number of sessions, was measured by adding the number of face-to-
face sessions a patient had with a BHC. With an alpha level of .05, assumed power of
.80, and 3 predicting variables, regression analyses should be able to detect a small to
medium effect size (Miles & Shevlin, 2001).
31


CHAPTER III
RESULTS
Participant Sample
In this sample, the BHCs had initial contact with 709 adult patients, and had
468 face-to-face sessions with patients. Of the 709 initial patient contacts, the sample
was predominantly female (75.2%, n = 533) and Caucasian (78%, n = 533) with a
mean age of 42.6 years (SD = 13.74, range = 18-84). The patient characteristics are
presented in Table 3.1. The majority of the sample had at least one psychological
diagnosis (81%, n = 573), with the most common diagnoses being depression (47%, n
= 333) and anxiety (24.7%, n = 175). Additionally, 39% (n = 276) of the patients in
this sample had two or more comorbid psychological diagnoses. Eighty-one (11.4%)
of the patients had a sleep disorder, fifty-five (7.8%) were diagnosed with bipolar or
another mood disorder, and forty-five (6.3%) of the patients used tobacco or
cigarettes. See Table 3.2 for the complete list of the specific patient psychological
diagnoses. The medical diagnoses of patients were as follows: pain (24.7%, n =
175), diabetes (13.1%, n = 93), hypertension (10.7%, n = 76), heart disease (4%, n =
29), and obesity (7.3%, n = 52). Of those diagnosed with a chronic illness, 12.7% (n
= 90) had two or more chronic illnesses. See Table 3.3 for a complete list of the
medical diagnoses in this sample.
32


Table 3.1: Participant characteristics by gender and overall sample
Characteristic Male (n = 176) Female (n = 533) Overall Sample (A=709)
Age (years) 45.06 13.07 Range: 18-80 41.81 13.87 Range: 18-84 42.62 13.74 Range: 18-84
Race Caucasian African American Hispanic 75% (n = 132) 13.1% (n = 23) 2.3% (n = 4) 79% (n = 421) 14.1% (n = 75) 4.1% (n = 22) 78% in = 553) 13.8% in = 98) 3.7% in = 26)
# of Psych Diagnoses Zero One Two Three Four 19.9% (n = 35) 43.8% (n = 77) 29% (n = 51) 7.4% (n = 13) 0% (n = 0) 18.3% (n = 101) 41.3% (n = 220) 28% in = 151) 10.1% (n = 54) 1.3% \n = 7) 19.2% in = 136) 41.9% \n = 297) 28.5% in = 202) 9.5% in = 67) 1% \n = 1)
# of Chronic Illnesses Zero One Two Three Four 39.8% (n = 70) 41.5% (n = 73) 14.8% (n = 26) 3.4% (n = 6) .6% (n = 1) 55% (n = 293) 34.1% {n = 182) 10.7% in = 57) .2% in = 1) 0% in = 0) 51.2% in = 363) 36% in = 255) 11.7% in = 83) .1% in = 7) .1% \n = 1)
33


Table 3.2: Patient psychological diagnoses by gender and overall sample
Diagnosis Male (n = 176) Female (n = 533) Overall Sample (jV=709)
Depression 45.5% (n = 80) 47.5% (n = 253) 47% in = 333)
Bipolar and Other Mood Disorders 6.3% (n = 11) 8.3% (n = 44) 7.8% in = 55)
Anxiety 27.3% (n = 48) 26.8% (n = 143) 26.9% in = 191)
Panic 3.4% (n = 6) 7.9% {n = 42) 6.8% in = 48)
Sleep Disorder 10.8% (n = 19) 11.6% {n = 62) 11.4% (n = 81)
Stress 4% (n = 7) 5.3% {n = 28) 4.9% (n = 35)
Posttraumatic Stress Disorder 1.7% (n = 3) 2.3% in = 12) 2.1% in = 15)
Tobacco/ Cigarette 5.7% (n = 10) 6.7% (n = 35) 6 .3% in = 45)
Alcohol Abuse 6.3% (n = 11) 5.3% in = 28) 5.5% in = 39)
34


Table 3.3: Patient medical diagnoses by gender and overall sample
Diagnosis Male (n = 176) Female in = 533) Overall Sample (iV=709)
Respiratory 10.2% (n = 18) 7.3% in = 39) 8% in = 57)
Heart Disease 8.5% (n = 15) 2.6% in = 14) 4% in = 29)
Hypertension 13% (n = 23) 9.9% in = 53) 10.7% in = 76)
Congestive Heart Failure 2.3% (n = 4) 2.3% in = 12) 2.3% in = 16)
Diseases of the Digestive System 2.8% (n = 5) 4.7% in = 25) 4.2% in = 30)
Pain 29% (n = 51) 23.3% in = 124) 24.7% in = 175)
Head Pain 1.7% (n = 3) 2.4% in = 13) 2.3% in = 16)
Diabetes 2 0.5% (n = 36) 10.7% in = 57) 13.1% in = 93)
Obesity 5.1% (n = 9) 8.1% in = 43) 7.3% in = 52)
Hepatitis C 4% {n = 7) .1% in = 6) 1.8% in = 13)
Kidney Disease 1.1% in = 2) .56% in = 3) .71% in = 5)
35


Medical Provider Characteristics
The majority of the patient referrals that resulted in face-to-face contact with a
BHC came from MDs (73%, n = 344) and 41.3% (n = 142) of the MD referrals were
in the Portland clinic. Nurse practitioners and MDs referred patients through a warm
handoff at approximate rates (28.8%, n = 36; 28.7%, n = 98). Overall, the most
common type of referral was the traditional referral (37.2%, n = 175) followed by the
warm handoff (28.6%, n = 134). Of the referrals from nurse practitioners, 31.2% (n =
39) of them were a traditional referral. Most of the referrals from nurse practitioners
came from the Dixie clinic (44%, n = 55). Refer to tables 3.4 and 3.5 for a complete
list of the medical provider characteristics.
Table 3.4: Clinic location of referrals by medical provider
Clinic Nurse Practitioner (n = 122) MD (n = 333) Other ( = 4) Overall Sample iN = 459)
Portland 26.2% (n = 32) 42.3% (n = 141) 25% (n = 1) 37.9% in = 174)
Fairdale 1.6% (n = 2) 16.2% ( = 54) 0% (n = 0) 12.2% in = 56)
Broadway 15.6% (n = 19) 17.7% (n = 59) 0% (n = 0) 17% in = 78)
Iriquois 13.9% (n = 17) 19.5% (n = 65) 50% (n = 2) 18.3% in = 84)
Dixie 42.6% (n = 52) 4.2% (n = 14) 25% {n = 1) 14.6% in = 67)
36


Table 3.5: Type of referral by medical provider
Referral Type Nurse Practitioner (n = 117) MD (n = 332) Other ( = 4) Overall Sample (N = 453)
Collateral 11.9% (n = 14) 17.8% (n = 59) 25% (n = 1) 16.3% (n = 74)
Contact
Consult 1.7% (n = 2) 1.8% (n = 6) 0% (n = 0) 1.7% (n = 8)
with MD
Follow up 21.4% (n = 25) 11.7% (n = 39) 0% (n = 0) 14.1% (n = 64)
Referral 33.3% (n = 39) 39.2% (n = 130) 50% (n = 2) 37.7% (n = 171)
Warm 31.6% {n = 37) 29.5% (n = 98) 25% (n = 1) 30% (n = 136)
Handoff
Behavioral Health Consultant
Characteristics
Most of the BHCs saw patients at one primary clinic location and for only one
session (84.8%). Only two of the BHCs saw patients in two different clinics. Of the
465 patients in this sample, 347 (84.8%) met with a BHC for one session, 43 (10.5%)
saw a BHC for two appointments, 17 (4.2%) had three appointments, and two (.5%)
had four sessions. See tables 3.6, 3.7, and 3.8 for additional characteristics by BHC.
37


Table 3.6: Clinic location of referrals by BHC
BHC Number (Years of Training)
Clinic 1(0) (n = 64) 2(2) 3(2) 4(2) 5(3) (n = 58) (n = 87) (n = 78) (n = 94) Mean Years Training = 2 6(3) (n = 84) Overall Sample (N= 465)
Portland 100% 3.4% 23% 0% 100% 0% 38.7%
(n = 64) (n = 2) (n = 20) (n = 0) (n = 94) (n = 0) (n = 180)
Fairdale 0% 96.6% 0% 0% 0% 0% 12%
( = 0) (n = 56) (n = 0) ( = 0) ( = 0) ( = 0) (n = 56)
Broadway 0% 0% 0% 100% 0% 0% 16.8%
(n = 0) (n = 0) (n = 0) ( = 78) (n = 0) ( = 0) ( = 78)
Iriquois 0% 0% 0% 0% 0% 100% 18.1%
(n = 0) 0 = 0) (n = 0) (n = 0) ( = 0) (n = 84) (n = 84)
Dixie 0% 0% 77% 0% 0% 0% 14.4%
(n = 0) 0 = 0) (n = 67) (n = 0) (n = 0) ( = 0) (n = 67)
38


Table 3.7: Type ofpatient referral by BHC
BHC Number (Years of Training)
Referral 1(0) 2(2) 3(2) 4(2) 5(3) 6 (3) Overall Sample
Type (n = 64) {n = 58) (n = 83) ( = 76) (n 94) (n = 84) (N = 459)
Mean Years Training = 2
Collateral 17.2% 5.2% 3.6% 1.3% 51.1% 9.5% 16.1%
Contact ( = 11) (n = 3) (n = 3) (n= 1) (n =48) ( = 8) ( = 74)
Consult 1.6% 0% 3.6% 0% 4.3% 0% 1.7%
with MD (n= 1) (n = 0) (n 3) (n = 0) ( = 4) (n = 0) (n = 8)
Follow Up 2.5% 1.7% 19.3% 25% 10.6% 11.9% 13.9%
( = 8) (n = 1) ( = 16) (n = 19) (n = 10) (n = 19) (n = 64)
Referral 9.4% 72.4% 28.9% 52.6% 20.2% 47.6% 37.3%
(n = 6) In = 42) (n =24) (n = 40) in = 19) (n =40) ( =171)
Warm 59.4% 20.7% 44.6% 21.1% 13.8% 31% 30.9%
Handoff (n = 38) (n = 12) ( = 37) (n 16) (n = 13) (n = 26) (n = 142)
39


Table 3.8: Number of patient sessions by BHC
BHC Number (Years of Training)
Number
of
Sessions
1(0) 2(2) 3(2) 4(2) 5(3) 6(3) Overall Sample
(n = 57) (n = 53) (/i = 71) (n = 67) (n = 86) (n = 75) (# = 409)
Mean Years Training = 2
One 89.5% 90.6% 64.8% 83.6% 91.9% 89.3% 84.8%
(n = 51) (n = 48) (n = 46) (n = 56) Si' ii (n = 67) (n = 347)
Two 5.3% 7.5% 23.9% 11.9% 4.7% 9.3% 10.5%
(n = 3) (n = 4) (n = 17) (n = 8) (n = 4) (n = 7) {n = 42)
Three 5.3% 1.9% 9.9% 4.5% 2.3% 1.3% 4.2%
(n = 3) (n = 1) (n ~ 7) (n = 3) ( = 2) ( = 1) (n = 17)
Four 0% 0% 1.4% 0% 1.2% 0% .5%
(n = 57) (n = 57) ( = 57) ( = 57) (n = 57) (n = 57) ir> ll
40


System Characteristics
The majority of patient contacts occurred in the Portland clinic (37.8%, n =
181) with the fewest occurring at the Fairdale clinic (11.9%, n =57). Of the referrals
in the Fairdale clinic, most of them came from MDs (96.5%, n = 55). Referrals from
MDs accounted for 72.5% (n = 333) of all of the referrals but in the Dixie clinic, the
majority of the patient referrals were from nurse practitioners (77.6%, n = 52).
Warm handoffs occurred with the greatest frequency in the Dixie clinic (38.1, n =
24). Refer to Tables 3.9 and 3.10 for a complete list of the system characteristics.
Table 3.9: Referral source by clinic location
Referral Source Portland (n = 174) Fairdale in = 56) Broadway in = 78) Iriquois in = 84) Dixie in = 67) Overall Sample (N = 459)
NP 18.4% {n = 32) 3.6% in = 2) 24.4% (n = 19) 20.2% in = 17) 77.6% in = 52) 26.6% in = 122)
MD 81% in = 141) 96.4% in = 54) 75.6% (n = 59) 77.4% in = 65) 20.9% in = 14) 72.5% in = 333)
Other .5% in = 1) 0% (n o) 0% (/I = 0) 2.4% in =2) 1.5% (n = l) .87% (n = 4)
41


Table 3.10: Type of referral by clinic location
Referral Type Portland (n = 180) Fairdale (n = 56) Broadway in = 73) Iriquois in = 84) Dixie in = 63) Overall Sample iN = 456)
Collateral Contact 35.6% (n 64) 1.8% (n = 1) 1.4% in = 1) 9.5% in = 8) 0% (n = 0) 16.2% in = 74)
Consult with MD 2.8% ( = 5) 0% (n = 0) 0% ( =0) 0% ( = 0) 4.8% in = 3) 1.8% in = 8)
Follow Up 10.6% (n = 19) 1.8% (n = 1) 24.7% in = 18) 11.9% in = 10) 23.8% in = 15) 13.8% in = 63)
Referral 15.6% (n = 28) 75% (n = 42) 53.4% in = 39) 47.6% in = 40) 33.3% in = 21) 37.3% in = 170)
Warm Handoff 35.6% (n = 64) 21.4% (n = 12) 20.5% in = 15) 31% (n = 26) 38.1% in = 24) 30.9% in = 140)
Hypothesis One
The first hypothesis for the present study stated that greater than or equal to
50% of the patients referred to a BHC would have at least one documented chronic
illness. A frequency distribution was conducted in order to test this hypothesis and
48.8% (n = 346) of the sample had at least one chronic illness.
42


Hypothesis Two
The second hypothesis for this study stated that the number of comorbid
psychological diagnoses would be positively related to the length of the behavioral
health session in minutes. In order to test this hypothesis, a bivariate correlation was
conducted which indicated that there was a significant relationship between the
variables, r = .12, p (one-tailed) < .01, N = 459.
Hypothesis Three
The third hypothesis stated that the number of comorbid chronic illnesses
would be positively related to the length of the behavioral health session in minutes.
A bivariate correlation indicated no significant relationship between these variables,
r = .03, p (one-tailed) = .26, N = 459.
Hypothesis Four
The fourth hypothesis for the present study stated that success versus failure in
adhering to the integrated model length of session (< 30 minutes) parameter will be
predicted by the number of comorbid psychological diagnoses and number of
comorbid chronic illness diagnoses when controlling for the effects of the number of
years of clinical training that the BHC has received. To analyze this dichotomous
dependent variable (failure to adhere to the model = 0, adherence to the model = 1),
43


hierarchical logistic regression was conducted. When the variable years of training
was forced into the equation first, it significantly contributed to the prediction of
adherence to the integrated model, (x2 = 28.52, df= 1, N = 468,p < .001). When all
predictor variables were entered, the overall logistic regression model was significant
(X2 = 34.40, df= 3, N = 468, p < .001) and accounted for 7.1% of the variance (Cox &
Snell, 1989). Years of training continued to contribute significantly to the equation
(OR = 1.8, p < .001) as well as the total number of psychological diagnoses (OR =
.78, p < .05). Refer to table 3.11 for the odds ratios for each variable.
Table 3.11: Logistic regression analysis for variables predicting success/failure to
adhere to the model length of session parameter
Variable P SE Odds Ratio P 95% Cl
Years of Training .58 .11 1.78 .000 [1.44, 2.20]
Chronic Illness Total -.14 .14 .87 .338 [.660, 1.15]
Psych Total -.25 .11 .78 .028 [.621, .97]
Constant .71 .32 2.02 .029
44


Hypothesis Five
The fifth hypothesis stated that adherence to the integrated model number of
follow-up sessions parameter (< 4 sessions) will be predicted by the number of
comorbid psychological diagnoses and number of chronic illness diagnoses when
controlling for the effects of the number of years of clinical training that the BHC has
received. However, a frequency distribution revealed that there were no cases that
failed to adhere to the model, therefore; the logistic regression was not conducted.
Supplementary Analysis
A supplementary analysis was used to further analyze hypothesis five.
Hierarchical multiple regression was conducted to investigate whether the total
number of psychological diagnoses and total number of chronic illnesses predicted
the actual number of face-to-face appointments a patient had with a BHC while
controlling for the BHC completed number of years of training. The means, standard
deviations, and intercorrelations can be found in Table 3.12. When all of the
variables were entered into the model, they did not significantly predict the number of
follow up visits a patient would have with a BHC, F(3, 405) = 1.64,/? = .180. See
Table 3.13 for the results of the multiple regression analysis.
45


Table 3.12: Means, standard deviations, and intercorrelations for number of follow
up appointments and predictor variables (N = 409)
Variable M SD 1 2 3
Number of visits 1.20 .54 -.04 .102* .002
Predictor variable
1. Number of years training 2.11 .97 - .22 -.02
2. Psych total 1.38 .94 - -.01
3. Chronic illness total .78 .77 -
*p < .05
46


Table 3.13: Multiple regression analysis for variables predicting number of patient
follow up visits with a BHC
Variable B SE P P 95% Cl
Model 1
Constant 1.24 .06 .00 [1.12, 1.37]
Completed # -.02 .03 -.04 .46 [-.10, .03]
Yrs of Training Model 2
Constant 1.17 .08 .00 [1.01, 1.32]
Completed # -.02 .03 -.04 .44 [-.07, .03]
Yrs of Training
Psych Total .06 .03 .10 .04 [.003, .112]
Chronic Illness Total .001 .03 .002 .97 [-.07, .07]
Note, R2 = .001 for Step 1; Ai?2 = .011, p = .113 for Step 2
47


CHAPTER IV
DISCUSSION
The overarching purpose of this study was to examine the role of multiple
psychological and chronic illness comorbidities in Behavioral Health Consultant
adherence to the PCBH model metrics in federally qualified health centers.
Integrated primary care models are being implemented with greater frequency and
have shown positive outcomes on patient care (Bryan, et al., 2009; Butler, et al.,
2008; Cummings, Cummings, & ODonohue, 2009). However, most of the research
on integrated care has focused on specific diagnoses such as depression and anxiety
and little is known about how patient comorbidites, both psychological and physical,
effect integrated care (Butler, et al., 2008).
Sample Characteristics
The current sample was largely Caucasian and female ranging in age from 18
to 84 with a mean age of 42.62 years. The majority of the sample had at least one
psychological diagnosis and 28.5% of the sample had two comorbid psychological
diagnoses while 9.5% had three. The most prevalent psychological diagnosis was
depression followed by anxiety. Nearly half of the sample had at least one chronic
illness while 12.7% had two or more comorbid chronic illnesses.
48


The 47% rate of depression in this sample was much higher than what has
been reported by other studies of patients in medical settings or the general adult
population (Dobmeyer, Rowan, Etherage, & Wilson, 2003; Katon, et al., 1990;
Kessler et al., 2003; Kessler et al., 2005; Olfson et al., 2000). Compared to one
particular primary care sample of patients at an Air Force base clinic, the current
sample was similar as far as general demographic characteristics, but did differ
slightly in some areas (Bryan, et al., 2009). Specifically, the present study sample
was slightly older, had a larger majority of female patients, had a higher prevalence of
depression, and a lower prevalence of sleep disorders. However, the rate of anxiety
symptoms was very similar between the two samples, as well as the prevalence of
comorbid psychological conditions (Bryan et al., 2009). Compared to the general
adult population, this sample had a much higher overall prevalence of psychological
and medical conditions and an especially higher rate of depression (Kessler, et al.,
2003; Kessler et al., 2005; Wang, et al., 2005). Contrary to previous study findings
(Kessler et al., 2005), depression was more prevalent than reported levels of anxiety
in this sample. This finding may be due to system-wide efforts to more effectively
identify depression coupled with the actual high rates of depression concomitant with
many chronic illnesses (Chapman, et al., 2005; Katon, 2003; Katon, et al., 2007;
Rudisch & Nemeroff, 2003).
49


In the current study sample, the number of patient sessions with a BHC
followed a pattern in which the majority of patients were seen for only one session,
which is similar to that in another study of patients in an integrated primary care
setting (Bryan, et al., 2009). While the number of patients with fewer sessions with a
BHC occurred with even greater frequency in the current sample, the overall
distribution of sessions was similar with fewer patients having two or more sessions.
This distribution is consistent with the PCBH model where patients are only seen for
1-4 sessions with the greater number of sessions occurring the least (Robinson &
Reiter, 2007).
Previous research has shown that patients who are referred to specialty mental
health treatment by their healthcare provider, are more likely to follow up with
treatment (Ledoux, et al., 2009). In the present study sample, there were greater
numbers of patient referrals from MDs than from nurse practitioners and very few
referrals from other providers. Nurse practitioners and MDs generally used the same
type of traditional referral modality with only slight variation between clinic
locations. One goal of the PCBH model is to increase access to care for patients in
primary care; therefore, it is suggested that there should at least be an equal number
of warm-hand offs as traditional referrals in an integrated setting to ensure that the
PCP is introducing BHC services as a regular part of patient care (Robinson & Reiter,
2007; Strosahl & Robinson, 2008). In this sample, the healthcare providers referred
50


patients to the BHCs through a traditional method more frequently, but did utilize
warm handoffs for approximately one-third (30%) of the referrals. In this study, it is
unclear whether the source or type of referral was related to whether the patient
subsequently received follow-up treatment sessions.
Hypothesis One
It was hypothesized that at least 50% of the current sample would have a
chronic illness. Although the prevalence of chronic illnesses in this sample was
48.8%, the result of the analysis did not support the a priori hypothesis. This is still a
significant number of patients with medical conditions that can greatly impact overall
functioning and wellbeing. Even if the BHC is specifically targeting the mental
health concerns of the patient within a session, the patients chronic illnesses may
contribute appreciably to the patients symptoms and course of treatment (Hoffman,
et al., 1996; Katon, 2003; Katon, et al., 2007; Lett, et al., 2004). It follows, then, that
BHCs who have additional specialized health psychology training may be better able
to identify and conceptualize the role of behavioral factors related to chronic illnesses
and, hence, develop a strong biopsychosocial approach to treatment (Belar, 2008b;
Newton, Woodruff-Borden, & Stetson, 2006; ODonohue, Cummings, & Cummings,
2009). Furthermore, this high rate of chronic illnesses coupled with the high
prevalence of psychological comorbidities suggest that the complex needs of this
51


primary care sample require the high level of integration espoused by the fourth
quadrant of the Four Quadrant Clinical Integration Model (Mauer, 2008). This
provides support for the decision to implement a closely integrated model such as the
PCBH model in these particular primary care settings.
Hypothesis Two
It was hypothesized that the number of comorbid psychological diagnoses
would be positively related to the length of the behavioral health session. This
hypothesis was supported by the present study results. In this sample, over 80% of
the patients had at least one psychological diagnosis and over one-third had two or
more documented psychological diagnoses, indicating a high level of mental health
need. These results indicate that as patients present with increased psychological
complexities, it may be more difficult for the BHC to work within the integrated
model. It has been noted that one of the biggest struggles for some BHCs working in
an integrated primary care model is time management (Dobmeyer, et al., 2003).
However, even when patients present with a high level of complex needs, the BHC
must limit the problem focus in order to maintain 15-30 minute sessions (Strosahl,
2005) and achieve the mission of the PCBH model of providing population-based
care (Robinson & Reiter, 2007). Through additional training, BHCs can further
52


develop the skills needed to effectively work within this model and overcome this
obstacle (Dobmeyer et al., 2003; Robinson & Reiter, 2007).
Hypothesis Three
Hypothesis three stated that the number of comorbid chronic illnesses would
be positively related to the length of the behavioral health session. This hypothesis
was not supported by the results of this study as it was found that the number of
comorbid chronic illnesses was not significantly related to the length of the
behavioral health session. While nearly half of the sample had a chronic illness, only
12.8% had comorbid (ie., >2) chronic illnesses which may indicate that it may be
useful to consider patients psychological symptoms as being more complex than
their medical conditions in terms of time spent with the BHC. Conversely, another
possible explanation for this is that the BHC may not have fully recorded all
documented comorbid chronic illnesses reported in the medical chart since the
clinical focus of the sessions was on the presenting mental health issues. Therefore,
the number of chronic illness diagnoses in our sample may have been underreported
similar to other studies of integrated primary care models (Bryan, et al., 2009). The
chronic illnesses may also not be the primary concern to the patient and/or neither the
patient nor the referring PCP may view the BHC as a professional to help with the
behavioral components of their medical diagnoses.
53


Hypothesis Four
In this sample, behavioral health provider adherence to the integrated primary
care model length of session metric was predicted by the number of psychological
diagnoses, as well as the number of years of training the BHC had completed. It
appears that patient presentations with comorbid psychological diagnoses may be
more complex and may require highly trained BHCs to stay within the model
parameter. Surprisingly the number of chronic illnesses was not a significant
predictor in this model. One possible explanation for this it that in this integrated
setting, patients were mostly referred for specific psychological concerns instead of
co-occurring medical conditions. Therefore there was a stronger focus on comorbid
psychological diagnoses in the session.
It is unclear from this analysis what particular aspects of a BHCs training are
important to work within this model. However, many proponents of integrated
primary care assert that integrated care is not simply taking traditionally trained
mental health specialists and placing them in primary care settings (Bluestein &
Cubic, 2009; Robinson & Reiter, 2007; Strosahl, 2001). As mentioned previously,
BHCs in working in this model must have specialized training in health psychology
to better understand the complex needs of primary care patients (Belar, 2008b;
Bluestein & Cubic, 2009). Furthermore, one goal of the PCBH model is to achieve
population based care in order to have a greater impact on the whole population
54


instead of just on a few patients (Robinson & Reiter, 2007). To successfully
accomplish this, a BHC must focus on providing brief effective treatment to many
patients instead of intensive care to a few (Strosahl, 2005). Therefore, it may be
necessary for BHCs working in the PCBH model to have specific training in
population-based care approaches in order to adhere to this model metric more
closely (Robinson & Reiter, 2007).
Hypothesis Five
Finally, it was hypothesized that the number of patient sessions with a BHC
would be predicted by the number of psychological and chronic illness diagnoses
when controlling for the number of years of training the BHC had received. Contrary
to what was hypothesized, the number of patient sessions with a BHC was not
predicted by the number of psychological and chronic illness diagnoses. In this
sample, there were only two patients that were seen for four sessions and none that
attended more than four sessions. Similar to other studies of this model (Bryan, et al.,
2009), the BHCs in this sample were able to demonstrate model fidelity in relation to
the number of sessions with a patient (Robinson & Reiter, 2007). In the present
study/program evaluation, the BHCs had substantial supervision and oversight with
respect to model fidelity. Certainly, this may have contributed to this finding as well.
55


The current study supports the importance of integrated primary care models
to help improve access to appropriate care for a sample of patients with complex
psychological and medical needs. The high prevalence of psychological comorbidites
and chronic illnesses in this sample bolster the claim that primary care is the de facto
mental health system (Kessler & Stafford, 2008b; Regier, et al., 1993). Additionally,
this study highlights the need for BHCs to have specialty training in order to
effectively work within integrated primary care models. Often, psychologists in
primary care settings have training in traditional mental health service delivery which
is necessary but not wholly sufficient for treating patients needs in primary care
settings (Bluestein & Cubic, 2009; McDaniel, et al., 2002; Robinson & Strosahl,
2009). When traditionally trained psychologists are placed in primary care settings
without developing the appropriate skills necessary to work in these models of service
delivery, this can create formidable barriers in implementing integrated models of
care (Dobmeyer, et al., 2003; ODonohue, Cummings, & Cummings, 2009).
Core areas of knowledge and particular skills have been identified for those
psychologists to work effectively within integrated models (Bluestein & Cubic, 2009;
Dobmeyer, et al., 2003; McDaniel, et al., 2002; Strosahl, 2005). First, a psychologist
must have strong broad and general training as a psychologist in order to be prepared
to treat the wide range of patient complexities across the life span that are
encountered in primary care settings (Dobmeyer, et al., 2003). Ideally, psychologists
56


should also have specialty training in health psychology which includes biological,
cultural, affective and cognitive factors that may affect a patients health (Bluestein &
Cubic, 2009; McDaniel, et al., 2002). Since chronic illnesses and many medical
diagnoses also have behavioral components, BHCs must be able to conceptualize and
treat comorbid medical and psychological presentations. It has been proposed that
traditional mental health training that does not include training in health psychology
is insufficient for those working in medical settings (Belar, 2008a, 2008b; Dobmeyer,
et al., 2003; McDaniel, et al., 2002). Finally, it is essential that primary care
psychologists develop knowledge, skills and attitudes that are consistent with the
primary care model of service delivery and the prevalent culture within those clinic
settings (Bluestein & Cubic, 2009).
One additional competency that has been proposed is the ability to work
effectively in an interdisciplinary setting such as primary care (Dobmeyer, et al.,
2003). The development of key interdisciplinary team skills is critical in integrated
primary care and fosters the coordination and communication between providers that
is a key component for successful integration (Bluestein & Cubic, 2009). As the
healthcare system moves away from the isolated treatment models that psychologists
have been trained in and moves toward more collaboration among treatment
providers, psychologists must receive training in medical settings to gain experience
in multidisciplinary team settings (McDaniel, et al., 2002). Community health
57


centers (CHCs) provide numerous training opportunities for psychologists who want
to work as a health psychologist (DeLeon, et al., 2007). In these settings, trainees
gain a better understanding of the medical field and other professions while learning
how to effectively communicate with professionals from other disciplines and work
within an integrated primary care model.
Limitations
As with any clinically based data collection, there are a number of limitations
to the study. The data collected to assess psychological and medical diagnoses was
collected in two ways. First, when available, the medical chart was reviewed and key
data points were recorded regarding psychological and medical comorbidities.
Medical chart availability was variable and therefore, patient self-report was often
utilized. Therefore, given the time and chart access limitations, diagnostic
information about the patients was not collected in an entirely systematic fashion.
Additionally, in this particular setting, the BHCs were instructed to target only the
patients presenting psychological concerns in the session. Therefore, there was less
focus on medical diagnoses which could have resulted in underreporting of chronic
illnesses and medical diagnoses since these were not the focus of the visit.
Furthermore, the sample characteristics and prevalence of psychological and medical
58


diagnoses cannot be generalized to the primary care population since this information
was only collected on those patients who were referred to a BHC.
The BHCs recording information did so somewhat retrospectively when
compiling notes and patient information at the end of their clinic day. Therefore,
some data points may have been missed. Omissions of this nature are typically
considered normally distributed error. Such omissions may also result in a
conservative estimate of comorbid diagnoses. However, time spent with patients is
considered to be a fairly accurate data point as it was a required notation by each
BHC.
The primary care model of service delivery implemented in this setting is a
brief treatment model based on the PCBH model (Robinson & Reiter, 2007; Strosahl,
1998) that does not allow the BHC to spend a lot of time (e.g., 15-30 minutes) with a
patient. Therefore, BHCs may have truncated visits in an effort to adhere to tenets of
the model. Nonetheless, questions regarding factors related to model adherence
remain empirical and are addressed hypothesis four and five in the present study.
Future Directions
Future research should focus on accurately documenting all psychological and
medical conditions of patients who are referred to BHCs in an integrated primary care
setting. Accurate documentation of the presenting problems of these patients may
59


help to fully understand how diagnoses and comorbid conditions may affect
adherence to an integrated model. Future research might also investigate whether
there is an even greater likelihood of patients receiving mental health treatment when
a healthcare provider in an integrated setting refers them. Additionally, researchers
may want to consider whether the type of medical provider predicts the patients
compliance with treatment recommendations. Finally, future research should fully
evaluate clinical and organizational outcomes in settings where the PCBH model is
implemented. This research is important to inform psychology training programs and
will hopefully emphasize the importance of increased training in health psychology as
well as integrated primary care models of care.
60


REFERENCES
Anderson, G., & Horvath, J. (2004). The growing burden of chronic disease in
America. Public Health Reports, 119(3), 263-270.
Belar, C. (2008a). Changing educational needs of psychologists: do we need more
medical knowledge, basic science and more psychological science? Journal of
Clinical Psychology in Medical Settings, 75(1), 12-17.
Belar, C. (2008b). Clinical health psychology: A health care specialty in
professional psychology. Professional Psychology Research and Practice,
39(2), 229-233.
Berenson, R., & Horvath, J. (2003). Confronting the barriers to chronic care
management in Medicare. Health Affairs, W37-W53.
Blount, A. (2003). Integrated primary care: Organizing the evidence. Families,
Systems, & and Health, 21, 121-133.
Blount, A., Schoenbaum, M., Kathol, R., Rollman, B., Thomas, M., O Donohue, W.,
& Peek, C. (2007). The economics of behavioral health services in medical
settings: A summary of the evidence. Professional Psychology Research and
Practice, 38(3), 290-297.
61


Bluestein, D., & Cubic, B. (2009). Psychologists and primary care physicians: a
training model for creating collaborative relationships. Journal of Clinical
Psychology in Medical Settings, 16(1), 101-112.
Bryan, C., Morrow, C., & Appolonio, K. (2009). Impact of behavioral health
consultant interventions on patient symptoms and functioning in an integrated
family medicine clinic. Journal of Clinical Psychology, 65(3), 281-293.
Butler, M., Kane, R., McAlpine, D., Kathol, R., Fu, S., Hagedorn, H., & Wilt, T.
(2008). Integration of mental health/substance abuse and primary care. (Rep.
No. 173). Prepared by the Minnesota Evidence-Based Practice Center
(Contract number 290-02-0009). AHRQ Publication No. 09-E003. Rockville,
MD: Agency for Healthcare Research and Quality.
Callaghan, G., Gregg, J., Ortega, E., & Berlin, K. (2005). Psychosocial interventions
with type 1 and 2 diabetes patients. In W. ODonohue, M. Byrd, N.
Cummings, & D. Henderson (Eds.), Behavioral integrative care: treatments
that work in the primary care setting (pp. 329-346). New York, NY: Brunner-
Routledge.
Chapman, D., Perry, G., & Strine, T. (2005). The vital link between chronic disease
and depressive disorders. Preventing Chronic Disease, 2(1), A14.
62


Ciechanowski, P., Katon, W., & Russo, J. (2000). Depression and diabetes: impact of
depressive symptoms on adherence, function, and costs. Archives of Internal
Medicine, 160(21), 3278-3285.
Ciechanowski, P., Katon, W., Russo, J., & Hirsch, I. (2003). The relationship of
depressive symptoms to symptom reporting, self-care and glucose control in
diabetes. General Hospital Psychiatry, 25(4), 246-252.
Cigrang, J., Dobmeyer, A., Becknell, M., Roa-Navarrete, R., & Yerian, S. (2006).
Evaluation of a collaborative mental health program in primary care: effects
on patient distress and health care utilization. Primary Care and Community
Psychiatry, 11(3), 121-127.
Collins, C., Hewson, D., Munger, R., & Wade, T. (2010, May). Evolving models of
behavioral health integration in primary care. New York, NY: Retrieved
from the Milbank Memorial Fund website: http://www.milbank.org/reports/
10430EvolvingCare/10430EvolvingCare.html
Cox, D. R. & Snell, D. J. (1989). The analysis of binary data (2nd edition). London:
Chapman and Hall.
Cummings, N., Cummings, J., & ODonohue, W. (2009). We are not a healthcare
business: our inadvertent vow of poverty. Journal of Contemporary
Psychotherapy, 39(1), 7-15.
63


DeLeon, P., Kenkel, M., & Belar, C. (2007, May). Community health centers can
reduce health disparities and train the next generation of psychologists.
Monitor on Psychology, 38(5). Retrieved from http://www.apa.org/monitor/
Dobmeyer, A., Rowan, A., Etherage, J., & Wilson, R. (2003). Training psychology
interns in primary behavioral health care. Professional Psychology: Research
and Practice, 34(6), 586-594.
Engel, G. L. (1977). The need for a new medical model: A challenge for biomedicine.
Science, 796(4286), 129-196.
Felker, B., Barnes, R., Greenberg, D., Chaney, E., Shores, M., Gillespie-Gateley, L.,
... Morton, C. (2004). Preliminary outcomes from an integrated mental health
primary care team. Psychiatric Services, 55(4), 442-444.
Fisher, E., Glasgow, R., Hayman, L., Orleans, C., Ponkshe, P., & Spring, B. (2008,
October). The role of behavior & environment in health, health care and
prevention. A Statement of the Society of Behavioral Medicine. Retrieved
from https://www.sbm.org/policy/role_behavior.pdf.
Foreyt, J., & Goodrick, G. (1993). Evidence for success of behavior modification in
weight loss and control. Annals of Internal Medicine, 119(1 Part 2), 698-701.
Gatchel, R., & Oordt, M. (2003). Clinical health psychology and primary care:
Practical advice and clinical guidance for successful collaboration.
Washington, DC: American Psychological Association.
64


Goldman, W., McCulloch, J., & Cuffel, B. (2003). A four-year study of enhancing
outpatient psychotherapy in managed care. Psychiatric Services, 54(1), 41-49.
Goodie, J., Isler, W., Hunter, C., & Peterson, A. (2009). Using behavioral health
consultants to treat insomnia in primary care: a clinical case series. Journal of
Clinical Psychology, 65(3), 294-304.
Gray, G., Brody, D., & Johnson, D. (2005). The evolution of behavioral primary care.
Professional Psychology: Research and Practice, 36(2), 123-129.
Gregg, J., Callaghan, G., Hayes, S., & Glenn-Lawson, J. (2007). Improving diabetes
self-management through acceptance, mindfulness, and values: A randomized
controlled trial. Journal of Consulting and Clinical Psychology, 75(2), 336-
343.
Guck, T., Guck, A., Brack, A., & Frey, D. (2007). No-show rates in partially
integrated models of behavioral health care in a primary care setting. Families
Systems and Health, 25(2), 137-146.
Hedrick, S., Chaney, E., Felker, B., Liu, C., Hasenberg, N., Heagerty, P., ... Katon,
W. (2003). Effectiveness of collaborative care depression treatment in
veterans' affairs primary care. Journal of General Internal Medicine, 18(1), 9-
16.
65


Hoffman, C., Rice, D., & Sung, H. (1996). Persons with chronic conditions: their
prevalence and costs. The Journal of the American Medical Association,
276(18), 1473-1479.
Hunkeler, E., Katon, W., Tang, L., Williams, J., Kroenke, K., Lin, E.,... Unutzer, J.
(2006). Long term outcomes from the IMPACT randomized trial for
depressed elderly patients in primary care. British Medical Journal,
332(7536), 259-262.
Katon, W. (2003). Clinical and health services relationships between major
depression, depressive symptoms, and general medical illness. Biological
Psychiatry, 54(3), 216-226.
Katon, W., & Ciechanowski, P. (2002). Impact of major depression on chronic
medical illness. Journal of Psychosomatic Research, 53(4), 859-863.
Katon, W., Lin, E., & Kroenke, K. (2007). The association of depression and anxiety
with medical symptom burden in patients with chronic medical illness.
General Hospital Psychiatry, 29(2), 147-155.
Katon, W., Robinson, P., Von Korff, M., Lin, E., Bush, T., Ludman, E., ...Walker,
E. (1996). A multifaceted intervention to improve treatment of depression in
primary carc. Archives of General Psychiatry, 53(10), 924-932.
66


Katon, W., Roy-Byrne, P., Russo, J., & Cowley, D. (2002). Cost-effectiveness and
cost offset of a collaborative care intervention for primary care patients with
panic disorder. Archives of General Psychiatry, 59(12), 1098-1104.
Katon, W., Russo, J., Heckbert, S., Lin, E., Ciechanowski, P., Ludman, E.,...Von
Korff, M. (2010). The relationship between changes in depression symptoms
and changes in health risk behaviors in patients with diabetes. International
Journal of Geriatric Psychiatry, 25(5), 466-475.
Katon, W., Von Korff, M., Lin, E., Lipscomb, P., Russo, J., Wagner, E., & Polk, E.
(1990). Distressed high utilizers of medical care: DSM-III-R diagnoses and
treatment needs. General Hospital Psychiatry, 12(6), 355-362.
Kessler, R., Berglund, P., Dernier, O., Jin, R., Koretz, D., Merikangas, K.,...Wang, P.
(2003). The epidemiology of major depressive disorder: results from the
National Comorbidity Survey Replication (NCS-R). The Journal of the
American Medical Association, 289(23), 3095-3105.
Kessler, R., Chiu, W., Dernier, O., & Walters, E. (2005). Prevalence, severity, and
comorbidity of 12-month DSM-IV disorders in the national comorbidity
survey replication. Archives of General Psychiatry, 62(6), 617-627.
Kessler, R., Dernier, O., Frank, R., Olfson, M., Pincus, H., Walters, E., et al. (2005).
Prevalence and treatment of mental disorders, 1990 to 2003. New England
Journal of Medicine, 352(24), 2515-2523.
67


Kessler, R., & Stafford, D. (2008a). Collaborative medicine case studies: Evidence in
practice. New York, NY: Springer.
Kessler, R., & Stafford, D. (2008b). Primary care is the de facto mental health
system. In R. Kessler, & D. Stafford (Eds.), Collaborative medicine case
studies: Evidence in practice (pp. 9-21). New York, NY: Springer.
Kroenke, K., & Mangelsdorff, A. (1989). Common symptoms in ambulatory care:
incidence, evaluation, therapy, and outcome. The American Journal of
Medicine, 86(3), 262-266.
Kroenke, K., Spitzer, R., Williams, J., Monahan, P., & Lowe, B. (2007). Anxiety
disorders in primary care: prevalence, impairment, comorbidity, and detection.
Annals of Internal Medicine, 146(5), 317-327.
Ledoux, T., Barnett, M., Garcini, L., & Baker, J. (2009). Predictors of Recent Mental
Health Service Use in a Medical Population: Implications for Integrated Care.
Journal of Clinical Psychology in Medical Settings, 16(4), 304-310.
Lett, H., Blumenthal, J., Babyak, M., Sherwood, A., Strauman, T., Robins, C.,
.. .Newman, M. (2004). Depression as a risk factor for coronary artery disease:
evidence, mechanisms, and treatment. Psychosomatic Medicine, 66(3), 305-
315.
68


Levant, R. (2005). Graduate education in clinical psychology for the twenty-first
century: Educating psychological health care providers. Journal of Clinical
Psychology, 61(9), 1087-1090.
Mauer, B. (2006). Behavioral Health/Primary Care Integration: The Four Quadrant
Model and Evidence Based Practices. Rockville, MD: National Council for
Community Behavioral Healthcare.
Mauer, B. (2008). Behavioral health/primary care integration and the person-centered
healthcare home. Washington, DC: National Council for Community
Behavioral Healthcare.
McDaniel, S., Belar, C., Schroeder, C., Hargrove, D., & Freeman, E. (2002). A
training curriculum for professional psychologists in primary care.
Professional Psychology: Research and Practice, 33(1), 65-72.
Miles, J., & Shevlin, M. (2001). Applying regression & correlation: A guide for
students and researchers. Thousand Oaks, CA: Sage Publications Ltd.
Miller, B., Mendenhall, T., & Malik, A. (2009). Integrated primary care: An
inclusive three-world view through process metrics and empirical
discrimination. Journal of Clinical Psychology in Medical Settings, 16(1), 21-
30.
69


Mokdad, A., Marks, J., Stroup, D., & Gerberding, J. (2004). Actual causes of death in
the United States, 2000. The Journal of the American Medical Association,
291(10), 1238-1245.
Narrow, W., Rae, D., Robins, L., & Regier, D. (2002). Revised prevalence estimates
of mental disorders in the United States: using a clinical significance criterion
to reconcile 2 surveys' estimates. Archives of General Psychiatry, 59(2), 115-
123.
Newton, T., Woodruff-Borden, J., & Stetson, B. (2006). Integrating mind and body:
Graduate psychology education in primary behavioral health care. Journal of
Clinical Psychology in Medical Settings, 13(1), 3-11.
O'Donohue, W., Byrd, M., Cummings, N., & Henderson, D. (2005). Behavioral
integrative care: Treatments that work in the primary care setting. New York,
NY: Brunner-Routledge.
ODonohue, W., Cummings, N., & Cummings, J. (2009). The unmet educational
agenda in integrated care. Journal of Clinical Psychology in Medical
Settings, 16(1), 94-100.
Ogden, C., Carroll, M., McDowell, M., & Flegal, K. (2007, November). Obesity
among adults in the United Statesno statistically significant chance since
2003-2004 [Data brief]. Hyattsville, Maryland: National Center for Health
Statistics.
70


Olfson, M., Shea, S., Feder, A., Fuentes, M., Nomura, Y., Gameroff, M., Weissman,
M. (2000). Prevalence of anxiety, depression, and substance use disorders in
an urban general medicine practice. Archives of Family Medicine, 9(9), 876-
883.
Packard, E. (2007, September). Postgrad growth area: Primary-care psychology.
gradPSYCH, 5(3), 14-17.
Peek, C. (2009). Integrating care for persons, not only diseases. Journal of Clinical
Psychology in Medical Settings, 16(1), 13-20.
Price, D., Beck, A., Nimmer, C., & Bensen, S. (2000). The treatment of anxiety
disorders in a primary care HMO setting. Psychiatric Quarterly, 77(1), 31-45.
Regier, D., Narrow, W., Rae, D., Manderscheid, R., Locke, B., & Goodwin, F.
(1993). The de facto US mental and addictive disorders service system:
Epidemiologic catchment area prospective 1-year prevalence rates of
disorders and services. Archives of General Psychiatry, 50(7), 85-94.
Robinson, P. & Reiter, J. (2007). Behavioral consultation and primary care: A guide
to integrating services. New York, NY: Springer.
Robinson, P. & Strosahl, K. (2009). Behavioral health consultation and primary care:
lessons learned. Journal of Clinical Psychology in Medical Settings, 16(1),
58-71.
71


Rosal, M., Ockene, J., Ma, Y., Hebert, J., Merriam, P., Matthews, C., & Ockene, I.
(2001). Behavioral risk factors among members of a health maintenance
organization. Preventive medicine, 33(6), 586-594.
Rothman, A., & Wagner, E. (2003). Chronic illness management: what is the role of
primary care? Annals of Internal Medicine, 138(3), 256-261.
Roy-Byme, P., Craske, M., Stein, M., Sullivan, G., Bystritsky, A., Katon, W., ...
Sherbourne, C. (2005). A randomized effectiveness trial of cognitive-
behavioral therapy and medication for primary care panic disorder. Archives
of General Psychiatry, 62(3), 290-298.
Rudisch, B., & Nemeroff, C. (2003). Epidemiology of comorbid coronary artery
disease and depression. Biological Psychiatry, 54(3), 227-240.
Simon, G. (2003). Social and economic burden of mood disorders. Biological
Psychiatry, 54(3), 208-215.
Simon, G., Katon, W., Lin, E., Rutter, C., Manning, W., Von Korff, M.,... Young, B.
(2007). Cost-effectiveness of systematic depression treatment among people
with diabetes mellitus. Archives of General Psychiatry, 64(1), 65-72.
Simon, G., Katon, W., VonKorff, M., Unutzer, J., Lin, E., Walker, E., ... Ludman, E.
(2001). Cost-effectiveness of a collaborative care program for primary care
patients with persistent depression. American Journal of Psychiatry, 158(10),
1638-1644.
72


Simon, G., VonKorff, M., & Barlow, W. (1995). Health care costs of primary care
patients with recognized depression. Archives of General Psychiatry, 52(10),
850-856.
Stetson, B., Carrico, A., Beacham, A., Ziegler, C., & Mokshagundam, S. (2006).
Feasibility of a pilot intervention targeting self-care behaviors in adults with
diabetes mellitus. Journal of Clinical Psychology in Medical Settings, 13(3),
239-249.
Strik, J., Lousberg, R., Cheriex, E., & Honig, A. (2004). One year cumulative
incidence of depression following myocardial infarction and impact on
cardiac outcome. Journal of Psychosomatic Research, 56(1), 59-66.
Strosahl, K. (1996a). Confessions of a behavior therapist in primary care: The
odyssey and the ecstasy. Cognitive and Behavioral Practice, 3(1), 1-28.
Strosahl, K. (1996b). Primary mental health care: A new paradigm for achieving
health and behavioral health integration. Behavioral Healthcare Tomorrow, 5,
93-96.
Strosahl, K. (1998). Integrating behavioral health and primary care services: The
primary mental health care model. In A. Blount (Ed.), Integrated primary
care: The future of medical and mental health collaboration (pp. 139-166).
New York: Norton.
73


Strosahl, K. (2001). The integration of primary care and behavioral health: Type II
change in the era of managed care. In N. Cummings, W. ODonohue, S.
Hayes, & V. Follette (Eds.), Integrated behavioral healthcare:
Positioning mental health practice with medical/surgical practice (pp. 45-69).
San Diego, CA: Academic Press.
Strosahl, K. (2005). Training behavioral health and primary care providers for
integrated care: A core competencies approach. In W. ODonohue, M. Byrd,
N. Cummings, & Henderson, D. (Eds), Behavioral integrative care:
Treatments that work in the primary care setting (pp. 15-52). New York, NY:
Brunner-Routledge.
Strosahl, K., & Quirk, M. (1994). The trouble with carve outs. Business and Health,
12(7), 52.
Strosahl, K., & Robinson, P. (2008). The primary care behavioral health model:
Applications to prevention, acute care and chronic condition management.
In R. Kessler & D. Stafford (Eds), Collaborative Medicine Case Studies:
Evidence in practice (pp 85-95). New York, NY: Springer.
Uebelacker, L., Wang, P., Berglund, P., & Kessler, R. (2006). Clinical differences
among patients treated for mental health problems in general medical and
specialty mental health settings in the National Comorbidity Survey
Replication. General Hospital Psychiatry, 28(5), 387-395.
74


U.S. Public Health Service. (1999). Mental health: A report of the Surgeon General.
Retrieved from: http://www.surgeongeneral.gov/librarv/mentalhealth/home.
html
U.S. Department of Health and Human Services. (2000). Healthy people 2010:
Understanding and improving health. Washington, DC: U.S. Government
Printing Office.
Wang, P., Berglund, P., & Kessler, R. (2000). Recent care of common mental
disorders in the United States .Journal of General Internal Medicine, 15(5),
284-292.
Wang, P., Lane, M., Olfson, M., Pincus, H., Wells, K., & Kessler, R. (2005). Twelve-
month use of mental health services in the United States: results from the
National Comorbidity Survey Replication. Archives of General Psychiatry,
62(6), 629-640.
World Health Organization (2008). Integrating mental health into primary care: A
global perspective. Geneva, Switzerland: WHO Press.
Yates, W., Mitchell, J., Rush, A., Trivedi, M., Wisniewski, S., Warden, D., ...
Gaynes, B. (2007). Clinical features of depression in outpatients with and
without co-occurring general medical conditions in STAR* D: confirmatory
analysis. Primary Care Companion to the Journal of Clinical Psychiatry,
9(1), 7-15.
75


Young, A., Klap, R., Sherbourne, C., & Wells, K. (2001). The quality of care for
depressive and anxiety disorders in the United States .Archives of General
Psychiatry, 5#(1), 55-61.
Young, A., Klap, R., Shoai, R., & Wells, K. (2008). Persistent depression and anxiety
in the United States: Prevalence and quality of care. Psychiatric Services,
59(12), 1391-1398.
76