HEALTH LITERACY AND CLINICAL PREVENTIVE BEHAVIOR AMONG OLDER AFRICAN AMERICANS IN ALLEGHENY COUNTY

by

Kristin L. Macfarlane

BS, BA, University of Pittsburgh, 2012

Submitted to the Graduate Faculty of

Behavioral and Community Health Sciences

Graduate School of Public Health in partial fulfillment

of the requirements for the degree of

Master of Public Health

University of Pittsburgh

2012

ii

UNIVERSITY OF PITTSBURGH

GRADUATE SCHOOL OF PUBLIC HEALTH

This essay is submitted

by

Kristin Macfarlane

on

December 20, 2012

and approved by

Essay Advisor:Steven M. Albert, PhD, MSPH ______________________________________Professor and ChairDepartment of Behavioral and Community Health SciencesGraduate School of Public HealthUniversity of Pittsburgh

Essay Reader:Jeanine Buchanich, PhD ______________________________________Research Assistant ProfessorDeputy Director, Center of Occupational Biostatistics and EpidemiologyDepartment of BiostatisticsGraduate School of Public HealthUniversity of Pittsburgh

iii

Copyright © by Kristin L. Macfarlane

2012

The public health relevance of this essay is to contribute to the body of research that seeks to

elucidate the causal link between health literacy and health outcomes. This study also addresses

the national issue of low clinical preventive services utilization among older adults.

Low functional health literacy is associated with negative health outcomes, but the causal

link as to why this is the case is not well understood. Clinical preventive service use has been

identified as a possible mechanism, but there is a dearth of research concerning this topic. For

older adults, who face higher risk for concomitant illnesses and higher risk for disability, clinical

preventive service use is a key component of retaining health and functional independence.

These services, however, are highly underutilized by older adults.

This study utilizes secondary, cross-sectional baseline data collected from the Boosting

Minority Involvement (BMI) study, a community-based cohort study intended to increase

minority participation in research. Participants included ninety-one African American men and

women aged 60 years and older, living in low-income community settings. Health literacy was

measured with the Short Test of Functional Health Literacy in Adults (S-TOFHLA) and clinical

preventive behavior was assessed as a percentage of clinical tests in which the participant

engaged as identified by the Prevention in Practice Report:“10 Keys” of Healthy Aging survey.

iv

Steven M. Albert, PhD

HEALTH LITERACY AND CLINICAL PREVENTIVE BEHAVIOR AMONG OLDER AFRICAN AMERICANS IN ALLEGHENEY COUNTY

Kristin L. Macfarlane, MPH

University of Pittsburgh, 2012

Multivariate linear regression analysis showed that health literacy did not significantly

predict percent clinical preventive behavior, controlling for covariates. The number of

comorbidites one experienced emerged as the only significant predictor of clinical preventive

behavior. A Spearman’s Rho correlation revealed that functional health literacy and number of

comorbidities were positively associated with percent clinical preventive behavior. Comorbidity

index was not related to one’s level of health literacy.

While this study is a good first step, further research is necessary to elucidate the

relationship between functional health literacy and clinical preventive behavior in an effort to

establish a tenable avenue for intervention to improve health outcomes.

v

TABLE OF CONTENTS

1.0 FUNCTIONAL HEALTH LITERACY.............................................................................1

1.1 HEALTH LITERACY IN THE UNITED STATES..........................................2

1.2 MEASURING HEALTH LITERACY................................................................3

1.3 HEALTH OUTCOMES........................................................................................4

1.4 VULNERABLE POPULATIONS........................................................................5

1.5 THEORY & BEHAVIOR.....................................................................................7

1.6 CLINICAL PREVENTIVE SERVICES UTILIZATION AMONG

OLDER ADULTS.................................................................................................9

1.7 THE ROLE OF THE PUBLIC HEALTH PROFESSIONAL........................10

1.8 PURPOSE OF THIS RESEARCH PROJECT.................................................12

2.0 METHODS..........................................................................................................................14

2.1 DATA COLLECTION........................................................................................14

2.2 MEASURES.........................................................................................................15

2.2.1 Percent Clinical Preventive Behavior...........................................................15

2.2.2 S-TOFHLA......................................................................................................16

2.2.3 Covariates........................................................................................................17

2.3 STATISTICAL ANALYSIS...............................................................................17

3.0 RESULTS............................................................................................................................20

vi

3.1 BASELINE CHARACTERISTICS...................................................................20

3.2 DESCRIPTIVE STATISTICS...........................................................................22

3.3 PERCENT CLINICAL PREVENTIVE BEHAVIOR.....................................24

3.3.1 Predictors of percent clinical preventive behavior......................................24

3.3.2 Correlation across percent clinical preventive behavior............................26

3.4 S-TOFHLA SCORE............................................................................................27

3.4.1 Predictors of S-TOFHLA score.....................................................................27

3.4.2 Correlation between comorbidity index and S-TOFHLA score................29

4.0 DISCUSSION......................................................................................................................30

4.1 HEALTH LITERACY AND CLINICAL PREVENTIVE BEHAVIOR.......30

4.2 COMORBIDITY INDEX AND CLINICAL PREVENTIVE BEHAVIOR...31

4.3 STUDY LIMITATIONS.....................................................................................33

4.4 FUTURE DIRECTIONS.....................................................................................34

APPENDIX A: INSTRUMENTS...............................................................................................37

APPENDIX B: DATA ANALYSIS............................................................................................41

BIBLIOGRAPHY........................................................................................................................42

vii

LIST OF TABLES

Table 1. Characteristics of study population, N=91......................................................................21

Table 2. Percent clinical preventive behavior by gender...............................................................23

Table 3. Multivariate linear regression on percent clinical preventive behavior...........................24

Table 4. Correlation across percent clinical preventive behavior..................................................26

Table 5. Mulitvariate linear regression on S-TOFHLA score.......................................................28

Table 6. Correlation of comorbidity index and S-TOFHLA score................................................29

Table A.1. Prevention in Practice Report: “10 Keys” of Healthy Aging survey...........................37

Table A.2. Selection of items used to assess health literacy from the S-TOFHLA......................40

Table B.1. Tests of normality........................................................................................................41

viii

PREFACE

The author wrote this essay in an attempt to contribute to the theoretical underpinnings of

the link between health literacy and health outcomes and as an exercise in applying the concepts

learned as a student of the University of Pittsburgh’s Graduate School of Public Health. The

author’s interest in both senior health and educational intervention to improve health spawned

the selection of this specific topic

The author would like to thank Dr. Steven, M. Albert for his constant mentorship and

support and Dr. Jeanine Buchanich for her statistical expertise and guidance. The author would

also like to acknowledge her husband, Brandon Lenoir, and her mother, Sandra Macfarlane, for

their support and their contribution as proofreaders.

ix

1.0 FUNCTIONAL HEALTH LITERACY

This project explores the pivotal role health literacy plays in the preventive health of elderly

African Americans. For the purpose of this study, 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” (United States Department of Health

and Human Services [DHHS], 2011; Ratzan & Parker, 2000; Nielsen-Bohlman, Panzer, &

Kindig, 2004). This definition encompasses not only one’s ability to acquire information through

speaking, writing, and communication but also one’s ability to comprehend and successfully

apply information to one’s own health management.

Health literacy underlies one’s ability to perform routine tasks in the healthcare system,

such as following medical instructions, taking prescription medication appropriately, and

completing health insurance forms (Williams, Parker, & Baker, 1995; Baker, Parker & Williams,

1996; Powers, 1998; Kutner, Greenberg, Jin, & Paulson, 2006). Seemingly straightforward

undertakings like communicating one’s symptoms to a health provider and comprehending

appointment slips become difficult when a person functions below a proficient level of health

literacy, and can result in adverse health outcomes (Nielsen-Bohlman et. al., 2004).

Without an understanding of the language of health, Americans may find it difficult to

make decisions such as where to seek services and which treatment options to pursue. Arguably,

one’s level of health literacy impacts one’s ability to navigate today’s complex health care

system and to actively participate in one’s own health care, which is increasingly necessary as

1

the delivery of health related services shifts to a model of consumerism in which decision-

making power is delegated to individuals (McCray, 2012; Parker, 2000). Unfortunately, limited

health literacy is a problem in the United States (Berkman et al., 2011).

1.1 HEALTH LITERACY IN THE UNITED STATES

According to the National Assessment of Adult Literacy study, which dedicated a portion of the

study to the examination of health-specific literacy skills, only 12% of adults 16 and older in the

United States have a proficient level of health literacy, meaning that these individuals have the

optimal skill set to execute complex activities like finding abstract information to solve problems

and synthesizing and analyzing multiple pieces of information from a document (Kutner et al.,

2006). Participants were classified into one of four performance levels based upon their

assessment score as follows (ordered highest to lowest): proficient, intermediate, basic, and

below basic. Most Americans, 53%, have intermediate health literacy, which indicates that they

are able to complete moderately challenging tasks such as making simple inferences about

complex information and use it to solve problems. The results also show that 22% of Americans

function at a basic level where they are able to understand and use information that is short and

simple for everyday tasks, and 14% percent have below a basic level, meaning that their skills

range from nonliterate to understanding only simple forms of information (Kutner et al., 2006)

Overall, as level of health literacy decreased in the NAAL study, so did levels of self-

reported health, which indicates the 88% of Americans who fall below the proficient level of

literacy may have difficulties comprehending and applying health information to decisions that

affect their health (Kutner et al. 2006). We know that low functional health literacy contributes

2

to poor health outcomes, health disparities and higher healthcare costs and we can infer from the

NAAL study that a majority of Americans experience suboptimal health outcomes and a larger

economic burden as a result of limited literacy skills (Agency for Healthcare Research and

Quality [AHRQ], 2011; Berkman et al., 2004; Howard, Sentell, & Gazmararian, 2006; Friedland,

1998; Lee, 1999). Increasing functional health literacy levels may therefore be an advantageous

mechanism through which to enhance the well-being of the American public.

As such, numerous organizations have identified increasing health literacy as a national

priority. Healthy People, a nationwide health-promotion and disease prevention program,

recognized improving consumer health literacy as an objective of both 2010 and 2020 (DHHS,

2000; DHHS, 2011). The Institute of Medicine (IOM), the National Institutes of Health (NIH)

and the American Medical Association (AMA) have also promoted health literacy research and

intervention development as a priority (Berkmen et al., 2011). The United States Department of

Health and Human Services (DHHS) also released a National Action Plan to Improve Health

Literacy, which emphasizes a multidisciplinary approach to making health information

accessible and understandable (DHHS, 2010).

1.2 MEASURING HEALTH LITERACY

The Short Test of Functional Health Literacy in Adults (S-TOFHLA), is a common tool used by

researchers to evaluate one’s level of health literacy (Baker, Williams, Parker, Gazmararian, &

Nurss, 1999). This tool is a shortened version of the original 67-item Test of Functional Health

Literacy in Adults (TOFHLA), and is commended for its brevity, high degree of validity and

reliability, and high degree of correlation to other health literacy tests (Baker et al., 1999). S-

3

TOFHLA is generally the tool of choice as it takes nearly half the time to complete when

compared to the TOFHLA. It is superior to tests such as the Rapid Estimate of Adult Literacy in

Medicine (REALM) because it is able to test comprehension within context as opposed to simply

testing the ability to pronounce words in isolation (Baker et al., 1999)

The S-TOFHLA integrates common vernacular and situations that occur in the healthcare

setting into questions, and classifies participants according to their prospective level of health

literacy as follows (highest to lowest): adequate, marginal, inadequate. Those who have an

adequate level of health literacy are able to complete and understand most of the situations and

materials they encounter in the health setting. Marginal health literacy indicates that one may

have difficulty with complex tasks, written materials and medical instructions. Those with

inadequate health literacy may find it difficult to perform simple tasks such as reading and

interpreting prescription labels and appointment slips (AMA, 1999; Parker, 2000).

1.3 HEALTH OUTCOMES

Health literacy is an important determinant of health and has even been suggested to play a more

important role in predicting health than age, level of education, race, income and employment

status (Berkman et al., 2011). Studies have shown that low health literacy is linked to poorer

understanding and management of health conditions, lower adherence to medication, less

preventive care, and lower levels of participation in health-related decision making, which

certainly contribute to adverse health outcomes (Gazmararian et al., 2003; Williams et al., 1998a;

Williams et al. 1198b; Scott et al., 2002; Berkman et al., 2004). Low health literacy is also

associated with higher hospitalization rates and emergency room visits, higher risk of mortality,

4

and poorer overall physical and mental health (Baker et al., 2004; Wolf et al., 2005). The causal

link for why inadequate health literacy is related to such negative health outcomes, however,

remains elusive.

Among the mechanisms proposed by the current literature are disease knowledge, self-

efficacy, self-management skills, risk behaviors, and appropriate/timely use of clinical

preventive services, and differences in healthcare access (Schillinger et al., 2002; Baker, 2002;

Scott, et al., 2002; Wolf, Gazmararian & Baker, 2007; Gazmararian, Williams, Baker & Peel,

2003). In a literature review of health literacy, healthcare use and health outcomes, The Agency

for Healthcare Research and Quality (AHRQ) called for further research to determine

mechanisms as to how exactly health literacy impacts health outcomes (Berkman et al., 2004).

1.4 VULNERABLE POPULATIONS

Enhancing health literacy has been proposed as a means to reduce or eliminate racial and

socioeconomic health disparities, which is a major national goal (Sudore et al., 2006; Saha ,2006;

Howard, Sentell & Gazmararian, 2006, Sentell & Haplin, 2006). Racial minorities, seniors, low-

income individuals, and those with less than a high school education experience a

disproportionate burden of morbidity and mortality in the United States (AHRQ, 2011). Many of

these subpopulations also exhibit lower levels of functional health literacy and may be

particularly vulnerable to poor health outcomes for this reason (AHRQ, 2004; AHRQ, 2011).

These populations include seniors, racial minorities (specifically African American [AAs], and

Hispanic populations), those living below the poverty level, those who have less than high school

equivalency, and those who do not speak English as a first language (Kutner et al., 2006;

5

Berkman et al., 2011). Since functional health literacy has proven to be a better predictor of

health status and health behavior than demographic and socioeconomic variables, this may be a

tenable avenue to closing gaps in health inequalities (Scott et al. 2002; Williams et al., 1998a;

Bennet et al., 1998; Lindau et al., 2002).

Seniors comprise a particularly vulnerable sub-population as limited health literacy is

prevalent among older adults and appears concurrently with an increased risk of acquiring

multiple chronic diseases and related disabilities that accompany an increase in age (Baker,

Gazmararian, Sudano, & Patterson, 2000; Williams et al., 1995; Weiss, Reed & Kligman, 1995;

Kramarow, Lubitz, Lentzner, & Gorina, 2007; Hoffman, Rice & Sung ,1996). Elderly patients

with chronic diseases and limited health literacy have less knowledge of their disease and worse

self-management skills than those that operate at a proficient level (Williams, et al 1998a). Those

with limited literacy also experience more difficulty in interpreting health messages and taking

prescription medication properly, which can seriously impact one’s health status (Ratzan &

Parker, 2000).

When faced with managing a chronic condition and/or multiple comorbidities, it is

essential to have the knowledge and skills required to maintain health and prevent adverse

outcomes. Considering that health literacy generally declines as one ages (Baker, Gaznararuabm

Sydabim, & Patterson, 2000; Benson & Foreman, 2002), it is likely that these individuals

experience greater difficulty in managing their health and therefore suffer from worse health

outcomes than their younger counterparts.

The identification of these subgroups is important when considering the fact that health

literacy may play a role in attenuating health disparities by acting as a direct correlate of health

outcomes (Sudore et al., 2006; Saha, 2006). By modifying levels of health literacy within

6

disproportionately burdened subgroups, poor health outcomes may be reduced concurrently with

health inequalities, but we must first understand the link between health literacy and health

outcomes within these groups.

1.5 THEORY & BEHAVIOR

Most of the proposed mechanisms linking health literacy to health outcomes, as listed at the end

of section 1.2, are either constructs found in behavioral change and education theories (i.e.

knowledge and self-efficacy), or are actual behaviors (see Glanz, Rimer, &Viswanath, 2008).

General behavioral modification, therefore, offers a more comprehensive mechanism to explore

this relationship, and it is worth examining the role of health literacy in prompting behavioral

change as a means of attenuating negative health outcomes.

Health literacy is associated with health behaviors (Cho, Lee, Arozullah, Critteden,

2008), defined by Gochman (1997) as “those personal attributes such as beliefs, expectations,

motives, values, perceptions, and other cognitive elements; personality characteristics, including

affective and emotional states and traits; and overt behavior patterns, actions, and habits that

relate to health maintenance, to health restoration, and to health improvement” (p. 169).

Although this definition is daunting as many components appear to contribute to one’s health

behavior, theory succeeds in simplifying these factors into constructs that are used to explain

how behavioral change works.

Common substantiated theories used to describe behavioral modification are the health

belief model, the theory of reasoned action, the transtheoretical model, and social cognitive

theory (Glanz et al., 2008). Many behavioral interventions combine constructs from multiple

7

theories. Theory-based interventions are more effective in promoting and sustaining behavioral

change than those that are not (Doak, Doak, Leanord & Root, 1996). Thus, health literacy

interventions that utilize theory to promote behavioral modification will increase the likelihood

of success. Before constructing an intervention based on theoretical constructs, however, we

must identify behaviors that are substantially impacted by health literacy.

There is evidence that low health literacy is associated with lower use of services that are

primarily delivered within a clinical setting, which includes screening tests, counseling, and

immunizations (Scott et al., 2002). Clinical preventive services are essential to maintaining and

improving health by detecting disease in earlier, more treatable stages and preventing the onset

and/or progression of illness (Benson & Aldrich, 2012). These services can also reduce the risk

of chronic disease, which is currently the primary cause of mortality, in addition to reducing

healthcare costs (Maciosek et al., 2010; Benson & Aldrich, 2012; Cohen & Neumann, 2009). For

older adults, who face a higher risk for concomitant illnesses and higher risk for disability,

clinical preventive service use is a key component of remaining healthy and independent

(Kramarow, Lubitz, Lentzner & Gorina, 2007; Hoffman et al., 1996,).

Despite the established benefits and the fact that most of these services are covered by

Medicare and Medicaid, clinical preventive services are highly underutilized by older adults,

which leads us to the question of what role health literacy may play in the decisions to engage in

these preventive clinical behaviors (Centers for Medicare & Medicaid Services, 2011; Cho et al.,

2008; Centers for Disease Control and Prevention [CDC],2012; CDC, 1998).

8

1.6 CLINICAL PREVENTIVE SERVICES

UTILIZATION AMONG OLDER ADULTS

Older adults are the fastest growing age group in the United States as the baby boomers are

transitioning into their retirement years, making the role of clinical preventive services

increasingly important to keeping this population healthy and functionally independent (CDC,

2009). Only 25% of those aged 50-64 and fewer than 40% of those 65 and older are up to date on

the core clinical preventive services recommended by the U.S. Preventative Services Task Force

(USPSTF), which encompass a number of screening tests and immunizations (DHHS 2010;

CDC 2012). Increasing the proportion of older adults who are up to date on these services has

been deemed a priority by Healthy People 2020; increasing health literacy among seniors may be

an effective way to address this issue (Tebo, 2011; DHHS, 2011).

Functional health literacy is associated with an increased likelihood of engaging clinical

preventive behaviors including cancer screening and vaccinations ( Scott, et al., 2002; Lindau, et

al., 2001; Friedman, Corwin, Dominick, & Rose, 2009, Baker et al., 1997). It is therefore no

surprise that higher literacy is also associated with earlier diagnosis of disease, which decreases

the risks related to limited treatment options (Bennett et al., 1998, CDC, 2012). Clinical

preventive services are also important for the health maintenance of those who already have a

chronic condition, as individuals with hypertension or diabetes need to monitor their blood

pressure or blood glucose on a regular basis (Benson & Aldrich, 2011).

Those with lower health literacy, however, may experience literacy-related barriers to

preventive service utilization such as poor understanding of screening, negative attitudes about

clinical services, limited understanding of risk factors and the role of preventive behavior, and

fear or embarrassment of discussing clinical services with health professionals (Berkman et al.

9

2004; American Medical Association [AMA], 1999; Lindau et al., 2001; Rogers, Wallace &

Weiss, 2006; Friedman et al., 2009; Baker et al.,1996, CDC, 1999; Dolan et at., 2004). Another

concern is that many older adults are unaware of USPSTF recommendations (AHRQ, 2004).

Thus, it would appear that improving health literacy may increase use of clinical preventive

services by combating the barriers described above. There is, however, an overall dearth of

research concerning the relationship between health literacy and clinical preventive services

(AHRQ, 2011).

The CDC acknowledges that, “one challenge is to educate older adults about the value of

preventive services and motivate them to get the clinical preventive services they need” (Benson

& Aldrich, 2012). If people are able to access and apply health-related information, they can

acquire the knowledge, positive attitudes, motivation, and/or the self-efficacy necessary to make

the decision to seek and utilize these clinical services. This is where the role of public health

professionals comes into play.

1.7 THE ROLE OF THE PUBLIC HEALTH

PROFESSIONAL

The public is constantly bombarded with complex health information presented as verbal and

written guidelines, regulations, educational materials, and medical instructions from which

individuals need to pull and apply information that is relevant to their own lives. One must

construct a knowledge base that serves as a foundation to evaluate information quality, analyze

risks and benefits, calculate dosages, interpret test results, and seek additional relevant health

10

information. This is a lot to ask of a population in which half of adults find it difficult to

understand and use health information (AMA, 1999, Kutner et al., 2006).

Moreover, many sources of health-related information, such as educational materials and

health messages, are written above the reading level of the average American or are presented in

a way that is not accessible (Davis et al., 1990; Meade, Diekmann, & Thornhill, 1992). While

much of this discussion has focused on the individual, public health professionals must also take

responsibility for recognizing and helping those with limited functional health literacy. This is

particularly important because it is not uncommon for those with poor literacy skills to conceal

difficulties to avoid shame, embarrassment, or stigmatization (Parikh et al., 1996; Baker et al.,

1996). Health professionals need to find and act on the most effective avenues to create a more

health literate nation.

DHHS’s National Action Plan to Improve Health Literacy (NAPIHL) encourages solving

this problem through multiple disciplines that operate at many levels of the socio-ecological

model. Systematic goals for improvement include increasing healthcare access and

communication, as well as supporting the partnering of stakeholders and the development of

policy changes. The NAPIHL also highlights the role of public health professionals in

development and dissemination of health information that is easy to understand, useable, and

culturally and linguistically relevant. Of primary importance is the construction of actionable

health messages; otherwise the message can become irrelevant or confusing to the public. DHHS

supports an increase in adult education efforts as well an increase in research that focuses on the

development and dissemination of effective health literacy interventions. Evidence-based health

literacy interventions are largely lacking (AHRQ, 2011)

11

1.8 PURPOSE OF THIS RESEARCH PROJECT

This study examines the relationship between functional health literacy and clinical preventative

behavior among African Americans (AAs) aged 60 and older and who live in low income

housing in Allegheny County. This study targets demographics that typically possess lower

levels of health literacy, lower rates of clinical prevention service utilization, and poor health

outcomes.

Studies have shown that functional health literacy is related to health outcomes, but the

causal link is not definitively understood. Clinical preventive service utilization has been

proposed as a possible mediating factor as these services play a role in preventing onset and

progression of disease as well as detecting illness in its early treatable form, although research is

limited.

We know that older adults who engage in clinical preventive behaviors have a higher

probability of remaining healthy and functionally independent (Cranksaw, Rabiner, & O’Keefe

2003). If health literacy is identified as a substantial contributing factor, it may be a viable

method of increasing preventive service utilization, thus increasing positive health outcomes.

Studying components that contribute to individual clinical preventive services is not

generalizable to all clinical preventive services. For example, poor knowledge and

misperceptions about prostate cancer screening do not necessarily translate as barriers to blood

pressure screening. Older adults are susceptible to multiple chronic conditions, thus

comprehensive approaches to disease prevention and management may be more valuable than

targeting individual diseases and the relevant protective behaviors. As such, examining the

knowledge, motivation and intentions behind overall clinical preventive behaviors may be more

relevant to developing effective interventions. Although most studies observe health literacy in

12

relation to each individual clinical procedure or test, this research assumes holistic approach and

evaluates total clinical preventive behavior.

This method considers whether or not individuals with higher health literacy are more

likely to engage in general clinical preventive behavior, which suggests the use of skills like

information seeking and informed decision-making. These skills also underlie behavioral

constructs such as knowledge and motivation relative to individual clinical tests/procedures. To

the author’s knowledge, this is the first study that investigates overall clinical preventive

behavior in relation to functional health literacy.

In an effort to contribute to the body of research that influences health literacy

interventions, this study tests the following hypotheses:

Hypothesis 1

1Ho: Health literacy does not predict clinical percent behavior, holding

covariates constant.

1Ha: Health literacy does predict clinical percent behavior, holding covariates

constant.

Hypothesis 2

2Ho: Health literacy is not positively associated with percent clinical prevention

behavior.

2Ha: Health literacy is positively associated with percent clinical prevention

behavior.

13

2.0 METHODS

2.1 DATA COLLECTION

This study utilizes secondary data collected from the Boosting Minority Involvement (BMI)

study conducted at the University of Pittsburgh for which Steven M. Albert, Chair of the

Department of Behavioral and Community Health Sciences at the University of Pittsburgh

Graduate School of Public Health, granted access and permission for use. The objective of BMI

was to increase research participation of senior African Americans (AAs) living in low income

communities in health assessments. Recruitment sites included four community centers and four

low-income senior housing buildings sponsored by Allegheny County, which were located in

Pittsburgh, PA. These areas were targeted due to a high concentration of AAs and a

geographical placement within low income areas.

Eligible participants lived in county-sponsored senior housing or attended community

centers. The study sample for this research included AA adults aged 60 years and older.

Participants were excluded if they were blind, rendering them unable to complete the S-

TOFHLA. The BMI protocol was approved by the University of Pittsburgh Institutional Review

Board prior to implementation and all participants provided written informed consent.

14

2.2 MEASURES

Although BMI was designed as a cohort study with measurements taken at baseline, 6 months,

and 12 months, this research uses only baseline data and is therefore cross-sectional in nature.

Assessments were conducted via in-person interviews and all measures were determined by self-

report.

2.2.1 Percent Clinical Preventive Behavior

Percent clinical preventive behavior was assessed by selecting clinical-specific items from the

Prevention in Practice Report: “10 Keys” to Healthy Aging survey, which was a key component

of an educational initiative to promote healthy aging in McKeesport, Pennsylvania (Robare et al.,

2011; Newman et al., 2010). The items used for this study are highlighted in yellow in Table

A.1 in Appendix A. This survey was designed by the Center for Aging and Population Health-

Prevention Research Center (CAPH-PRC) within the University of Pittsburgh’s Graduate School

of Public Health and was based on the 10 dimensions necessary to maintain health in older

adults, which were founded on epidemiological, clinical, and laboratory research. The clinical

dimensions that were used included prevention of bone loss and muscle weakness, control of

blood pressure, regulation of blood sugar, participation in cancer screening, regular

immunizations, and lowering cholesterol.

Percent clinical preventive behavior was calculated for each individual by dividing the

sum of clinical preventive behaviors by the number of clinical tests for which one was eligible.

Answers to each item were dichotomized (yes or no) to indicate if the person received the test. A

percentage score as opposed to a sum is more appropriate due to variations in clinical test

15

relevance as there are gender specific tests (i.e. mammography) and eligibility requirements (i.e.

immunizations).

Clinical preventive behaviors used to calculate an overall percentage consisted of

knowing one’s blood pressure, blood sugar, LDL cholesterol; receipt of an annual influenza

vaccination and pneumococcal vaccination within one’s lifetime; acquisition a bone mineral

density test, colon cancer screening test within one’s lifetime; and attainment of a prostate

specific antigen (PSA) test for males, and for females, a mammogram within the past year and a

pap test or pelvic exam in the past 2 years. Although some items ask for test result rather than

whether the test was done, awareness of the results suggests that the relevant clinical test was

employed.

2.2.2 S-TOFHLA

The S-TOFHLA was used to assess participants’ level of health literacy (Baker et al.,

1999). This instrument utilizes a modified Cloze procedure for 36 items; every fifth to seventh

word of a health-related passage is deleted and replaced with a blank space. The participant

must select a word to fill the blank from a multiple choice bank of four answers for which there

is only one correct answer. Table A.2 in Appendix A displays a number of S-TOFHLA items

used to assess health literacy as an example. Items for which the participant selected the correct

answer were coded as “1” while incorrect or incomplete answers received a “0.” Correct

answers were summed to generate a score for each individual on a scale that ranges from 0-36

and classifies each participant’s level of health literacy as follows: 0-16 items correct =

inadequate, 17-22 items correct = marginal, and 23-36 items correct = adequate.

16

2.2.3 Covariates

Gender, age, and education (dichotomized into <high school and >high school) were included as

explanatory variables as these variables have been demonstrated to be related to preventive

health service use among seniors (CDC, 1998). A comorbidity index was calculated by

summing self-reported illness, which included cerebrovascular disease, coronary heart disease,

congestive heart failure, hypertension, diabetes mellitus, depression, COPD or asthma,

osteoporosis, depression, macular degeneration, glaucoma, peripheral neuropathy, Parkinson’s

disease, and history of cancer. The comorbidity index was an important component for which to

adjust because of its relationship to clinical prevention behaviors. For example, a person with

diabetes may be more likely than others to keep track of their blood sugar than other participants.

Likewise, a person with a previous cancer diagnosis may engage in cancer screening as a result

of a higher perceived susceptibility.

2.3 STATISTICAL ANALYSIS

Prior to statistical analysis, the data were cleaned to detect, correct, and remove inconsistencies.

Cleaning, descriptive statistics and statistical analyses were performed using IBM SPSS

Statistics 19.

A Shapiro-Wilk test was used to assess the normality of percent clinical prevention

behavior and S-TOFHLA score to facilitate the selection of statistical analysis methods. This

test revealed that neither variable came from a normally distributed population (see Table B.1 in

Appendix B).

17

Multivariate linear regression was employed to test the hypothesis that health literacy

predicts percent clinical preventive behavior. Predictor variables were forced into models based

on sound theoretical reasons for their selection (see Studenmund & Cassidy, 1987). I chose this

method because normality is not required for linear regression analysis and because this method

was the most theoretically justifiable model (Field, 2009). Two models were employed, the first

of which included age, education, gender and S-TOFHLA scores predictors of clinical preventive

behavior. The second model included comorbidity index to account for the association between

particular conditions and relevant clinical preventive services.

Non-parametric correlation analysis was used to explore the independent relationship

between S-TOFHLA and comorbidity index across percent preventive behavior. Correlation

differs from linear regression analysis because it tests for association between two variables; one

cannot be used to predict the other (Field, 2009). Due to the non-normal distribution of percent

clinical preventive behavior, a Spearman’s rho correlation was used to detect significant,

monotonic associations that may be overlooked in linear regression analysis. A one-tailed test

was employed due to theoretical justification to suspect a positive relationship between each

variable (S-TOFHLA score and comorbidity index) and percent clinical preventive behavior.

The result of regression and correlation analysis between S-TOFHLA score and percent

clinical preventive behavior spawned an exploratory analysis of the relationship between

comorbidity index and S-TOFHLA score. Number of comorbidities may be a moderator of the

relationship between health literacy and preventive service use. Patients with health conditions

have been shown to access health-related information, thus allowing for the modification of

one’s health literacy (Castleton et al., 2010). An individual with multiple illnesses may have

greater clinical health-related knowledge or self-efficacy due to more exposure to clinical

18

services and the healthcare setting out of necessity. Therefore, multivariate linear regression

analysis was used to explore comorbidity index as a predictor of health literacy. Two models

were employed to predict S-TOFHLA. The first model included age, gender, education and

comorbidity index. The second model included percent clinical preventive behavior to control

for the possibility that increased exposure to the clinical setting may also influence health

literacy.

The use of a nonparametric correlation test was necessary due to the non-normal

distribution of S-TOFHLA score. A Spearman’s rho correlation was used to detect any

significant relationship between comorbidity index and S-TOFHLA score that may not be

apparent in regression analysis. A one-tailed test was employed due to theoretical justification to

suspect a positive relationship between comorbidity index and S-TOFHLA.

19

3.0 RESULTS

3.1 BASELINE CHARACTERISTICS

Baseline characteristics of the 91 AA participants are presented in Table 1. The sample was

predominantly female (80.2%); 18.7% of the population had less than a high school level of

education, 35.2% had a high school equivalency, and 46.2% of participants had educational

experience beyond high school. Median S-TOFHLA score and percent clinical preventive

behavior were reported because these variables were not normally distributed. The median S-

TOFHLA score was 34 on a scale of 0-36 points. Most participants had an adequate level of

health literacy (92.3%). Only 3.3% of participants had an inadequate level of health literacy and

4.4% had a marginal level of health literacy. The median percent clinical preventive behavior

was 62.5% out of a possible 100%, with a range of 0-89%. The mean comorbidity index was 3.2

self-reported illnesses per participant. Most participants had hypertension (78%) and/or arthritis

(68.1%). Congestive heart failure (7.7%), depression (6.6%) and Parkinson’s disease (2.2%)

were the least reported illnesses.

20

Table 1. Characteristics of study population, N=91

Characteristic N (%)Age (mean years + S.D.) 72.4 + 7.4; range: 60-95Gender

Female 73 (80.2)Male 18 (19.8)

EducationAsfdasf< High school 17 (18.7)

High school equivalent 32 (35.2)Any vocational school 4 (4.4)Any college 35 (38.5)Graduate or professional school 3 (3.3)

S-TOFHLA Score (mean + S.D.) 31.6 + 7; median 34; range 0-36Health literacy level

Inadequate 3 (3.3)Marginal 4 (4.4)Adequate 84 (92.3)

Percent clinical prevention behavior 57.8 + 20.1; median 62.5; range 0-89(mean + S.D.)

Comorbidity index (mean + SD) 3.2 + 1.9; range 0-10Comorbidities

Cerebrovascluar diseaseG

10 (11.0)Diabetes mellitus 31 (34.1)Hypertension 71 (78.0)Coronary heart disease 22 (24.0)Congestive heart failure 7 (7.7)Macular degeneration 12 (13.2)Glaucoma 15 (18.5)Depression 6 (6.6)Arthritis 62 (68.1)Osteoporosis 13 (14.3)COPD or asthma 15 (16.5)Peripheral neuropathy 12 (13.2)Parkinson’s disease 2 (2.2)History of cancer 13 (14.3)

21

3.2 DESCRIPTIVE STATISTICS

Descriptive statistics for each item used to calculate percent clinical preventive behavior are

presented by gender in Table 2. A larger percentage of females (74%) knew their last blood

pressure reading when compared with their male counterparts (50%). A higher percentage of

women (60.3%) also had a bone mineral density test compared to men (38.9%) A higher

percentage of males received an annual influenza vaccine (88.9%) and pneumococcal vaccine

(94.1%) compared to females (65.8% for each). Men had a slightly higher percentage (72.2%) of

colon cancer screening when compared to women (60.3%) The percentage of men and women

who knew their last blood sugar reading and who knew their LDL cholesterol was similar at

about 33% and 67% respectively. A two-tailed t-test was run for each of the items listed above to

detect differences in test/procedure acquisition by gender. Knowing one’s blood pressure was

the only item that approached statistical significance (p< .10). Gender differences, however, were

not present at the standard level of alpha=.05 for any one item.

For gender-specific tests, 94.4% of men had a PSA test within the last year, 72.3% of

women received a mammogram within the past year and 69.6% of women had a papanicolaou

test or pelvic exam within the past two years. As a whole, men had a slightly higher median

percent clinical preventive behavior (62.5) than women (55.6), although a Mann-Whitney U test

revealed that this difference was not statistically significant.

22

Table 2. Percent clinical preventive behavior by gender

Clinical prevention behavior (timeframe of receipt of clinical test specified by instrument)Receipt of test and/or knowledge of test result

Male (%) Female (%) Total (%)

Last blood pressure readingYes 50.0 74.0 69.2No 50.0 26.0 30.8

Last blood sugar readingYes 33.3 32.9 33.0No 66.7 67.1 67.0

LDL cholesterolYes 11.1 12.3 12.1No 88.9 87.7 87.9

Usually receive influenza vaccine (annual)Yes 88.9 65.8 70.3No 11.1 34.2 29.7

Pneumococcal vaccine (ever)Yes 94.1 65.8 71.1No 5.9 34.2 28.9

Bone mineral density test (ever)Yes 38.9 60.3 56.0No 61.1 39.7 44.0

Colon cancer screening (ever)Yes 72.2 60.3 62.6No 27.8 39.7 37.4

Received prostate specific antigen (PSA) tests Test (within past year) 94.4

Yes 5.6No

Received mammogram (within past year)Yes 72.6No 27.4

Received pap test or pelvic exam(within past two years)

Yes 69.6No 31.0

Total clinical prevention behavior (percent) 60.1+16.2 57.2 + 21 57.8 + 20.1(mean + S.D, median) median 62.5

62.5median 55.6 median 62.5

23

3.3 PERCENT CLINICAL PREVENTIVE BEHAVIOR

3.3.1 Predictors of percent clinical preventive behavior

Linear regression analysis was used to test the hypothesis that S-TOFHLA predicts percent

clinical preventive behavior, adjusting for covariates. Two models were employed, both of which

incorporated age, education, and gender as demographic characteristics shown to be related to

clinical preventive behavior (CDC, 1998). Comorbidity index was added as a predictor in the

second model. Table 3 displays the results of both models.

Table 3. Multivariate linear regression on percent clinical preventive behavior

Model 1a Model 2b

Predictor Variable

Coefficient (B-value)

Standardized Coefficient

P-value Coefficient (B-value)

Standardized Coefficient

P-value

S-TOFHLA score

.410 .143 .224 .431 .150 .185

Age .248 .091 .406 .243 .089 .397Education 5.711 .112 .316 4.503 .088 .413Gender -5.915 -.118 .306 -6.689 -.133 .230Comorbidity

index3.11** .288 .006

Constant 27.159 .344 20.501 .459F .873 2.343p-value .484 .048*R .198 .348R2 .039 .121Adjusted R2 -.006 .069Dependent: percent clinical preventive behaviora Predictors: (constant), S-TOFHLA score, age, education, genderb Predictor’s (constant), S-TOFHLA score, age, education, gender, comorbidity index*p<.05, p<.01

Model 1 does not significantly explain the variation in percent clinical preventive

behavior. Additionally, neither S-TOFHLA score, age, education, nor gender significantly

predicted percent clinical preventive behavior when controlling for the respective covariates.

24

When comorbidity index is added, the F statistic (2.343) is significant, (p=.048, p< .05),

specifying that this model is better at predicting the outcome of percent clinical preventive

behavior than using the mean as a best guess. Model 2 significantly predicts 12.1% of the

variance of percent clinical preventive behavior (R2=.121). Generalizability of the model is

assessed by calculating the difference between R2 and the adjusted R2 (.121-.069=.052), which

indicates that 5.2% less of the variance of the percent clinical preventive behavior would be

explained had this model been derived from a population when compared with this sample.

Thus, this model is fairly generalizable.

The only statistically significant predictor in Model 2 is comorbidity index (p=.006,

p< .01), while holding all other variables constant, suggesting that this variable may play a more

substantial role in predicting percent clinical preventive behavior than S-TOFHLA score, age,

gender and education. A positive relationship exists; an increase of one medical condition is

associated with an increase in 3% in preventive behavior (B-value=3.111), which is a relatively

large magnitude. This model indicates that participants that have a larger number of illnesses

engage in a higher percentage of clinical preventive behavior.

It is worth noting that even though S-TOFHLA score is not significant at alpha=.05, this

variable approaches significance in Model 2 and has a higher standardized coefficient (Beta

= .288), and a lower p-value (.185) than education, gender and age. This trend is also present in

Model 1 and suggests that clinical preventive behavior increases as health literacy increases.

Health literacy may therefore play a more important role as a predictor than other demographic

covariates, which is consistent with the findings of other studies (Scott et al. 2002; Williams et

al., 1998a; Bennet et al., 1998; Lindau et al., 2002). However, a lack of statistical significance

25

leads us to accept the null hypothesis that health literacy does not predict clinical percent

behavior, holding covariates constant.

3.3.2 Correlation across percent clinical preventive behavior

Correlation analysis was used to test the hypothesis that S-TOFHLA score is positively

associated with percent clinical preventive behavior. Comorbidity index was added to the

analysis as a result of its relationship to preventive behavior, which was discovered in regression

analysis (Section 3.3.1). The results of Spearman’s rho correlation analysis across percent

clinical preventive behavior and S-TOFHLA score is summarized in Table 4.

Table 4. Correlation across percent clinical preventive behavior

VariableSpearman’s rho correlationcoefficient, P-value(non-parametric)

S-TOFHLA score .221, .018*Comorbidity index .229, .029*

(one-tailed test), *=p<.05, **p<.01

Percent clinical preventive behavior is significantly correlated with S-TOFHLA score,

(rs=.221, p<.05). The coefficient of determination is small (rs2=.049) and accounts for 4.9% of

the proportion of variance in the ranks that two variables share. We reject the null hypothesis

and conclude that health literacy and percent preventive behavior have a positive, monotonic

relationship, meaning that as one increases, so does the other.

Comorbidity index is also significantly correlated with S-TOFHLA score (rs =.229,

p<.05) and accounts for 5.2% of the proportion of variance in the ranks that the two variables

share (rs2=.052). This relationship is also positive, but the correlation coefficient of comorbidity

index and percent clinical preventive behavior (.229) is slightly larger than that of the S-

26

TOFHLA score and percent clinical preventive behavior (.221). The magnitude of both

coefficients is relatively small, but the number of illnesses one has appears to have a somewhat

larger association with clinical preventive behavior when compared to health literacy.

3.4 S-TOFHLA SCORE

3.4.1 Predictors of S-TOFHLA score

The results presented in section 3.3 initiated an exploratory analysis of the relationship between

comorbidity index and S-TOFHLA score. Linear regression analysis was used to test the

hypothesis that the number of comorbidities one experiences predicts S-TOFHLA score,

adjusting for covariates. Two models were employed, both of which incorporate age, education,

and gender as demographic characteristics shown to be related to health literacy (Kutner et al.,

2006). Percent clinical preventive behavior was added as a predictor in the second model

because a significant correlation between S-TOFHLA score and clinical preventive behavior was

detected in section 3.3.2. The results are displayed in Table 5.

27

Table 5. Multivariate linear regression on S-TOFHLA score

Model 1a Model 2b

Predictor Variable

Coefficient

Standardized Coefficient

P-value Coefficient Standardized Coefficient

P-value

Comorbidity index

-.076 -.020 .836 -.223 -.060 .562

Age -.058 .095 .544 -.068 -.073 .474Education 2.551 .143 .152 2.283 .128 .211Gender 6.317** .363 .000 6.507** .374 .000Percent clinical

preventive behavior

.048 .137 .185

Constant 20.076* .027 18.685* .040F 4.575 4.05p-value .002 .002R .419* .439*R2 .175 .192Adjusted R2 .137 .145Dependent: S-TOFHLA scorea Predictors: (constant), comorbidity index, age, education, genderb Predictors: (constant), comorbidity index, age, education, gender, % clinical preventive behavior*p<.05, **p<.01

Both Model 1 and Model 2 significantly explain the variation in S-TOFHLA score at

alpha=.05. When percent clinical preventive behavior is added to Model 2, the proportion

variation of S-TOFHLA score explained by the model jumps from 17.5% in Model 1 to 19.2% in

Model 2, specifying that Model 2 is a slightly better predictor of S-TOFHLA score than Model 1.

Generalizability of the model is assessed by calculating the difference between R2 and the

adjusted R2 (.192-.145=.047) , which indicates that only 4.7% less of the variance of the percent

clinical preventive behavior would be explained had this model been derived from a population

as opposed to a sample. This model is therefore fairly generalizable.

Gender is the only significant independent predictor of S-TOFHLA score (p=.000,

p< .01). In light of limitations in sample size and distribution, trends among predictor variables

approaching significance were examined. Percent clinical preventive behavior approaches

significance (.185), and has a higher standardized coefficient (Beta=.137) than all other

28

explanatory variables and thus may have more of an effect on health literacy than comorbidity

index, age, and education. Comorbidity index had the smallest standardized coefficient (Beta=

-.060) and a high p-value and therefore does not appear to affect S-TOFHLA score. We

therefore accept the null hypothesis that comorbidity index does not predict health literacy.

3.4.2 Correlation between comorbidity index and S-TOFHLA score

Non-parametric correlation analysis was used to explore the independent relationship between

comorbidity index and S-TOFHLA score. Due to the non-normal distribution of S-TOFHLA

score, a Spearman’s rho correlation was used to detect general associations that may have been

overlooked in linear regression analysis, which is presented in Table 6.. Table 3 showed that

percent clinical preventive behavior was positively associated with S-TOFHLA score and, thus,

was not included in Table 6

Table 6.Correlation of comorbidity index and S-TOFHLA score

IndependentVariable

Spearman’s rho correlationcoefficient, P-value(non-parametric)

Comorbidity index .068, .261

(one-tailed test), *p<.05, **p<.01

Comorbidity index was not statistically associated with S-TOFHLA score. A very small

positive trend does exist (rs=.068), but we conclude that the number of illnesses one has is not

positively associated with one’s level of health literacy.

29

4.0 DISCUSSION

4.1 HEALTH LITERACY AND CLINICAL

PREVENTIVE BEHAVIOR

This study found that functional health literacy does not predict the percentage of

relevant clinical prevention behaviors in which one engages, adjusting for covariates. It appears,

however, that health literacy approaches significance as a predictor. Small sample size may have

limited the study from finding statistical significance. Even so, there is still an 18.5% chance of

obtaining the observed result of the S-TOFHLA score as a predictor variable by chance even if

no effect exists. The relationship is not statistically significant, but the trend is consistent with

the findings of other studies for individual preventive procedures like cancer screening and

immunizations (Peterson et al., 2011; Scott et al., 2002; Lindau et al., 2001).

Although we cannot conclude that health literacy predicts clinical preventive behavior,

these variables are positively correlated, meaning that as one variable increases, so does the

other. The magnitude of the relationship, however, is relatively small (rs =.221), and the analysis

did not control for explanatory variables. A limitation of this statistical test is that it is indicative

only of association and does not express causality; it is not possible to determine the direction of

the relationship in a correlation. In order to elucidate the foundation for this positive correlation,

clinical preventive behavior was also tested as a predictor of health literacy. Although percent

30

clinical preventive behavior was not a significant predictor, the trend once again reflected a

positive association.

The positive regression coefficients may indicate that each partially contributes to the

other. Those who are exposed to one or more clinical preventive behaviors, whether as a result

of contact with health professionals or a heightened awareness of health-related risks, have the

opportunity to increase health-related knowledge at that particular appointment. This potential

increase in health literacy and/or more experience with clinical services may then cause

individuals to engage in more clinical preventive behaviors. Therefore, a cycle may exist where

each variable positively affects the other, which would account for the presence of an

association. Multivariate linear regression appeared to be the best fit for this study, although a

significant, non-parametric correlation suggests the possibility that health literacy and clinical

preventive behavior possess a relationship that may be more accurately expressed by a more

complex regression equation.

One study, however, cannot completely confirm or refute the association between health

literacy and percent clinical preventive behavior, and further research is needed to detangle this

obscure relationship.

4.2 COMORBIDITY INDEX AND CLINICAL

PREVENTIVE BEHAVIOR

Number of comorbidities was the only significant predictor of percent clinical preventive

behavior, while controlling for covariates. It was also independently, positively associated with

clinical preventive behavior in a correlation analysis. When studying the use of clinical tests in a

31

senior population, we must consider the fact that individuals may already have a moderate to

high level of contact with health care professionals due to a higher prevalence of health issues.

This sample had a mean comorbidity index of 3.2, indicating that many individuals were

managing multiple health conditions. Perhaps these individuals were exposed to the healthcare

setting more often out of necessity, and consequently had a greater opportunity for

communication about clinical preventive behavior and health management with their health care

professionals. In fact, studies have documented that physician recommendation or

encouragement may play a substantial role in one’s behavioral intention and decision to engage

in clinical preventive procedures, which may overshadow the effect of health literacy on clinical

preventive behavior if one exists (Guerra, Dominguez, & Shea, 2005, Lewis & Jenson, 1996;

Aiken, West, Woodward, & Reno, 1994; Drake et al., 2004). Another possibility is that the

particular conditions exhibited by this sample were closely related to the specific clinical tests

and procedures used to calculate percent clinical preventive behavior. For example, a number of

clinical procedures are applicable to those with a history of cancer, heart disease and diabetes

mellitus (i.e. cancer screening, and knowing one’s blood pressure and blood sugar), but are not

necessarily as relevant to Parkinson’s and arthritis patients. Hypertension was the most prevalent

condition within the sample and diabetes mellitus ranked third, which lends support to this

explanation.

Comorbidity index neither predicted nor was associated with health literacy. While this

variable predicted percent clinical preventive behavior, it was not likely due to its influence over

health literacy in this study. This reduces the possibility that increased exposure to healthcare

due to necessity might increase exposure to medical terminology and information as previously

theorized.

32

4.3 STUDY LIMITATIONS

This study has a number of limitations. First, this study is restricted to AA men and women aged

60 and older, living in a low-income, community setting and is not representative of the general

population. The results may not be generalizable to other ethnicities, younger individuals, or to

different income levels and living situations. There are limitations in data collection as this study

utilizes secondary data that were not collected for the purpose of this analysis. Additionally, the

cross-sectional nature of the data does not allow for the assessment of temporality or causality.

Characteristics of the sample also act as a limitation. This group had a relatively large

proportion of individuals that had adequate health literacy (92.3%), which contributed to a highly

skewed distribution. Given that only 7 out of 91 participants had below adequate health literacy

and only 3 of those 7 fell into the inadequate category, it is difficult to detect any effects even if

they exist. This could be due to self-selection bias as those with lower health literacy skills may

have opted to not participate in this study. It could also be the result of the relatively high

education levels of this sample (71.3% > high school).

It should be noted that this is the first attempt to calculate overall percent clinical

preventive behavior. As such, internal consistency and reliability have not been tested. A caveat

emerges in the use of the “Prevention in Practice: 10 Keys to Healthy Aging survey.” Questions

that cover blood pressure, blood sugar and LDL cholesterol ask the participants for the result of

the test as opposed to whether or not they received the test. It may be more difficult to recall an

actual number, which may affect the accuracy of the answer. We can, however, infer that

knowledge of these numbers conveys either a receipt of the test or a sense of desire to manage

one’s health, which are both clinically-related preventive behaviors. Future studies can phrase

question to explicitly capture the utilization of services.

33

It is also important to acknowledge that there are items on the survey that address tests

and procedures that do not match U.S. Preventive Services Task USPSTF guidelines (see Moyer,

2012; USPTF, 2002). Guidelines, however, are constantly changing and are sometimes different

across multiple national organizations, which is not only confusing to the general population, but

also impedes the design of a consistent instrument that assesses holistic clinical preventive

behavior. The CAPH-PRC is part of the network for Centers for Disease Control and Prevention

(CDC) Prevention Research Centers and is therefore a reputable resource in the field of healthy

aging As such, clinical items on the survey were deemed relevant to healthy aging and were

included in the analysis regardless of USPSTF recommendation

Lastly, although the sample size is adequate according to the number of predictors (Field,

2009), it was relatively small considering the high variability within S-TOFHLA score and

percent clinical preventive behavior observations. This sample size therefore may not be

adequate for drawing robust conclusions.

4.4 FUTURE DIRECTIONS

Although this study establishes a positive association between health literacy and clinical

preventive behavior and comorbidity index as a significant predictor of percent clinical

preventive behavior, further research is needed to detangle the relationship of these three

variables. The previously described limitations may have reduced the ability to detect effects. It

is notable that other studies have encountered a similar issue of a large proportion of individuals

with high functional health literacy (see Wolf et al., 2007). We, therefore, cannot make

34

definitive conclusions from this study and further research is necessary to clarify the relationship

between health literacy and health outcomes.

Future research should ensure the collection of an adequate sample of those that fall into

inadequate and marginal levels of health literacy. Additionally, the collection of demographics of

those who choose not to participate may be useful in elucidating any self-selection bias and may

provide a foundation for including these individuals in future studies. Most importantly,

longitudinal cohort studies should be employed to detect causality between health literacy,

preventive behaviors and comorbidities. Most of the current health literacy research utilizes

cross-sectional data, which does not provide an adequate understanding of how health literacy is

related to health behavior or health outcomes (Dewalt et al., 2004).

Current research focuses on the identification of constructs related to specific clinical

preventive procedures/tests, which are not generalizable to all clinical preventive behaviors. This

helps us understand barriers that underlie interventions targeted to particular preventive services,

but loses the scrutiny of how people work as health consumers. There is a need for the

development of instruments that measure holistic preventive behavior in order to gain a greater

understanding of how increasing health literacy might impact behavioral change and health

outcomes.

The current health care system has become consumer-centric, and individuals need to

make more proactive decisions regarding their own health. Providing health information is not

enough. Health professionals need to equip the population with the appropriate skills to seek

health-related information and apply that information to make informed decisions. Identifying

how health literacy may influence a person’s decision to engage in general preventive behaviors,

especially among populations who are disproportionately affected by negative health outcomes,

35

is necessary to developing effective and comprehensive interventions. More theoretically-based

studies should address the effect of literacy on behavioral change to further explore the

relationship of health literacy to health outcomes.

36

APPENDIX A

INSTRUMENTS

A.1 PERCENT CLINICAL PREVENTIVE BEHAVIOR

Table A.1 displays items from the Prevention in Practice Report: “10 Keys” to Healthy Aging

survey, which was used to calculate percent clinical preventive behavior. Only clinical-specific

items, which are highlighted in yellow, were used in the analysis.

Table A.1 Prevention in Practice Report: “10 Keys” to Healthy Aging survey

10 Keys to Healthy Aging

10 Key Self RatingOne answer for each question.

1

1

Systolic Blood Pressure

1. What was your last blood pressure? ________/________ Don’t Know

1.1 Are you on blood pressure medication? Yes No Don’t Know

1.2 Do you have Hypertension? Yes No Don’t Know

2

2

Smoking 2. Do you smoke (cigarettes, cigars, pipes, etc)? Yes No Sometimes2.1 If yes, do you want to quit? Yes No Sometimes

37

Table A.1 continues onto page 38

Prostate Cancer (men)

Breast CancerScreenings:

Cervical (women)

Colon Cancer (Men and Women)

3. Have you had a Prostate Specific Antigen (PSA) test in the past year?

Yes No Don’t Know

3._ Have you had a mammogram within the past year? Yes No Don’t Know

Does Not Apply

3._ Have you had a Pap test or Pelvic exam in the past 2 years?

Yes No Don’t Know Does Not Apply

3._ Have you ever been screened for Colon Cancer? Yes No Don’t Know

4

4

Immunizations 4. Do you usually get a yearly flu shot? Yes No

4.1 Have you ever had a pneumonia shot? Yes No Don’t Know

5

5

Blood Glucose 5. What was your last blood sugar reading? __________mg/dl Don’t Know5.1 Do you have Diabetes?

Yes No Don’t Know5.2 Are you on prescribed medication for Diabetes?

Yes No Don’t Know

6

6

LDL Cholesterol 6. What is your LDL cholesterol level? ___________ mg/dl Don’t Know

6.1 Are you on prescribed cholesterol lowering medication?

Yes No Don’t Know

7

7

Physical Activity 7. How many hours a week do you spend performing household chores, preparing meals, and shopping? _________ hours Don’t Know

7.1 Not including the above, how many hours a week do you exercise?

__________ hours Don’t Know

38

Table A.1 continues onto page 39

8

8

Bone Loss &Muscle Weakness

8. Do you think that your height has decreased?Yes No Don’t Know

8.1 Have you ever had a bone density test?Yes No Don’t Know

8.2 Are you on prescribed medication for bone loss?Yes No Don’t Know

8.3 Are you able to walk a ¼ mile ?Yes No Sometimes

9

9

Social Contact 9. Do you have social contact in the community at least once a week?

Yes No Sometimes

1

10

Depression 10. Do you have a good outlook on life?Yes No Sometimes

10.1 Have you been diagnosed with Depression?Yes No Don’t Know

10.2 Are you on medication for Depression?Yes No Don’t Know

39

A.2 HEALTH LITERACY

Table A.2 displays a selection of items from the Short Test of Functional Health Literacy in

Adults. The participants must select the correct answer to fill each blank from a bank of four

questions. For example, the correct answer to ‘Your doctor has send you to have a __________

x ray’ is ‘a. stomach.’ All other answers are considered incorrect.

Table A.2 Selection of items used to assess health literacy from the S-TOFHLA

Your doctor has sent you to have a ___________ x-ray.a. stomachb. diabetesc. stitchesd. germs

You must have an ______________ stomach when you come for _________.a. asthma a. isb. empty b. amc. incest c. ifd. anemia d. it

The x-ray will _________ from 1-3 __________ to do.a. take a. bedsb. view b. brainsc. talk c. hoursd. look d. diets

40

APPENDIX B

DATA ANALYSIS

B.1 DISTRIBUTION

Table B.1 displays tests of normality for S-TOFHLA score and percent clinical preventive

behavior. The Shapiro-Wilk test shows that neither variable comes from a normally distributed

population (p<.05.) These tests help facilitate the selection of statistical tests.

Table B.1 Tests of normality

VariableKolmogorov-Smirnova Shapiro-Wilk

Statistic df Sig. Statistic df Sig.S-TOFHLA score .281 91 .000 .559 91 .000

Percent clinical preventive behavior

.159 91 .000 .933 91 .000

a. Lilliefors Significance Correction

41

BIBLIOGRAPHY

Agency for Healthcare Research and Quality. (2004). Health literacy interventions and outcomes: An updated systematic review (AHRQ Publication No. 04-E007-2). Rockville, MD: Agency for Healthcare Research and Quality. Retrieved November 1, 2012 from http://www.ahrq.gov/downloads/pub/evidence/pdf/literacy/literacyup.pdf

Agency for Healthcare Research and Quality. (2011). 2010 National healthcare disparities report (AHRQ Publication No. 11-0005). Rockville, MD: Agency for Healthcare Research and Quality. Retrieved September 12, 2012 from http://www.ahrq.gov/qual/nhdr10/nhdr10.pdf

Aiken, L.S., West, S.G., Woodward, C.K., & Reno, R.R. (1994). Health beliefs and compliance with mammography: screening recommendations in asymptomatic women. Health Psychology, 13, 122-129.

American Medical Association. (1999). Health literacy : report of the Council on Scientific Affairs. Journal of the American Medical Association, 281(6), 522-557.

Baker, D. W. (2006). The meaning and the measure of health literacy. Journal of General Internal Medicine, 21(8), 878-883.

Baker, D. W., Gazmararian, J. A., Sudano, J., & Patterson, M. (2000). The association between age and health literacy among elderly persons. Journals of Gerontology Series B Psychological Sciences and Social Sciences, 55(6), S368-S374. doi: 10.1093/geronb/55.6.S368

Baker, D.W., Gazmararian, J.A., Williams, M.V., Scott, T., Parker, R.M., Green, D., . . . Peel, J. (2002) Functional health literacy and the risk of hospital admission among Medicare managed care enrollees. American Journal of Public Health, 92(8), 1278-1283.

Baker, D. W., Gazmararian, J. A., Williams, M. V., Scott, T., Parker, R. M., Green, D., . . . Peel, J. (2004). Health literacy and the use of outpatient physician services by Medicare managed care enrollees. Journal of General Internal Medicine, 19(3), 215-220.

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(12), 791-798.

42

Baker, D. W., Parker, R. M., Williams, M. V., Clark, W. S., & Nurss, J. (1997). The relationship of patient reading ability to self-reported health and use of health services. American Journal of Public Health, 87(6), 1027-1030.

Baker, D.W., Parker, R.M., Williams, M.W., Pitkin, K., Parikh, N.S., Coates, W., & Imara, M. (1996). The health care experience of patients with low health literacy. Archives of Family Medicine, 5(6), 329-334.

Baker, D.W., Williams, M.V., Parker, R.M., Gazmararian, J.A., & Nurss, J. (1999). Development of a brief test to measure functional health literacy. Patient Education and Counseling, 38(1), 33-42.

Bennett, C.L., Ferreira, M.R., Davis, T.C., Kaplan, J., Weinberger, M., Kuzel, T., . . .Sartor, O. (1998). Relation between literacy, race, and stage of presentation among low-income patients with prostate cancer. Journal of Clinical Oncology, 16, 3101-3104.

Benson, J.G., & Forman, W.B. (2002). Comprehension of written health care information in an affluent geriatric retirement community: Use of the test of functional health literacy. Journal of Gerontology, 48(2), 93-97.

Benson, W.F., & Aldrich, H., (2012). CDC focuses on need for older adults to receive clinical preventive services. Atlanta, GA: Centers for Disease Control and Prevention. Retrieved October 8, 2012 from http://www.chronicdisease.org/nacdd-initiatives/health-aging/meeting-records.

Berkman, N. D., DeWalt, D. A., Pignone, M. P., Sheridan, S. L., Lohr, K. N., Lux, L., . . . Bonito, A.J. (2004). Literacy and health outcomes (Publication No. 04-E007-2). Rockville, MD: Agency for Healthcare Research and Quality. Retrieved September 5, 2012 from http://archive.ahrq.gov/downloads/pub/evidence/pdf/literacy/literacy.pdf

Berkman, N.D., Sheridan, S.L., Donahue, K.E., Halpern, D.J., Biera, A., Crotty, K., . . . Viswanathan, M. (2011). Health literacy interventions and outcomes: An updated systematic review (AHRQ Publication No. 11-E006). Rockville, MD: Agency for Healthcare Research and Quality. Retrieved Nov. 3, 2012 from http://www.ahrq.gov/downloads/pub/evidence/pdf/literacy/literacyup.pdf

Castleton, K., Fong, T., Wang-Gillam, A., Waqar, M.A., Jeffe, D.B., . . .Govindan, R. (2011). A survey of internet utilization among patients with cancer. Support Care Cancer, 19, 1183-1190.

Centers for Disease Control and Prevention. (1999). Scientific and technical information simply put. Atlanta, GA: Centers for Disease Control and Prevention. Retrieved on October 17, 2012 from http://www.ashg.org/pdf/CDC%20-%20'Scientific%20&%20Technical%20Information%20-%20Simply%20Put'%20%20(1999).pdf

Centers for Disease Control and Prevention. (1998). Influenza and pneumococcal vaccination levels among adults aged >65 years – United States. Morbidity and Mortality Weekly Report, 47(38), 797-802.

43

Centers for Disease Control and Prevention. (2012). Clinical preventive services - aging. Retrieved October 2, from http://www.cdc.gov/aging/services/index.htm

Center for Aging and Population Health. (2011). The “10 keys” to healthy aging resource guide. Pittsburgh, PA: University of Pittsburgh, A CDC Prevention Research Center. Retrieved October 2, 2012 from http://www12.edc.gsph.pitt.edu/CHA_OAEP/documents/ResourceGuide_062911.pdf

Centers for Medicare & Medicaid Services. (2011). Prevention – general information. Retrieved on November 28, 2012 from http://www.cms.gov/PrevntionGenInfo/20_prevserv.asp

Cho, Y.I., Lee, S.D., Arozullah, A.M., Crittenden, K.S. (2008). Effects of health literacy on health status and health service utilization amongst the elderly. Social Science & Medicine, 66(8), 1809-1806.

Cohen, J.T., & Neumann, P. (2009). The cost savings and cost-effectiveness of clinical preventive care. (Research Synthesis Report No. 18). Princeton, NJ: The Robert Wood Johnson Foundation. Retrieved October 2, 2012 from http://www.statereforum.org/sites/default/files/report_final.pdf

Crankshaw, E., Rabiner, D., & O’Keeffe, J. (2003). An overview of programs and initiatives sponsored by DHHS to promote healthy aging: A background paper for the blueprint on aging for the 21st century technical advisory group (TAG) meeting. Washington, DC: Office of Disability, Aging and Long-Term Care Policy, U.S. Department of Health and Human Services.

Davis, T., Crouch, M., Willis, G., Miller, S., & Abehou, D. (1990). The gap between patient reading comprehension and the readability of patient education materials. Journal of Family Practice, 31, 533-538.

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 Interanal Medicine, 19, 1228-1239.

Drake, B.F., Keane, T.E., Mosley, C.M., Adams, S.A., Elder, K.T., Modayil, M.V., . . . Hebert, J.R. (2006). Prostate cancer disparities in South Carolina: Early detection, special programs, and descriptive epidemiology. Journal of the South Carolina Medical Association, 102(7), 241-249.

Doak, L., Doak, C., Leonard, G., & Root, J. (1996). Applying theory in practice, in teaching patients with low literacy skills, second edition. Philadelphia, PA: J.B. Lippincott Company.

Dolan, N.C., Ferreira, M.R., Davis, T.C., Fitzgibbon, M.S., Rademaker, A., Liu, D., . . . Bennet, C.L. (2004). Colorectal cancer screening knowledge attitudes and beliefs among veterans: Does literacy make a difference? Journal of Clinical Oncology, 22(13), 2617-2622.

44

Field, A.(2009). Discovering Statistics Using SPSS: Third Edition. Thousand Oaks, CA: SAGE Publications.

Friedman D.B., Corwin, S.J., Dominick, G.M., & Rose, I.D. (2009). African American men’s understanding and perceptions about prostate cancer: why multiple dimensions of health literacy are important in cancer communication. Journal of Community Health. 34, 449-460.

Gazmararian, J.A., Williams, M.W., Baker, D.W., & Peel, J. (2003). Health literacy and patient knowledge of chronic disease. Patient Education and Counseling, 51(3), 267-275.

Gochman, D. S. (1997). Handbook of health behavior research I, personal and social determinants. New York: Plenum Press, p. 167.

Glanz , K., Rimer, B., & Viswanath, K. (2008). Health Behavior and Health Education: Theory, Research, and Practice. San Francisco, CA: Jossey-Bass Wiley.

Guerra, C.E., Dominguez, F., Shea, J.A. (2005). Literacy and knowledge, attitudes, and behavior about colorectal cancer screening. Journal of Health Communication, 10, 651-663.

Howard, D.H., Sentell, T., & Gazmararian, J.A. (2006). Impact of health literacy on socioeconomic and racial differences in health in an elderly population. Journal of General Internal Medicine, 21(8), 857-861.

Hoffman, C., Rice, D., & Sung, H.Y. (1996). Persons with chronic conditions: Their prevalence and costs. Journal of the American Medical Association, 267, 1473-1479.

Kutner, J., Greenberg, E., Jin, Y., Paulsen, C. (2006). The health literacy of America’s adults: Results from the 2003 national assessment of adult literacy (NCES 2006-483). Washington, DC: Department of Education, National Center for Education Statistics.

Kramarow, E., Lubitz, J.D., Lentzner, H., & Gorina, Y. (2007). Trends in the health of older Americans, 1970–2005. Health Affairs, 26(5), 1417-1425.

Lee, P.P. (1999). Why literacy matters: Links between reading ability and health. Archives of Ophthalmology, 117(1), 100-103. doi:10-1001/pubs.Ophthalmol.-ISSN-0003-9950-117-1-esa8449.

Lewis, S.F. & Jenson, N.M. (1996). Screening signoidoscopy: Factors associated with utilization. Journal of General Internal Medicine. 22, 542-544.

Lindau, S.T., Tomori, C., McCarville, M.A., & Bennett, C.L. (2001). Improving rates of cervical cancer screening and pap smear follow-up for low-income women with limited health literacy. Cancer Investigation, 19(3), 316–323.

Maciosek, M.V., Coffield, A.B., Flottemesch, T.J., Edwards, N.M., & Solberg, L.I. (2010). Greater use of preventive services in us health care could save lives at little or no cost. Health Affairs (Millwood), 29(9), 1656-1660.

45

Meade, C.D., Diekmann, J., Thornhill, D.G. (1992). Readability of American Cancer Society patient education literature. Oncology Nursing Forum, 19, 51-55.

Moyer, V.A. (2012). Screening for prostate cancer: U.S. Preventive Services Task Force recommendation statement. Annals of Internal Medicine, 157(2), 120-134.

Nielsen-Bohlman, L., Panzer, A. M., & Kindig, D. A. (2004). Health literacy: A prescription to end confusion. Washington, DC: National Academies Press.

Newman, A.B., Bayles, C.M., Milas, C.N., TcTigue, K.W., Robare, J.F., Tayle, C.A., . . . Kuller, L.K. (2010.) The 10 Keys to Healthy Aging: Findings From an Innovative Prevention Program in the Community. Journal of Health Aging. 22(5): 547-556.

McCray, A.T. (2012). Promoting health literacy. Journal of the American Medical Informatics Association, 12(2), 152-162.

Parker, R.M. (2000). Health literacy: a challenger for American patients and their health care providers. Health Promotion International, 15(4), 277-283.

Parikh, N.S., Parker, R.M., Nurss, J.R., Baker, D.W., & Williams, M.V. (1996.) Shame and health literacy: The unspoken connection. Patient Education and Counseling, 27(1), 33–39.

Peterson, P.N., Shetterly, S.M., Clarke, C.L., Bekelman, D.B., Chan, P.S., Allen, L.A., . . . Masoudi, F.A. (2011). Health literacy and outcomes among patients with heart failure. Journal of the American Medicine Association. 305(16), 1695-1701. doi:10.1001/jama.2011.512.

Ratzan, S.C., & Parker, R.M. (2000). Introduction. In: National Library of Medicine Current Bibliographies in Medicine: Health Literacy (NLM Pub. No. CBM 2000-1). Bethesda, MD: National Institutes of Health, U.S. Department of Health and Human Services.

Robare, J.F., Bayles, C.M., Newman, A.B., Williams, K., Milas, C., Boudreau, R., . . . Kuller L.H. (2011). The “10 keys” to healthy aging: 24-month follow-up results from an innovative community-based prevention program. Health Education and Behavior, 38(4), 397-388.

Rogers, E.S., Wallace L.S., & Weiss. B.D. (2006). Misperceptions of medical understanding in low-literacy patients: implications for cancer prevention. Cancer, Culture and Literacy, 13(3), 225-339.

Saha, S. (2006). Improving literacy as a means to reducing health disparities. Journal of General Internal Medicine, 21(8), 893-895.

Scott, T.L., Gazmararian, J.A., Williams, M.V., & Baker, D.W. (2002). Health literacy and preventive health care use among Medicare enrollees in a managed care organization. Medical Care, 40(5), 395-404.

46

Schillinger, D., Grumback, K., Piette, J, Wang, F., Osmond, D., Daher, C., . . . Bindman, A.B. (2002). Association of health literacy with diabetes outcomes. The Journal of the American Medical Association. 288(4), 475-482. doi:10.1001/jama.288.4.475.

Sentell, T.L., & Halpin, H.A. (2006) The importance of adult literacy in understanding health disparities. Journal of General Internal Medicine, 21, 862-868.

Smith, R.A., Cokkinides, V., & Brawley, O.W. (2008). Cancer screening in the United States, 2008: A review of current American Cancer Society guidelines and cancer screening issues. A Cancer Journal for Clinicians, 58(3), 161-179.

Studenmund, A.H., & Cassidy, H.J. (1987). Using econometrics: a practical guide. Boston: Little Brown.

Sudore, R.L., Yaffe, K. Satterfield, S. Harris, T.B., Mehta, K.M., . . . Schillinger, D. (2006). Limited literacy and mortality in the elderly: the health, aging, and body composition study. Journal of General Internal Medicine, 21(8), 806-812.

Tebo, G., (2011). Healthy People 2020: Preparing for a new decade, [PowerPoint slides], Retrieved from Office of Disease Prevention and Health Promotion, Department of Health and Human Services, website: www.hrsa.gov/advisorycommittees/shortage/Meetings/20110118/MeetingMaterials/hlthppl20.ppt

Weiss, B.D., Hart, G., McGee, D.L., & D’Estelle, S. (1992). Health status of illiterate adults: Relation between literacy and health status among persons with low literacy skills. American Board of Family Practice, 5(3), 257-264.

White, S., Chen, J., & Atchison, R. (2008). Relationship of preventive health practices and health literacy: A national study. American Journal of Health Behavior, 32(3), 227-242.

Williams, M.V., Baker, D.W., Honig, E.G., Lee, T.H., & Nowlan, A. (1998a). Inadequate literacy is a barrier to asthma knowledge and self-care. CHEST Journal, 114(4), 1008-1015.

Williams, M.V., Baker, D.W., Parker, R.M., & Nurss, J.R. (1998b). 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(2), 166-172. doi:10-1001/pubs.Arch Intern Med.-ISSN-0003-9926-158-2-ioi70066.

Williams, M.V., Davis, T., Parker, R.M., & Weiss, B.D. (2002). The role of health literacy in patient- physician communication. Archives of Family Medicine, 34(5), 383–389.

Williams, M.W., Parker, R.M., Baker, D.W., Perikh, N.S., Pitkin, K., Coates, W.C. & Nurss, J.R. (1995). Inadequate functional health literacy among patients at two public hospitals. The Journal of the American Medical Association, 247(21), 1677-1682.

47

Weiss, B.D., Reed, R.L., & Kligman, E.W. (1995). Literacy skills and communication methods of low-income older persons. Patient Education and Counseling, 25(2),109-119.

Wolf, M.S., Davis, T.C., Cross, J.T., Tomori, C., Green, K.M., & Bennett, C.L. (2004). Health literacy and patient knowledge in a Southern HIV clinic. International Journal of STD & AIDS, 15(11), 747-752.

Wolf, M.S., Davis, T.C., Arozullah, A., Penn, R., Arnold,C., & Bennett, C.L. (2005). Relationship between literacy and HIV treatment knowledge among individuals enrolled in HAART regimens. AIDS Care, 17(7), 863-873.

Wolf, M.S., Gazmararian, J.A., & Baker, D.W. (2007). Health literacy and health risk behaviors among older adults. American Journal of Preventive Medicine, 32(1), 19-24.

United States Department of Health and Human Services. (2000). Healthy People 2010. Washington, DC: U.S. Government Printing Office.

United States Department of Health and Human Services. (2011). Healthy People 2020 summary of objectives, health communication and health IT. Retrieved on September 2, 2012 from http://www.healthypeople.gov/2020/topicsobjectives2020/pdfs/HealthCommunication.pdf

United States Department of Health and Human Services, Office of Disease Prevention and Health Promotion. (2010). National action plan to improve health literacy. Washington, DC: Author.

United States Preventive Services Task Force. (2008). Screening for colorectal cancer: U.S. Preventive Services Task Force recommendation statement. Annals of Internal Medicine, 149(9), 627-637.

United States Preventative Services Task Force. (2002). Screening for prostate cancer, recommendation and rationale. Annals of Internal Medicine, 137(11), 915-916.

48