QUALITY OF LIFE OF PERSONS WITH DIABETES: UNDERSTANDING THE
EFFECTS OF HEALTH LITERACY, SELF-EFFICACY AND KNOWLEDGE OF
CHRONIC ILLNESS AND DISABILITY
By
Enik˝ C. R´k
o
a

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Rehabilitation Counselor Education
2011

ABSTRACT
Abstract
QUALITY OF LIFE OF PERSONS WITH DIABETES: UNDERSTANDING
THE EFFECTS OF HEALTH LITERACY, SELF-EFFICACY AND
KNOWLEDGE OF CHRONIC ILLNESS AND DISABILITY
By
Enik˝ C. R´k
o
a

Livneh’s (2001) quality-of-life-based model of psychosocial adaptation to chronic illness and
disability (CID) provides a framework to conceptualize the process through which individuals negotiate challenges due to altered health. Drawing on this model, the current study
examined the relationship between three process variables, health literacy, self-efficacy and
knowledge of CID and quality of life outcomes among 126 individuals with diabetes. Four
outcomes were studied, employment, physical and psychological well-being and social relationships. I hypothesized that these process variables predict quality of life. I also postulated
that health literacy affects quality of life indirectly, through knowledge of CID and diabetes
management self-efficacy. This study also investigated the prevalence of low health literacy
in this population and tested whether health literacy skills are related to disparities in health
outcomes based on race/ethnicity, socio-economic status (SES), and age.
Findings suggest that low health literacy is not prevalent in this group. Four in five
(81.7%) individuals with diabetes have adequate health literacy. In examining whether
health disparities in race/ethnicity, SES and age are reflected in health literacy scores, current findings support significant race/ethnicity effects on both reading and numeracy and on
overall health literacy as well. There were no differences in health literacy scores based on
SES and age. There was a lack of association between health literacy and the quality of life
domains investigated. In terms of employment, individuals who worked, performed slightly

higher on the health literacy test, but the differences were not statistically significant. Results were different in relation to diabetes knowledge and diabetes management self-efficacy.
Findings suggest an unadjusted effect of self-efficacy and diabetes knowledge on employment.
In the adjusted model, physical health and the level of interference of diabetes with activities
of daily living (ADLs) are the main statistically significant independent effects that explain
the odds of one being employed. Although high levels of self-reported diabetes management
self-efficacy is strongly associated with employment, the causal pathways for this relationship
is likely to be through physical health and the lack of interference of diabetes with ADLs. In
relation to physical health, the effect of diabetes management self-efficacy holds in the adjusted model as well, supporting the presence of an independent non-redundant effect. Level
of understanding of diabetes and its treatment positively affects physical health. Results
of regression analysis indicate that higher levels of diabetes management self-efficacy and a
better understanding of diabetes and its treatment is associated with better psychological
well-being and social relationships as well.
The structural and predictive utility of Livneh’s model was partially supported by these
findings. Empirical evidence from this research substantiates the relevance of diabetes management self-efficacy and the level of understanding of diabetes and its treatment on different
quality of life domains. The current findings have applicability for vocational rehabilitation.
This research evidence supports the importance to involve vocational rehabilitation customers who have diabetes in diabetes knowledge and self-efficacy training. Clinical interventions that teach factual knowledge and health management behaviors could help individuals
improve their understanding of their medical condition and enhance self-efficacy, which in
turn has benefits on several quality of life domains.

For Lea and Zsolt

iv

ACKNOWLEDGMENTS

Several individuals assisted me in writing this dissertation, and I would like to thank them
for their involvement and contribution to this research project. Foremost, I would like
to express my sincere gratitude to my advisor, Dr.Michael Leahy for his involvement in
my professional development. I would like also to thank to the other three committee
members for their professional support and guidance, Dr.John Kosciulek, Dr.Tenko Raykov
and Dr.Mark Reckase. My work experience at Project Excellence during my doctoral training
helped me improve my knowledge of statistical methods, and of the research projects directly
tied to program evaluation in the public vocational rehabilitation system and I would like to
take this opportunity to thank Dr.Virginia Thielsen for her excellent leadership with these
projects. I also would like to thank to the staff of the Diabetes Management Program at
Ingham Regional Medical Center in Lansing and the Center for Diabetes at Saint Joseph
Mercy Hospital in Ann Arbor, and to the leadership of Michigan Rehabilitation Services, for
allowing me to collect data through their institutions.

v

Contents

TABLE OF CONTENTS

List of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
List of Figures

viii

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 Introduction
1.1 Statement of the Problem . . . . . . . . . . .
1.2 Conceptual Framework . . . . . . . . . . . . .
1.2.1 Livneh’s Model . . . . . . . . . . . . .
1.2.2 Measurement Model . . . . . . . . . .
1.3 Purpose of the Study . . . . . . . . . . . . . .
1.4 Research Questions and Research Hypotheses
1.4.1 Research Questions . . . . . . . . . . .
1.4.2 Research Hypotheses . . . . . . . . . .
1.5 Definition of Terms . . . . . . . . . . . . . . .

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.

ix

.
.
.
.
.
.
.
.
.

1
8
10
10
12
14
15
15
16
18

2 Literature Review
2.1 Understanding Health Literacy . . . . . . . . . . . . . . . . . . . .
2.1.1 Definitions and Prevalence of Health Literacy . . . . . . . .
2.1.2 Quantifying Health Literacy . . . . . . . . . . . . . . . . . .
2.2 Health Literacy, Health Outcomes and Psychosocial Well-being . . .
2.3 Self-Efficacy, Knowledge of Chronic Condition and Health Literacy
2.4 Narrowing the Focus to Diabetes . . . . . . . . . . . . . . . . . . .
2.4.1 Self-Management of Diabetes . . . . . . . . . . . . . . . . .

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

.
.
.
.
.
.
.

21
22
22
25
28
31
36
40

3 Method
3.1 Research Question and Hypotheses
3.2 Participants and Procedure . . . .
3.3 Variables and Instruments . . . . .
3.4 Data Analysis . . . . . . . . . . . .

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

43
43
46
50
60

.
.
.
.
.
.

61
62
62
65
67
68
70

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

4 Results
4.1 Univariate Findings According to Outcome Measures
4.1.1 Employment . . . . . . . . . . . . . . . . . . .
4.1.2 Physical Health . . . . . . . . . . . . . . . . .
4.1.3 Psychological Well-being . . . . . . . . . . . .
4.1.4 Social Relationships . . . . . . . . . . . . . .
4.2 Findings in Relation to the Research Questions . . .
vi

.
.
.
.

.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

4.2.1
4.2.2
4.2.3
4.2.4

4.2.5
4.2.6

Research Question 1: The Prevalence of Low Health Literacy in Individuals with Diabetes . . . . . . . . . . . . . . . . . . . . . . . . . . .
Research Question 2: The Relationship of Health Literacy Skills with
Employment Outcomes . . . . . . . . . . . . . . . . . . . . . . . . . .
Research Question 3: The Relationship of Health Literacy Skills with
Social Functioning, Psychological Well-being and Physical Health . .
Research Question 4: The Relationship of Health Literacy Skills with
Disparities in Health Outcomes According to Race/Ethnicity, SES,
and Age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Research Question 5: The Effect of Diabetes Knowledge and Diabetes
Management Self-efficacy on the Quality of Life Domains . . . . . . .
Research question 6: The effect of Diabetes Knowledge and Diabetes
Management Self-efficacy on the Relationship between Health Literacy
and the Quality of Life outcomes studied . . . . . . . . . . . . . . . .

5 Discussion
5.1 Summary of Results . . . . . . . .
5.2 Limitations . . . . . . . . . . . . .
5.3 Findings in the Light of Theory . .
5.4 Contributions to Prior Research . .
5.5 Implications for Practice . . . . . .
5.6 Future Research Recommendations
5.7 Conclusions . . . . . . . . . . . . .
Appendices
Appendix A
Appendix B
Appendix C
Appendix D

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

.
.
.
.
.
.
.

.
.
.
.

70
71
72

72
74

81

.
.
.
.
.
.
.

82
82
86
89
92
97
98
100

.
.
.
.

102
103
106
109
111

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

vii

LIST OF TABLES
List of Tables

3.1

Sample Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

49

3.2

Health Literacy in the Participating Sample . . . . . . . . . . . . . . . . . .

53

4.1

Wages and Hours Worked . . . . . . . . . . . . . . . . . . . . . . . . . . . .

62

4.2

Comparison of Working and Not Working Individuals on the Predictors of
Primary Interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

64

Pearson Correlation Coefficients for Main Independent Measures and Physical
Health Outcome Measure . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

67

Pearson Correlation Coefficients for Main Independent Measures and Psychological Well-being Outcome Measure . . . . . . . . . . . . . . . . . . . . . .

68

Pearson Correlation Coefficients for Main Independent Measures and Social
Relationship Outcome Measure . . . . . . . . . . . . . . . . . . . . . . . . .

69

4.6

Scores on the Health Literacy Scale by Race/Ethnicity . . . . . . . . . . . .

73

4.7

Scores on the Health Literacy Scale by Race/Ethnicity (2 groups) . . . . . .

73

4.8

Scores on the Health Literacy Scale by SES

. . . . . . . . . . . . . . . . . .

74

4.9

Scores on the Health Literacy Scale by Age Categories . . . . . . . . . . . .

74

4.3

4.4

4.5

4.10 Predicting Employment Outcome Using Logistic Regression

. . . . . . . . .

76

4.11 Predicting Physical Health Using Ordinary Least Square Regression . . . . .

77

4.12 Predicting Psychological Well-being Using Ordinary Least Square Regression

79

4.13 Predicting Social Relationships Using Ordinary Least Square Regression

80

viii

. .

LIST
List of Figures OF FIGURES

1.1

1.2

A Model Depicting the Structure, Content, and Process of Psychosocial Adaptation to CID (Livneh, 2001). For interpretation of the references to color in
this and all other figures, the reader is referred to the electronic version of this
dissertation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

11

Conceptual Model of the Relationships Between Health Literacy and Quality
of Life Outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

13

ix

Chapter 1
Introduction
Health literacy, the ability to read, understand, and act on health information is linked to
health outcomes. Low health literacy is prevalent in adults and has far reaching negative
consequences for several indicators of health (Baker, 2006; Berkman et al., 2004; DeWalt,
Berkman, Sheridan, Lohr, & Pignone, 2004; Lincoln et al., 2006; Sudore et al., 2006; Wolf,
Gazmararian, & Baker, 2005). A systematic review of relevant articles published between
1980-2003 found that the odds of a poor outcome is 1.5 to 3 times higher for individuals
with low health literacy (DeWalt et al., 2004). Adequate health literacy in turn is associated
with better health outcomes (Carmona, 2006) in individuals with different disabilities and
chronic conditions (Mancuso & Rincon, 2006; Weiss, Francis, Senf, Heist, & Hargraves,
2006). It is also believed, that health literacy has the potential to reduce health disparities by
race, SES and educational attainment (Fiscella, Franks, Gold, & Clancy, 2000; Saha, 2006).
Understanding and properly acting on health care information is important for effective
chronic disease management. These competencies eventually lead to the maintenance or
improvement of health.
1

While the general education level of the population never before has been this high
(Cheeseman-Day & Newburger, 2002; Stoops, 2004), averaging for adult Americans at above
high school level (Kirsch, Jungeblut, Jenkins, & Kolstad, 2002), many individuals struggle
with low literacy and low health literacy. Two waves of large scale surveys found similar results in investigating adult literacy levels in the United States. The National Adult Literacy
Survey (NALS), conducted in 1992 on a random representative sample of adults found that
large proportions of the population have low or marginal literacy skills (Kirsch et al., 2002).
Data indicates that about 40 million of American adults are functionally illiterate and 50
million have marginal literacy skills. Individuals with low literacy experience difficulties synthesizing and integrating information from lengthy texts or performing mathematical tasks,
involving two or more sequential operations, which leads to challenges in daily functioning,
particularly in tasks that involve problem solving (Kirsch et al., 2002).
The National Assessment of Adult Literacy (NAAL, 2003) as a follow-up to NALS, also
found that many adults struggle due to low or moderate levels of literacy (Cutilli & Bennett,
2009; Kutner, Greenberg, Jin, & Paulsen, 2006). About one-third of survey participants have
no more than basic literacy skills that permit only simple, everyday literacy tasks (Kutner
et al., 2006). Low literacy is pervasive in spite of population-level average increase in formal
education, suggesting a disconnect between the number of years of education attained and
literacy skills.
Data on the state of health literacy is available from research studies involving smaller
samples. These studies suggest that health literacy skills are comparable to general literacy
skills. Large proportions of adults have low health literacy skills. Marginal or inadequate
health literacy was found at 36-48% of the samples studied, with sample sizes ranging from

2

200 to 3,260 individuals (Baker, Parker, Williams, & Clark, 1998; Baker, Williams, Parker,
Gazmararian, & Nurss, 1999; Baker et al., 2002, 2007). The prevalence and far reaching
negative consequences of poor health literacy in adults alerted researchers to study this issue
and remedy these gaps (Carmona, 2006).
Lower levels of health literacy impacts a broad spectrum of health behaviors, and ultimately health status. Poor health literacy was found to be associated with poor knowledge
of disease, decreased likelihood for using preventive health care services, higher rates of
hospitalization, more need for emergency care, less medication adherence, and worse healthmanagement (Baker et al., 1998, 2002; Kripalani, Paasche-Orlow, Parker, & Saha, 2006).
Low health literacy also creates communication barriers between the individual and the
health care provider. Research suggests that patients with low health literacy encounter
difficulties in communicating with health care personnel, and find it difficult to follow the
doctor’s instruction or interpret prescription labels (Davis et al., 2006; Houts et al., 1998;
Houts, Witmer, Egeth, Loscalzo, & Zabora, 2001; Kripalani & Weiss, 2006; Parker & Gazmararian, 2003; Schillinger, Bindman, Wang, Stewart, & Piette, 2004; Sudore et al., 2009).
Underlying mechanisms and connecting pathways between literacy and health outcomes are
complex and not well understood, but studies have repeatedly found a direct link between
health literacy and health outcomes.
Health literacy was found to be associated with certain demographic characteristics. Low
levels of health literacy are more prevalent in minorities, older and less educated individuals, women, and individuals with low SES (Sudore et al., 2006). Studies on functional
literacy skills found that socioeconomically disadvantaged individuals, members of certain
racial/ethnic groups and the elderly are particularly at risk due to these limitations (Kirsch

3

et al., 2002). As far as general literacy skills, the NALS found that older adults were more
likely to have limited literacy in comparison to other age groups. This survey also found that
individuals at lower literacy level were more likely to receive food stamps than individuals
at higher literacy level, and almost half of those at lowest literacy level live in poverty. In
terms of race/ethnicity, the NALS found that all minorities perform lower on health literacy
tests than the White majority. They also show disparities in health status when compared
to the White majority (Kirsch et al., 2002).
Quality of life is ”viewed as a broad and multidimensional construct that encompasses
several life domains” (Livneh, 2002, p.154). Its core domains include psychological wellbeing, physical well-being, social and interpersonal well-being, financial and material wellbeing, employment or productivity and functional ability (Bishop, 2005). Livneh (2001)
grouped these domains into three major areas of functioning, the intrapersonal (health,
functional abilities, and psychological well-being), the interpersonal (functioning amongst
family and peers) and the extrapersonal (aspects of functioning pertinent to learning, work,
housing, and recreation).
The World Health Organization (WHO) defines quality of life as ”individuals’ perception
of their position in life in the context of culture and value systems in which they live and in
relation to their goals, expectations, standards and concerns” (World Health Organization,
1998). WHO highlights the multidimensional nature of the construct, and distinguishes dimensions that include physical health, psychological and social well-being and environmental
integration. Quality of life is the result of a subjective evaluation and it is contingent on
one’s culture and social milieu.
Health is a fundamental aspect of one’s quality of life. Several studies investigated direct

4

and indirect pathways of health literacy to health outcomes in groups of individuals with
various types of chronic illnesses (Berkman et al., 2004; DeWalt et al., 2004; Wolf et al.,
2005). In contrast, the relationship of health literacy and employment outcome was less
studied. To a limited extent, the National Adult Literacy Survey (Kirsch et al., 2002)
gathered information on the socio-economic characteristics of the sample and explored the
relationship of general literacy level and employment. It was found that individuals with
higher level of literacy were more likely to be employed, worked more hours and had higher
wages than individuals with lower literacy level.
There is a large body of literature on health literacy (Zorn, Allen, & Horowitz, 2004)
especially in the medical field. Outcomes of a bibliometric analysis identified a dramatic
increase in the number of studies exploring this topic in the past couple of decades (Bankson,
2009). Most publications were reported in the Journal of General Internal Medicine, AIDS
Education and Prevention, Journal of School Health, Patient Education and Counseling,
Journal of Health Communication, Adult Learning, American Journal of Health Education,
College Student Journal, Health Promotion International and in the American Journal of
Health Behavior. The lack of studies on this topic in the vocational rehabilitation literature
suggests that there is a gap in existing research examining the association of health literacy
with employment outcomes in people with chronic illness and disability.
The relationship of health literacy and work is an area of research with potential for new
knowledge particularly in regard to the vocational rehabilitation field. The benefits of work
on the well-being of the individual cannot be sufficiently stressed (Bishop, Chapin, & Miller,
2008; Marrone & Golowka, 1999). Employment is fundamental for societal integration and
economic self-sufficiency. Work carries not only financial benefits, but also expands one’s so-

5

cial networks (Marrone & Golowka, 1999). While several factors could potentially moderate
the effects of unemployment (e.g., coping, hardiness, social support, education, or ethnicity,
and the social, environmental or economic climate in which the individual lives), generally
the lack of employment has adverse health and psychological effects (Dooley, Fielding, &
Levi, 1996). An unemployed individual is at risk of developing unhealthy behaviors, such as
alcohol consumption, or smoking, that may increase the risk of disease and mortality (Dooley,
Catalano, & Hough, 1992). Unemployment more than doubles the risk of developing depression symptoms (Dooley, Catalano, & Wilson, 1994; Marrone & Golowka, 1999). It can be
a major stressor leading to the loss of financial resources and psychosocial assets such as
status, social support, meaning in life and time structure (Dooley et al., 1996). To study
employment and factors that could influence employment outcomes is very important for
individuals regardless of disability status.
Historically persons with disabilities had less opportunities for employment than persons without disabilities. The employment gap between person with and without disabilities
persists in spite of improved vocational rehabilitation services (Erickson & Lee, 2008; Rehabilitation Research and Training Center, 2010). The 2010 Annual Disability Statistics
Compendium, based on data gathered with the 2009 American Community Survey, reports
that 12% of the U.S. population has a disability. In terms of employment, 74% of adults
without disabilities are working (51% work full-time), but only 35% of persons with disabilities are employed, and only a small proportion of individuals with disabilities (20.0%) are
working full-time (Rehabilitation Research and Training Center, 2010). There is a gap of
39% in employment between persons with and without disabilities. The majority of persons
with disabilities are affected by unemployment. To decrease the gap between employment

6

outcomes of people with and without disabilities, more research is needed to identify determinants of employment, especially for persons with CID. Several studies investigated
potential determinants of employment, but the associations between health literacy and employment outcomes and the underlying mechanisms connecting these constructs have not yet
been explored in vocational rehabilitation research. Therefore, a study that investigates the
connection and specific mechanisms relating health literacy to employment outcomes would
be of benefit in understanding factors that influence employment outcomes for people with
disabilities.
Adequate health literacy is particularly important for persons with disabilities, as large
proportions of these people are expected to manage challenges that occur due to altered
health or due to changes in the health care system. These individuals need to quickly identify first signs of changing health status and seek out appropriate resources to help remedy
issues and properly adjust. For example some individuals with CID are often expected to
follow a regular medical regimen (Baker, 1999). Many individuals with CID are in ongoing
contact with health care providers and the health care system. Low health literacy carries
the risk for poor self care and poor medical care. A better understanding of factors that
impact quality of life outcomes could help design more interactive and effective rehabilitation interventions based on individualized self-management training to effectively maintain
or potentially improve one’s health status. Many disabilities and chronic conditions require
ongoing self-care and lifestyle modification. Understanding factors that could impact successful adaptation to such changes is fundamental to improve the quality of life of these
individuals.
The aim of this study is to explore the associations and mechanisms that underlie the

7

relationship between health literacy and quality of life outcomes. Of primary interest for
this study is employment. Studying these relationships could help better understand the
determinants of work in persons with CIDs. A better understanding of these issues could
have major implications for employment outcomes for these individuals. Such knowledge
could be used to design interventions for these individuals, particularly in regard to medical
self-management and treatment planning. Furthermore, this study examines the relationship
of health literacy with three other quality of life domains, psychological well-being, physical
health and social functioning. These domains are also reflective of an individual’s overall
well-being. Increasing health literacy influences not only better self-management but also
disease prevention, or stabilization of chronic conditions that in turn could increase the
likelihood to retain or gain employment.

1.1

Statement of the Problem

Individuals who have the skills to understand and act upon health care information are
more successful in negotiating the health care environment. Such skills lead to successful
management of one’s health, and ultimately improve one’s physical well-being. Physical and
emotional health, social functioning and employment are important domains of quality of
life. Persons with different disabilities are expected to systematically manage their health,
and being competent in accessing, processing and using health information is important
for their functioning. Most prior research published predominantly in the medical field explored the relationship between health literacy and health outcomes. A large number of
studies found that low literacy adversely affects health outcomes. It has been found that
individuals with low health literacy have less medical knowledge, have difficulty understand8

ing basic health information, experience increased hospitalization and use of preventive care
services (Carmona, 2006; Kripalani et al., 2006). Low health literacy ultimately leads to
worse control of one’s chronic condition (Kripalani et al., 2006).
Pathways between health literacy and employment outcomes have not been studied, yet
they are highly likely to resemble the associations of health literacy with health outcomes.
Similarly, health literacy and its connection to the psychological well-being and social functioning of individuals have been also studied to a limited extent. Work is an important
aspect of an individual’s life. Employment is very important for persons with chronic illness
and disability. It enables individuals to access financial resources, and health benefits. Its
psychological benefits cannot be sufficiently stressed. Historically persons with disabilities
had significantly lower employment rates than persons without disabilities. Understanding
variables that can influence and predict employment outcomes is very important and has
major implication for the vocational aspects of these peoples lives.
The goal of this present study is to increase understanding of the relationship between
health literacy and quality of life. This study examines pathways and associations between
health literacy and employment outcomes. In addition it also explores the relationship
among health literacy and psychological and social well-being. The current research will
also examine associations between disability knowledge and self-efficacy and quality of life
outcomes, and the impact of these constructs on direct associations between health literacy
and quality of life outcomes. The study specifically tests whether these two constructs
mediate the connection between health literacy and employment.

9

1.2

Conceptual Framework

This study is grounded in two conceptual models. The overall theoretical framework utilized to conceptualize the study draws on Livneh’s model on psychosocial adaptation to
CID (Livneh, 2001). In addition, for the purposes of this current study, a measurement
model was created to illustrate the hypothesized relationships among the major constructs
investigated.

1.2.1

Livneh’s Model

Livneh’s theoretical framework is a comprehensive conceptualization of the adaptation process that a person experiences when acquiring CID (Livneh, 2001). The model presents
constructs that are typically involved in the adaptation process that unfolds as the person
acquires a disability, and illustrates possible pathways and potential end results to this process. This quality-of-life based model of adaptation to the onset of CID, a simplified version
of which is illustrated in Figure 1.1, describes the adaptation process that is contingent
on specific factors and circumstances of the person, his/her environment and the type of
disability or chronic condition. This model was slightly adapted to the purpose of current
study.
The current study utilizes Livneh’s model to illustrate the ongoing adaptation process
that individuals with disabilities are experiencing while encountering challenges in their
everyday life. For the purposes of current research, adjustment to onset of disability is
replaced by adjustment to changes that are experienced due to altered health status or
other life circumstances that directly impact the person’s health. In the context of this
research disability is already present and the individual is expected to successfully negotiate
10

Figure 1.1: A Model Depicting the Structure, Content, and Process of Psychosocial Adaptation to CID (Livneh, 2001). For interpretation of the references to color in this and all
other figures, the reader is referred to the electronic version of this dissertation.
changing health conditions to reverse, diminish or prevent adverse effects. In the context
of this adjustment model, the main variable of interest to this study, health literacy, is a
characteristic of the individual and encompasses functional health literacy knowledge and
skills. This literacy level impacts quality of life outcomes and has both direct and indirect
effect on these outcomes. Health literacy, as a knowledge and skills indicator, classifies under
the group of variables that Livneh labels as Contextual Influences. With this group, health
literacy, alongside general education level is grouped under Sociodemographic Characteristics.
Knowledge of chronic condition and self-efficacy are hypothesized to have a mediator role in
the relationship of health literacy and the outcome variables studied (i.e., employment, health
11

status, psychological well-being and social relationships). Knowledge of CID is also grouped
under the Sociodemographic Characteristics. Self-efficacy, the other mediator studied, is a
personality characteristic, and this variable is classified under the measures identified within
the group of Personality and Psychological Attributes. In Livneh’s model specific quality of
life domains are presented as outcomes of adaptation. Quality of life has been studied prior
by several researchers and identified as being the ultimate goal of rehabilitation (Crewe,
1980). In the current study four distinct outcomes are investigated. Of primary interest is
employment, but physical, psychological well-being and social relationships are also examined
in relation to the predictors studied.

1.2.2

Measurement Model

A measurement model was built to represent the postulated relationship of variables studied
in this research (Figure 1.2). This model illustrates the hypothesized pathways that connect
health literacy to the major quality of life outcomes, and includes both direct and indirect
pathways that were planned to be investigated. In this model health literacy is the first component, and it is viewed as characteristic of the individual that will impact the effectiveness
with which the person could manage his/her medical condition, through the management of
health care information and his/her ability to navigate the health care system (Baker, 2006).
Health literacy is also hypothesized to have an influence on other quality of life domains,
such as employment, psychological well-being and social relationships.
Mediators represent another integral part of this model. Self-efficacy and knowledge of
chronic condition are the two constructs that will be investigated within this category. In
the context of this model, knowledge of chronic condition pertains to factual knowledge that
12

Figure 1.2: Conceptual Model of the Relationships Between Health Literacy and Quality of
Life Outcomes

a person acquires about his/her disability or chronic condition. This specialized knowledge
pertains to one’s disability. Acquisition of such knowledge enables the person to gather
and be able to effectively utilize information that is pertinent to his/her health and wellbeing. The hypothesis in this model and in the relationship of the independent variable
and this mediator can be conceptualized so that health literacy directly impacts quality
of life outcomes, but also influences these outcomes indirectly, through factual knowledge
pertinent to one’s disability, and through the individual’s self-efficacy. As illustrated in this
measurement model, during analysis specific demographic characteristics will be accounted
13

for, as it has been found in prior studies that there are significant differences in health
literacy in individuals of different age groups, educational level and of various racial/ethnic
background (Kirsch et al., 2002). There are also disparities among individuals of different
racial/ethnic backgrounds, age and SES on different outcomes as well (e.g., on health status
or employment outcomes).
Two main pathways connecting health literacy to quality of life outcomes is illustrated
in this measurement model. The first indicates direct effect of health literacy on the quality
of life domains. A second pathway is illustrated through the mediated effects. In this relationship the effect of health literacy is assessed through the mediation of knowledge of CID
and self-efficacy on quality-of-life outcomes. Both independent and combined mediational
effects were planned to be tested during statistical analysis.

1.3

Purpose of the Study

The purpose of this study was to investigate the prevalence of low health literacy in individuals with diabetes, and to examine associations between health literacy and certain domains
of quality of life. This study also examines the relationship between knowledge of CID and
self-efficacy and quality of life. The outcome of main interest is vocational functioning. In
addition, this study investigates the impact of health literacy on other important domains
of quality of life, specifically health status, social relationships and psychological well-being.
Knowledge gained through this research could increase our understanding of health literacy and its associations to certain quality of life domains previously less studied. The
most important addition will be in expanding knowledge of the relationship between health
literacy and employment outcomes. Furthermore, this research could expand understand14

ing of the relationship of knowledge of CID, and self-efficacy with quality of life outcomes.
These findings could add to efforts invested in understanding the discrepancy in employment
outcomes of persons with and without disabilities. This knowledge could be used in practice
settings to improve employment outcomes of persons with disabilities through the implementation of psychoeducational interventions aimed at improving self-management of one’s
disability or chronic condition. To exclude confounding effects due to unique characteristics
of different disabilities, one group of persons with disabilities were recruited to participate,
people with diabetes. This conditions is highly prevalent in the adult population and individuals with diabetes are expected to be constantly involved in the management of their
chronic condition, lending this group to a suitable study population.

1.4

Research Questions and Research Hypotheses

1.4.1

Research Questions

The specific research questions that this research sought to answer include:
1. How prevalent is low health literacy in individuals with diabetes?
2. Are health literacy skills related to employment outcomes?
3. Are health literacy skills related to social functioning, psychological well-being and
health status in this group of individuals with diabetes?
4. Are literacy skills related to disparities in health outcomes according to race/ethnicity,
SES, or age?
5. How does knowledge of chronic conditions and self-efficacy influence specific domains of
15

quality of life (i.e., physical health, employment outcome, emotional and psychological
well-being and social functioning) in people with diabetes?
6. What is the influence of knowledge of chronic conditions and self-efficacy on the relationship between health literacy and specific domains of quality of life (i.e., physical
health, employment outcome, emotional and psychological well-being and social functioning)?

1.4.2

Research Hypotheses

Stemming from these questions the specific research hypotheses examined are the following:

Hypothesis 1: Direct effects

1. Health literacy has a direct positive effect on employment outcome.
2. Health literacy has a direct positive effect on social well-being.
3. Health literacy has a direct positive effect on psychological well-being.
4. Health literacy has a direct positive effect on physical well-being.

Hypothesis 2: Testing for knowledge of chronic condition as a mediator

1. Knowledge of chronic condition partially mediates the direct effect of health literacy
on employment outcome.
2. Knowledge of chronic condition partially mediates the direct effect of health literacy
on social well-being.
16

3. Knowledge of chronic condition partially mediates the direct effect of health literacy
on psychological well-being.

4. Knowledge of chronic condition partially mediates the direct effect of health literacy
on physical well-being.

Hypothesis 3: Testing for self-efficacy as a mediator

1. Self-efficacy partially mediates the direct effect of health literacy on employment outcome.

2. Self-efficacy partially mediates the direct effect of health literacy on social well-being.

3. Self-efficacy partially mediates the direct effect of health literacy on psychological wellbeing.

4. Self-efficacy partially mediates the direct effect of health literacy on physical well-being.

Hypothesis 4: Combined effects of knowledge of chronic condition and self-efficacy on the
relationship of health literacy and quality of life outcomes

1. The combined mediation of self-efficacy and knowledge of chronic condition is stronger
than their individual effects on the relationship between health literacy on employment
outcome.

2. The combined mediation of self-efficacy and knowledge of chronic condition is stronger
than their individual effects on the relationship between health literacy on social wellbeing.
17

3. The combined mediation of self-efficacy and knowledge of chronic condition is stronger
than their individual effects on the relationship between health literacy on psychological
well-being.
4. The combined mediation of self-efficacy and knowledge of chronic condition is stronger
than their individual effects on the relationship between health literacy on physical
well-being.

1.5

Definition of Terms

Quality of life is the outcome measure studied in this research. Of the several quality-oflife domains discussed in the literature (Bishop, 2005; Crewe, 1980; Livneh, 2001; World
Health Organization, 1998) the current study investigated employment, psychological wellbeing, physical health and social relationships. The predictor of main interest is health
literacy. Diabetes management self-efficacy and knowledge of chronic condition are predictors/mediators that were also studied in relation to the quality of life outcomes. The next
sections provide definitions of these constructs.
Health literacy is a multifaceted and complex construct that currently has several conceptualizations (Baker, 2006). For the purpose of current research, health literacy was defined as
the ”degree to which individuals have the capacity to obtain, process, and understand basic
health information and services needed to make appropriate health decisions” (Ratzan and
Parker, 2000, p.vi). This is a widely adopted definition of health literacy. It has been used
in studies conducted by the Institute of Medicine and the Department of Health and Human
Services as well (Carmona, 2006; Department of Health and Human Services, 2000). Health
literacy consist of skills and abilities that enable the individual to navigate the health sys18

tem, and is contingent on prior knowledge of vocabulary and conceptual knowledge of health
and healthcare in particular (Baker, 2006). These skills include reading fluency, prose literacy (the ability to read and understand text), document literacy (the ability to locate and
use information in documents), quantitative literacy (the ability to apply arithmetic operations and use numerical information in printed material) and is strongly contingent on prior
knowledge of vocabulary (familiarity with individual meanings of words).
Quality of life is a multidimensional construct (Bishop, 2005; Crewe, 1980; Livneh, 2001;
World Health Organization, 1998) that includes psychological or emotional well-being, physical well-being, social and interpersonal well-being, financial and material well-being, employment or productivity and functional ability (Bishop, 2005). For the purpose of this research,
WHO’s conceptualization of the construct is used (World Health Organization, 1998). According to this definition quality of life is the ”individuals’ perception of their position in life
in the context of culture and value systems in which they live and in relation to their goals,
expectations, standards and concerns” (World Health Organization, 1998). Similarly to others, WHO also identifies distinct dimensions of quality-of-life that include physical health,
psychological well-being and social relationships and environmental integration. Most importantly, quality of life is the result of a subjective evaluation and it is contingent on one’s
culture and social milieu.
In WHO’s definition, descriptors of physical health include the lack of pain and discomfort, adequate energy and lack of fatigue, level of endurance, and the ability to control pain.
The ability to maintain good physical health has a positive effect on one’s quality of life.
Psychological well-being means the presence of positive feelings, self-esteem, good cognitive
functioning (thinking, memory, concentration and decision making ability), and a positive

19

body image. Positive emotions have a favorable influence on one’s quality of life. The social
relationships domain refers to personal relationships, sexual activity, and the amount and
quality of social support received by the person. Optimal social functioning is fundamental
for quality of life. Lastly, the environment domain describes physical safety and security,
home environment, financial resources, health and social care, opportunities to acquire new
information and skills, participation in recreation and leisure, exposure to pollution and
noise, and the quality traffic and transportation experienced by the person. The better
these conditions are, the higher one’s quality of life is likely to be.
Self-efficacy is a psychological construct introduced by Bandura (1977). Self-efficacy denotes the individual’s confidence in his/her skills and abilities to purposefully execute specific
behaviors in order to be able to reach specific goals (Bandura, 1977). Self-efficacy in the
health context is the individual’s confidence in own abilities and skills to reach desired health
outcomes. It identifies the person’s confidence to perform behaviors such as medication adherence, compliance with treatment plan, certain regimens of diet, exercise or preventive
behaviors (Sarkar, Fisher, & Schillinger, 2006).
Knowledge of chronic conditions denotes the factual knowledge a person acquires pertinent to his/her condition, and it is unique to each condition (Gazmararian, Williams, Peel,
& Baker, 2003; Williams, Baker, Parker, & Nurss, 1998). This specialized knowledge pertaining to one’s own disability enables the person to gather and utilize information that is
relevant to his/her condition.

20

Chapter 2
Literature Review
First mentioned in 1974 in a paper titled Health Education as Social Policy (Ratzan &
Parker, 2000), health literacy became the focus of intense research in the past decades,
particularly in the medical field. Health literacy was studied in relation to several indicators
of health. Studies have found that reading ability and numeracy in the health context and
patient physical and psychological well-being are significantly related. This section of the
paper provides an analysis of influential studies that were conducted to explore the nature
of these relationships and underlying mechanisms.
This chapter begins with a discussion of theoretical and operational definitions of health
literacy. Subsequent sections will cover knowledge of CID and self-efficacy, the other two
predictors of main interest to this research. Seminal research investigating the relationships
among knowledge of chronic conditions, self-efficacy and health literacy with the quality of
life outcomes of focus to this study will be discussed next. The selection if this population
was not a random choice, and the literature review will present the rationale for that as well.

21

2.1
2.1.1

Understanding Health Literacy
Definitions and Prevalence of Health Literacy

General Literacy Defined
The expert panel which convened to design the National Adult Literacy Survey, defined
general literacy as the ability to use ”print and written information to function in society,
to achieve one’s goals, and to develop one’s knowledge and potential” (Kirsch, Jungeblut,
Jenkins, and Kolstad, 2002, p.2). To measure the construct, the NALS Committee developed
three scales specifically to assess prose (ability to read and comprehend textual information),
document (ability to read documents) and quantitative literacy (ability to perform arithmetic
operations). This definition is independent of context, and pertains to generic written and
numerical information processing skills.
Health Literacy Defined
Research on the topic of health literacy increased dramatically in the past decades. As
a byproduct of this, as Baker (2006), a prominent researcher of this topic noted, significant confusion and debates occurred about its definition. There are several definitions of
health literacy currently available but there is a lack of agreed upon definition of the construct (Baker, 2006). Issues with definition led to disagreement regarding its measurement
(Baker, 2006).
The most widely accepted definition of health literacy comes from the introduction section
of the Current Bibliographies in Medicine series published on health literacy (Ratzan &
Parker, 2000). According to this definition health literacy is the ”degree to which individuals
have the capacity to obtain, process, and understand basic health information and services
22

needed to make appropriate health decisions” (Ratzan and Parker, 2000, p.vi.). Other
major agencies and institutions also adopted this definition. It was used by the Institute of
Medicine, in the Health Literacy: A Prescription to End Confusion report on health literacy
and in several projects completed through the Department of Health and Human Services
Healthy People 2010 (Carmona, 2006; Department of Health and Human Services, 2000).
According to the most circulated definition currently available, health literacy is conceptualized as an individual capacity. It represents a combination of cognitive skills needed
to perform basic reading and numerical tasks, that ultimately enables the individual to
efficiently navigate the health care environment. This set of individual capacities include
reading fluency, the ability to read and understand text (prose literacy), the ability to locate
and use information in documents (document literacy) and the ability to apply arithmetic
operations and use numerical information in printed material (quantitative literacy). Health
literacy relays on prior knowledge of vocabulary (knowing what individual words mean) and
on the conceptual knowledge of health and health care in particular (Baker, 2006).
A more comprehensive conceptualization of the construct identifies three types of health
literacy (Nutbeam, 2000). Functional literacy is described as reading skills that enable an
individual for effective daily functioning. Communicative or interactive literacy describes the
ability to understand and extract information communicated orally through interactions with
others and applying this information to new situations. Critical literacy is the individual’s
ability to critically analyze the information that is used to have greater control over life events
and situations (Nutbeam, 2000). This definition goes beyond functional literacy skills, and
incorporates communication and critical thinking skills. All these conceptualizations delimit
health literacy to individual capacity.

23

Other conceptualizations of the construct incorporate factors external to the individual.
In this sense health literacy depends on both the characteristics of the person and the
health care system. Health literacy may vary based on the medical problem, the health care
provider and based on the system providing the care. One approach (Adkins & Corus, 2009)
for example, using a sociocultural perspective, provides a multifaceted and complex view of
health literacy. Health literacy in this context is defined as ”the ability to derive meaning
from different forms of communication by using a variety of skills to accomplish healthrelated objectives” (Adkins and Corus, 2009, p.202). Thus health literacy becomes rather a
public act than a individual act, primarily because incorporates skills and competencies that
assist people in navigating the social milieu. These researchers argue that communication,
negotiation, interactions between the patient and the environment, and the ability of the
person to successfully navigate the healthcare system requires more than cognitive abilities
of decoding, processing and effectively using health information.
An expert panel from the Institute of Medicine concluded that health literacy is an
achieved level of knowledge or proficiency that depends upon an individual’s capacity and
motivation to learn and also on the resources provided by the health care system. It encompasses cultural and conceptual knowledge, oral literacy (including speaking and listening
skills), print literacy (including writing and reading skills) and numeracy. Literacy in the
health context has three basic functions, that include obtaining health information and
services, processing and understanding health information and making appropriate health
decisions. Health literacy reaches beyond the individual’s cognitive capacity, being rather
socially constructed between the patient and the provider (Adkins & Corus, 2009).
Low literacy is prevalent among American adults. The 1993 Adult Literacy Survey

24

identified that large proportions of American adults have low reading skills, that impair
their ability to function well in their everyday life (Kirsch et al., 2002). This study found
that 90 million American adults have marginal or lower level of health literacy, placing them
at high risk of being able to successfully negotiate the health care environment. Functional
health literacy may be significantly worse, because this skill is context specific.
Low health literacy is also highly prevalent in the adult population of the U.S. As reported
in previous research, 36-48% of the samples studied were identified as having marginal or
inadequate health literacy (Baker et al., 1998, 1999, 2002).
Disparities in general literacy and health literacy skills exist based on socioeconomic
status, age and racial/ethnic background. Low literacy is more prevalent in older individuals.
Almost half (44%) of the 65 and older age group has low literacy. Low literacy is also
more prevalent in minorities, females, individuals with less education and lower SES (Kirsch
et al., 2002). The National Adult Literacy Survey found that African Americans, Native
Americans and Hispanic Americans were overrepresented at low or marginal literacy level.
There are disparities in health literacy among different racial/ethnic and age groups, and
similarly health and health care disparities among these groups (Fiscella et al., 2000). A
study indicated that reducing disparities in health literacy could possibly reduce disparities
noted in health status among these groups (Saha, 2006).

2.1.2

Quantifying Health Literacy

Currently there are several conceptualizations of the health literacy construct. At one end
there is functional health literacy involving a set of cognitive skills and at the other extreme
health literacy is viewed as a social construct that transcends cognitive skills and include
25

social networking or characteristics of the healthcare system. Most studies investigated
functional health literacy, primarily due to the lack of availability of a measure for the
broader construct of health literacy (DeWalt & Pignone, 2005). There are several measures
developed to quantify health literacy (Baker, 2006). Currently the most commonly used tests
are the Test of Functional Health Literacy in Adults Short Form (Baker et al., 1999) and
the Rapid Estimate of Adult Literacy in Medicine (Davis et al., 1993). These instruments
have good psychometric properties and as a result many studies used them to operationalize
functional health literacy.
The Rapid Estimate of Adult Literacy in Medicine (REALM) is the most common objective measure of adult literacy in medical settings. It is a valid and reliable instrument
that measures reading ability. The full-scale REALM consists of 66-items. This test requires individuals to read and pronounce various medical terms and conditions in a threeto five-minute time frame. The number of correct pronunciations forms the basis of this
assessment. Depending on scores attained, individuals are classified into four groups, that
include: 1. third grade level or less, 2. four through six grade level, 3. seven or eight grade
level and 4. nine grade level or above. Individuals at six grade level or below have inadequate literacy, those at seven or eight grade level have marginal literacy and the ones at
high school or higher reading level have adequate health literacy. REALM has a short form
as well, The Rapid Estimate of Adult Literacy in Medicine-Revised the REALM-R (Bass,
Wilson, & Griffith, 2003). This is a brief, 8-item measure that has high internal consistency
(Cronbach’s alpha of 0.91).
The Test of Functional Health literacy in Adults (TOFHLA) is a sentence completion test
based on the cloze procedure where individuals are asked to add missing words to sentences

26

from a list of options. The full-format of this test takes below 22 minutes to complete. This
test measures comprehension of prose and numerical information. The TOFHLA consists
a total of sixty-seven items, of which 17 are numeracy and 50 are reading comprehension
questions. Individuals are grouped into inadequate, marginal, or adequate literacy level.
Recently a shorter version of this instrument, the Short Test of Functional Health literacy in
Adults (S-TOFHLA) was developed (Baker et al., 1999). This version consists of 36 reading
comprehension items and 4 numeracy items and takes less than 12 minutes to complete.
Just as in the long version, here as well individuals are grouped into three levels of health
literacy, inadequate, marginal and adequate level of health literacy.
The Spearman correlation between REALM and TOFHLA is high (r=0.80, Baker, 2006)
and both measures are highly correlated with general vocabulary tests (e.g., Wide Range
Achievement Test).
Health Activities Literacy Scale (HALS) is a newer instrument that was developed by
the Educational Testing Services (Baker, 2006). This instrument is considered more comprehensive than either the REALM or the TOFHLA. It includes items that measure prose,
quantitative and document literacy. Literacy skills are measured in regard to five healthrelated areas (health promotion, health protection, disease prevention, health care, health
maintenance and systems navigation).
Subjective assessments of health literacy were also used in some studies, where the physician’s perception of patient’s literacy level was used as an indication of the client’s health
literacy level.
The 2003 National Assessment of Adult Literacy survey included items with health content to measure the ability to purposefully use information related to health care. The items

27

were designed to fit the general literacy assessments such as prose literacy (i.e., skills to
search and comprehend organized texts), document literacy (i.e., skills to search and comprehend noncontinuous texts in various formats), and quantitative literacy (i.e., skills to
perform numeric computations of numbers embedded in printed materials).

2.2

Health Literacy, Health Outcomes and Psychosocial Well-being

There are both direct and indirect relationships between health literacy, health outcomes and
psychological well-being. While several studies have found empirical support for the direct
link between literacy, education, health literacy and health status, researchers still struggle to
understand the indirect connections among these constructs (Baker, 1999). Understanding
these mechanisms is fundamental to impact patient outcomes. The examination of factors
that intervene between literacy and quality of life outcomes to mediate or moderate these
relations represent topics of future research.
Health literacy was found to be related to health status, participation in preventive care
and health promotion, frequency of hospitalization, the use of emergency care, adherence
to treatment plan, illness complications and time to recovery (Baker et al., 1998, 2002;
Kalichman, Ramachandran, & Catz, 1999). Low health literacy is linked to adverse health
outcomes (Berkman et al., 2004; Wolf et al., 2005). Research found that individuals with
low health literacy have less medical knowledge, have difficulty understanding basic health
information, have worse treatment compliance, and do not benefit as much from preventive
care services (Carmona, 2006; Kripalani et al., 2006) as do individuals who have adequate
28

literacy skills. Low health literacy is associated with worse control of one’s chronic condition (Kripalani et al., 2006). A study showed that low health literacy is associated with higher
prevalence of certain chronic conditions such as diabetes mellitus and heart failure (Wolf et
al., 2005).
Significant associations were found between several health status indices and literacy
level. Patients with inadequate literacy were more likely to be hospitalized than patients with
adequate literacy after adjusting for age, gender, race, self-reported health, socioeconomic
status, and health insurance (Baker et al., 1998). This study, based on a sample of 958
individuals using logistic regression found that the odds of being hospitalized was two times
higher for individuals with low literacy. The adjusted odds ratio was 1.69 after controlling
for demographic measures, health insurance and self-reported general health status in the
month prior to enrollment in the study (Baker et al., 1998).
A study examining the effect of health literacy on mortality, found that inadequate health
literacy is a significant predictor of all-cause mortality among community-dwelling elderly
persons. The hazard ratios for all-cause mortality, after adjusting for demographics, socioeconomic status, and baseline health, were 1.52 and 1.13 for participants with inadequate and
marginal health literacy, respectively, in comparison to participants with adequate health
literacy (Baker et al., 2007).
Other studies examined health literacy in relation to treatment adherence. One study
involving a sample of 117 HIV patients found that in the first year after enrollment into
the treatment program 71% of the sample missed at least one dose (Golin et al., 2002).
Among reasons why doses were missed 15% said this was due to too many pills and 8% said
they were confused about instructions on dosage. Surprisingly, this study found no bivariate

29

associations between health literacy and treatment adherence. Another study on adherence
to HIV treatment found that there is a significant association between health literacy, education and treatment follow-through (Kalichman et al., 1999). Results of multiple logistic
regression indicated a significant and independent association between treatment adherence
and literacy with control for age, ethnicity, income, HIV symptoms, substance abuse, social
support, emotional distress, and attitudes toward primary care providers. Patients with low
literacy missed doses for reasons that include confusion and depression. Health literacy was
also a significant predictor of good glycemic control, with an odds ratio (OR) of 3.97 in a
sample of persons with diabetes (Schillinger et al., 2003).
A study investigating the relationship between health literacy and functional health status in a sample of 2,923 community-dwelling older adults found that inadequate health
literacy adversely affects physical function and mental health (Wolf et al., 2005). Chronic
conditions, health risk behaviors, and sociodemographic characteristics were accounted for
in the regression model. Results suggest that inadequate health literacy increases the odds
of experiencing challenges with activities of daily living (OR=2.83) and pain that interferes
with normal work activities (OR=2.01). These individuals also more likely report on activity limitations due to physical health (OR=1.79). This study confirmed the associations
between health literacy and poor general health.
Studies on health literacy and its relationship to mental health found that low literacy
is associated with depression (Gazmararian, Baker, Parker, & Blazer, 2000). Individuals
with inadequate health literacy had 2.7 times the odds of being depressed in comparison to
individuals with adequate health literacy. However, after adjusting for demographics, health
status measures, social support, and health behaviors, the computed odds ratio was smaller

30

and not statistically significant. Another study, investigating the relationship of literacy
and emotional well-being, found a that lower health literacy was associated with elevated
depression and poorer social support (Kalichman & Rompa, 2000).
Evidently, many of this studies point toward the significant association between health
literacy and health outcomes. People who have the skills to read, comprehend and act on
health information, have better health outcomes than individuals who have limitations in
these skills. But the relationships are not always straightforward. Other variables, such
as disease knowledge and self-efficacy can intervene in between literacy and outcome. More
exactly, as some studies suggest, health literacy could manifest its impact indirectly, through
knowledge of one’s condition and self-efficacy.

2.3

Self-Efficacy, Knowledge of Chronic Condition and
Health Literacy

Self-efficacy is defined as one’s belief in own abilities to carry out tasks to reach specific
goals (Bandura, 1977). According to Bandura, these expectations of personal mastery affect
both initiation and persistence of coping behavior. ”The strength of people’s convictions
in their own effectiveness is likely to affect whether they will even try to cope with given
situations” (Bandura, 1977, p.193). Self-efficacy has been studied extensively regarding
health-related behavior.
Self-efficacy is situation and task-specific (van der Bilj, van Poelgeest-Eeltink, & ShortridgeBaggett, 1999). Individuals can feel efficacious in one situation, but less efficacious in a
different situation. Individuals with diabetes for example perform specific self-management
31

tasks such as eating an appropriate diet, getting proper exercise, checking ones blood glucose
levels, taking oral medications and/or insulin, and, often, balancing the amount of medication or insulin respective to the amount of food intake, amount of exercise, and varying
blood glucose levels on a daily basis (Wallston, Rothman, & Cherrington, 2007). Behaviors
required for diabetes management differ significantly from behaviors needed to manage other
disabilities and chronic conditions.
Self-efficacy is important in initiating and sustaining actions, and as such it has major relevance in several aspects of human behavior, including health-promoting behaviors
in individuals with different disabilities and chronic conditions (Allegrante & Marks, 2003;
Johnson et al., 2006; Nakahara et al., 2006; Sarkar et al., 2006). These behaviors include
several aspects of the treatment plan, from changing one’s lifestyle (diet, exercise), to more
treatment-specific self-care (such as glucose management, or taking one’s medication) in the
case of persons with diabetes. Self-efficacy influences health-related goals and choices, and
the amount of effort invested in attaining goals (Wallston et al., 2007). Ultimately, selfefficacy influences health outcomes. A review of studies on self-efficacy and diabetes found
that it was associated with successful management of diabetes (Wallston et al., 2007). Another study (Johnson et al., 2006) found that self-efficacy mediated the relationship between
providers and medication adherence. More positive interactions foster adherence self-efficacy
(the belief that one can follow a treatment regimen) which ultimately affected adherence to
medication in a sample of HIV infected patients.
An examination of the causal relationship between psychosocial factors and glycemic
control in a group of diabetic patients at baseline, 6 months and 12 months following baseline found that self-efficacy directly reinforced adherence, and adherence was associated with

32

glycemic control (Nakahara et al., 2006). This study also showed that self-efficacy is a mediator of psychosocial variables on glycemic control (psychosocial factors examined included
social support and self-efficacy). Nakahara’s study, involving a structural equation modeling
analysis, showed that self-efficacy has both direct and indirect effect on glycemic control in
patients with diabetes. An earlier study also showed through multiple regression analysis
that self-efficacy is associated of diabetes self-care behavior (Glasgow et al., 1989). What
this study also found was that self-efficacy differed respective to the specific area of self-care.
Efficacy in adherence to a specific diet or a physical exercise program was lower that for
medication adherence or glucose testing.
A study of the relationship of self-efficacy and self management behaviors in a sample
of individuals who have diabetes found a significant association between self-efficacy scores
and self-management behaviors such as diet, exercise and foot care, but found no associations between self-efficacy and medication adherence (Sarkar et al., 2006). This study also
found, that adjustment of health literacy scores have no impact on the relationship between
self-efficacy and self-management, which led the researchers conclude that self-management
interventions based on self-efficacy could be effective with groups of different health literacy
levels.
Another study on a group of individuals with osteoarthritis found that self-efficacy is
relevant in the management of the condition, and impacts one’s general well-being and
health status by acting as an important mediator of arthritis-related outcomes (Allegrante
& Marks, 2003).
Health literacy is related to one’s knowledge of chronic condition. Knowledge of chronic
illness and disability pertains to the factual knowledge and individual acquires regarding the

33

conditions. There are instruments available to measure conceptual disease-specific knowledge
(e.g., asthma, hypertension, diabetes, and heart failure) but there is no current measure of
the general public’s conceptual knowledge about health and illness (Baker, 2006).
A study on 653 Medicare managed care enrolles found that levels of chronic disease
knowledge were low, even for participants with adequate literacy (Gazmararian et al., 2003).
Specifically, this study found that patients with inadequate health literacy were less likely
to correctly answer 8 of the 20 asthma questions, 5 of the 11 diabetes questions, 4 of the 16
hearth failure questions, and 8 of the 25 hypertension questions than those with adequate
literacy skills (Gazmararian et al., 2003). In the group of individuals who have diabetes,
smaller proportion knew the role of medication in regulating their blood sugar in comparison
to those with adequate literacy (50% respectively 68%). There was a direct correlation between mean knowledge scores and literacy levels. Furthermore, at multivariate analysis this
study found that health literacy had a significant association with chronic disease knowledge
after controlling for demographic variables, disease duration, and prior educational exposure
to information about the chronic disease (Gazmararian et al., 2003). This study concludes,
that poor health literacy is related to poor knowledge of chronic illness and disability.
An earlier study involving participants with hypertension and diabetes also found that
individuals with low health literacy are less likely than patients with adequate literacy to
know essential information about their chronic condition (Williams et al., 1998). In this study
there was a strong correlation between functional health literacy and patients’ knowledge
of their illness. Patients with diabetes and hypertension with low health literacy answered
correctly fewer knowledge questions, that patients with adequate health literacy. This study
specifically found that while only 50% of those with inadequate literacy knew the symptoms

34

of hypoglycemia, 94% of those with adequate functional health literacy knew those symptoms (Williams et al., 1998). Findings also showed, while 92% of patients with hypertension
and adequate literacy levels knew that a blood pressure reading of 160/100 millimeters of
mercury (mmHg) was high, only 55% of those in the lowest reading level were aware of
that. Chronic disease knowledge is related to demographic measures such as age, gender,
education, and race (Gazmararian et al., 2003).
Another study examining both the effect of self-efficacy and chronic disease knowledge,
investigating the impact of health literacy on longitudinal asthma outcomes found a statistically significant relationship between low health literacy and worse quality of life, worse
physical function, and more emergency department utilization (Mancuso & Rincon, 2006).
When examining the effect of literacy in the context of other variables, health literacy did
not remain statistically significant in predicting quality of life. Although the magnitude of
its effect on emergency department utilization remained the same as in bivariate analysis, its
effects on quality of life and functional status became attenuated when demographic characteristics, such as age and education were entered in the analysis. This study found that
disability severity and self-efficacy did not change the effect of literacy on outcomes, but
when asthma knowledge was introduced, there were significant changes in health literacy
regression coefficients (Mancuso & Rincon, 2006). The study also found that patients with
low literacy had less asthma-related health knowledge. Unlike other studies, that showed
a link between self-efficacy and client outcomes, this study found no relationship between
health literacy and self-efficacy, but detected an association between health literacy and
asthma-knowledge.
Understanding the mechanisms through which self-efficacy and chronic disease knowledge

35

can influence these different quality of life outcomes and its relationship to health literacy is
very important for self-care, health management and ongoing monitoring of one’s functional
status, as that is expected of persons with many types of CIDs. The hypothesized relationship of this study states that health literacy affects quality-of-life outcomes directly, but
also indirectly through self-efficacy and chronic disease knowledge. Health-literacy influences
one’s level of self-efficacy, the belief that one can cause a desired outcome and these together
positively impact desired quality-of-life outcomes. Similarly, health literacy also acts indirectly on outcomes through the effect on chronic disease knowledge. Previous paragraphs
presented studies that examined direct and indirect pathways from literacy and health outcomes. This study aims at examining the relationship of this variable on other quality of
life outcomes. The primary focus is on employment, but the relationship to psychological
well-being and social functioning will also be examined during this study.

2.4

Narrowing the Focus to Diabetes

Diabetes represent a major public health concern. The literature has noted that there is an
emerging ”diabetes epidemic” internationally (Wild, Roglic, Green, Sicree, & King, 2004).
The constant need for care and the costs involved with diabetes increasingly urge researchers,
healthcare providers and policy makers to invest more in better understanding the condition
in order to prevent or more effectively treat it (American Diabetes Association, 2008; Centers
for Disease Control and Prevention, 2008). In the United States in 2007 the estimated direct
and indirect cost combined spent on diabetes was 174 billion dollars (Centers for Disease
Control and Prevention, 2008).
Diabetes Mellitus is a chronic condition characterized by high levels of blood glucose
36

(sugar) that results from deficits in insulin production, insulin action or a combination of
these two (Hornichter, 1995). There are two major types of diabetes, type 1 or juvenile-onset
and type 2 or adult-onset diabetes.
Persons living with type 1 or juvenile-onset diabetes cannot produce insulin in the body.
In this condition the body’s immune system destroys the beta cells of the pancreas. These
cells are responsible for insulin production. As a result the body cannot produce the insulin,
resulting in a deficit in the metabolism of sugar (Centers for Disease Control and Prevention,
2008). This type is also known as insulin-dependent diabetes. The person has to inject insulin
into his/her system regularly to ensure normal metabolism of glucose. This type of diabetes
typically occurs in early childhood, but may occur at any age. As many as 5-10% of all
diabetes cases are of this type (Centers for Disease Control and Prevention, 2008). There
are certain risk factors that may lead to this condition, but the mechanisms of why and how
it develops are still not well understood (Centers for Disease Control and Prevention, 2008).
In non-insulin dependent or type 2 diabetes, which occurs in 90-95% of cases, the body
develops insulin resistance. The body cells cannot efficiently use insulin, as a result excess
sugar accumulates in the body. What typically happens as the disease unfolds, is that
the body needs more insulin and the pancreas fails to produce the right amount (Centers
for Disease Control and Prevention, 2008). There are certain risk factors that contribute
to the development of this condition, that include age, race-ethnicity, obesity, genetics or
family history, deficient glucose metabolism and lack of physical activity. Type 2 diabetes
occurs in adulthood, and is disproportionately high in older adults. This condition is less
likely to occur in young children or adolescents. In individuals with signs of prediabetes, the
development of type 2 diabetes is preventable, especially with lifestyle interventions (Centers

37

for Disease Control and Prevention, 2008).
A more rare form of diabetes is gestational diabetes, that occurs in some pregnant women
and manifests in glucose intolerance. There are other rare types of diabetes as well, that have
similar symptoms and may lead to further complications if not properly treated (Centers for
Disease Control and Prevention, 2008).
Diabetes is a highly prevalent chronic condition in the United States and internationally (Centers for Disease Control and Prevention, 2008; Wylie-Rosett, 2009) and the prevalence of this condition will likely increase in the following decades (Wild et al., 2004). A
recent study on the global prevalence of diabetes using data from year 2000 to project the
prevalence of the condition for year 2030, suggest that during these decades diabetes will
double across the globe. These computations were extrapolated on 191 countries that are
members of the World Health Organization using data from certain countries and the United
Nations population estimates (Wild et al., 2004).
The prevalence of diabetes is correlated with age (Cowie et al., 2010). Higher rates of
older individuals have diabetes, and the 65 or older age group report the highest rates of
diabetes. Approximately 20% of older adults report having diabetes, and only three percent
of the 18-44 age bracket have the condition (Centers for Disease Control and Prevention,
2009; Cowie et al., 2010). The prevalence of diabetes is disproportionately higher in older
adults in comparison to individuals of other age categories. As far as health literacy skills,
a study found that individuals with low health literacy are more likely to report having
diabetes (Wolf et al., 2005).
Studies also have found that diabetes is higher is certain racial-ethnic groups (Cowie et
al., 2010). This study in particular found that the prevalence of diabetes is more then two

38

times as high as in non-Hispanic African Americans and Mexican Americans than in individuals of Non-Hispanic White racial/ethnic background. Type 2 diabetes is more prevalent
in the ethnic minorities including Non-Hispanic African Americans, Hispanic/Latino Americans, Asian Americans, and Native Americans, and Native Hawaiians and Other Pacific
Islander groups than it is in Non-Hispanic White Americans (Centers for Disease Control
and Prevention, 2008). In terms of health outcomes of persons with diabetes, it was found
that African American individuals are more likely than European Americans to experience
poor long term diabetic outcomes such as diabetic retinopathy, amputations, and kidney
disease (Sequist et al., 2008).
Within the last decades diabetes become a major cause of morbidity and mortality in the
United States (Fowler, 2010). Data collected by the Center of Disease Control and Prevention
based on the examination of 72,507 death certificates found that diabetes was identified as
the seventh leading cause of deaths in the United States in 2006 (Centers for Disease Control
and Prevention, 2008). Diabetes, as a major cause of mortality is presumably higher than
what was reported actually (for less than half of deceased individuals is diabetes mentioned
in the death certificate, CDC, 2008).
The results of the 2004-2006 National Health Interview Survey (NHIS) indicate that
nearly 24 million people in the United States have diabetes. Based on the most recent data
available from the National Center for Health Statistics (NCHS), collected with the National
Health Interview Survey, 1997-2009, the prevalence of diabetes doubled between 1997 and
2009 among adults aged 18 or older, from 5.1% in 1997 to 10.1% in early 2009 (Centers for
Disease Control and Prevention, 2009). From 2008 to 2009, in only one year the rate of
diabetes increased from an estimate of 8.2% to 10.1% in adults 18 or older. An examina-

39

tion of estimates of diagnosed and undiagnosed diabetes combined reach significantly higher
numbers. As many as 23.6 million people or 7.8% of the population have diabetes, with 5.7
million undiagnosed individuals (Centers for Disease Control and Prevention, 2008).
A study investigating global prevalence of diabetes found that this condition is highly
prevalent in other countries as well (Wild et al., 2004). These alarmingly high and continually
growing figures point toward the importance and immediacy of diabetes as a major health
issue.

2.4.1

Self-Management of Diabetes

Diabetes can lead to serious complications if not properly treated. These complications
may include hearth disease, vision loss, kidney disease, amputations of lower extremities,
neuropathy (loss of sensation) or dental disease (Hornichter, 1995). Based on recent statistics
reported by the Center for Disease Control and Prevention diabetes accounts for considerable
proportion of individuals who develop these conditions. Hearth disease death rates and
the risk of strokes for persons with diabetes is 2 to 4 times higher than for individuals
without diabetes (Centers for Disease Control and Prevention, 2008). Data from 2003-2004
indicate, that 3 out of 4 individuals with diabetes have high blood pressure, 130/80 mmHg
or more (Centers for Disease Control and Prevention, 2008). In adults (20-74 years old)
diabetes is the leading cause for developing blindness and kidney failure (Centers for Disease
Control and Prevention, 2008).
Living with diabetes poses challenges to the individual on a day-by-day basis. The
management of type 1 diabetes requires regular intake of insulin through injection or a
pump. A control of type 2 diabetes can be accomplished by a strict meal plan and exercise.
40

In case of excess weight, loosing some of it is part of treatment goals. Medication is also
essential in the treatment of diabetes. Recent data indicates that 14% of adults with diabetes
take insulin only, 13% take both insulin and oral medication, 57% are using oral medication
only, and 16% do not take either insulin or oral medication (Centers for Disease Control and
Prevention, 2008). The quality of life of a person with diabetes can be significantly improved
with careful management of the condition. Self-management can improve, stabilize health
status and lower the risk of complications (Centers for Disease Control and Prevention,
2008).
Prior studies found that low literacy is prevalent in individuals with diabetes (Powell,
Hill, & Clancy, 2007; Williams et al., 1998). One study investigating literacy level in relation
to health management and health knowledge in people with diabetes found that literacy was
not associated with readiness to take diabetes health actions, but literacy was associated
with knowledge of chronic disease, so that individuals with low literacy had less knowledge
of diabetes (Powell et al., 2007). These researchers concluded that individuals with low
literacy are also willing to participate in self management of their chronic condition.
Individuals who live with diabetes are required to be actively involved in management of
their condition to maintain or improve their functioning. Control of the condition enables
individuals to fully participate in several areas of life, including work and social life. It is
important to acquire a foundational knowledge of the disability to be able to understand
risk behaviors that may worsen the condition and healthy behaviors that could stabilize
or possibly improve their health. A basic understanding of the condition and behavioral
modifications can impact self-efficacy of the person.
Understanding the relationship of health literacy, knowledge of diabetes and health lit-

41

eracy and their impact on quality of life outcomes is fundamental for disease management.
The ability to acquire and effectively use healthcare information, and to have a basic understanding of one’s chronic condition is important for the person be able to initiate and
sustain healthy lifestyle-modifying behaviors. Involving clients in goal-setting and training
in diabetes self-management can help patients to establish and obtain specific behavioral
goals (DeWalt et al., 2009). It is important to provide educational materials that are adequate to the person’s literacy level to optimize learning and health management benefits for
individuals (Powell et al., 2007; Wallace et al., 2009). This knowledge and self-efficacy will
empower people with diabetes and will likely motivate them to become active participants
and decision makers in their own health management process.

42

Chapter 3
Method
The current study examined the nature of health literacy in persons with diabetes and
explored relationships between health literacy and quality-of-life. Using Livneh’s (2001)
model of psychosocial adaptation, this study tested associations between health literacy and
employment, physical health and psychosocial well-being. This study also examined the
influence of disability knowledge and self-efficacy on the direct relationships between health
literacy and the four distinct quality of life dimensions studied. While the primary focus was
to study predictors of employment, psychological and physical health and social relationships
were also investigated in relation to the three main independent variables, health literacy,
knowledge of diabetes and diabetes management self-efficacy.

3.1

Research Question and Hypotheses

This study sought to answer the following research questions:

1. How prevalent is low health literacy in individuals with diabetes?
43

2. Are health literacy skills related to employment outcomes?
3. Are literacy skills related to disparities in health outcomes according to race/ethnicity,
SES, or age?
4. How does knowledge of chronic condition and self-efficacy influences specific domains of
quality of life (i.e., physical health, employment outcome, emotional and psychological
well-being and social functioning)?
5. What is the influence of knowledge of chronic conditions and self-efficacy on the relationship between health literacy and specific domains of quality of life (i.e., physical
health, employment outcome, emotional and psychological well-being and social functioning)?
The specific research hypotheses examined were as follows:

Hypothesis 1: Direct effects

1. Health literacy has a direct positive effect on employment outcome.
2. Health literacy has a direct positive effect on social well-being.
3. Health literacy has a direct positive effect on psychological well-being.
4. Health literacy has a direct positive effect on physical well-being.

Hypothesis 2: Testing for knowledge of chronic condition as a mediator

1. Knowledge of chronic condition partially mediates the direct effect of health literacy
on employment outcome.
44

2. Knowledge of chronic condition partially mediates the direct effect of health literacy
on social well-being.
3. Knowledge of chronic condition partially mediates the direct effect of health literacy
on psychological well-being.
4. Knowledge of chronic condition partially mediates the direct effect of health literacy
on physical well-being.
Hypothesis 3: Testing for self-efficacy as a mediator
1. Self-efficacy partially mediates the direct effect of health literacy on employment outcome.
2. Self-efficacy partially mediates the direct effect of health literacy on social well-being.
3. Self-efficacy partially mediates the direct effect of health literacy on psychological wellbeing.
4. Self-efficacy partially mediates the direct effect of health literacy on physical well-being.
Hypothesis 4: Combined effects of knowledge of chronic condition and self-efficacy on the
relationship of health literacy and quality of life outcomes
1. The combined mediation of self-efficacy and knowledge of chronic condition is stronger
than their individual effects on the relationship between health literacy on employment
outcome.
2. The combined mediation of self-efficacy and knowledge of chronic condition is stronger
than their individual effects on the relationship between health literacy on social wellbeing.
45

3. The combined mediation of self-efficacy and knowledge of chronic condition is stronger
than their individual effects on the relationship between health literacy on psychological
well-being.
4. The combined mediation of self-efficacy and knowledge of chronic condition is stronger
than their individual effects on the relationship between health literacy on physical
well-being.

3.2

Participants and Procedure

The population of interest in this study was persons with diabetes. One hundred and twentysix individuals participated in this study. Participants were recruited from Michigan. Three
different data collection methods had been used: in-person individual and group interviews,
mail survey and online survey. The interviews lasted for 30 minutes on average. The majority
of participants were recruited from diabetes education and support groups of two major hospitals located in Lansing and Ann Arbor (i.e., the Diabetes Management Program at Ingham
Regional Medical Center in Lansing, and the Center for Diabetes at the Saint Joseph Mercy
Hospital in Ann Arbor) and from current and former customers of Michigan Rehabilitation
Services (MRS). MRS is the public vocational rehabilitation agency in the State of Michigan
that provides vocational rehabilitation services to individuals with different types of disabilities, including diabetes. To recruit participants, I first contacted professionals (medical
doctors, rehabilitation counselors, and community support group leaders) that work in the
direct service of persons with diabetes to ask for permission to advertise the study in their
institution. The study was approved by both hospitals’ Institutional Review Board (IRB)
and by the Michigan State University’s IRB. The leadership of MRS also issued a formal
46

letter in which permission was granted to collect data through their agency.
To recruit individuals from education and support groups, I visited these classes and
introduced the study to class attendees. Individuals who manifested an interest to participate
were first presented the informed consent (see Appendix A and B). Individuals who gave
consent, were directed to complete the survey. Participants were also given the option to set
up times at their convenience to complete the tests or to take a test packet home, and later
return it in the stamped self-addressed envelope provided to them. The consent was created
in compliance with the requirements and guidelines for research on human subjects requested
by the Michigan State University’s IRB. Saint Joseph Mercy Hospital’s IRB requested to
modify the consent to include their specific guidelines. As a result a slightly different version
of the consent form was used with participants from the Saint Joseph Mercy Hospital. Both
versions of the consent are enclosed in the appendix.
I used advertisements and the AWARE database to identify and recruit participants
through MRS. The AWARE database is the major data storage program of all current
and former MRS clients. Test packets were sent out in two waves to a random sample of
current and former customers with diabetes (25.5 % response rate). The packets consisted
of a modified version of the advertisement flyer, the informed consent, the instrument and
a stamped self-addressed envelope. The flyer was also posted at the agency, informing
participants about the study (purpose of the study and duty of participants, eligibility,
benefits of participation, payment, and researcher contact information). The flyer could be
found in Appendix C.
Zoomerang online survey software tool was used to reach out to potential participants
through the internet. The link was distributed to a sample of patients attending diabetes

47

educational classes held at the Center for Diabetes at the Saint Joseph Mercy Hospital in
Ann Arbor but only twelve participants opted to complete the survey online.
The questionnaire consisted of six parts: TOFHLA - Full version (Parker, Baker, Williams,
& Nurss, 1995), Perceived Diabetes Self-Management Scale (Wallston et al., 2007), Diabetes
Knowledge Test (Gazmararian et al., 2003), World Health Organization Quality of Life Brief
version or WHOQOL-BREF (World Health Organization, 1998), Employment and Demographic questions (see Appendix D). There was a preestablished order in the administration
of the assessment instruments. The first instrument administered was the health literacy
test, starting with the Numeracy section. Next the WHOQOL-BREF measure and the
employment items were administered. The self-efficacy and diabetes knowledge measures
followed next. Lastly, background and demographic questions aimed at collecting information about the participants’ educational background, health insurance information and
disability information.
All interview and mail survey respondents received 10 dollars for participation. A raffle
was used with the online survey format, and of the 12 online respondents 2 individuals
received payment.
Table 3.1 provides a description of the sample in terms of gender, race/ethnicity, education, and employment status.
As far as race-ethnicity the majority self-identified as European American (61.1%) and
another large proportion as African American (27.0%). A smaller group of 7.8% indicated
that they belong to another racial/ethnic group (Hispanic/Latino, Native American or Multiracial). The average age of participants was 54.66 years (SD= 11.41, Min.=21 , Max.=82).
More females than males participated (70 versus 54). In terms of education, almost all par-

48

Table 3.1: Sample Characteristics
N

Percent

Female

70

55.6

Male

54

42.9

White

77

61.1

Black

34

27.0

Other

10

7.9

High School/GED or Less

48

38.1

Some College No Degree

40

31.7

College Degree or More

37

29.4

Yes

56

44.4

No

67

53.2

Gender

Race/Ethnicity

Education

Working Now

Ns vary due to missing data.

ticipants completed at least high school. There was a small number of 5 individuals who
did not complete high school, so the ”Some high school, no degree” and ”High School or
High School Equivalency (General Educational Development or GED)” were collapsed into
a single category. Of this sample, 38.1% had a high school degree or its equivalent, another
31.7% had some college education, while another 29.4% had a college degree or more education. Less than half, or 44.4% were employed, and slightly more than half of the sample
reported not working at the time of the data collection (53.2%).
Specific exclusion criteria was applied to exclude any confounding effects due to certain
49

individual characteristics. As it has been a standard procedure in other research studies (Arozullah et al., 2006), individuals younger than 18, or persons with severe visual
problems, blindness, deafness, dementia or other cognitive impairments were excluded from
participation.
I performed power analysis using G*Power 3 (Faul, Erdfelder, Lang, & Buchner, 2007) to
determine the required sample size. The family of tests to be used are F-tests, and multiple
regression tests in particular. The results of this a priori or prospective power analysis
indicated that N=150 is needed to have an 80% of correctly rejecting the null hypothesis
with a moderate effect size (input parameters: power = 0.80, the population effect size to
be detected was set to a medium effect or d=0.15, the prespecified significance level or alpha
was established at .05). Due to timelines and budgetary constrains, and the low response
rate in participation, data collection was concluded with N=126 responses.

3.3

Variables and Instruments

Independent Variables

Assessment of Health Literacy

Health literacy is the independent variable of main interest for the current study. Functional health literacy is defined as the degree to which individuals have the capacity to obtain,
process, and understand basic health information and services needed to make appropriate
health decisions.
Health literacy was measured using the Test of Functional Health literacy in Adults, or
TOFHLA (Parker et al., 1995). While the short form can be completed in slightly more than
50

10 minutes, the authors of the test recommend using the long form for research purposes.
Items of the TOFHLA simulate real life situations a person may encounter in the health
care environment. The instrument uses material from health care settings. The reading
section measures the person’s ability to read and understand information related to health
and health care. The numeracy component measures the individual’s ability to understand
and utilize basic mathematical knowledge needed to function in the health care environment
(e.g., interpret prescription labels, or determine if blood glucose is normal based on a chart
provided). The two sections combined consist of 67 items (50 reading comprehension items
and 17 numeracy items). The test is based on modified Cloze method, where sections are
left blank and the person is asked to select from multiple choices (4 answer choices for each
missing word) the answer he finds to best complement the sentence. It is a knowledge test
of health and utilization of health information.
This test yields a continuous measure, which in several studies was used as a categorical variable. This transformation is accomplished by grouping individuals into inadequate,
marginal and adequate skill level at pre-specified cut-points of the continuous scale. Final
scores on the reading section are computed by summing the raw points accumulated and
they may range from 0 to 50. Scores on the numeracy section are weighted, by multiplying
raw points with a constant of 2.941, and this yields final scores ranging between 0 and 50.
The combined final scores results from the sum of the scores on the two subtests, and range
between 0-100. Cut scores are at 60, and 75 (Nurss, Parker, Williams, & Baker, 2001; Parker
et al., 1995). Scores between 0-59 represent inadequate health literacy. At this level individuals will often misread the simple materials, such as prescription bottles and appointment
slips. Scores between 60-74 indicate marginal health literacy. It is assumed that the individ-

51

ual can read and understand simple information, but there are difficulties with processing
more complex text. A score between 75-100 denotes adequate literacy. People at this level
can properly function in the health care environment, but may encounter some challenges
with complex material. TOFHLA subscale and total scores were highly skewed in this sample. To normalize the data, transformations were pursued, but these transformations yielded
results that were similar to what was found in the original data. Natural log was used for
the reading subscale, and square root for the numeracy and total health literacy scores. As
such, the results on the original data are reported in relation to the health literacy variable.
TOFHLA was widely used in previous studies (Baker et al., 1998; Kalichman & Rompa,
2000; Mancuso & Rincon, 2006; Williams et al., 1998). The measure was found to correlate with the Wide Range Achievement Test-Revised (WRAT-R), a widely used instrument
to assess basic skills in reading, spelling and arithmetics (r=.74). TOFHLA also strongly
correlates with the Rapid Estimate of Adult Literacy in Medicine (r=.84). Prior studies
found that the measure has high internal consistency with a Cronbach alpha of .98 for all
items combined (Parker et al., 1995; Nurss et al., 2001). In the present study, the computed
Cronbach’s alpha was 0.74 for the numeracy subscale, 0.80 for the readings subscale, and
0.85 for all items combined. Numeracy and reading scores were correlated (r=.65, p<0.001).
When the three categories were created at pre-established cut-scores, I found that only
a small percentage (13.5%) of the participating sample has marginal and inadequate level of
health literacy skills. Table 3.2 displays results obtained on the current sample.
Dependent Variables
Assessment of Employment Outcome
Several survey items were used to assess employment outcome. Participants were asked to
52

Table 3.2: Health Literacy in the Participating Sample
N

Percent

Inadequate

6

4.8

Marginal

11

8.7

Adequate

103

81.7

6

4.8

Missing
Ns vary due to missing data.

report on their past and current employment status. They were asked to report if they ever
worked on a paid position, or whether they are currently working. Furthermore, the type of
work was also reported (part-time or full-time), and participants were also asked to provide
their job title. Data on satisfaction with current employment, information on the average
number of hours worked in a week, and hourly wages were also collected. Participants were
asked to report if they have any benefits through employment, such as health insurance.
They were also asked if they are currently on welfare. These items yielded a comprehensive
picture of participants’ current employment situation. All these items were based exclusively
on self-reports.

Assessment of Physical, Social and Psychological domains of QOL

The World Health Organization Quality of Life-BREF (WHOQOL-BREF) was used to
assess the quality-of-life domains of interest to this study (World Health Organization, 1998).
This test has a total of 26 items that measures four dimensions of quality of life, physical
health, psychological well-being, social relationships and environmental domain. WHOQOLBREF is the short form of WHOQOL-100, a 100 item quality of life instrument developed by
the World Health Organization. The short version was developed to have a brief, convenient,
53

more concise yet accurate instrument to assess these specific constructs.
Physical Health includes facets such as pain and discomfort, energy and fatigue, sleep
and rest. These items inquiry about the control the person has over pain, the ease with
which relief from pain can be achieved, and about the presence and nature of unpleasant
physical sensations. Control over these feelings has a favorable impact on quality of life.
Physical health items also ask about energy level, enthusiasm and endurance, the ability to
sleep and rest so that the person is able to perform daily activities and social functions. ”To
what extent do you feel that physical pain prevents you from doing what you need to do?”
is a sample item and answer choices include Not at all, A little, A moderate amount, Very
much and An extreme amount.
Items in the Psychological domain ask about positive feelings, self-esteem, thinking,
memory, concentration and decision making ability, body image and appearance. The more
a person is able to experience positive feelings, the better his/her self-esteem and body image
is. These positive feelings have a beneficial effect on quality of life outcomes. ”How much
do you enjoy life?” is a sample item, and answers include, Not at all, A little, A moderate
amount, Very much and An extreme amount.
The Social Relationships domain questions pertain to personal relationships, sexual
activity, and the amount and quality of social support received by the person. If the person
has fulfilling relationships, his/her life quality will likely be better. ”How satisfied are you
with your personal relationships?” is one sample item, the answer to which include the
following categories, Very dissatisfied, Dissatisfied, Neither satisfied nor dissatisfied, Satisfied
and Very satisfied.
The Environment domain consists of items pertinent to physical safety and security,

54

home environment, transportation, financial resources, health and social care, opportunities
to acquire new information and skills, participate in recreation and leisure, and the ability
to avoid pollution and noise. A favorable combination of these conditions is associated with
better quality of life. For most of the questions there is a 5-point Likert scale. A sample
question reads as ”How satisfied are you with your access to health services?” and the answer
choices to this question are Very dissatisfied, Dissatisfied, Neither satisfied nor dissatisfied,
Satisfied and Very satisfied.
Domain scores for the WHOQOL-BREF are calculated by taking the mean of all items included in each domain and multiplying by a factor of four. These scores are then transformed
to a 0-100 scale. At the development of the short instrument, four domain confirmatory factor analysis model was employed to assess its dimensionality and factor structure. The four
domain structure was confirmed, presenting the four facets of the test as described above
(physical health, psychological well-being, social relationships and environmental functioning). Cronbach alpha values for the individual domains were calculated to be .82 for Physical
Health, .75 for Psychological Well-being, .66 for Social Relationships, and .80 for the Environment scale (World Health Organization, 1998). These numbers support that the scales
have good internal consistency (World Health Organization, 1998).
For the purpose of the current study only the first three domains were studied, Physical
health, Psychological Well-being, and Social Relationships. The measure possessed adequate
internal consistencies for all three subscales. Cronbach alpha values were .89 for Physical
Health, .87 for Psychological Well-being, and .80 for Social Relationships in the participating
sample of this study.

55

Mediator Variables
Assessment of Self-Efficacy
Self-efficacy is the individual’s confidence in own skills and abilities to undertake specific
behaviors to obtain a desired outcome (Bandura, 1977). Management of different types of
disabilities require different behaviors. Researchers have developed instruments to assess
self-efficacy in relation to the unique features of several disabilities. Self-efficacy in the
health context pertains to the individual’s belief in own abilities to execute recommended
behaviors that will lead to desired health outcomes. Such behaviors include medication
adherence, compliance with treatment regimens, diet, exercise, and preventive health care.
The Perceived Diabetes Self-Management Scale (PDSMS) was used to measure selfefficacy in relation to managing diabetes (Wallston et al., 2007). Research indicates that
there is high degree of variability in the extent to which persons perceive themselves efficacious in managing their own health when living with diabetes (Wallston et al., 2007).
The test is unidimensional and was created to assess self-efficacy in persons with diabetes.
The PDSMS consists of 8 items. Response options for the PDSMS items range from 1 =
Strongly Disagree to 5 = Strongly Agree. The total score range from 8 to 40 with higher
scores indicating higher levels of self-efficacy in managing diabetes. ”It is difficult for me
to find effective solutions for problems that occur with managing my diabetes” is a sample
item and the person is asked to respond on the 5-point scale. The measure was not found
to be correlated with age, time since onset of disability, or education which makes it usable with different age groups or individuals of various socio-economic backgrounds. This
measure was positively correlated with self-reported self-care activities, and negatively correlated with glycemic control. These correlations support the validity of the measure. The
56

internal consistency of this instrument (Wallston et al., 2007) was found to be good, with a
Cronbach’s alpha of .83.
In the current sample the mean score on the scale was 27.36 (SD=6.29, Min.=12, Max.=40).
As far as internal consistency, Cronbach alpha was computed to be .87, indicating acceptable internal consistency of the items of the scale. There were no significant mean-level
differences among individuals of different racial/ethnic backgrounds, or between males and
females. However, individuals with less education (High school or less) scored significantly
lower than the other two groups (i.e., ”Some College No Degree” and ”College Degree or
More”), F(2,121)=10.66, p<.001.
Assessment of Knowledge of Chronic Condition
Knowledge of chronic condition is another construct that is unique to different CIDs.
The chronic disease knowledge questions were drawn on a study (Gazmararian et al., 2003)
that used knowledge questions from earlier developed disease knowledge scales for diabetes,
hypertension, asthma, and congestive hearth failure. The purpose of these scales is to capture
the mastery of essential concepts that are typically communicated to patients with these
disabilities. The instrument for diabetes consists of 11 questions that ask individuals about
their knowledge of normal blood glucose levels, symptoms, and complications. Educational
material used by the American Diabetes Association (Gazmararian et al., 2003) was used to
develop questions for the diabetes scale. Response choices to items are true or false. Scores
are computed by the number of correctly answered items. Mean score in the current sample
on this scale was 8.85 (SD=1.26, Min.=6, Max.=11). There were no race/ethnicity, gender
or education effects noted on outcomes of this scale. The calculated Cronbach’s alpha of .28
evidenced poor overall internal consistency for the measure. This figure indicates that the
57

Diabetes Knowledge test was unreliable for participants of the current research.
Due to issues with the psychometric properties concerning the Diabetes Knowledge test,
a subjective self-assessment item was also included in the analysis. This item assesses the individual’s general sense of understanding diabetes and its treatment. Individuals were asked
to rate their level of understanding of their condition on a 7-point Likert scale, by responding
to the question ”How would you rate your understanding of diabetes and its treatment?” on
the scale ranging between 1=Poor and 7=Excellent (M=5.13, SD=1.43, Min=1, Max=7).
Significant race/ethnicity and education effects were noted on this scale. A comparison between European Americans and Individuals with Other Racial/Ethnic backgrounds found a
statistical significant difference in the two groups on these ratings t(70)=2.11, p<.05. European Americans scored higher (M=5.40, SD=1.24) than individuals of other racial-ethnic
backgrounds (M=4.80, SD=1.66). In terms of education effects, post hoc comparisons with
Bonferroni multiple comparison test indicate that the mean score of individuals with ”High
School degree or less” is significantly lower than the average for the individuals with ”College
degree or more” education (M=4.79, SD=1.41 versus M=5.51, SD=1.24).
Control Variables
Background and demographic information were collected as well. Participants were asked
to provide information regarding their age, gender, race/ethnicity and educational attainments. Most of this information regarding demographics was discussed at the section describing the sample.
In addition, two more measures were introduced as potentially impacting the four distinct
quality of life outcomes of persons with diabetes. Subjectively rated level of interference of
diabetes with activities of daily living was one to be controlled, as a potential influencing
58

factor on employment, physical and psychological well-being, as well as social functioning.
Participants were asked to respond to the question ”How often does your diabetes prevent
you from doing your normal daily activities (ex. work)?” by giving ratings on a 7-point scale,
from 1=Never to 7=Frequently ((M=2.44, SD=1.64, Min=1, Max=7). An examination of
race/ethnicity effects indicate that European Americans reported less interference of diabetes with activities of daily living than participants from other racial groups, t(80)=-2.09,
=.04. No gender or education effects were noted on this measure. In terms of employment,
individuals who worked at the time of data collection reported significantly less interference
of diabetes in their daily functioning than non-working individuals t(117)=3.12, p=.002,
d=0.56.
Time since diagnosis was also examined. Individuals diagnosed within the past 5 years
appeared to be significantly different from individuals diagnosed with diabetes more than 5
years ago. Individuals diagnosed <=5 years ago were more likely to be employed, χ2 (1, N =
120) = 2.90, p=.09, 53.8% of individuals diagnosed 5 years or less time ago worked, however
of those diagnosed more than 5 years ago only 38.2% worked. Furthermore, individuals diagnosed five or less than 5 years ago reported better physical health (M=61.61, SD=21.29) than
individuals diagnosed more than 5 years ago (M=52.72, SD=23.74), t(120)=-2.14, p=.03
(two-tailed), d=.39. Time since diagnosis also appears to be related to participants’ perception of their social relationships, with individuals with 5 years of shorter time since diagnosis reporting better social relationships (M=62.74, SD=22.80 versus M=54.23, SD=25.87),
t(119)=1.86, p=.06.

59

3.4

Data Analysis

During univariate analysis a series of chi-square, Student t-tests and one-way ANOVAs were
used to examine the associations between demographic factors and the variables of main
interest. T-tests and one-way ANOVAs examined the relationship of continuous variables.
Categorical variables were examined using unadjusted chi-square statistics. Effect sizes were
also calculated during these analyses. A thorough comparison of scores of males and females,
of individuals of different age categories, those who have the disability for longer time, versus
those who acquired the disability more recently, and also individuals of different educational
attainments were studied on both outcome measures, and on the main predictors as well.
Regression analyses were planned to test the independent effect of health literacy on
quality of life outcome measures, after adjusting for covariates, that include, age, gender,
race/ethnicity, and education. Logistic regression was used on categorical outcomes, and
multiple linear regression on the continuous dependent variables. Binary logistic regression
is a suitable method to examine the association of health literacy with the categorical yes-no
type employment outcome measure. For a continuous outcome measure such as physical or
emotional well-being ordinary least square regressions were employed.
To test for mediator effects, instead of the traditional method to test mediation effect
developed by Baron and Kenny (Baron & Kenny, 1986) the bootstrap approach developed
by Preacher and Hayes (Preacher & Hayes, 2004) was planned to be used. This method is
suitable to examine the independent and joint mediation effect of diabetes management selfefficacy and diabetes knowledge. I hypothesized that diabetes management self-efficacy and
diabetes knowledge mediates the relationship between health literacy and specific quality of
life outcomes and these statistical methods are appropriate to test these hypotheses.

60

Chapter 4
Results
This study had several goals. One goal was to investigate the nature of health literacy, in
particular the prevalence of low health literacy in individuals with diabetes. Furthermore,
this study also sought to explore whether health literacy skills are related to disparities in
health outcomes in individuals of different race/ethnicity, SES, or age. Drawing on Livneh’s
model of psychosocial adaptation to CID the current study examined the relationship of
health literacy, diabetes knowledge and perceived diabetes management self-efficacy with
four distinct quality of life outcomes, physical health, psychological well-being, employment
and social relationships. It was hypothesized that these three independent measures have
a significant direct association with the quality of life outcome measures. Furthermore,
it was hypothesized that health literacy affect these outcomes indirectly, through diabetes
knowledge and diabetes management self-efficacy. Particularly, the hypothesis stated that
diabetes knowledge and diabetes management self-efficacy mediate the relationship of health
literacy and quality of life (i.e., physical health, psychological well-being, social relationships,
and employment).

61

Table 4.1: Wages and Hours Worked
N

Min

Max

Median

Mean

SD

Hours worked in a week on last job

114

6.00

75.00

36.25

34.72

13.63

Weekly earnings on last job

101 43.74 4,500.00

Hours worked in a week on current job

59

5.00

65.00

Weekly earnings on current job

52 45.00 4,500.00

392.50 579.59 649.70
28.33

29.91

15.38

366.39 567.65 737.13

Univariate data analyses were performed to describe the variables of interest in this
particular group, and to examine the interrelationship of main independent and dependent
measures. Logistic and multiple linear regression were employed to examine the significance
of main predictors in relation to the four different quality-of-life measures examined in the
study. SPSS.19 for Windows was used for all data analysis.

4.1

Univariate Findings According to Outcome Measures

4.1.1

Employment

Almost all individuals (n=120 or 95.2%) reported that they held paid employment in the
past. Less than half of the participating sample (56 or 44.4%) said that they are currently
employed either part-time or full-time in a paid position. Of the 56 individuals employed,
about half (n=26) said they were employed full-time, and another 30 said that they have
part-time employment. Wages and hours worked are summarized in Table 4.1.
Hours worked in a week varied from 6 to 75, with a mean of 34.72 (SD=13.63) and a
median of 36.26 for the last job held. As far as current employment, weekly work hours varied
62

between 5 and 65 hours, with an average of 29.91 (SD=15.38) and a median of 28.33 hours
a week. There was significant variation in terms of earnings as well. Weekly earnings varied
between $43.74 and $4,500.00 as far as last employment, and from $45.00 and $4,500.00
for the current job, with means of $579.59 (SD=649.70) and median of $392.50 respectively
$567.65 (SD=737.13), Med=366.39.
Of the individuals employed at the time of data collection (n=56) only 26 or 46.4% said
that they have benefits on the job (such as health insurance) and 36 individuals said that
they were satisfied or very satisfied with their employment. When prompted to report on
their outlook on finding employment in the future, 40.4% of all participants said that they
are either very or fairly confident that they can find employment in the future.
Univariate statistical analysis found that employed and not employed participants differ
significantly in several aspects. An examination of demographic characteristics indicated
that gender and employment status were related, χ2 (1, N = 121) = 6.46, p=.01, 35% of
women worked, while almost 60% of male participants said they did. Race/ethnicity and
employment status were also associated, χ2 (1, N = 119) = 3.12, p=.08, only 40.3% of
European Americans worked, but 57.1% of individuals from other racial/ethnic background
worked. Education and employment were associated as well, individuals with more education
were more likely to work than individuals who had high school degree, or GED or less than
that, phi=0.29, p=.005. While only 27.1% of the individuals with ”High School Degree, GED
or Less” worked in the present, 59.5% of the individuals with ”Some College, No Degree”
and 54.1% of participants with a ”College Degree or More” said they were employed.
In this study time since diagnosis was also examined as having a possible association with
employment. There were two groups, and the 5 years elapsed since diagnosis was used as a

63

Table 4.2: Comparison of Working and Not Working Individuals on the Predictors of Primary
Interest
Working
Variables
Age

d

t

df

N

M

Not Working
SD

N

M

SD

-.24

1.32 120 55 53.24

9.83 67 55.99 12.60

TOFHLA numeracy

.07

-0.37 120 55 39.51

7.95 67 38.88 10.33

TOFHLA reading

.04

-0.22 118 56 47.02

3.42 64 46.88

TOFHLA total

.07

-0.37 117 55 86.51 10.32 64 85.70 13.25

Diabetes Knowledge
Understanding Diabetes
Interf. of Diab with ADLs

3.59

-.42

2.32 119 55

8.56

1.24 66

9.09

1.25

.02

-0.10 121 56

5.16

1.51 67

5.13

1.38

-.56

3.05 118 55

1.96

1.37 65

2.86

1.78

6.11 66 26.31

6.39

Self-Efficacy

.38

-2.11 120 56 28.71

Physical Health

.77

-4.23 121 56 65.75 23.39 67 49.16 20.08

Psychological Well-Being

.41

-2.28 121 56 66.26 20.07 67 58.03 19.76

Social Relationships

.24

-1.30 119 56 60.79 27.00 65 54.81 23.52

Ns vary due to missing data.

cut-point. Individuals were grouped into the 5 years or less, and more than 5 years groups
based on the length of time that elapsed since they were told that they have diabetes. This
variable was also associated with employment outcome. Individuals diagnosed 5 or less than
5 years ago were more likely to be employed than individuals diagnosed more than 5 years
ago, χ2 (1, N = 120) = 2.90, p=.09, 53.8% of individuals diagnosed 5 years or less time ago
worked, however of those diagnosed more than 5 years ago only 38.2% worked.
Table 4.2 compares averages for individuals employed and not employed on the main
continuous variables of interest. While individuals not employed were slightly older on
average than individuals who were working, a t-test comparing the mean age of individuals
64

employed and not employed found no difference between the two groups. When comparing
health literacy scores of employed and not employed individuals I found that mean scores
of the working group was slightly higher than the average of non-employed group, but these
differences were not statistically significant. Health literacy was found not to be related to
employment at univariate level.
Statistically significant differences were noted at univariate level between the two groups
on diabetes knowledge, diabetes management self-efficacy and self-assessed interference of
diabetes with activities of daily living. Individuals not working scored higher (M=9.09,
SD=1.25) on the Diabetes Knowledge test than individuals who are working (M=8.56,
SD=1.24), t(119)=2.32, p=.02 (two-tailed), d=-.42. Individuals not working reported higher
level of interference of diabetes with activities of daily living (M=2.86, SD=1.78) than individuals who were working (M=1.96, SD=1.37), t(118)=3.05, p=.003 (two tailed), d=-.56. Individuals who were working had higher level of diabetes management self-efficacy (M=28.71,
SD=6.11) than individuals who were not working (M=26.31, SD=6.39), t=-2.11,p=.04 (twotailed), d=.38.
Individuals with diabetes who were working reported better physical health, t(121)=4.23, p=<.001, d=.77, and better psychological functioning t(121)=-2.28, p=.02, d=.41, but
there was no statistically significant difference in their subjective sense of the quality of their
social relationships, t(119)=-1.30, p=.24, d=.24.

4.1.2

Physical Health

Physical health was another quality of life dimension of primary interest for the study. Using
a series of t-tests, ANOVA, and correlations, the univariate associations between physical
65

health and other important predictor and control variables were tested.
One-way ANOVA showed a significant association between race/ethnicity and physical
health, F (2,120)=3.71, p=.03. Post-hoc tests (Tamhane) revealed a statistically significant
difference between White and African American individuals, with European Americans reporting better physical health (M=59.52, SD=21.62) than individuals of African American
racial/ethnic background (M=47.79, SD=22.84). There was no statistically significant difference between males and females on this measure. However the physical health of individuals
of different levels of education was significantly different. Individuals with more education
reported better physical health. Individuals with college or more education scored highest
(M=63.84, SD=23.28), followed by those with some college education (M=56.96, SD=23.39)
and individuals with high school or GED, or even less than high school education reported
having the worst physical health (M=49.63, SD=21.02). In terms of time since diagnosis and
physical health there was a statistically significant univariate association. Individuals diagnosed five or less than 5 years ago reported better physical health (M=61.61, SD=21.29)
than individuals diagnosed more than 5 years ago (M=52.72, SD=23.74), t(120)=-2.14,
p=.03 (two-tailed), d=.39.
Bivariate correlations (Table 4.3) showed a strong positive association between diabetes
management self-efficacy and physical health (r=.61, p<.001), and level of understanding
of diabetes and its treatment with physical health (r=.36, p<.001). There was a significant
negative correlation between interference of diabetes with activities of daily living and physical health (r=-.57, p<.001). Knowledge of diabetes and health literacy were not associated
with self-rated physical health.

66

Table 4.3: Pearson Correlation Coefficients for Main Independent Measures and Physical
Health Outcome Measure
Variables

1

Physical Health

2

3

-

Interf. of Diab with ADLs

-.57**
.36**

-.21
.14

.30

6

-

-.07

5

-

Understanding Diabetes

4

Diabetes Knowledge
Diabetes Management Self-Efficacy

.61**

TOFHLA total

.13

-

-.48 .43 -.04

-

-.29 .20 -.04 .13

-

** p<.01, * p<.05. Ns vary from 117 to 125.

4.1.3

Psychological Well-being

A univariate statistical analysis for gender, race, education effects on psychological health
found no significant relationships between these demographic characteristics and the subjective evaluation of psychological well-being. While the mean phycological well-being scores of
males (M=63.85, SD=20.69) was somewhat higher than the average for females (M=60.51,
SD=19.31) these differences were not statistically significant. In terms of race/ethnicity, European Americans (M=62.80, SD=20.62) had higher mean scores than African Americans
(M=59.31, SD=18.35), while the Other races category (M=64.25, SD=21.76) scored higher
than both African Americans and European Americans, but none of these differences were
statistically significant. Participants with higher levels of education had higher mean scores
on psychological well being. Individuals with ”College degree or more” rated their psychological health highest (M=65.16, SD=22.83), followed by individuals with ”Some college, no
degree” (M=64.53, SD=18.75), while mean scores of participants with ”High School Degree,
or GED or less education” were the lowest (M=56.61, SD=18.21), but these effects were not
67

Table 4.4: Pearson Correlation Coefficients for Main Independent Measures and Psychological Well-being Outcome Measure
Variables

1

Psychological Well-being

2

3

-

Interf. of Diab with ADLs

-.53**
.37**

-.21
.14

.30

6

-

-.05

5

-

Understanding Diabetes

4

Diabetes Knowledge
Diabetes Management Self-Efficacy

.65**

TOFHLA total

.04

-

-.48 .43 -.04

-

-.29 .20 -.04 .13

-

** p<.01, * p<.05. Ns vary from 121 to 125.

statistically significant, F(2,121)=2.53, p=.08. Time since diagnosis, or whether the person
was diagnosed within the past 5 years, or more than 5 years ago is not associated with
psychological well-being t(120)=-1.42, p=.16.
Bivariate correlations capturing the association of psychological well-being with the continuous measures of major importance are presented in Table 4.4. Diabetes management
self-efficacy and psychological well-being are positively correlated (r=.65, p<.001), similarly
to level of understanding of diabetes and its treatment with psychological well-being (r=.37,
p<.001). Subjectively rated level of interference of diabetes with activities of daily living is
negatively related to psychological well-being.

4.1.4

Social Relationships

Univariate analysis of differences in self-assessed social relationships according to demographic variables found no significant differences in relation to race/ethnicity, gender or
education level. Average scores of social relationships were highest for individuals of Other
68

Table 4.5: Pearson Correlation Coefficients for Main Independent Measures and Social Relationship Outcome Measure
Variables

1

Social Relationships

2

3

-

Interf. of Diab with ADLs

-.43**
.44**

-.21
.14

.30

6

-

.13

5

-

Understanding Diabetes

4

Diabetes Knowledge
Diabetes Management Self-Efficacy

.47**

TOFHLA total

-.02

-

-.48 .43 -.04

-

-.29 .20 -.04 .13

-

** p<.01, * p<.05. Ns vary from 119 to 125.

racial ethnic background (M=66.67, SD=27.32), followed by White Americans (M=59.27,
SD=24.23), while African Americans (M=52.21, SD=25.65) scored lower in this measure,
but the differences were not statistically significant, F(2,118)=1.56, p=.21. Comparisons
of average scores for males and females showed very similar findings, (M=57.31, SD=24.57
for males and M=58.49, SD=25.75 for females). Individuals with more education rate their
social relationships higher (M=59.23, SD=28.85 for those with ”College degree or more”,
M=58.23, SD=23.58 for those with ”Some college, no degree”, and M=55.43, SD=23.65 for
the individuals with ”High school degree, or GED or less education”) but these differences
were not statistically significant. Similarly to the other outcomes, time elapsed since diagnosis appears to have an impact on the ratings individuals with diabetes assign to the
quality of their social relationships. Individuals diagnosed within the last 5 years rate their
social relationships more favorably (M=62.74, SD=22.80) than individuals who have been
diagnosed more than 5 years ago (M=54.23, SD=25.87), t(119)=-1.86, p=.06.
Correlations between the independent variables of main interest and social relationships
69

are presented in Table 4.5. Individuals who experiences high level of interference of diabetes
with their activities of daily living, reported less satisfaction with their social relationships
(r=-.43, p<.01). Diabetes management self-efficacy was positively associated with social relationships (r=.47, p<.01), and similarly, the level of understanding of diabetes and its treatment was positively associated with social relationships (r=.44, p<.01). Diabetes Knowledge
and Health Literacy however was not related to satisfaction with one’s social relationships.

4.2

Findings in Relation to the Research Questions

There were two primary goals of the current study : 1) To investigate the nature of health
literacy in individuals with diabetes; 2) To examine the associations between health literacy,
self-efficacy, diabetes knowledge and different quality of life outcomes, and to determine
whether the associations between health literacy and quality of life outcomes, if there are such
associations, are mediated by diabetes management self-efficacy and diabetes knowledge.

4.2.1

Research Question 1: The Prevalence of Low Health Literacy in Individuals with Diabetes

Research question 1 asked about the prevalence of low health literacy in individuals with
diabetes. Health literacy was measured with the Test of Functional Health Literacy in Adults
(TOFHLA). Using pre-defined cut scores to group individuals into three different levels of
health literacy development, findings suggest that a large proportion of participants have
adequate health literacy level. Table 3.2 presents the distribution of the sample at the three
different levels of health literacy. A small group of six individuals representing 4.8% of the
studied sample had inadequate health literacy. Another 11 individuals (8.7%) have marginal
70

health literacy skills. A large proportion of the participating group (n=103, or 81.7%) had
adequate health literacy. This finding suggests that 4 out of 5 individuals have adequate
health literacy skills in this group of individuals with diabetes.

4.2.2

Research Question 2: The Relationship of Health Literacy
Skills with Employment Outcomes

Research question 2 asked whether health literacy skills are associated with employment
outcomes. A dichotomous measure was created to operationalize employment outcome.
Successful employment outcome was defined as full or part-time employment that was performed for a monetary compensation. A total of 20 individuals reported being retired. These
individuals were removed from the analysis. Data for the remaining 106 individuals were
included for the data analysis. An independent sample t-test examined the association of
health literacy with employment outcome. The outcome of this analysis indicated that there
is no association between employment outcome and health literacy. Individuals who worked
performed slightly higher on the health literacy test, but the differences were not statistically significant. As such, no further testing was pursued employing the logistic regression
to examine the effect of health literacy on employment in the context of other important
predictor and control variables, as it was originally planned.

71

4.2.3

Research Question 3: The Relationship of Health Literacy
Skills with Social Functioning, Psychological Well-being and
Physical Health

Research question 3 examined whether health literacy skills are related to social functioning, psychological well-being and physical health. Bivariate correlations were computed to
examine the association of health literacy with social functioning, psychological well-being
and physical health status. Findings indicate that health literacy was not associated with
any of these three quality of life outcomes. Health literacy and physical health are not associated (r=.14, ns), neither health literacy and psychological well-being (r=.04, ns), nor
health literacy and social relationships (r=-.02, ns).

4.2.4

Research Question 4: The Relationship of Health Literacy
Skills with Disparities in Health Outcomes According to
Race/Ethnicity, SES, and Age

Research question 4 asked whether literacy skills are related to disparities in health outcomes according to race/ethnicity, SES, and age. Table 4.6 presents means and standard
deviations of health literacy scores separately for the three racial/ethnic groups. One-way
ANOVAs support significant race/ethnicity effects on both subscales and on the total health
literacy scale. Mean differences on the numeracy, F(2,118)=3.20, p=.04, as well as reading,
F(2,117)=3.05, p=.05 and on the total scales were statistically significant, F(2,115)=3.54,
p=.03. This evidence supports the hypothesis that health literacy skills are related to discrepancies in health outcomes based on racial/ethnic background.
72

Table 4.6: Scores on the Health Literacy Scale by Race/Ethnicity
European American
Variables

African American

Other

M

SD

M

SD

M

SD

TOFHLA numeracy

40.95

8.11

36.33

11.64

37.40 6.60

TOFHLA reading

47.61

3.16

45.91

3.79

46.89 2.37

TOFHLA total

88.47

10.23

82.03

15.02

84.89 7.37

Table 4.7: Scores on the Health Literacy Scale by Race/Ethnicity (2 groups)
European American
Variables

All Others

M

SD

M

SD

Cohen’s d

TOFHLA numeracy

40.95

8.11

36.58 10.62

0.48

TOFHLA reading

47.61

3.16

46.12

3.53

0.45

TOFHLA total

88.47

10.23

82.66 13.68

0.50

For the purpose of regression analyses all other racial/ethnic groups were combined ( see
Table 4.7). A comparison of individuals of European American background with the group
of individuals of all other races found that European Americans scored significantly higher
than all others combined on the health literacy measures.
An examination of health literacy scores based on SES found no evidence in support of
health disparities that were noted in relation to SES. The different SES categories, operationalized as annual household income do not appear to present disparities in health literacy.
While the averages in higher SES groups are slightly higher (see Table 4.8), the differences
are not significantly different neither for the health literacy subscales nor the total scale.
While other studies suggest that older adults have lower health literacy, the current
findings suggest no differences in health literacy scores among individuals from different age
groups. Table 4.9 displays scores on the subscales and on the main scale for different age
73

Table 4.8: Scores on the Health Literacy Scale by SES
$20,000 or less
Variables

$20,001 - $40,000 $40,001 or more

M

SD

M

SD

M

SD

TOFHLA numeracy

38.24

11.58

40.56

7.59

39.57

7.61

TOFHLA reading

46.52

4.01

47.67

2.82

47.66

2.34

TOFHLA total

84.83

15.05

87.96

9.48

87.20

9.06

Table 4.9: Scores on the Health Literacy Scale by Age Categories
49 or less
Variables

50 - 59
M

SD

60 or more

M

SD

TOFHLA numeracy

40.64

8.99

37.39 10.37 40.18 7.91

TOFHLA reading

46.75

3.55

46.55

TOFHLA total

87.34 11.80 84.07 13.71 87.56 9.44

4.24

M

SD

47.56 2.11

categories. This evidence indicates comparable levels of health literacy in adults, regardless
of age category.

4.2.5

Research Question 5: The Effect of Diabetes Knowledge
and Diabetes Management Self-efficacy on the Quality of
Life Domains

Research question 5 asked if knowledge of chronic conditions and self-efficacy influence the
investigated domains of quality of life (i.e., physical health, employment outcome, psychological well-being and social relationships) in people with diabetes. To answer this question,
logistic and ordinary least square regression analyses was employed.
To test the effect of knowledge of diabetes and diabetes management self-efficacy on
74

employment a logistic regression model was constructed. Variables with a significant univariate relationship with the outcome (p < .25) were considered for the full model (Hosmer
& Lemeshow, 2000). Both knowledge of diabetes and diabetes management self-efficacy were
associated with employment outcome at univariate level (individuals who were employed reported higher level of self-efficacy but were less knowledgeable of diabetes than those who did
not work). Furthermore, the level of interference of diabetes with activities of daily living,
and time since diagnosis were also included in the model, as they were identified as important
variables to be associated with employment at univariate level. I also adjusted to physical
and psychological health, and diabetes intrusiveness in ADLs (because there were significant
group mean differences with individuals who worked, reporting better physical and psychological health, and saying they experience less intrusion of the condition of diabetes in their
activities of daily living).
In sum, several patient characteristics were associated with employment status in univariate analysis. Patients who were male, better educated, diagnosed with diabetes within
the past 5 years, had higher levels of diabetes management self-efficacy, higher levels of selfrated physical and psychological health, and reported less interference of diabetes with their
daily living activities, were more likely to be employed.
A number of covariates that appeared to affect employment were included in the regression model (Table 4.10). Due to differences between physical and psychological health
between employed and not employed individuals, the model accounted for these effects. After
adjusting for differences in age, gender, race/ethnicity, self-reported physical and psychological health, self-efficacy, diabetes knowledge, interference of diabetes with ADLs it was found
that diabetes management self-efficacy and diabetes knowledge have no direct relationship

75

Table 4.10: Predicting Employment Outcome Using Logistic Regression
Variables

OR

Sig.

95% CI

Diabetes Management Self-Efficacy

0.92 0.24

0.81-1.05

Diabetes Knowledge

0.82 0.41

0.50-1.33

Physical Health

1.08 0.00

1.04-1.13

Psychological Well-Being

0.95 0.06

0.90-1.00

Interference of diabetes with ADLs

0.60 0.02

0.39-0.93

Time Since Diagnosis

0.54 0.31

0.16-1.79

Age

1.02 0.48

0.96-1.08

Female

0.24 0.02

0.07-0.83

Other Racial/Ethnic Group

6.38 0.01 1.64-24.82

High School or Less

0.38 0.18

Some College no Degree

2.89 0.16 0.65-12.79

Constant

0.09-1.59

35.19 0.24

Hosmer and Lemeshow Test statistic = 8.06, df=8, p=.43
Nagelkerke R-Square or Pseudo R-square = 0.55

with employment. Physical health and the level of interference of diabetes with ADLs are
significant predictors of employment outcome.
While there was an unadjusted effect of self-efficacy and diabetes knowledge on employment, in the adjusted model self-reported physical health and the level of interference of
diabetes with ADLs are the main statistically significant independent effects that explain
the odds of one being employed. The better the persons health and the least diabetes symptoms interfere with ADLs, the better the odds of one being employed is. Although high levels
of self-reported diabetes management self-efficacy is strongly associated with employment,
the causal pathways for this relation is likely to be through physical health and through the
76

Table 4.11: Predicting Physical Health Using Ordinary Least Square Regression
Variables

B

SE B

β

p-value

Diabetes Management Self-Efficacy

1.41

0.33

0.39

0.00

Understanding Diabetes

2.64

1.31

0.16

0.05

Interference of diabetes with ADLs

-4.91

1.08 -0.36

0.00

Time since diagnosis

-1.96

3.29 -0.04

0.55

0.14

0.02

0.73

3.20 -0.04

0.59

3.39

0.00

0.99

Age

0.05

Female

-1.73

Other race

0.05

High School Degree or less

-1.14

3.91 -0.02

0.77

Some college no degree

-7.64

3.79 -0.16

0.05

Constant

19.26 13.99

0.17

Note: R2 = .54 (R2 − Adjusted = .50).

lack of interference of diabetes with ADLs.
To construct the multiple regression model for the physical health univariate associations were carefully examined. Positive correlation was found between diabetes management
self-efficacy and physical health (r=.61, p<.001), and level of understanding of diabetes and
its treatment with physical health (r=.36, p<.001) (Table 4.3). A significant negative correlation was identified between interference of diabetes with activities of daily living and
physical health (r=-.57, p<.001). Diabetes knowledge and health literacy were not associated with self-rated physical health, as a result these two variables were not entered into the
multivariable model.
When I examined the effect of self-efficacy on physical health, I found a significant univariate relationship (Table 4.11), that holds even after adjusting for basic demographics,
77

psychological well-being and social relationships. The statistically significant zero-order correlation between diabetes management self-efficacy and physical health (r = .61, p<.001)
persisted after controlling for the other variables in the model, indicating an independent
non-redundant effect of self-efficacy on self-reported physical health.
Self-efficacy has an independent direct effect on health, and possibly has an indirect effect
on employment through physical well-being. As it has been found, self-efficacy is associated
with health. Health and lack of interference of diabetes in daily living activities in turn
affects the individual’s ability to work.
In order to be able to test the relationship of self-efficacy and diabetes knowledge on
psychological well-being, an ordinary least square regression model was examined (Table 4.12). Variables that showed a significant association with psychological health at univariate level were selected for the regression model. Bivariate correlations capturing the
association of psychological well-being with the continuous measures of major importance
presented in Table 4.4 indicate a significant correlation between diabetes management selfefficacy, level of understanding of diabetes and its treatment, and subjectively rated level of
interference of diabetes with activities of daily living with psychological well-being.
As a result, these variables were entered into the regression model. An inspection of
results of the regression analysis (Table 4.12) reveals that the significant univariate effect
of diabetes management self-efficacy remains significant in the multiple regression model
as well. Diabetes management self-efficacy has a significant direct effect on psychological
well-being in the adjusted model as well. Individuals with higher levels of diabetes management self-efficacy report better psychological well-being. Also, the lack of interference of
diabetes with daily living activities influences psychological well-being positively, and this

78

Table 4.12: Predicting Psychological Well-being Using Ordinary Least Square Regression
Variables

B

SE B

β

p-value

Diabetes Management Self-Efficacy

1.53

0.28

0.48

0.00

Understanding Diabetes

1.85

1.05

0.13

0.08

0.94 -0.27

0.00

0.12

0.09

0.22

2.78 -0.04

0.54

Interference of diabetes with ADLs
Age

-3.24
0.15

Female

-1.72

Other race

2.83

2.88

0.07

0.33

High School Degree or less

1.67

3.39

0.04

0.62

3.31 -0.05

0.56

Some college no degree

-1.92

Constant

10.07 11.80

0.40

Note: R2 = .52 (R2 − Adjusted = .49).

direct relationship holds in the multiple regression model as well. The level of understanding
of diabetes and its treatment as a subjectively reported knowledge component also has a
positive association with psychological health, at p=.08 statistical significance level.
To examine the relationship of diabetes management self-efficacy and diabetes knowledge
on social relationships, similarly, I tested an ordinary least square regression model. While
the relationship between diabetes knowledge was found not to be related to the quality of
social relationships, the subjectively rated level of understanding of diabetes and its treatment is positively correlated to this quality of life outcome. Consequently, this measure was
introduced in the model. Also, perceived diabetes self-management scores were significantly
related with individuals’ perception of their social relationships. In turn, interference of diabetes with activities of daily living has a significant negative correlation with the outcome,
supporting that individuals who’s condition interferes with their daily functioning rate their
79

Table 4.13: Predicting Social Relationships Using Ordinary Least Square Regression
Variables

B

SE B

β

p-value

Diabetes Management Self-Efficacy

0.91

0.41

0.23

0.03

Understanding Diabetes

5.37

1.66

0.30

0.00

Interference of diabetes with ADLs

-4.02

1.36 -0.27

0.00

Time since diagnosis

-4.12

4.13 -0.08

0.32

0.18

0.09

0.28

4.03 -0.02

0.84

Age

0.19

Female

-0.80

Other Race

4.01

4.27

0.08

0.35

High School Degree or less

7.31

4.90

0.14

0.14

Some college no degree

0.06

4.75

0.00

0.99

Constant

3.27 17.60

0.85

Note: R2 = .38 (R2 − Adjusted = .32).
social relationships lower. Time since diagnosis also appears to be related to participants’
perception of their social relationships. Individuals diagnosed 5 or less than 5 years ago have
better social relationships.
Regression results are displayed in Table 4.13. The significant univariate associations
between diabetes management self-efficacy, level of understanding of diabetes and its treatment, and interference of diabetes with activities of daily living and social relationships hold
after controlling for demographics and other relevant variables. Individuals who feel efficient
in managing their diabetes, rate higher their level of understanding of diabetes and its treatment, and consider that diabetes interfere less with their condition give a better rating to
their social relationships.

80

4.2.6

Research question 6: The effect of Diabetes Knowledge and
Diabetes Management Self-efficacy on the Relationship between Health Literacy and the Quality of Life outcomes
studied

Research Question 6 asked whether diabetes knowledge and diabetes management selfefficacy influence the relationship between health literacy and the investigated domains of
quality of life (i.e., physical health, employment outcome, emotional and psychological wellbeing and social functioning). The answer to this question was not pursued, due to the lack
of significant direct association between health literacy and the four quality of life domains
examined.

81

Chapter 5
Discussion

5.1

Summary of Results

This study had several goals. One goal was to investigate the nature of health literacy, in
particular the prevalence of low health literacy in individuals with diabetes. Furthermore,
this study also sought to explore whether health literacy skills are related to disparities
in health outcomes according to race/ethnicity, SES, or age. The study also examined
the relationship of health literacy, diabetes knowledge and perceived diabetes management
self-efficacy with four distinct quality of life outcomes, physical health, psychological wellbeing, employment ant social relationships. It was hypothesized that the three independent
measures have a significant direct association with the quality of life outcome measures. It
was also postulated that health literacy affects these outcomes indirectly, through diabetes
knowledge and diabetes management self-efficacy. Particularly, the hypothesis stated that
diabetes knowledge and diabetes management self-efficacy mediate the relationship of health
literacy and quality of life (i.e., physical health, psychological well-being, social relationships,
82

and employment).
In relation to the 1st research question it was found that low health literacy is not
prevalent in individuals with diabetes. Of the current sample, only six individuals (4.8%)
had inadequate and 11 individuals (8.7%) had marginal health literacy skills. Four in five
individuals with diabetes (81.7%) had adequate health literacy skills.
The most important finding, pertains to the 2nd research question. This question
examined the association of health literacy with employment. Information on employment
status of participants indicates that half of the participating sample (56 or 44.4%) works
either full- or part-time. There was great variation in the number of weekly work hours,
and hourly wages. Of the employed group less than half, 46.4% had benefits on the job
(such as health insurance) and 64.3% stated that they were satisfied or very satisfied with
their employment. As far as future employment 40.5% of all participants felt confident it is
possible to obtain a job.
This study hypothesized, that health literacy is a significant determinant of employment
outcomes. The current findings refuted this hypothesis. While this data reported a small
difference between mean health literacy scores of employed and not employed individuals,
with employed individuals performing higher on the health literacy test, these differences
were not statistically significant.
The 3rd research question asked whether health literacy skills are related to social
functioning, psychological well-being and physical health in this group of individuals with
diabetes. Current findings suggest that there are no associations among health literacy skills
with any of these three quality of life outcomes.
The 4th research question investigated the relationship of health literacy to disparities

83

in health outcomes based on race/ethnicity, SES and education. Current findings support
significant race/ethnicity effects on both subscales and on the total health literacy scale.
However, there were no differences in health literacy scores based on SES and age. While
the averages in higher SES groups are slightly higher, the differences are not statistically
significant neither for the health literacy subscales nor the total scale. While prior research
found lower health literacy in older adults, the current findings suggest no differences in
health literacy scores among individuals from different age groups. This evidence indicates
comparable levels of health literacy in adults, regardless of age category. This parallels
education attainments in different age categories. In all three age categories (<=49, 50-59,
>=60) the distribution of individuals for the three education levels were comparable (almost
1/3 of each group had ”College Degree or more” education).
The 5th research question examined the associations between diabetes knowledge
and diabetes management self-efficacy and the quality of life outcomes studied in present
research. In terms of employment outcomes, several patient characteristics were associated with successful employment in univariate analysis. Patients who were male, better
educated, diagnosed with diabetes within the past 5 years, had higher levels of self-efficacy,
higher levels of self-rated physical and psychological health, and reported less interference
of diabetes with their daily living activities, were more likely to be employed. In the multivariable model, after adjusting for differences in age, gender, race/ethnicity, self-reported
physical and psychological health, self-efficacy, diabetes knowledge, interference of diabetes
with ADLs I found that physical health and the level of interference of diabetes with ADLs
are significant predictors of employment outcome, but neither diabetes management selfefficacy nor diabetes knowledge are significant in predicting employment.

84

While there was an unadjusted effect of self-efficacy and diabetes knowledge on employment, in the adjusted model self-reported physical health and the level of interference of
diabetes in ADLs are the main statistically significant independent effects that explain the
odds of one being employed. The better the persons health and the least diabetes symptoms
interfere with ADLs, the better the odds of one being employed is. Although high levels of
self-reported diabetes management self-efficacy is strongly associated with employment, the
causal pathways for this relation is likely to be through physical well-being and through the
lack of interference of diabetes with ADLs.
In relation to physical health only the effect of self-efficacy was tested, because knowledge of diabetes and health literacy were not associated with self-rated physical health at
univariate level. Level of understanding of diabetes and its treatment positively correlated
with physical health (r=.36, p<.001) while interference of diabetes with ADLs showed a negative correlation with this outcome. The effect of diabetes management self-efficacy holds
in the adjusted model as well, supporting the presence of an independent non-redundant
effect. Level of understanding of diabetes and its treatment positively affects physical health
(β=.16, p=.05).
A joint analysis of the relationship of diabetes management self-efficacy with physical
health and employment, suggests that self-efficacy has an independent direct effect on health,
and possibly has an indirect effect on employment through physical well-being. Diabetes
management self-efficacy is positively associated with health. Health and lack of interference
of diabetes in daily living activities in turn affects the individual’s ability to work.
Diabetes management self-efficacy, level of understanding of diabetes and its treatment,
and subjectively rated level of interference of diabetes with activities of daily living were sig-

85

nificantly correlated with psychological well-being (Table 4.4) at univariate level. These
associations hold in the adjusted model as well. Individuals with higher levels of diabetes
management self-efficacy and better understanding of diabetes and its treatment (p=.08) report better psychological well-being. A person who reports that diabetes does not interfere
with ADLs also gives a better rating to his/her psychological well-being.
In relation to the last outcome measure, social relationships it was found that the
significant univariate associations with diabetes management self-efficacy, level of understanding of diabetes and its treatment, and interference of diabetes with ADLs hold after
controlling for demographics and other relevant variables. Individuals who felt efficient in
managing their diabetes, rate higher their level of understanding of diabetes and its treatment, and consider that diabetes interfere less with their condition give a better rating to
their social relationships. While time since the person was told he/she has diabetes is a
significant factor at univariate level (with individuals diagnosed within the past 5 years reporting better social relationships), in the multivariable model this factor is not significant
any longer.
As far as the 6th research question of the study, it could not have been answered, due
to findings that answered prior questions. Because of the lack of direct relationship between
health literacy with any of the four distinct quality of life domains, the mediation test was
no longer necessary.

5.2

Limitations

This research has several limitations that need to be considered when discussing current
findings. One drawback of the study is related to sample characteristics. This study involves
86

a convenience sample, that causes limitations in generalization of results to the entire populations of individuals with diabetes. It is difficult to judge the representability of the current
sample and consequently the extent to which these findings could be generalized to the population of persons with diabetes. The study aimed at investigating the population of persons
with diabetes, but accomplished to recruit a convenience sample, that is different in certain
aspects. A recently published report provides the most current data on the demographic
profile of persons with diabetes living in the United States (Centers for Disease Control and
Prevention, 2011). This data suggests that there is a difference in the prevalence of diabetes
by age, gender and race/ethnicity. Certain age and racial/ethnic groups are at higher risk of
developing diabetes. Diabetes occurs in higher rates in males, individuals of older age, and
certain racial/ethnic groups. Center for Disease Control and Prevention (2011) reports that
the risk of diagnosed diabetes is 18% higher among Asian Americans, 66% higher among
Hispanics, and 77% higher among non-Hispanic African Americans than it is in White Americans. The current study however has higher rates of female participants, in spite of evidence
indicating higher prevalence of diabetes in males. Reasons that could explain this discrepancy could be that females are more willing to favorably respond to invitations to participate
in research studies and also they might be more willing to seek out diabetes education and
support groups. Participants were recruited from diabetes educational and support groups,
and through regular mail contacts. Participant contact information from the MRS database
was selected in a manner to ensure representability of the participating sample (in terms
of race/ethnicity, education and age). In spite of these efforts, the sample was particularly
skewed in terms of participant educational attainment. As a result, there are severe limitations in terms of generalizing of current findings beyond the current sample. It is possible

87

that through support groups as well was easier to motivate better educated individuals to
complete the instruments. Instead of generalizing current findings to the entire population
of persons with diabetes living in the United States, current findings could be used to draw
conclusion on a subset of this population, specifically persons with diabetes who have higher
levels of education overall. Convenience sampling resulted in an oversampling of females and
better educated individuals. As such, it becomes of critical importance to consider these
characteristics when thinking about these results in relation to the population.
The size of the sample also warrants replication before generalizing these findings. While
data collection spanned over eight months, it was a significant challenge to obtain the desired
number of participants, in spite of the fact that participation was monetarily compensated.
TOFHLA takes approximately 22 minutes to complete, and the rest of the packet can also
take another 10-15 minutes, which could explain why some individuals refused to participate
and stay in class for an additional 30-35 minutes. Due to these challenges in recruiting
participants for the study the sample size was slightly lower than what was suggested by the
a priori power analysis. As a result, regression analysis were conducted with the available
sample. The stability of current results might not be optimal, as a result these findings must
be treated with caution and the study replicated with larger samples. More compensation or
other recruitment methods should be employed to better motivate individuals to participate.
Prior studies noted that the relation of health literacy to health outcomes and use
of healthcare is different in various patient populations and in different healthcare systems (Baker et al., 1998). This study controls for patient population by narrowing its focus
on diabetes only, but the generalizations of findings will likely be limited to individuals with
diabetes who live only in the United States, highly likely due to differences in the way the

88

healthcare system is structured and accessed in other countries.
Another limitation stems from the fact that many individuals attended diabetes education
classes in the past or at the time of data collection (85.7%). The other small proportion
(14.3%) never had exposure to diabetes education materials. As a results, this content could
have impacted results on the diabetes knowledge test to some extent.
As already stated, health literacy measured within this study aimed at operationalizing
functional literacy, specifically reading and computational ability. These skills are intraindividual characteristics, yet several studies already noted that the construct of health literacy
is broader and transcends individual capacity. Some studies provide a more comprehensive
view and conceptualization of this construct, that incorporates culture, and the medical
system with which the persons comes into contact (Adkins & Corus, 2009). Due to the complexity of this definition, there are no psychometrically sound measures currently available
to measure all these aspects. Due to limitations in the measures, in this research functional
health literacy was operationalized that is conceptualized as a characteristic within the individual. Other studies, with more comprehensive measurement instruments would possibly
bring more accurate results.

5.3

Findings in the Light of Theory

Livneh’s (2001) quality of life based model illustrating psychosocial adaptation to CID presented in Figure 1.1 was partially supported by the data. This model was adopted to the
purpose of this study so that adaptation to ongoing challenges that individuals with disabilities encounter in their daily living was examined, rather than adaptation to the recently
acquired disability or chronic illness. In the modified model I conceptualized the ongoing
89

adaptation process in response to changes and altered health when someone is living with
a chronic illness and disability. I postulated that process variables influence significantly
several quality of life outcomes. The measurement model specifically discussed the variables
examined in this study, and the hypothesized relationships among them.
The model was partially supported by several findings from this research. This study
created empirical evidence to support the significance of diabetes management self-efficacy,
level of understanding of diabetes and its treatment, and the lack of interference of diabetes
symptoms on ADLs in predicting distinct quality of life domains in this sample of individuals
with diabetes.
No significant relationship was detected between health literacy with any of the four
distinct quality of life domains. It is possible that there is no relationship between these
constructs. In contrast, if these constructs are associated, but this study could not detect
this association, there could be several reasons for it. One such reason could be that the
relationship might be different from what was tested in this research. Other reasons for
not being able to detect such relationships could be due to issues that occurred due to
characteristics of the sample or the measurement instruments. In this study there was a
restriction on range of the health literacy measure. As it was described in detail, this measure
was very skewed, and most individuals performed really well on the instrument. Many
individuals who participated had education beyond high-school and was really challenging
to enroll individuals from lower education levels to participate, and these limitations yielded
a very skewed data, consequently influencing the results when examining the effects of health
literacy on the different quality of life domains.
The current data supports the hypothesized relationships among several predictor and

90

outcome variables. Individuals’ sense of self-efficacy in regard to the management of their
condition has a significant direct effect at both univariate level, and in the adjusted multiple
regression model as well on physical health, psychological well-being and social relationships.
Self-efficacy was associated with employment as well at univariate level. Self-efficacy is a
process variable in Livneh’s model that influences quality of life outcomes, and the current
data supports this hypothesis. While self-efficacy showed a univariate association with employment, this relationship became non-significant in the adjustment model, when physical
health was entered as a potential predictor of employment outcome. This finding suggest
that self-efficacy has an indirect effect on employment, through physical health. Individuals
who have a stronger sense of self-efficacy in managing their diabetes have better health,
which in turn leads to better employment outcomes.
Diabetes knowledge was associated with some outcomes at univariate level, but in the
multivariable model this measure was not significant in predicting quality of life outcomes.
Many individuals have comparable levels of diabetes knowledge, possibly due to their involvement with diabetes education classes. Most information contained in the diabetes knowledge
test is communicated to people in diabetes education classes provided through the hospitals,
and most individuals master this knowledge without any difficulty. The majority of individuals recruited through mail as well participated in diabetes education classes in the past.
Differences on diabetes knowledge were tested between those who did and those who did not
participate in these classes (using an independent sample t-test), and none was found when
the group means were compared. The diabetes knowledge test had inadequate psychometric properties, and its questionable reliability significantly decreases trust in these results.
In turn, the one-item measure asking individuals to identify their level of understanding

91

of diabetes and its treatment was used as an alternative knowledge component in addition
the 11-item diabetes knowledge test. This variable was found to be associated with certain
quality of life outcomes, suggesting that the better individuals understand their condition
the better they rank their quality of life, across physical health, psychological well-being and
social relationships. A subjective assessment of one’s own level of understanding of diabetes
and its treatment was associated at univariate level and in the adjusted model as well with
physical health, psychological well-being, and social relationships.

5.4

Contributions to Prior Research

In this study it was found that low health literacy is not prevalent in persons with diabetes.
Inadequate functional health literacy effects only a small proportion of individuals with
diabetes. In comparison to several other studies’ participants, the current sample overall
performed better on TOFHLA, suggesting better functional health literacy skills. Sample
characteristics and instrumentation could explain these findings to some extent.
Several other studies (Baker et al., 1998; Gazmararian, et al., 2003; Williams, et al.,
1998) recruited samples from individuals from disadvantaged backgrounds in terms of low
health literacy (i.e., elderly, predominantly minorities, individuals with less education). In
contrast, participants for this study came from diabetes educational and support groups
and from consumer groups of the public vocational rehabilitation agency. This current
sample approximates better the general population. One way this sample might be skewed in
comparison to other samples and the general population as well is the high level of education
of the individuals in the group. Almost 1/3 of current participants, or 31.7% had some college
education, without acquiring a degree, and 29.4% had college degree or more education. Only
92

38.1% had high school degree, GED or less. As a cross-reference for educational attainment,
individuals with diabetes who exited the public vocational rehabilitation in fiscal year in
2009 in Michigan were examined. A total of 221 individuals with diabetes were rehabilitated
during FY2009. An examination of the level of education of these individuals reveals that
the education level of the current sample is considerable higher that the education level of
the MRS consumers’ who have diabetes. In this group of 221 individuals 69.2% had high
school degree, GED or less, 16.7% had some college without earning any degree, and 14.0%
had a Bachelor or higher degree. The racial/ethnic composition of MRS customers closed
in FY2009 who have diabetes (n=221) is as follows: 50.7% White, 44.3% African American,
and the other 5.0% were of some other race/ethnicity. This study’s sample however had
61.1% European American, and 27.0% African American. Evidently, there were some marked
differences between this sample, other studies sample, and the customers from MRS in terms
of educational and ratial/ethnic background.
An examination of samples used by other seminal studies on the impact of health literacy
on health outcomes reveals that these other studies often times collected data on samples
from vulnerable populations as far as low health literacy. For example, 92% of the sample of
one study consisted of individuals of African American racial background (Baker et al., 1998),
and 86.4% had 12 years of education, while only 13.6% had > 12 years of education. Other
studies (Gazmararian et al., 2003; Baker et al., 2004) recruited participants from the 65
years or older age category. In this study, multivariate analysis results indicated that health
literacy was independently related to disease knowledge. Also, in another study (Williams
et al., 1998) the sample of 114 individuals with diabetes had low education level (68% less
than 12 years of education) and almost 90% belong to some minority group. These are the

93

same groups that are at-risk groups of having low health literacy, such as elderly, low SES
groups, females and minorities.
One working hypothesis for the current study was that health literacy is a significant
predictor of employment outcomes. The current empirical findings refuted this hypothesis.
While current data indicated a small difference between mean literacy scores of employed and
not employed individuals, with employed individuals performing higher on the health literacy
test, these differences were not statistically significant. According to current findings, there is
no statistical dependence between health literacy and employment. This finding is revealing
and adds new knowledge to the current vocational rehabilitation literature. Furthermore,
could raise additional research questions on intermediary constructs that could reveal indirect
associations between functional health literacy and employment outcomes.
In lack of prior research examining the relationship of functional health literacy and
employment, the impact of education and general literacy skills on employment outcomes
of persons with disabilities could provide a context to discuss current findings. While the
National Adult Literacy Survey (Kirsch et al., 2002), in examining the relationship of general
literacy level and employment found that individuals with higher level of literacy are more
likely to be employed, they also worked more hours and had higher wages, my research
specifically looking at the associations of literacy in the health context found no associations
between such skills and employment outcomes. It is possible that general literacy skills
matter rather than literacy in the health context when it comes to employment outcomes.
While it is possible for health literacy and employment not to be associated, skewness in
the data might have interfered with the power of the current study to accurately test these
associations. Further studies, with a better distribution in the health literacy construct must

94

be pursued to help answer these questions with more certainty.
Prior research, examining the associations between education and employment in individuals with disabilities is inconclusive. A recently conducted comprehensive meta-analysis
discussed determinants of employment for persons with disabilities (Saunders, Leahy, McGlynn, & Estrada-Hernandez, 2009). This study integrated outcome research produced during
25 years in rehabilitation counseling. Researchers examined nearly 200 different independent
variables in relation to obtaining or re-gaining employment for persons with disabilities. Of
primary interest to current research were findings in relation to the impact of education
on employment. While there were no studies looking at health literacy in relation to employment, 22 studies were identified that investigated the relationship of education and
employment. Findings were mixed, some studies concluded that college education was an
advantage, while others said the opposite, concluding that college education was not associated with employment. Another groups of studies found no relationship between education
and successful employment.
General literacy, health literacy and educational attainment are interrelated. More studies, looking at each construct simultaneously could possibly examine their impact, relative
importance or possible interaction effect on employment outcomes of persons with disabilities. These references mentioned above can provide a context for discussion, without the
certainty of conclusive evidence. Samples and constructs are not identical, which creates
serious challenges when trying to draw conclusions on these matters.
The findings indicating no association of health literacy with physical health and psychological well being were somewhat unexpected. These results contradict findings from other
research, but are in alignment with some other studies’ findings. Many studies indicated

95

a significant direct association between health literacy and health outcomes. Poor health
literacy was found to be associated with poor knowledge of disease, decreased likelihood for
using preventive health care services, higher rates of hospitalization, more need for emergency care, less medication adherence, and worse health-management, and communication
problems with health care providers (Baker et al., 1998, 2002; Kripalani et al., 2006; Davis
et al., 2006; Houts et al., 2001, 1998; Schillinger et al., 2004; Kripalani & Weiss, 2006;
Parker & Gazmararian, 2003; Sudore et al., 2009). In terms of psychological well-being,
prior research found a significant association between inadequate health literacy and depression (Gazmararian et al., 2000). The odds of experiencing depression was 2.7 for with
individuals with inadequate health literacy when compared to individuals with adequate
health literacy. Depression is only one dimension of psychological well-being, while the current study used the WHO’s Brief Quality of Life Survey instrument, where the psychological
well-being domain encompasses more than the lack of depression. In this definition psychological well-being includes the presence of positive feelings and self-esteem, good cognitive
functioning and positive body image. A solid reason for such contrasting findings may very
much be the differences in the underlying constructs (e.g., hospitalization as a measure of
physical health, versus physical health defined as the lack of pain and discomfort, fatigue,
having energy, the ease with which relief from pain can be achieved). The lack of full correspondence in operationalization of outcomes could led to differences in findings in comparison
to other studies’ results.
Diabetes management self-efficacy was found to have a significant impact on each quality
of life domain. The current data suggests an indirect connection between self-efficacy and
employment through physical health, and a direct association with physical health, psy-

96

chological well-being and social relationships. This finding aligns to other studies findings,
where it has been shown that the belief in one’s abilities to carry out health management
behaviors has favorable effect on several individual outcomes (Johnson et al., 2006). Higher
levels of self-efficacy positively impact health management behaviors such as diet, exercise,
blood glucose monitoring, and foot care (Sarkar et al., 2006) in the population of persons
with diabetes.

5.5

Implications for Practice

Rehabilitation counseling interventions are targeted toward improved client outcomes across
different life domains, such as physical health, psychological well-being, social and interpersonal well-being, financial and material well being, employment and productivity, and
functional ability. A better understanding of factors that impact quality of life outcomes
could help design more effective rehabilitation interventions based on individualized selfmanagement training to effectively maintain or potentially improve ones quality of life.
Individuals with chronic illness and disability are in ongoing contact with health care
providers and the health care system to some degree because of the need to manage their
medical condition. These individuals need to quickly identify first signs of changing health
status and seek out appropriate resources to help remedy issues and properly adjust.
Level of interference of diabetes with daily living activities had a significant negative
impact on all quality of life outcomes studied. Individuals need to develop knowledge and
skills to reduce this interference, and effectively manage their health. As current and prior
research supports, a healthy sense of self-efficacy is fundamentally important, for someone
to optimally negotiate challenges related to own condition.
97

Diabetes management self-efficacy was found to have an independent direct effect on
health, and possibly an indirect effect on employment through health and self-management
behaviors. Diabetes management self-efficacy also has a significant positive impact on psychological well-being and social relationships. The ability to self-manage diabetes is associated with individuals confidence in own skills to carry out diabetes self-management
activities, such as eating an appropriate diet, getting proper exercise, foot care, checking
blood glucose and taking oral medication and insulin. Given that this construct has such a
major impact on several quality of life domains, it is important to help individuals develop
an optimal level of self-efficacy. Self-efficacy influences health-related choices, health-related
goals, self-management behaviors and the amount of effort and perseverance invested to
carry out these tasks (Sarkar et al., 2006). Vocational rehabilitation agencies could develop
group or individual training activities to help individuals develop or improve their diabetes
management self-efficacy, by teaching health management behaviors and providing positive
feedback to individuals if tasks are completed successfully.

5.6

Future Research Recommendations

Evidently there is plenty of opportunity to expand or replicate these research findings. One
recommendation would be to replicate this study with a novel, more up to date health
literacy instrument. TOFHLA has been criticized for testing reading and computational
skills only. Previous researchers found that the health aspect and health context is not
sufficiently covered in the instrument. A possible future research project could be to develop
a new health literacy instrument, that goes beyond individual capacity and incorporates
characteristics of the healthcare system as well.
98

Non-experimental and experimental studies could be employed, to reexamine current findings. Variables such as health literacy, diabetes knowledge, and self-efficacy could be experimentally manipulated and examined in relation to the quality of life outcomes herein investigated. Experimental manipulation would allow for more certainty regarding the causality
of these relationships. This way confounds could be more rigourously controlled. Longitudinal studies could be a possibility, before and after diabetes education classes, to assess the
impact of the knowledge and skills acquired on diabetes knowledge, and on self-management
of health. Longitudinal studies could also be effective is self-efficacy enhancement programs.
Additional studies examining the mechanism connecting health literacy, diabetes management self-efficacy and diabetes knowledge to quality of life outcomes would be of great
benefit to understanding determinants of quality of life of this population. Could have great
potential to study what specific self-management behaviors are initiated as a result of improved self-efficacy and diabetes knowledge.
Health literacy was found not to be directly associated with any of the four different quality of life outcomes investigated in this study. While a new measure could significantly help
improve operationalizing this construct, these improved measures could be used to replicate
these findings. If similar results are found, studies could be conceptualized and carried out to
examine possible indirect relationships between health literacy and the investigated quality
of life domains.
Also, it is important to reach out to groups that are hesitant to participate, and recruit
them to be able to get a comprehensive picture of these relationships and generalize the
findings to the whole population.

99

5.7

Conclusions

This study investigated the prevalence of low health literacy and examined whether health
literacy skills are related to disparities in health outcomes based on race/ethnicity, SES,
and age. Livneh’s (2001) quality-of-life-based model was used to examine the relationship
between health literacy, self-efficacy and knowledge of CID and four distinct quality of life
domains. Four in five (81.7%) respondents had adequate health literacy. In examining
whether health disparities in race/ethnicity, SES and age are reflected in health literacy
scores, current findings support significant race/ethnicity effects on both reading and numeracy and on overall health literacy as well. There were no differences in health literacy
scores based on SES and age. There was a lack of association between health literacy and
the quality of life domains investigated. In terms of employment, individuals who worked,
performed slightly higher on the health literacy test, but the differences were not statistically
significant. Interesting results emerged pertaining associations between diabetes knowledge
and diabetes management self-efficacy and quality of life. There was an unadjusted effect of
self-efficacy and diabetes knowledge on employment. In the adjusted model, physical health
and the level of interference of diabetes with activities of daily living (ADLs) are the main
statistically significant independent effects that explain the odds of one being employed. Although high levels of self-reported diabetes management self-efficacy is strongly associated
with employment, the causal pathways for this relationship is likely to be through physical
health and the lack of interference of diabetes with ADLs. In relation to physical health, the
effect of diabetes management self-efficacy holds in the adjusted model as well, supporting
the presence of an independent non-redundant effect. Level of understanding of diabetes
and its treatment positively affects physical health. These results suggest that individuals
100

with higher levels of diabetes management self-efficacy, and those who have a better understanding of diabetes and its treatment report better psychological well-being. Diabetes
management self-efficacy, level of understanding of diabetes and its treatment, and lack of
interference of diabetes with ADLs are significant predictors of social relationships.
Results of current research, tentative to some extent due to limitations stemming from
sampling and instrumentation issues, expand our understanding of the relationship between
the three process variables studied and the quality of life outcomes. The current sample
represents a sub-group of individuals with diabetes, where females and individuals with more
education were overrepresented. As such, the results of this research should be confined to
this sub-group and should not be generalized to all individuals with diabetes.
Findings from this research have theoretical and practical applications. Livneh’s theory
was partially supported by current findings, and these outcomes could also be used in clinical
practice. This data supports the importance to design psychoeducational interventions to
improve self-management of one’s disability or chronic condition while receiving vocational
rehabilitation services. These interventions should incorporate diabetes knowledge and selfefficacy training, which competencies seem to considerably impact different quality of life
domains.

101

Appendices

APPENDICES

102

Appendix A
LETTER OF INFORMED CONSENT (USED WITH ALL BUT PARTICIPANTS
RECRUITED THROUGH SAINT JOSEPH MERCY HEALTH SYSTEM)

103

Informed Consent

Purpose of research: You are being asked to participate in a research study that aims
to investigate the influence of health literacy on the quality of life of persons with diabetes.
Health literacy, the ability to understand, process and act on health information is very
important for health outcomes. This study investigates the role of this knowledge on other
quality of life domains, specifically employment, and psychosocial functioning. The importance and immediacy of this study cannot be sufficiently stressed given the high prevalence
of diabetes and the high risk of complications if it is not properly treated. A better understanding of factors that influence health management and quality of life of individuals
with diabetes is important. Knowledge gained from this study could be used in designing
interventions for more efficient management of diabetes and prevention of complications.
What you will do: All you need to do is to complete a series of questionnaires that will take
approximately 30 minutes. The test packet includes the Test of Functional Health Literacy
for Adults, the World Health Organization Quality of Life Survey, a Perceived Diabetes
Self Management Scale, A Diabetes Knowledge test, and a brief section asking about your
employment and demographic data.
Confidentiality: The data for this project are being collected anonymously. Neither the
researchers nor anyone else will be able to link data to you. Each survey will be assigned a
case number, and names or other personal identifiers will not be asked in the questionnaires.
Data will be stored for a period of 3 years after the completion of this research on password
protected computers on MSU campus. Data will remain confidential including both paper
and computer-stored records.
Risks and benefits: Your participation in this study may contribute to the understanding
104

of factors that could influence health management of diabetes, and ultimately impact quality
of life of people living with this condition. There are no foreseeable risks associated with
participation in this study. You are free at any time to stop participating in this study.
There will be no negative consequences in withdrawing participation.
Costs and compensation for being in the study: You will be paid $10 for your participation. Taking part is voluntary: Please note that your participation in this project is
voluntary and you have the right to skip any question(s) you do not want to answer. In
addition, you may discontinue at any time.
Contact information for questions and concerns: If you have concerns or questions
about this study please contact Dr. Michael J. Leahy through regular mail at Michigan State
University, 463 Erickson Hall, East Lansing, MI 48824, or email at leahym@msu.edu or by
phone at (517) 432-0605. If you have questions or concerns about your role and rights as a
research participant, would like to obtain information or offer input, or would like to register
a complaint about this study, you may contact, anonymously if you wish, the Michigan
State University’s Human Research Protection Program at 517-355-2180, Fax 517-432-4503,
or e-mail irb@msu.edu or regular mail at 207 Olds Hall, MSU, East Lansing, MI 48824.
Documentation of informed consent: Your signature below means that you voluntarily
agree to participate in this research study.

Signature

Date

105

Appendix B
LETTER OF INFORMED CONSENT (USED WITH PARTICIPANTS RECRUITED
THROUGH SAINT JOSEPH MERCY HEALTH SYSTEM)

106

St. Joseph Mercy Health System
Consent to Participate in Research
You are being asked to participate in a research study that aims to investigate the influence
of health literacy on the quality of life of persons with diabetes. Health literacy, the ability
to understand, process and act on health information is very important for health outcomes.
This study investigates the role of this knowledge on other quality of life domains, specifically
employment, and psychosocial functioning. The importance and immediacy of this study
cannot be sufficiently stressed given the high prevalence of diabetes and the high risk of
complications if it is not properly treated. A better understanding of factors that influence
health management and quality of life of individuals with diabetes is important. Knowledge
gained from this study could be used in designing interventions for more efficient management
of diabetes and prevention of complications.

If you decide to participate, you will be asked to complete a series of questionnaires that will
take approximately 30 minutes. You will be paid $10 for your participation.

The data for this project are being collected anonymously. Neither the researchers nor
anyone else will be able to link data to you. Each survey will be assigned a case number,
and names or other personal identifiers will not be asked in the questionnaires. Data will
be stored for a period of 3 years after the completion of this research on password protected
computers at Michigan State University. Data will remain confidential including both paper
and computer-stored records.

Your participation in this study may contribute to the understanding of factors that could
influence health management of diabetes, and ultimately impact quality of life of people
107

living with this condition. There are no foreseeable risks associated with participation in
this study.

Please note that your participation in this project is voluntary and you have the right to
skip any question(s) you do not want to answer. You may discontinue your participation at
any time. You will not be penalized or lose benefits that you would normally be entitled to
if you refuse to participate or decide to stop participating in the study.

If you agree to participate, you must be given a signed copy of this document and a written
summary of the research.

You may contact Eniko Rak, Principal Investigator, by phone at 517-402-0897 any time
you have questions about the research. You may contact the SJMHS Institutional Review
Board Coordinator, at 734-712-5470 if you have questions about your rights as a research
participant.

Signing this document means that the research study has been described to you, you have had
all your questions about the study answered, and that you voluntarily agree to participate.

Printed Name of Participant

Signature of Participant/Date

Signature of Person Providing Information/Principal Investigator/Date

108

Appendix C
ADVERTISEMENT FLYER

109

110

Appendix D
INSTRUMENTS

111

TOFHLA (N)
Please answer questions 1 through 17 on the blank line provided after the question:
These are directions you or someone else might be given at the hospital. Please read each
direction to yourself then answer the questions that follow about what it means.
GARFIELD IM
16 Apr 93
FF941858 Dr. LUBIN, MICHAEL
PENICILLIN VK
250 MG 40/0
Take one tablet by mouth four times
a day
02

(4 of 40)

1. If you take your first tablet at 7:00 a.m., when should you take the next one?
________________________________________________________________________
2. And the next one after that?
________________________________________________________________________
3. What about the last one for the day, when should you take that one?
________________________________________________________________________
Have a look at this one:
GARFIELD IM
16 Apr 92
FF941861 Dr. LUBIN, MICHAEL
AMIOXICILLIN LIQ
125MG/5ML 15OML1/0
Refrigerate-Shake well; discard
after March 15, 1993
02 12

(1 of 1)

4. Could you take that medicine on July 10, 1993?
________________________________________________________________________

112

Here is another direction you might be given:
GARFIELD IM
16 Apr 93
FF941860 Dr. LUBIN, MICHAEL
METHOTREXATE
2.5MG

10/0

Take every third day.
08 11 14 31

(1 out of 10)

5. If you began taking your medicine Tuesday, when should you take it next?
________________________________________________________________________
6. What day would you take it after that?
________________________________________________________________________
7. Normal blood sugar is 60 – 150. Your blood sugar today is 160. If this were your
score, would your blood sugar be normal today?
________________________________________________________________________

Have a look at this one:

CLINIC APPOINTMENT
LOCATION: 3rd floor

CLINIC: Diabetic

DAY:

Thurs.

nd

DATE: April 2

am
HOUR: 10:20
pm

Issued by
YOU MUST BRING YOUR PLASTIC CARD WITH YOU

8. When is your next appointment?
________________________________________________________________________

113

9. Where should you go?
_______________________________________________________________________
Have a look at this one:
GARFIELD IM
16 Apr 93
FF941860 Dr. LUBIN, MICHAEL
TETRACYCLINE
250 MG

40/0

Important: Finish all this medication
unless otherwise directed by prescriber
02 03 04 11 31

(4 of 40)

10. How many of those pills should you take?
_______________________________________________________________________
Have a look at this one:
GARFIELD IM
28 Dec 92
FF941857 Dr. LUBIN, MICHAEL
PHENOBARBITAL
30 MG

90/2

After two refills or six months from date
of issue, this prescription can only be
refilled by authority of physician.
(2 refills)
01 08

(9 of 90)

11. How many times can you get that prescription refilled?
_______________________________________________________________________
12. When is the date of issue?
_______________________________________________________________________
13. When is six months from the date of issue?
_______________________________________________________________________
114

Have a look at this one:
GARFIELD IM
16 Apr 93
FF941862 Dr. LUBIN, MICHAEL
DOXYCYCLINE
100 MG

20/0

Take medication on empty stomach one
hour before or two to three hours after a
meal unless otherwise directed by your
doctor.
02 11

(0 of 20)

14. If you eat lunch at 12:00 noon, and you want to take this medicine before lunch, what
time should you take it?
_______________________________________________________________________
15. If you forgot to take it before lunch, what time should you take it?
_______________________________________________________________________
16. For clinic care, you only must apply once each six months. Let’s just say the last time
you came to the clinic was on July 12, 1992. When would you have to reapply for
financial aid?
_______________________________________________________________________
Have a look at this one:
You can get at no cost if after deductions your monthly income and
other resources are less than:
$ 581 for a family of one
$ 786 for a family of two
$ 991 for a family of three

$ 1,196 for a family of four
$ 1,401 for a family of five
$ 1,606 for a family of six.

17. Let’s say that after deductions, your monthly income and other resources are $1,129. And
let’s say you have three children. Would you have to pay for your care at that clinic?
_______________________________________________________________________

115

TOFHLA (R)
Instruction: Here are some other medical instructions that you or anybody might see around the
hospital. These instructions are in sentences that have some of the words missing. Where a word
is missing, a blank line is drawn, and 4 possible words that could go in the blank appear just
below it. I want you to figure out which on those 4 words should go in the blank, which word
makes the sentence make senses. When you think you know which one it is, circle the letter in
front of that word, and go on the next one. When you finish the page, turn the page and keep
going until you finish all the pages.
PASSAGE A

Your doctor has sent you to have a _____________ X-ray.
a. stomach
b. diabetes
c. stitches
d. germs

You must have an _______________ stomach when you come for __________.
a. is.
a. asthma
b. empty
b. am.
c. if.
c. incest
d. it.
d. anemia

The X-ray will __________ from 1 to 3 ___________ to do.
a. take
a. beds
b. view
b. brains
c. talk
c. hours
d. look
d. diets

THE DAY BEFORE THE X-RAY.

For supper have only a ____________ snack of fruit, _____________ and jelly,
a. little
a. toes
b. broth
b. throat
c. toast
c. attack
d. thigh
d. nausea

with coffee or tea.
116

After _______________, you must not __________ or drink
a. easy
a. minute,
b. ate
b. midnight,
c. during,
c. drank
d. before,
d. eat

anything at ____________ until after you have ___________ the X-ray.
a. ill
a. are
b. all
b. has
c. each
c. had
d. any
d. was

THE DAY OF THE X-RAY.

Do not eat ________________.
a. appointment.
b. walk-in.
c. breakfast.
d. clinic.

Do not __________, even ____________.
a. heart.
a. drive,
b. breath.
b. drink,
c. dress,
c. water.
d. dose,
d. cancer.

If you have any _____________, call the X-ray ________________ at 616-4500.
a. Department
a. answers,
b. Sprain
b. exercise,
c. Pharmacy
c. tracts,
d. Toothache
d. questions,

117

PASSAGE B
I agree to give correct information to ___________ if I can receive Medicaid.
a. hair
b. salt
c. see
d. ache

I ____________ to provide the county information to
a. agree
b. probe
c. send
d. gain

____________ any
a. hide
b. risk
c. discharge
d. prove

statements given in this _______________ and hereby give permission to
a. emphysema
b. application
c. gallbladder
d. relationship

the _________________ to get such proof. I _______________ that for
a. inflammation
a. investigate
b. entertain
b. religion
c. understand
c. iron
d. county
d. establish

Medicaid I must report any _____________ in my circumstances
a. changes
b. hormones
c. antacids
d. charges

within _________ (10) days of becoming __________ of the change.
a. three
a. award
b. one
b. aware
c. five
c. away
d. ten
d. await

118

I understand ___________ if I DO NOT like the _____________ made on my
a. marital
a. thus
b. this
b. occupation
c. that
c. adult
d. than
d. decision

case, I have the ____________ to a fair hearing. I can ___________ a
a. request
a. bright
b. refuse
b. left
c. wrong
c. fail
d. right
d. mend

hearing by writing or _____________ the county where I applied.
a. counting
b. reading
c. calling
d. smelling

If you __________ TANF for any family _____________ , you will have to
a. wash
a. member,
b. want
b. history,
c. cover
c. weight,
d. tape
d. seatbelt,

___________ a different application form. _____________ , we will use
a. relax
a. Since,
b. break
b. Whether,
c. inhale
c. However,
d. sign
d. Because,

the ____________ on this form to determine your __________________ .
a. lung
a. hypoglycemia.
b. date
b. eligibility.
c. meal
c. osteoporosis.
d. pelvic
d. schizophrenia.

119

PASSAGE C
It has been explained to __________ that during the course of the
a. my
b. me
c. he
d. she

_______________ or procedure, unforeseen conditions may be ________________
a. syphilis
a. revealed
b. hepatitis
b. depressed
c. colitis
c. directed
d. operation
d. notified

that necessitate an extension of the _______________ procedure (s) or
a. appendix
b. another
c. original
d. addict

different procedure (s) than those _________ forth in paragraph 2.
a. get
b. set
c. see
d. go

I therefore, ________________ and request that the above named
a. exercise
b. authorize
c. energize
d. pressurize

______________, his assistants or attending physicians ______________ such
a. perform
a. infection,
b. smear
b. pregnant,
c. insurance,
c. onset
d. physician,
d. stress

120

procedure as are necessary and _______________ in the exercise of professional judgment.
a. undesirable
b. emergency
c. desirable
d. diagnosis

The authority ______________ under this Paragraph 3 shall _______________
a. granted
a. pretend
b. treated
b. extend
c. tested
c. recede
d. proceed
d. X-rayed

to treating all conditions that ______________ treatment and are not known
a. reason
b. refer
c. require
d. relate

_______ the time the operation or ________________ is commenced.
a. cholesterol
a. us
b. menopause
b. be
c. gonorrhea
c. or
d. at
d. procedure

121

WHOQOL-BREF
Instructions: This questionnaire asks how you feel about your quality of life, health, or other
areas of your life. Please answer all the questions. If you are unsure about which response to give
to a question, please choose the one that appears most appropriate. This can often be your first
response.
Please keep in mind your standards, hopes, pleasures and concerns. We ask that you think about
your life in the last two weeks. Please read each question, assess your feelings, and circle the
number on the scale that gives the best answer for you for each question.
Very poor

2.

How would you rate
your quality of life?

How satisfied are you
with your health?

Neither poor
nor good

Good

Very good

1

2

3

4

5

Very
dissatisfied

1.

Poor

Dissatisfied

Neither
satisfied nor
dissatisfied

Satisfied

Very
satisfied

1

2

3

4

5

The following questions ask about how much you have experienced certain things in the last two
weeks.

Not at all

3.

4.

5.
6.

To what extent do you
feel that physical pain
prevents you from
doing what you need to
do?
How much do you
need any medical
treatment to function in
your daily life?
How much do you
enjoy life?
To what extent do you
feel your life to be
meaningful?

A little

A moderate
amount

Very
much

An
extreme
amount

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

122

Not at all

Slightly

A moderate
amount

Very
much

Extremely

7.

How well are you able
to concentrate?

1

2

3

4

5

8.

How safe do you feel
in your daily life?

1

2

3

4

5

9.

How healthy is your
physical environment?

1

2

3

4

5

The following questions ask about how completely you experience or were able to do certain
things in the last two weeks.
Not at all

15.

How well are you able
to get around?

Moderately

Mostly

Completely

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

Very poor

Do you have enough
10. energy for everyday
life?
Are you able to accept
11. your bodily
appearance?
Have you enough
12. money to meet your
needs?
How available to you
is the information that
13.
you need in your dayto-day life?
To what extent do you
14. have the opportunity
for leisure activities?

A little

Poor

Neither poor
nor well

Well

Very well

1

2

3

4

5

The following questions ask you to say how good or satisfied you have felt about various aspects
of your life over the last two weeks.
Very
dissatisfied
16.

How satisfied are you
with your sleep?

Dissatisfied

Neither
satisfied nor
dissatisfied

Satisfied

Very
satisfied

2

3

4

5

1

123

Dissatisfied

Neither
satisfied nor
dissatisfied

Satisfied

Very
satisfied

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

How satisfied are you
20. with your personal
relationships?

1

2

3

4

5

How satisfied are you
with your sex life?

1

2

3

4

5

How satisfied are you
22. with the support you
get from your friends?

1

2

3

4

5

How satisfied are you
23. with the conditions of
your living place?

1

2

3

4

5

How satisfied are you
24. with your access to
health services?

1

2

3

4

5

How satisfied are you
25. with your mode of
transportation?

1

2

3

4

5

Very
dissatisfied
How satisfied are you
with your ability to
17.
perform your daily
living activities?
How satisfied are you
18. with your capacity for
work?
How satisfied are you
19.
with your abilities?

21.

The follow question refers to how often you have felt or experienced certain things in the last
two weeks.
Never
How often do you
have negative feelings,
26. such as blue mood,
despair, anxiety,
depression?

Seldom

Quite often

Very often

Always

1

2

3

4

5

124

EMPLOYMENT INFORMATION
Did you do any paid work in the past?
Yes
No
What was your hourly wage on your last job? _________
How many hours were you working a week on your last job? __________
What is your current employment status?
Employed full-time
Employed part-time

Not working

What is your occupation? __________________________________________
On average, how many hours do you work in a week? ______________
What is your hourly pay? ____________
Do you have benefits?
Yes
No
If yes, please describe: _________________________________________
How do you feel about your current work?
Not at all satisfied
Not satisfied
Satisfied
Very satisfied
How do you feel about having a job in the next year?
Very confident
Fairly confident
Not very confident
Not at all confident
I do not want a job
Do not know
Welfare status
On welfare during past year
On welfare, but not during past year
Never on welfare

125

PDSMS

Disagree

Neither
agree,
nor
disagree

Agree

Strongly
agree

1

2

3

4

5

1

2

3

4

5

1

2

3

4

5

I am able to manage things
4. related to my diabetes as well as
most other people.

1

2

3

4

5

I succeed in the projects I
5. undertake to manage my
diabetes.

1

2

3

4

5

Typically, my plans for
6. managing my diabetes don’t
work out well.

1

2

3

4

5

No matter how hard I try,
7. managing my diabetes doesn’t
turn out the way I would like.

1

2

3

4

5

I’m generally able to accomplish
8. my goals with respect to
managing my diabetes.

1

2

3

4

5

Strongly
disagree
It is difficult for me to find
effective solutions for problems
1.
that occur with managing my
diabetes.
I find efforts to change things I
2. don’t like about my diabetes are
ineffective.
3.

I handle myself well with respect
to my diabetes.

126

CID KNOWLEDGE: DIABETES
True or False (Circle the right answer)
1. Normal blood sugar is between 70 and 140

T

F

2. If you feel thirsty, tired, and weak, it usually means your blood sugar is high

T

F

3. The best time to take insulin or diabetes pills is 15–30 min before a meal

T

F

4. Insulin and diabetes pills make your blood sugar go down

T

F

5. A person with diabetes check their feet for blisters or sore spots every day

T

F

6. When you exercise, your blood sugar goes down

T

F

7. If you feel shaky, sweaty, and hungry, it usually means your blood sugar is low

T

F

8. If you suddenly get sweaty, nervous and shaky, you should eat some form of
sugar

T

F

9. If diabetes is not well controlled, it can injure both kidneys and nerves

T

F

10. You should get your eyes checked every year

T

F

11. If you wake up in the morning, and you feel sick to your stomach and do not
want to eat, you should take half of the usual dose of medicine

T

F

127

BACKGROUND INFORMATION
Gender:

Female

Male

Race/Ethnicity
Non-Hispanic White
African American
Hispanic
Asian/Pacific Islander

Native American
Multiracial
I prefer not to answer

How old are you? ________ years old
What is the highest level of education that you have completed?
No formal education
High school degree or GED
Less than eight grade
Some college, no degree
Some high school, no degree
College degree or more
What is the total annual net income of all persons living in your household?
$20,000 or less
$80,001 - $100,000
$20,001 - $40,000
More than $100,000
$40,001 - $60,000
I don’t know
$60,001 - $80,000
I prefer not to answer
How many people live in your household? _________ people (including yourself)
Language:

Nonnative English speakers

Native English speakers

Which type of diabetes did your doctor say that you have?
type 1 (insulin-dependent or juvenile diabetes)
type 2 (non insulin-dependent or adult onset diabetes)
Other: ____________________
How long ago were you told by a doctor that you had diabetes?
Recently (less than 1 year ago)
More than 5 years ago
A few years ago (2 – 5 years)
At birth

How often does your diabetes prevent you from doing your normal daily activities (ex.
work)? Circle one number.
Never

2

3

4
128

5

6

Frequently

1
Did you ever go to diabetes patient education program?
Yes
No
Attending one right now

7

How would you rate your understanding of diabetes and its treatment? Circle one
number.
Poor
Excellent
1
2
3
4
5
6
7

Are you now taking diabetes pills?

Are you now taking insulin?

What is you height?

Yes

No

Yes

No

_______ feet _______ inches

How much do you weight?

________ pounds

Do you have health insurance?

Who pays for your insurance?
Employer
Military

Yes

No

Self

Medicare

Medicaid

What sources do you use for health information?
None
Newspaper
Magazines
Books
Brochures
Internet
Television

A little

Some

A lot

1
1
1
1
1
1

2
2
2
2
2
2

3
3
3
3
3
3

4
4
4
4
4
4

129

References

REFERENCES

130

References

American Diabetes Association. (2008). Economic costs of diabetes in the U.S. in 2007.
Diabetes Care, 31, 596-615.
Adkins, N. R., & Corus, C. (2009). Health literacy for improved health outcomes: Effective
capital in the marketplace. The Journal of Consumer Affairs, 43, 199-222.
Allegrante, J. P., & Marks, R. (2003). Self-efficacy in management of osteoarthritis. Rheumatic
Diseases Clinics of North America, 29, 747-768.
Arozullah, A. M., Lee, S. D., Khan, T., Kurup, S., Ryan, J., Bonner, M., et al. (2006). The roles
of low literacy and social support in predicting the preventability of hospital admission.
Journal of General Internal Medicine, 21, 140-145.
Baker, D. W. (1999). Editorials: Reading between the lines deciphering the connections between
literacy and health. Journal of General Internal Medicine, 14, 315-316.
Baker, D. W. (2006). The meaning and the measure of health literacy. Journal of General
Internal Medicine, 21, 878-883.
Baker, D. W., Gazmararian, J. A., Williams, M. V., Scott, T., Parker, R. M., Green, D., et al.
(2002). Functional health literacy and the risk of hospital admission among Medicare
managed care enrollees. American Journal of Public Health, 92, 1278-1283.
Baker, D. W., Parker, R. M., Williams, M. V., & Clark, W. S. (1998). Health literacy and the risk
of hospital admission. Journal of General Internal Medicine, 13, 791-798.
Baker, D. W., Williams, M. V., Parker, R. M., Gazmararian, J. A., & Nurss, J. R. (1999).
Development of a brief test to measure functional health literacy. Patient Education and
Counseling, 38, 33-42.
Baker, D. W., Wolf, M. S., Feinglass, J., Thompson, J. A., Gazmararian, J. A., & Huang, J.
(2007). Health literacy and mortality among elderly persons. Archives of Internal
Medicine, 167, 1503-1509.
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change.
Psychological Review, 84, 134-139.
Bankson, H. L. (2009). Health literacy: An exploratory bibliometric analysis, 1997-2007.
Journal of the Medical Library Association, 97, 148-150.
131

Baron, R. M., & Kenny, D. A. (1986). The moderator mediator variable distinction in social
psychological research: Conceptual, strategic, and statistical considerations. Journal of
Personality & Social Psychology, 51, 1173-1182.
Bass, P. F., Wilson, J. F., & Griffith, C. H. (2003). Brief report: A shortened instrument for
literacy screening. Journal of General Internal Medicine, 18, 1036-1038.
Berkman, N. D., DeWalt, D. A., Pignone, M. P., Sheridan, S. L., Lohr, K. N., Lux, L., et al.
(2004). Literacy and health outcomes. Evidence Report/Technology Assessment No. 87
(Technical Report). Research Triangle Park, NC: RTI International - University of North
Carolina Evidence-Based Research Center.
Bishop, M. (2005). Quality of life and psychosocial adaptation to chronic illness and acquired
disability: A conceptual and theoretical synthesis. Journal of Rehabilitation, 71 (2), 5-13.
Bishop, M., Chapin, M. H., & Miller, S. (2008). Quality of life assessment in the measurement of
rehabilitation outcome. Journal of Rehabilitation, 74 (2), 45- 55.
Carmona, R. H. (2006). Health literacy: A national priority. Journal of General Internal
Medicine, 21, 803-803.
Center for Disease Control. (2008). National diabetes fact sheet: General information and
national estimates on diabetes in the United States, 2007. (Technical Report). Atlanta,
GA: U.S. Department of Health and Human Services, Centers for Disease Control and
Prevention.
Center for Disease Control. (2009). National diabetes surveillance system. Centers for Disease
Control and Prevention: Data and trends [article online].
Center for Disease Control. (2011). National diabetes fact sheet: National estimates and general
information on diabetes and prediabetes in the United States, 2011. (Technical Report).
Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease
Control and Prevention.
Cheeseman-Day, J., & Newburger, E. C. (2002). The big payoff: Educational attainment and
synthetic estimates of work-life earnings (Technical Report). U.S. Census Bureau, U.S.
Department of Commerce Economics and Statistics Administration.
Cowie, C. C., Rust, K. F., Byrd-Holt, D. D., Gregg, E. W., Ford, E. S., Geiss, L. S., et al. (2010).
Prevalence of diabetes and high risk for diabetes using A1C criteria in the U.S.
population in 1988-2006. Diabetes Care, 33, 562-569.
Crewe, N. M. (1980). Quality of life: The ultimate goal in rehabilitation. Minnesota Medicine,
63, 586-589.
132

Cutilli, C. C., & Bennett, I. M. (2009). Understanding the health literacy of America: Results of
the National Assessment of Adult Literacy. Orthopedic Nursing, 28, 27-34.
Davis, T. C., Long, S. W. L., Jackson, R. H., Mayeaux, E. J., George, R. B., Murphy, P. W., et
al. (1993). Rapid Estimate of Adult Literacy in Medicine. Clinical Research and
Methods, 25, 391-395.
Davis, T. C., Wolf, M. S., Bass, P. F., Middlebrooks, M., Kennen, P., Baker, D. W., Bennett, C.
L., Durazo-Arvizu, R., Bocchini, A., Savory, S. & Parker, R. (2006). Low literacy
impairs comprehension of prescription drug warning labels. Journal of General Internal
Medicine, 21, 847-851.
DeWalt, D. A., Berkman, N. D., Sheridan, S., Lohr, K. N., & Pignone, M. P. (2004). Literacy
and health outcomes. A systematic review of the literature. Journal of General Internal
Medicine, 19, 1228-1239.
DeWalt, D. A., Davis, T. C., Wallace, A. S., Seligman, H. K., Bryant-Shilliday, B., Arnold, C.
L., et al. (2009). Goal setting in diabetes self-management: Taking the baby steps to
success. Patient Education and Counseling, 77, 218-223.
DeWalt, D. A., & Pignone, M. P. (2005). Reading is fundamental: The relationship between
literacy and health. Archives of Internal Medicine, 165, 1943-1944.
Department of Health and Human Services. (2000). Healthy people 2010. Understanding and
improving health and objectives for improving health. 2 vols. (Tech. Rep.). Washington,
DC: U.S. Government Printing Office (GPO) available at www.healthypeople.gov.
Dooley, D., Catalano, R., & Hough, R. (1992). Unemployment and alcohol disorder in 1910 and
1990: Drift versus social causation. Journal of Occupational and Organizational
Psychology, 65, 277-290.
Dooley, D., Catalano, R., & Wilson, G. (1994). Depression and unemployment: Panel findings
from the epidemiologic catchment area study. American Journal of Community
Psychology, 22, 745-765.
Dooley, D., Fielding, J., & Levi, L. (1996). Health and unemployment. Annual Review of Public
Health, 17, 449-465.
Erickson, W., & Lee, C. (2008). 2007 Disability status report: United States. (Technical Report).
United States. Ithaca, NY: Cornell University Rehabilitation Research and Training
Center on Disability Demographics and Statistics.
Faul, F., Erdfelder, E., Lang, A. G., & Buchner, A. (2007). G*Power 3: A flexible statistical
power analysis program for the social, behavioral, and biomedical sciences. Behavior
Research Methods, 39, 175-191.
133

Fiscella, K., Franks, P., Gold, M. R., & Clancy, C. M. (2000). Inequality in quality addressing
socioeconomic, racial, and ethnic disparities in health care. The Journal of the American
Medical Association, 283, 2579-2584.
Fowler, M. J. (2010). Diabetes: Magnitude and mechanisms. Clinical Diabetes, 28 (1), 42-47.
Gazmararian, J., Baker, D., Parker, R., & Blazer, D. G. (2000). A multivariate analysis of factors
associated with depression: Evaluating the role of health literacy as a potential
contributor. Archives of Internal Medicine, 160, 3307-3314.
Gazmararian, J., Williams, M., Peel, J., & Baker, D. (2003). Health literacy and knowledge of
chronic disease. Patient Education and Counseling, 51, 267-275.
Glasgow, R. E., Toobert, D. J., Riddle, M., Donnelly, J., Mitchell, D. L., & Calder, D. (1989).
Diabetes-specific social learning variables and self-care behaviors among persons with
type II diabetes. Health Psychology, 8, 285-303.
Golin, C. E., Liu, H., Hays, R. D., Miller, B. C. K., L. G., Ickovics, J., Kaplan, A., & Weger, N.
S. (2002). A prospective study of predictors of adherence to combination antiretroviral
medication. Journal of General Internal Medicine, 17, 756-765.
Hornichter, R. D. (1995). Diabetes mellitus. In M. Brodwin, F. Tellez, & S. K. Brodwin (Eds.).
Medical, psychosocial and vocational aspects of disability (1st ed., pp. 285-296). Athens,
GA: Elliott & Fitzpatrick, Inc.
Hosmer, D. W., & Lemeshow, S. (2000). Applied logistic regression. New York: Wiley.
Houts, P. S., Bachrach, R., Witmer, J. T., Tringali, C. A., Bucher, J. A., & Localio, R. A. (1998).
Using pictographs to enhance recall of spoken medical instructions. Patient Education
and Counseling, 35, 83-88.
Houts, P. S., Witmer, J. T., Egeth, H. E., Loscalzo, M. Z., & Zabora, J. R. (2001). Using
pictographs to enhance recall of spoken medical instructions II. Patient Education and
Counseling, 43, 231-242.
Johnson, M. O., Chesney, M. A., Goldstein, R. B., Remien, R. H., Catz, S., & Gore-Felton, C,
Charlebois, E., & Morin, S. F. & NIMH Healthy Living Project Team* (2006). Positive
provider interactions, adherence self-efficacy, and adherence to antiretroviral medications
among HIV-infected adults: A mediation model. AIDS Patient Care and STDs, 20, 258268.
Kalichman, S. C., Ramachandran, B., & Catz, S. (1999). Adherence to combination antiretroviral
therapies in HIV patients of low health literacy. Journal of General Internal Medicine,
14, 267-273.
134

Kalichman, S. C., & Rompa, D. (2000). Emotional reactions to health status changes and
emotional well-being among HIV-positive persons with limited reading literacy. Journal
of Clinical Psychology in Medical Settings, 7, 203-211.
Kirsch, I. S., Jungeblut, A., Jenkins, L., & Kolstad, A. (2002). Adult literacy in America: A first
look at the results of the National Adult Literacy Survey (NALS). (Technical Report).
Washington, DC: National Center for Education Statistics, U.S. Department of
Education.
Kripalani, S., Paasche-Orlow, M. K., Parker, R. M., & Saha, S. (2006). Advancing the field of
health literacy. Journal of General Internal Medicine, 21, 804-805.
Kripalani, S., & Weiss, B. D. (2006). Teaching about health literacy and clear communication.
Journal of General Internal Medicine, 21, 888-890.
Kutner, M., Greenberg, E., Jin, Y., & Paulsen, C. (2006). The health literacy of America’s
adults: Results from the 2003 National Assessment of Adult Literacy. (Technical Report).
Washington, DC: U.S. Department of Education, Office of Educational Research and
Improvement National Center for Education Statistics.
Lincoln, A., Paasche-Orlow, M. K., Cheng, D. M., Lloyd-Travaglini, C., Caruso, C., Saitz, R., et
al. (2006). Impact of health literacy on depressive symptoms and mental-health related
quality of life among adults with addiction. Journal of General Internal Medicine, 21,
818-822.
Livneh, H. (2001). Psychosocial adaptation to chronic illness and disability: A conceptual
framework. Rehabilitation Counseling Bulletin, 44, 151-160.
Mancuso, C. A., & Rincon, M. (2006). Impact of health literacy on longitudinal asthma
outcomes. Journal of General Internal Medicine, 21, 813-817.
Marrone, J., & Golowka, E. (1999). If work makes people with mental illness sick, what do
unemployment, poverty and social isolation cause? Psychiatric Rehabilitation Journal,
23, 187-193.
Nakahara, R., Yoshiuchi, K., Kumano, H., Hara, Y., Suematsu, H., & Kuboki, T. (2006).
Prospective study on influence of psychosocial factors on glycemic control in Japanese
patients with type 2 diabetes. Psychosomatics, 47, 240-246.
Nurss, J., Parker, R. M., Williams, M. V., & Baker, D. W. (2001). Test of Functional Health
Literacy in Adults (TOFHLA) Manual. Peppercorn Books and Press, Inc., Hartford, MI.
Nutbeam, D. (2000). Health literacy as a public health goal: A challenge for contemporary health
education and communication strategies into the 21st century. Health Promotion
International, 15, 259-268.
135

Parker, R. M., Baker, D. W., Williams, M. V., & Nurss, J. R. (1995). The Test of Functional
Health Literacy in Adults: A new instrument for measuring patients’ literacy skills.
Journal of General Internal Medicine, 10, 537-541.
Parker, R. M., & Gazmararian, J. A. (2003). Health literacy: Essential for health communication.
Journal of Health Communication, 8, 116-118.
Powell, C. K., Hill, E. G., & Clancy, D. E. (2007). The relationship between health literacy and
diabetes knowledge and readiness to take health actions. The Diabetes Educator, 33, 144151.
Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect effects
in simple mediation models. Behavior Research Methods, Instruments, & Computers, 36,
717-731.
Ratzan, S. C., & Parker, R. M. (2000). Introduction. in: National Library of Medicine Current
Bibliographies in Medicine: Health literacy. Bethesda MD: National Institutes of Health,
Department of Health and Human Services.
http://www.nlm.nih.gov/archive/20061214/pubs/cbmlhliteracy.pdf.
Rehabilitation Research and Training Center. (2010). Annual Disability Statistics Compendium:
2010 (Tech. Rep.). United States. Ithaca, NY: Cornell University Rehabilitation Research
and Training Center on Disability Statistics and Demographics.
Saha, S. (2006). Editorial: Improving literacy as a means to reducing health disparities. Journal
of General Internal Medicine, 21, 893-895.
Sarkar, U., Fisher, L., & Schillinger, D. (2006). Is self-efficacy associated with diabetes selfmanagement across race/ethnicity and health literacy? Diabetes Care, 29, 823-829.
Saunders, J. L., Leahy, M. J., McGlynn, C., & Estrada-Hernandez, N. (2009). Predictors of
employment outcomes for persons with disabilities: An integrative review of potential
evidence-based factors. Journal of Applied Rehabilitation Counseling, 37, 2, 3-20.
Schillinger, D., Bindman, A., Wang, F., Stewart, A., & Piette, J. (2004). Functional health
literacy and the quality of physician-patient communication among diabetes patients.
Patient Education and Counseling, 52, 315-323.
Schillinger, D., Piette, J., Grumbach, K., Wang, F., Wilson, K., Daher, C., et al. (2003). Closing
the loop: Physician communication with diabetic patients who have low health literacy.
Archives of Internal Medicine, 163, 83-90.

136

Sequist, T. D., Fitzmaurice, G. M., Marshall, R., Shaykevich, S., Sarfran, D. G., & Ayanian, J. Z.
(2008). Physician performance and racial disparities in diabetes mellitus care. Archives of
Internal Medicine, 168, 1145-1151.
Stoops, N. (2004). Current population reports: Educational attainment in the United States: 2003
(Tech. Rep.). U.S. Census Bureau, U.S. Department of Commerce Economics and
Statistics Administration.
Sudore, R. L., Landefeld, C. S., Perez-Stable, E. J., Bibbins-Domingo, K., Williams, B. A., &
Schillinger, D. (2009). Unraveling the relationship between literacy, language
proficiency, and patient-physician communication. Patient Education and Counseling,
75, 398-402.
Sudore, R. L., Yaffe, K., Satterfield, S., Harris, T. B., Mehta, K. M., Simonsick, E. M., et al.
(2006). Limited literacy and mortality in the elderly. The Health, Aging, and Body
Composition Study. Journal of General Internal Medicine, 21, 806-812.
van der Bilj, J., van Poelgeest-Eeltink, A., & Shortridge-Baggett, L. (1999). Psychometric
properties of the Diabetes Management Self-Efficacy Scale for patients with type 2
diabetes mellitus. Journal of Advanced Nursing, 30, 352-359.
Wallace, A. S., Seligman, H. K., Davis, T. C., Schillinger, D., Arnold, C. L., Bryant-Shilliday,
B., Freburger, J. K. & DeWalt, D. A. (2009). Literacy-appropriate educational materials
and brief counseling improve diabetes self-management. Patient Education and
Counseling, 75, 328-333.
Wallston, K., Rothman, R., & Cherrington, A. (2007). Psychometric properties of the Perceived
Diabetes Self-Management Scale (PDSMS). Journal of Behavioral Medicine, 30, 395401.
Weiss, B. D., Francis, L., Senf, J. H., Heist, K., & Hargraves, R. (2006). Literacy education as
treatment for depression in patients with limited literacy and depression. A randomized
controlled trial. Journal of General Internal Medicine, 21, 823-828.
World Health Organization. (1998). World Health Organization Quality of Life Brief: User
manual (Tech. Rep.). Geneva, Switzerland: World Health Organization.
Wild, S., Roglic, G., Green, A., Sicree, R., & King, H. (2004). Global prevalence of diabetes:
Estimates for the year 2000 and projections for 2030. Diabetes Care, 27, 1047-1053.
Williams, M. V., Baker, D. W., Parker, R. M., & Nurss, J. (1998c). Relationship of functional
health literacy to patients’ knowledge of their chronic disease. A study of patients with
hypertension and diabetes. Archives of Internal Medicine, 158, 166-172.

137

Wolf, M. S., Gazmararian, J. A., & Baker, D. W. (2005). Health literacy and functional health
status among older adults. Archives of Internal Medicine, 165, 1946-1952.
Wylie-Rosett, J. (2009). The diabetes epidemic: What can we do? Journal of the American
Dietetic Association, 109, 1160-1162.
Zorn, M., Allen, M. P., & Horowitz, A. M. (2004). Current bibliographies in medicine 2004-1:
Understanding health literacy and its barriers (Technical Report). Available at:
http://www.nlm.nih.gov/pubs/cbm/healthliteracybarriers.html.

138