|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
HEALTH LITERACY: THE VALIDATION OF A SHORT FORM HEALTH LITERACY
SCREENING ASSESSMENT IN AN AMBULATORY CARE SETTING
JOLIE NANCIBETH HAUN
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
O 2007 Jolie Nancibeth Haun
To my mom and family, my mentors, the veterans who serve our country,
my kitties, Teddy Bear Ballgame and Ishtar, who were always with me, and
Jamey Michael Vidal, the man who stood by me.
I want to thank God for allowing me to live in His loving light for lighting my path and
bearing my burdens. I want to acknowledge the spirits and angels that provided me with lessons
to keep me in tune and on track. I want to give thanks and honor my mom, Dian Sally Haun -
my biggest fan, a "great listener," and my confidant. I thank my dad, Joseph Henry Haun Jr., for
giving me the spirit to walk my path. I want to thank my sister, Alesia Forkush, for all those
home-made meals and walks on the bay and my sister, Angela Godbout, for helping me when I
needed it. I also want to thank to my brothers, Greg Bingham, Steven Forkush, and Edward
Godbout. I send love to my little sis' Jessica Haun and little bros' Zach and Keeland. I also thank
and send love to my soul brother Shaun Thomas in BDA. I want to thank my nieces, Alisa
Bingham, Jessica Grant, Crystal Prince, Christie Godbout, Katrina Forkush, Sabrina Forkush,
and nephews, Baron Forkush, Colton Forkush, Shean Godbout, Gregory Scott Bingham, Josh
Bingham, Ryan Bingham, and Shaun Bingham, for reminding me that I'm a role model and
keeping me motivated to better myself for the future of my family.
I want to honor my dear friend Tonya Francis, for staying with me in spirit, giving me
strength, for reminding me of how lucky I am and for the memory of her beautiful smile. I
want to give thanks for my friendship with Brandy Lynn Klopp (aka, Brandylion) the best
friend ever and my "person." I want to thank Chelsie Ford who was there in the beginning and
I'll never forget. I also thank Gwendolyn Ward for reminding me all women are beautiful and to
honor the divine feminine Goddess that I am. I thank my forever friend Robert Hibbet (the book
bag was good luck). I want to express my gratitude for the loving support of my God parents,
Jim and Hope Gordon. I thank Alan Bibbler for helping me grow up. I thank Robby Smith (aka
Captain Nasty) for showing me how to dream 'bigger' and for teaching me how to fly.
I want to thank the fine institution of the University of Florida, for allowing me the
opportunity to make my dreams come true and earn the right to be a "Gator." It is the faculty
and staff at this university and others like it, that provide the past and future generations with the
knowledge and skills to aspire to change the world for the better. It was the belief, wisdom,
guidance, and support provided by my mentors that made me the doctor I am today. I want to
express my eternal appreciation to John Graham-Pole for believing in me and bringing me into
the world of research and teaching me how to laugh and heal. I want to express appreciation for
my chair, Jill Varnes and my co-chair, Virginia Noland-Dodd, who helped make this proj ect
possible and provided me with support throughout. I also thank Barbara Rienzo for being a sage
and blessing me with her teachings. I thank Morgan Pigg Jr., for always being there for a special
talk. I want to express my gratitude for Delores James for giving me a new perspective. I thank
Laura Lang for giving me my work ethic and standards of excellence. I want to express gratitude
for Tristan Johnson my dear friend and mentor. I send love and light to Michael Samuels and
Mary Rockwood-Lane for their spiritual guidance through my invocation of the medicine woman
- to guide my healing practice. I thank Jill Sonke-Henderson for her passion for art and gift for
healing. I thank Patty Donaldson for working on this proj ect with me with motivation and
enthusiasm. I also thank JoAnne Mcleary for taking care of me throughout my doctoral program.
I want to express appreciation for the support of Gale H. and H. Charles Anderson. I could
not have achieved my goals without the help they both so graciously provided throughout the
years. Their role modeling provided me a vision to guide my aspirations and achieve my
accomplishments. And with great honor, appreciation, and all of my heart I thank Jamey
Michael Vidal. I am indebted to this man for his unconditional patience, love, and support. I
want to thank him for carrying me when I could not stand. We did it!
TABLE OF CONTENTS
ACKNOWLEDGMENTS .............. ...............4.....
LIST OF TABLES ................. ...............9..___ .....
LIST OF FIGURES ............ _...... ._ ...............11...
AB S TRAC T ............._. .......... ..............._ 12...
1 INTRODUCTION ................. ...............14.......... ......
Study Obj ectives............... ...............1
Research Questions............... ...............1
Delimitations ................. ...............18...............
Limitations ................. ...............18.................
Assum options .............. ...............18....
Definition of Terms .............. .....................19
Sum m ary ................. ...............20.......... ......
2 LITERATURE REVIEW ................. ...............21................
Defining Health Literacy .............. ....... .. ............2
General Literacy & Health Literacy Assessment .............. ...............22....
Rates of Health Literacy ................. ...............23.......... ....
Impact of Health Literacy ................... ....... ....... ...............25......
Risk Factors Associated with Low Health Literacy .............. ...............33....
Health Literacy and the Medical Encounter ................. ............. ......... ...........3
Identifying Patients with Inadequate Health Literacy Skills .............. .....................3
Patient Participation ................. ...............42.................
Barriers to Patient Participation ................. ...............42................
Patient Education Programs ................. ...............43........... ....
Health Literacy Skills .............. ...............44....
Patient Empowerment. .............. ... .. .......... .. ......... .............4
Creating a Shame-free Environment and Providing Resources .............. ....................4
Theoretical Framework............... ...............4
Systems Theory .............. ...............47....
Ecological Model................. ..... .... .........4
Transtheoretical Model & Stages of Change ................. ...............49...............
Sum m ary ................. ...............50.......... ......
3 METHODOLOGY .............. ...............55....
Research Design .............. ...............55....
Sample Population............... ...............5
Instrum ents .............. ...............56....
Data Collection ................. ...............59.................
D ata Analy sis............... ... ..............6
Research Question # 1.............. ...............61....
Research Question #2............... ...............64...
Research Question #3 ................ ...............65........... ....
Research Question #4............... ...............65...
Instrument Reliability ................. ...............66.......... ......
Sum m ary ................. ...............66.......... ......
4 RE SULT S .............. ...............71....
Research Questions............... ...............7
Research Question # 1 .............. ...............71....
Research Question # 2 ................ ...............75................
Research Question # 3 .............. ...............76....
Research Question # 4 ................ ...............77................
Summary ................. ...............77.................
5 SUMMARY, DISCUS SION, AND RECOMMENDATIONS ................. ......................89
Sum m ary ................. ...............89.......... ......
Discussion ................. ...............90.................
Study Findings ................. ...............91.................
Limitations ................. ...............99.................
Recom m endations.............. ................ .. ...... .. ..........10
Implications for the Role of Health Educators and Health Education ................... ....... 101
Implications for the Role of Providers and other Healthcare Professionals .................. 102
Putting Research into Practice through Policy and Systemic Change ..........................104
Recommendations for Future Research............... ...............10
Conclusion ................ ...............107................
A PARTICIPANT SURVEY .............. ...............109....
B TEST OF FUNCTIONAL HEALTH LITERACY IN ADULTS: SHORT FORM.............11 4
C RAPID ESTIMATE OF ADULT LITERACY IN MEDICINE ............_.. .........__.....120
D BRIEF HEALTH LITERACY SCREENING TOOL ............ ..... ._ ................ 121
E STUDY DATA COLLECTOR ELECTRONIC RECRUITMENT FLYER .....................122
F MEMORANDUM: PROPOSED HEALTH LITERACY SCREENING POSTING ..........123
LI ST OF REFERENCE S ................. ...............126................
BIOGRAPHICAL SKETCH ................. ...............134......... ......
LIST OF TABLES
2-1 General literacy and health literacy screening assessments. ......____ ...... .....__.........51
2-2 Scale for levels of TOFHLA/STOFHLA and general literacy skills ............... .... ........._..52
2-3 Levels for grade equivalent scale of the Rapid Estimate of Adult Literacy in
Medicine (REALM) ................. ...............52.................
2-4 Description of Stages of Change, Self-efficacy & Consciousness Raising. ......................53
3-1 Frequency and percentage of participants recruited at the eight VA ambulatory care
sites. ............. ...............68.....
3-2 Frequency and percentage of participant data collected by each of the volunteer data
collectors. .............. ...............68....
3-3 Administration time for the REALM, STOFHLA, BRIEF, & Participant Survey............69
3-4 Conceptual illustration of the sensitivity, specifieity, false positive, false negative,
positive predictive value, and negative predictive value. ............. .....................6
3-5 Definitions and equations of sensitivity, specifieity, false positive, false negative,
positive predictive value, and negative predictive value. ............. .....................7
4-1 Demographi cal di stributi on by age, gender, educati on, and ethni city ............... .... ...........79
4-2 Frequency of participants' self reported reading ability ....__ ................ ...............79
4-3 Participants' level of health literacy as measured by the BRIEF, REALM, &
STOFHLA. ........._._ ..... ._ ...............80...
4-4 Cross tabulation table for BRIEF and STOFHLA levels for sample data. ........................80
4-5 Cross tabulation table for BRIEF and REALM levels for sample data. ............................80
4-6 Pearson Product Moment correlation coefficients between the REALM, STOFHLA,
& BRIEF Item Scores. ............. ...............80.....
4-7 Areas under the receiver operating characteristic curve for BRIEF and BRIEF items
with REALM as the state variable. .............. ...............80....
4-8 Areas under the receiver operating characteristic curve for BRIEF and BRIEF items
with STOFHLA as the state variable. .............. ...............81....
4-9 Performance of BRIEF in detecting inadequate and marginal health literacy using
STOFHLA as the state variable. ............. ...............8 1....
4-10 Performance of BRIEF in detecting inadequate and marginal health literacy using
REALM as the state variable. ............. ...............82.....
4-11 Stepwise Model: Unstandardized Coefficients, Standardized Regression
Coefficients, t-test Statistics, and Partial r-squares. ....._.__._ .... ... ._ ........._._......83
4-12 Frequencies and proportions of participant responses to items about health literacy
and heath information. ............. ...............83.....
4-13 Frequencies and proportions of participant responses to items about their knowledge,
readiness & confidence related to patient health education ................. ......................83
LIST OF FIGURES
3-1 Analysis of Variance (ANOVA) Equation Model ................. ..............................70
3-2 Multiple Linear Regression Model ................ ...............70........... ...
4-1 Participants' level of health literacy as indicated by the BRIEF, REALM, &
STOFHLA. ........._._....... ...............84....
4-2 The individual BRIEF items for identification of inadequate health literacy using the
Receiver Operating Characteristic (ROC) Curves ................. ..............................85
4-3 The BRIEF health literacy assessment for identification of inadequate health literacy
using the Receiver Operating Characteristic (ROC) Curve ................. ............ .........85
4-4 The individual BRIEF items for identification of inadequate and marginal health
literacy using the Receiver Operating Characteristic (ROC) Curves .............. ................86
4-5 The BRIEF health literacy assessment for identification of inadequate and marginal
health literacy using the Receiver Operating Characteristic (ROC) Curve. ......................86
4-6 The individual BRIEF items for identification of inadequate health literacy using the
Receiver Operating Characteristic (ROC) Curves ....._._................. ............... ....87
4-7 The BRIEF health literacy assessment for identification of inadequate health literacy
using the Receiver Operating Characteristic (ROC) Curve ......__ ............. ...... .........87
4-8 The individual BRIEF items for identification of inadequate and marginal health
literacy using the Receiver Operating Characteristic (ROC) Curves .............. ................88
4-9 The BRIEF health literacy assessment for identification of inadequate and marginal
health literacy using the Receiver Operating Characteristic (ROC) Curve .......................88
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
HEALTH LITERACY: THE VALIDATION OF A SHORT FORM HEALTH LITERACY
SCREENING ASSESSMENT IN AN AMBULATORY CARE SETTING
Jolie Nancibeth Haun
Chair: Jill Varnes
Cochair: Virginia Noland-Dodd
Major: Health and Human Performance
Adverse health outcomes associated with low health literacy affect one in three Americans.
Low literacy also consumes health care resources equivalent to billions of dollars annually.
Many variables are associated with inadequate health literacy yet the exact relationships between
patient variables and health literacy outcomes are unclear. Understanding patient variables
associated with inadequate health literacy can provide insight into disparities in healthcare and
identify priority populations. To prevent personal and system loss, healthcare providers need an
efficient means of identifying patients' literacy skills. My research validated a brief screening
tool in an ambulatory clinical setting against two previously validated health literacy measures.
My study also identified patient variables associated with health literacy level. Finally my study
identified the stage of readiness for the sample participants related to their awareness of health
literacy and their utilization of patient health education.
My study was conducted with 378 veteran participants presenting in eight acute ambulatory
care clinics. The proposed BRIEF screening tool and two previously validated assessments were
significantly correlated with moderate positive correlations, suggesting the BRIEF is clinically
valid. A Principal Component Analysis suggests the BRIEF screen measures one distinct
construct "health literacy" accounting for 60% of score variance. The receiving operator
characteristic curve (ROC) analysis also suggests the BRIEF is a more sensitive measure of
inadequate health literacy than the individual BRIEF items. Findings suggest clinicians can ask
four brief questions to screen patients' health literacy needs and provide as a valid indicator to
alert ambulatory healthcare team members of their patients' health literacy needs. Upon
screening, clinicians can refer patients to an official evaluation and/or patient education
intervention and tailor their clinical practice to meet the individual needs of patients. A
multivariate analysis indicated demographic variables were j ointly associated with 34% of the
STOFHLA score variance. Univariate statistics suggest the majority of participants were
knowledgeable about health literacy and health education; confident in their ability to seek
support; and had received support related to health literacy and patient health education. My
findings and their implications for research and practice are discussed.
Health literacy, a primary indicator of one' s health status, is possessed by less than two-
thirds of Americans (Kirsch, Jungeblut, Jenkins, & Kolstad, 1993; Partnership for Clear Health
Communication). The Institute of Medicine (2004) estimates that this equates to approximately
90 million people who lack literacy skills needed to function in the health environment. This
health issue affects millions of people and costs billions of dollars in healthcare resources
annually. Findings of the 2005 National Assessment of Adult literacy (NAAL) survey show that
the literacy skills of American adults have not significantly changed over the last decade
(Schneider, 2006). Each year, inadequate health literacy results in approximately $73 billion in
unnecessary health care costs (Center on an Aging Society at Georgetown University, 1999).
Health literacy was conceptualized in 1974 in the context of health education (Simonds,
1974). Over the years the concept has grown into a Hield of study. Today, health literacy is
identified as a complex construct with many contributing variables such as: reading, seeking,
exchanging, understanding and using health information. Healthy People 2010 defines health
literacy 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" (U. S.
Department of Health and Human Services, 2000, p. 1 1-20). Additionally, the National Health
Education Standards (Joint Committee on National Health Education Standards, 1995, p. 5)
define health literacy as "the capacity of individuals to obtain, interpret, and understand basic
health information and services and the competence to use such information and services in ways
which enhance health." Thus, health literacy is not only the ability to read and comprehend
health materials, but also requires the application of listening, analyzing, and decision-making
skills in a diversity of health situations (Consumer Health Advisory Committee, 2000).
As individuals navigate through the healthcare process they use forms of communication,
writing, analysis and reading skills in order to be able to effectively interact with healthcare
providers and participate in decision making. These health literacy skills are critical in equipping
individuals with the tools they need to interact with healthcare providers and health information.
If individuals have difficulty with any of these skills, they are at increased risk of experiencing
poor health related outcomes (Partnership for Clear Health Communication, n.d.). Because
patients rely on these basic forms of communication and comprehension to progress through the
healthcare system, an adequate level of health literacy is critical for experiencing an optimal
Safeer and Keenan (2005) summarize the current status of health literacy in the medical
environment and their recommendations to rectify identified problems as follows:
Though most adults read at an eighth-grade level, and twenty percent of the population
reads at or below a fifth-grade level, most healthcare materials are written at a 10th-grade
level. Older patients are particularly affected because their reading and comprehension
abilities are influenced by their cognition and their vision and hearing status. Inadequate
health literacy can result in difficulty accessing healthcare, following instructions from a
physician, and taking medication properly. Patients with inadequate health literacy are
more likely to be hospitalized than patients with adequate skills. Patients understand
medical information better when spoken to slowly, simple words are used, and a restricted
amount of information is presented. For optimal comprehension and compliance, patient
education material should be written at a sixth-grade or lower reading level, preferably
including pictures and illustrations. All patients prefer reading medical information written
in clear and concise language. Physicians should be alert to this problem because most
patients are unwilling to admit that they have literacy problems (p. 463).
It is imperative that individuals who do not have an adequate level of health literacy be
identified and assisted in order to promote a high quality healthcare experience for the patient, as
well as prevent unnecessary healthcare costs. Identifying individuals in need of assistance is
essential for promoting effective and timely intervention.
Though health literacy screening is a logical solution to identifying patients with health
literacy needs, time constraints are a significant barrier to getting health professionals to assess
their patients' health literacy levels. Additionally, if clinicians did screen their patients, they
would then be responsible for intervening when necessary, which may require additional time.
Providing health professionals with a brief health literacy screening process and promoting the
utilization of patient health education referrals provides clinicians with a time efficient means of
screening patients and a viable option for patient education support.
While several health literacy screening assessments have been developed, many require
specialized materials, time, and scoring interpretation. The greatest weakness of current health
literacy assessments is their partial measurement of the larger holistic concept of functional
health literacy. Baker (2006) concludes that the Rapid Estimate of Adult Literacy in Medicine
(REALM) and the Test of Functional Health Literacy in Adults (TOFHLA) are "clearly
inadequate" (p. 880). Though these two tools are the most commonly used measures of health
literacy, Baker (2006) proposes neither adequately test "individual's capacities. Rather, the tests
measure selected domains that are thought to be markers for an individual's overall capacity" (p.
In efforts to create an effective brief screening tool, Chew and colleagues (2004) tested
screening questions to identify patients with inadequate or marginal health literacy skills. With a
sample of 332 preoperative patients, researchers provided participants with the Short Test of
Functional Health Literacy in Adults (STOFHLA) [shortened version of the original TOFHLA]
and 16 health literacy screening items. Findings indicated 4.5% of patients had inadequate health
literacy skills and 7.5% had marginal health literacy skills on the STOFHLA. Of the sixteen
health literacy screening items, three were found to effectively detect inadequate health literacy
skills: How often do you have someone help you read hospital materials; How confident are you
filling out medical forms by yourself; and How often do you have problems learning about your
medical condition because of difficulty understanding written information? (Chew, Bradley &
Boyko, 2004). Wallace and colleagues (2006) evaluated the three questions identified in Chew's
(2004) study to determine their accuracy in identifying patients with limited or marginal health
literacy skills. Their study concluded one screening question was accurate in detecting
individuals with limited and limited/marginal health literacy skills, How confident are you filling
out medical forms by yourself!
Baker (2006) asserts that the inconsistencies between Chew and Wallace's work indicate a
need for further research. Further, he and others suggest that additional research is needed to
assess health literacy. Parker and colleagues (2006) state that, "More research is needed .a
continued need to advance measuring individual skills. Efforts to advance population-level
measurement and indicators are also greatly needed" (p. 891-892). The overriding objective of
this study was to address this need by testing the validity of a brief four question health literacy
screening tool in an ambulatory care setting within the Veterans Administration (VA).
1. Validate the effectiveness of a health literacy BRIEF screening tool (4 screening questions)
in the identification of participants with low health literacy skills.
2. Identify relationships between health literacy level and selected patient variables.
3. Identify the degree of perceived knowledge, confidence and readiness related to health
literacy in an ambulatory VA patient population.
4. Identify the degree of confidence and readiness in an ambulatory VA patient population
related to patient health education.
1. Does the BRIEF screening tool correlate with the STOFHLA & REALM to identify
patients with inadequate health literacy in an ambulatory VA healthcare setting?
2. What patient variables are associated with health literacy level in an ambulatory VA
3. What is the patient' s level of knowledge, readiness, and confidence related to health
literacy in an ambulatory VA healthcare setting?
4. What is the patient' s level of confidence and readiness related to receiving patient health
education in an ambulatory VA healthcare setting?
1. The study was conducted with VA patients as opposed to other patient populations.
2. Respondents were recruited from a single ambulatory healthcare setting in North Florida to
provide a sample pool.
3. The study was delimited to the specificity of the items which comprised the assessment
4. The respondents were comprised of a sample of volunteers.
5. The study was comprised of paper and pencil assessments and a questionnaire and thus
data gathering was limited to the context of these instruments.
Because of the following limitations the results of this study will only be generalized to
VA patients in an ambulatory care setting:
1. This study used a relatively homogeneous convenience sample of VA patients presenting
for care in an ambulatory setting.
2. Patients were recruited based on the voluntary participation of the VA ambulatory
3. VA patients in an ambulatory care setting may be different from non-VA patients.
4. VA patients in an ambulatory care setting may be different from patients in a non-
ambulatory healthcare setting.
5. VA patients who volunteered to complete the assessment may be different from those who
did not volunteer to complete the assessment.
1. Observations are normally distributed and representative of the population.
2. The data sample was homogeneous.
3. Variation in data is due to chance.
4. Participant responses were valid and reliable.
5. The measures (Rapid Estimate of Adult Literacy in Medicine and the Test of Functional
Health Literacy in Adults) used to screen patients' health literacy level were adequate and
6. The time of data collection (May to August 2006) was adequate for the purposes of this
Definition of Terms
For the purposes of this study selected terms are defined as the following.
Health literacy "the capacity of individuals to obtain, interpret, and understand basic health
information and services and the competence to use such information and services in ways which
enhance health" (Joint Committee on National Health Education Standards, 1995, p. 5).
Health education learning experiences designed to facilitate voluntary actions, on the part of
the learner, conducive to making quality health decisions.
Health educator an individual who participates in the process of providing individuals and
communities with information and skills which promote the common obj ectives of personal and
public health education.
Patient educator an individual who participates in the process of providing individuals and
target groups (i.e. diabetics) with information and skills which promote the common obj ectives
of personal health and healthcare education.
Clinician or Provider any individual licensed to provide care to patients in the clinical
Support The process of providing patients with access, information, or skills related to using
health information to make health decisions.
Health literacy is a primary indicator of one' s health status. Approximately one in three
Americans, have less than adequate health literacy skills (Kirsch, Jungeblut, Jenkins, & Kolstad,
1993; Partnership for Clear Health Communication). An estimated 90 million people lack
literacy skills necessary to function in the healthcare environment (The Institute of Medicine,
2004). Clinicians need a quick efficient means of screening patients' health literacy level to
respond to patients' individual needs. The obj ective of this study was to test the validity of a
brief four question health literacy screening tool and identify predicting variables of an
individual's health literacy level.
Defining Health Literacy
Defining health literacy provides a descriptive context for understanding what skills are
required to function effectively throughout the healthcare process. The Council of Scientific
Affairs of the American Medical Association (1999) refers to functional health literacy as "the
ability to read and comprehend prescription bottles, appointment slips, and the other essential
health-related materials required to successfully function as a patient" (p. 552). This definition
implies not only a basis of knowledge, but also the ability to apply knowledge as an active
participant in one's own healthcare.
In addition to basic reading, writing, and communication skills, functional health literacy
also includes the capability to understand instructions on prescription drug bottles and medical
education materials, comprehend doctor's instructions, find information, and analyze health
information (Committee on Health Literacy, 2004). Other functional health literacy skills include
actively participating in health encounters; understanding and giving consent; advocating rights;
and the general ability to negotiate with complex healthcare systems (Committee on Health
Literacy, 2004). Not only are there many skills required to proficiently participate in one's
healthcare, but health literacy capabilities vary by context and setting and are not contingent on
years of education or general reading capability. An individual who is literate in general, may
have inadequate functional health literacy capabilities in the healthcare environment (Consumer
Health Advisory Committee, 2000). Baker (2006) suggests several factors contribute to the
measurement of one' s health literacy level including reading fluency; prior knowledge;
complexity of health information; oral complexity; culture; social norms; and barriers. Pawlak
(2005) proposes the following determinants of health literacy: age; genetics (Cognition &
Ability); language; race and ethnicity (Culture); education (Reading Level & Technologic
Competence); employment; socio-economic status; environment (Access to Care and
Technology). Because there are many contributing variables to being functionally health literate
anyone is likely to need assistance with the various capabilities required, regardless of
demographics or socio-economic status (U. S. Department of Health and Human Services, 2000).
General Literacy & Health Literacy Assessment
There are several measures designed to evaluate an individual's literacy level. Andrus &
Roth (2002) and Pawlak (2005) have identified the most popular assessments (Table 2-1).
Specific to health literacy, several assessments have been developed for clinical assessment of
patients' health literacy skills. The REALM and TOFHLA are the most commonly used
measures of health literacy; however "neither test is a comprehensive assessment of an
individual's capacities. Rather, the tests measure selected domains that are thought to be markers
for an individual's overall capacity .. [and are] clearly inadequate" (Baker, 2006, p. 880).
As educators and practitioners prepare to assist individuals with low health literacy, it is
important to have an understanding of the general levels of literacy. Sometimes general reading
assessments are based on criteria for four to five levels (on a scale from 0 to 500), which can
provide an indication of level of health literacy abilities (See Table 2-2). Individuals scoring at
level one have the ability to perform simple tasks and those who score at levels four and five are
able to perform long complex tasks requiring higher cognitive levels of analysis (National
Institute for Literacy).
Commonly, both general and health literacy assessments focus on a three point scale with
the first level being sub-average; mid-level being marginal; and the higher level being a
gradation of proficiency. Commonly, both general and health literacy assessments focus on a
three point scale with the first level being sub-average; mid-level being marginal; and the higher
level being a gradation of proficiency. Similarly, the TOFHLA, Test of Functional Health
Literacy in Adults (Parker, Baker, Williams, & Nurss, 1995; Nurss, Parker, Williams, & Baker,
2001), which is commonly used as an indicator for level of health literacy skills, divides scores
into three criterion levels: (1) inadequate (0 to 16), (2) marginal (17 to 22); and (3) adequate (23
to 36) as seen in Table 2-2. Individuals scoring at level one on the TOFHLA do not possess the
necessary skills to function efficiently when executing health literacy related tasks. Individuals
scoring at level 2 have marginal skills that allow for achieving simple tasks, but not complex
tasks; whereas those scoring at level three have the skills necessary to accomplish health literacy
related tasks proficiently.
However, more commonly, general literacy levels are indicated by academic grade levels
or equivalencies, with the numerical value correlating with the individual's grade level. For
example: if a fifth grader reads and applies skills on grade level, then the individual's score is a
"5"; if below level, then a "3" or "4" depending on their skills and abilities (Committee on
Health Literacy, 2004). Some health literacy scales are also correlated with grade levels. The
REALM, Rapid Estimate of Adult Literacy in Medicine (Davis et al., 1993) has three levels,
which correspond to the general reading grade levels with scores ranging from 0 to 66 (See Table
2-3). These levels allow for an interchangeable scaling system for evaluating general and health
Rates of Health Literacy
Though adequate health literacy is critical for functioning efficiently throughout the
healthcare process, at least 46% of the adult population, approximately 90 million people, lack a
sufficient foundation of basic skills to function successfully; including reading basic materials,
comprehension, and providing and seeking information (National Institute for Literacy, n.d.).
More conservative, yet consistent estimates suggest 40 to 48% of adult Americans struggle with
functional literacy tasks (Kirsh, Jungeblut, Jenkins & Kolstad, 1993; Andrus & Roth, 2002).
Fourteen % of adults in the US have a below basic level of prose literacy; 12% of US adults have
below basic document literacy; 22% of adults have below basic quantitative literacy (Kutner,
Greenberg & Baer, 2006). Older adults: 23% of those more than 64 years of age have below
basic prose literacy; 27% below basic document literacy; 34% below basic quantitative skills
(Kutner, Greenberg & Baer, 2006). Further, Kutner and colleagues (2006) suggest more than 1/3
of English-speaking patients and more than 1/2 primarily Spanish-speaking patients have low
If 90 million people are illiterate and unable to solve problems proficiently then logic
suggests they will have the same difficulties with health related information and materials
(National Institute for Literacy, n.d.). For example, recent data indicate that low health literacy
impacts nearly one in every three people living in the United States, approximately 90 million
people (Partnership for Clear Health Communication, n.d.), suggesting that inadequate low
literacy skills in the general population perpetuate inadequate health literacy skills.
In addition, the average individual reads at the 8th-9th grade level; however, most health
related materials are written at a higher reading level (Partnership for Clear Health
Communication, n.d.). This suggests that individuals with average reading capabilities will suffer
from inadequate levels of health literacy. Inadequate health literacy combines the difficulties of
low literacy skills and further complicates matters by adding complex health content. Due to the
compounded nature of this problem nearly half of the population is unprepared to effectively
engage in the healthcare process and therefore put themselves at risk of inadequate care, negative
outcomes, and unnecessary costs. Health literacy impacts a significant portion of the general
population, warranting maj or efforts in prevention and remediation.
Impact of Health Literacy
Partnership for Clear Health Communication (What is health literacy? n.d.) suggests,
People with low health literacy are often less likely to comply with prescribed treatment
and self-care regimens, fail to seek preventive care and are at higher (more than double)
risk for hospitalization, remain in the hospital nearly two days longer than adults with
higher health literacy, and often require additional care that results in annual healthcare
costs that are four times higher than for those with higher literacy skills (para. 2).
The impact of health literacy affects the health process in many ways. Pawlak (2005)
suggests, "Low literacy, an aging population, prevalence in chronic conditions, and a
complicated healthcare system influence and magnify health disparities in the United States"
(p. 174). Inadequate health literacy is associated with less health related knowledge; decreased
comprehension of medical information; poorer health status; poor compliance rates; infrequent
and delayed use of preventative services; increased hospitalization; increased use of emergency
services; increased healthcare costs; and inadequate management of chronic illness (Andrus &
Roth, 2002; Dewalt, Berkman, Sheridan, Lohr, & Pignone, 2004; Pawlak, 2005). The following
sections illustrate health outcomes resulting from inadequate health literacy.
Cost of inadequate health literacy. Providing the public with health literacy skills and a
sense of empowerment to use those skills is a humanitarian approach to increasing the quality of
healthcare. More importantly, increasing levels of health literacy is an economical approach to
decreasing healthcare costs and decreasing mis-utilization of the healthcare system. Ultimately,
low health literacy skills translate into as much as $58 billion a year in U. S. health system costs
(Partnership for Clear Health Communication, n.d.). The Center on an Aging Society at
Georgetown University (1999), estimates that low health literacy costs the nation at least $73
billion annually. The incurring costs of medical errors; unnecessary hospitalization; extended
hospital stays; medical non-adherence; and delayed onset of healthcare all contribute to the
exorbitant cost incurred annually by Americans. These data provide convincing evidence to
support the need to intervene with individuals with low and marginal health literacy to prevent
negative healthcare outcomes and unnecessary healthcare costs.
Hospitalization. The American Hospital Association reports the national average cost of
inpatient care is approximately $1000 per day, and the national average hospital-stay cost is
$6000. Hospitalization costs can dramatically increase with unnecessary extended hospital stays
resulting from low health literacy skills. In fact, one's level of health literacy has been found to
be significantly associated with an increased risk of hospitalization (Center on an Aging Society
at Georgetown University, 1999; Berkman et al., 2004). Clinical research findings suggest
patients with low health literacy were more likely to be hospitalized than those with adequate
health literacy skills (Baker, Parker, Williams, Clark, & Nurss, 1997; Baker, Parker, Williams, &
Clark, 1998). Medicare enrollees with low health literacy were more likely than enrollees with
adequate health literacy to use the emergency room and to be admitted as inpatients (Kutner,
Greenberg & Baer, 2006). In addition to incurring extra medical costs, with nursing shortages
and overcrowded hospitals, unnecessary hospitalization associated with low health literacy
contributes to inadequate healthcare for people who are in more critical need of hospitalization.
Health outcomes. Health literacy influences a variety of health outcomes. In general,
one' s level of health literacy skills is the strongest indicator of an individual's health status
(Partnership for Clear Health Communication, n.d.). Increased literacy levels are associated with
increased knowledge of health services and the likelihood of accessing those health services. For
example, individuals with low literacy skills are more likely to have not received appropriate
immunizations than individuals with adequate literacy. In addition, low literacy levels are also
significantly associated with higher levels of smoking (Hawthorne, 1996). Both immunizations
and smoking behaviors can have long-term effects on one's health. If 90 million people are
affected by inadequate health literacy and those individuals are more likely to have not received
necessary healthcare, such as immunizations or treatment for sexually transmitted diseases.
Those 90 million individuals not only endanger their own health, but also the health of the others
Health literacy has also been linked to the management of chronic diseases such as
diabetes. Individuals with low health literacy are less likely to stabilize their blood sugar level
than those individuals with adequate health literacy skills (Schillinger et al., 2002). Some studies
have found correlations between literacy and measures of disease, such as people with diabetes
(Kutner, Greenberg & Baer, 2006).
Mancuso and colleagues (2006) investigated the association between health literacy and
asthma outcomes and assessed how health literacy affects those outcomes using the Test of
Functional Health Literacy in Adults (TOFHLA). The mean age of study participants was 42
years, and 83% were women. Findings suggested (p < .05) that among respondents, over a two-
year period, a lower health literacy level was associated with decreased quality of life, physical
dysfunction, and increased utilization of emergency services for asthma, (Mancuso & Rincon,
2006). Low health literacy was also associated with poor longitudinal asthma outcomes.
Mancuso and colleagues (2006) suggest that efforts to improve asthma outcomes should focus on
improving literacy skills that promote comprehension and effective self-management. Inadequate
health literacy is a significant factor in the adverse health-related outcomes resulting from non-
Associations between health literacy levels and mental health have been investigated as
well. To examine these associations the maj ority of researchers have relied on the
instrumentation of one of two popular health literacy assessments, the Rapid Estimate of Adult
Literacy in Medicine (REALM) and the Test of Functional Health Literacy in Adults
(TOFHLA). Lincoln and colleagues (2006) found an association between low literacy and more
depressive symptoms and with more severe depressive symptoms in people with alcohol and
drug dependence. (Lincoln et al., 2006). Weiss and colleagues (2006) evaluated a literacy
intervention with adult patients identified as having depressive symptoms. The participants'
depression scores were similar in both the intervention and control groups at baseline. Scores
improved in both groups, but the improvement was significantly larger in the literacy
intervention group. These research findings suggest, "there may be benefit to assessing the
literacy skills of patients who are depressed, and recommending that patients with both
depression and limited literacy consider enrolling in adult education classes as an adjuvant
treatment for depression" (Weiss, Francis, Senf, Heist & Hargraves, 2006, p. 823).
Mortality has also been associated with low health literacy levels. In a five-year
prospective study from 1999 to 2004, Sudore and colleagues (2006a) assessed the association of
limited literacy with mortality among 2, 152 elders without functional difficulties or dementia.
Participants' mean age was 75.6 years, 48% were male, 38% were black, and 24% had limited
literacy. Compared to participants' with adequate literacy, participants with limited literacy, "had
a higher risk of death (19.7% vs 10.6%) with a hazard ratio (HR) of 2.03 (95% confidence
intervals [CI], 1.62 to 2.55). limited literacy remained independently associated with mortality
(HR 1.75; 95% CI, 1.27 to 2.41)" (Sudore et al., 2006a, p. 806). Sudore and colleagues (2006a)
suggest limited literacy is associated with a nearly 2-fold increase in mortality in the elderly. The
relationship between literacy and health is still being explored, but evidence suggests people with
low literacy are more likely to report having poor health, diabetes and heart failure, than those
with adequate literacy (Kutner, Greenberg & Baer, 2006).
Medication and care adherence. Recent studies assessed the relationship between
literacy and medical adherence. Li and colleagues (2000) suggest literacy level is a predictor of
medical adherence. Kalichman and colleagues (1999) found lower literacy skills were
significantly associated with increased potential for poor adherence among patients taking
medication for HIV infection. Patients with low literacy were less likely to adhere to their
medical regimen than patients with adequate health literacy skills (Kalichman, Ramachandran, &
Lindau, Basu and Leitsch (2006) investigated the influence of low literacy on cervical
cancer screening knowledge, and assessed the relationship between health literacy level and
racial disparities in cervical cancer. Patients with inadequate literacy were less likely to follow
up within one year; however findings were not statistically significant. Having less than a high
school education and having a physician-estimated low literacy level were significant predictors
of duration of time to follow-up. Among women with an abnormal Pap smear, those perceived
by their physician to have low literacy were significantly more likely to fail to present for follow-
Relationships between health literacy and medical adherence have been found in a wide
range of patient populations. In a sample of 197 participants with glaucoma, patients with low
literacy were less adherent to their medical regimen than those with adequate health literacy
skills (Muir et al., 2006). They reported a positive relationship between health literacy level and
the number of received refills (Muir et al., 2006). Muir and colleagues (2006) suggest
interventions designed for patients with inadequate health literacy skills may improve medical
Despite the growing amount of evidence indicating a relationship between health literacy
and health related outcomes, there are some researchers who provide evidence to suggest
otherwise. Though low health literacy has been associated with nonadherence to antiretroviral
therapy and higher HIV-RNA levels, Paasche-Orlow and colleagues (2006) findings suggest
"low literacy was not associated with a lower odds of adherence or virologic suppression in this
longitudinal analysis of HIV-infected patients with a history of alcohol problems" (p. 83 5).
Further evidence provided by Fang and colleagues (2006) suggests limited health literacy is not
significantly associated with self-reported adherence, but "was associated with incorrect answers
to questions on warfarin's mechanism (adjusted odds ratio [OR] 4.8 [1.3 to 17.6]), side-effects
(OR 6.4 [2.3 to 18.0]), medication interactions (OR 2.5 [1.1 to 5.5]), and frequency of
monitoring (OR 2.7 [1.1 to 6.7])" (p. 841).
A study of low income patients reported 42% of the patients did not understand their
medication instructions, 49% could not decipher Einancial aid forms, and 26% did not know
when their next appointment was scheduled to occur (Williams et al, 1995). The same study also
concluded that low literacy patients were Hyve times more likely to misinterpret their medication
prescriptions. A study of emergency care patients showed 81% of patients could not read and
understand the rights and responsibilities portion of their paperwork and 74% did not know they
qualified for free care (Baker, Parker, Williams, Clark, Nurss, 1997). In a sample of 251 adult
patients, Davis and colleagues (2006) found one-third of participants had low literacy.
Participants' comprehension of warning labels was associated with literacy level; and patients
with low literacy were 3.4 times less likely to interpret prescription medication warning labels
correctly (95% confidence interval: 2.3 to 4.9). These Eindings suggest patients with low literacy
have difficulty understanding prescription medication warning labels (Davis et al., 2006)
Proper medication management is a vital component of chronic disease control. In a
sample of patients with coronary heart disease, medication management was measured and
compared with participants' literacy level. Approximately half (50.7%) had inadequate literacy
skills, and 28.9% had marginal skills. Medical management was significantly associated with
literacy (P<.001), "patients with inadequate literacy skills had 10 to 18 times the odds of being
unable to identify all of their medications, compared to those with adequate literacy skills
(P<.05)" (Kripalani et al., 2006, p. 852). If an individual does not understand his/her medical
instructions or rights and responsibilities as a patient, he/she cannot optimally participate in the
healthcare process. Research indicates there is a connection between health literacy and the
ability to adhere to medical regimens and actively participate in the healthcare process as an
informed responsible patient. Though the relationships between literacy level, adherence, and
outcomes is not yet clear, current findings warrant further investigation.
Medical jargon and vocabulary comprehension. Physician-patient communication is a
critical component of healthcare, yet patients commonly do not understand medical instructions
and other communication with their providers. The importance of functional health literacy is
obvious, but the largest study of health literacy to date amplifies the need to be functionally
literate for comprehending medical language (Williams et al., 1995). In a public hospital setting
with predominantly indigent and minority patients, Williams and colleagues (1995) found 35%
of English-speaking patients and 62% of Spanish-speaking patients had inadequate or marginal
health literacy skills. Further, 42% of patients could not comprehend directions for taking
medication on an empty stomach; 26% were unable to understand appointment information; and
60% did not understand a standard informed consent document. Low health literacy is a maj or
contributor to the communication problems between patients and their providers; patients often
remember, "less than half of what the physician tried to explain" (Kripalani & Weiss, 2006, p.
888). Further, individuals with low health literacy "may have less familiarity with medical
concepts and vocabulary, and they ask fewer questions" (Kripalani & Weiss, 2006, p. 888).
Another study suggests patients are correct a mere 36% of the time when interpreting
medical language (Hadlow & Pitts, 1991). In a list of 50 medical terms, no one term was
correctly identified by all patient participants. The relationship between the use of technical
language and outcome is that the more technical language that is used, the less comprehension
and recall; providers are advised to use little technical language and check for patient
understanding when they are used (Thompson, 1998). In addition, with increased specialization
of providers, the more technical and difficult the terminology used with patients (Ruben, 1993).
As more providers become specialists, and as technology for diagnoses and treatments becomes
more complex, the knowledge gap between patients and providers widens. Thus, patients remain
unassertive, uninvolved and often providers' decisions are based on stereotypical impressions or
provider assumptions, influenced by the patients lack of communication (DiMatteo & Lepper,
1998). Patients need to be competent in communicating about technical terminology as well as
specific details pertaining to their illness. If they do not have this knowledge, they need to at
least be able to communicate that they do not understand.
Additionally, the informed consent process is an area of research which must be addressed
when determining the influence of health literacy on the healthcare process. The consideration of
health literacy is vital in the research protocol process. Patients need adequate health literacy
skills to knowledgeably consent to participating in a research protocol. Future research efforts in
health literacy should continue to evaluate the actions needed to provide a comprehensive
informed consent process that meets the requirements of risk to patients, as well as meeting the
needs of providing participants with a comprehensive informed consent process (Sugarman, &
Paasche-Orlow, 2006). Providing individuals with the appropriate health literacy skills to
function as proactive knowledgeable patients who can knowledgeably consent to participate in
research is vital for sustaining ethical research standards.
Risk Factors Associated with Low Health Literacy
Because there are many contributing variables to being functionally health literate, anyone
is likely to need assistance with the various capabilities required, regardless of demographical
status (U. S. Department of Health and Human Services, 2000). However, certain populations
such as the elderly, minorities, immigrants, and individuals with low socio economic status are
disproportionately affected by the negative outcomes of low and marginal health literacy skills
(Kirsch, Jungeblut, Jenkins, & Kolstad, 1993). For example, Sudore and colleagues (2006b)
analyzed the relationship between health literacy, demographics, and access to healthcare. Their
findings suggest, "After adjusting for socio-demographics, associations remained between
limited health literacy and being male, being black, and having low income and education,
diabetes mellitus, depressive symptoms, and fair/poor self-rated health (P<.02)" (p. 770).
Further, after adjusting for socio-demographics, health status, and co-morbidities, older
people with a sixth-grade reading level or lower were twice as likely to have any of the three
indicators of poor healthcare access (odds ratio=1.96, 95% confidence interval=1.34-2.88)
(Sudore et al., 2006b). Limited health literacy was prevalent and was associated with low
socioeconomic status, co-morbidities, and poor access to healthcare, suggesting, "It may be an
independent risk factor for health disparities in older people" (Sudore et al., 2006b, p. 770). The
maj ority of adults sixty years-old and over have inadequate or marginal literacy skills; half of
welfare recipients read below fifth grade level; and 40 50% of minorities have reading
problems (Kirsch et al., 1993). Low levels of health literacy can affect the healthcare process in
many ways and cumulating risk factors can have a negative additive effect. These demographic
factors are reviewed in the following sections.
Ethnicity. Minorities are historically known to experience disparities in healthcare.
Disparities in healthcare among minorities are a maj or current research initiative (Cooper &
Roter, 2003; U.S. Department of Health and Human Services, 2000; Cooper-Patrick et al., 1999).
It seems clear that ethnicity and cultural background can influence patient-provider
communication and patients' level of functional health literacy in at least three ways; (1) they
may have different languages or dialects, (2) preferred styles of communication may differ; and
(3) people from different cultures have different explanatory models for health and illness.
The dominance demonstrated by most providers, as discussed previously creates
interpersonal distance between the patient and provider because the provider often assumes the
role of dominance due to socio-educational backgrounds. Commonly, minority patients remain
"silent" and thus the communication is broken down; especially when patient and providers
come from different cultures. When people from different cultures interact in health settings,
there is often a lack of acceptance and failure to respect one another. Interactions with mixed
cultures within the health setting often result in negative and distorted perceptions (Cline &
McKenzie, 1998). Ultimately, cultural differences and language barriers preventing quality
communication need to be minimized or minority patients will continue to receive less than
optimal care, with ineffective and often inappropriate communication, which will perpetuate
poor health outcomes.
Previous research demonstrates the role of literacy skills and minority status in
comprehension of the consent process. Sudore and colleagues report that in a sample of 204
participants (mean age of 61 years), 40% had limited literacy, and that "lower literacy (P=.04)
and being black (P=.03) were associated with requiring more passes through the consent process.
Not speaking English as a primary language was associated with requiring more passes through
the consent process in bivariate analyses (P<.01)" (Sudore et al., 2006c, p. 867).
Previous research also suggests ethnicity and cultural differences strongly influence
communication and literacy (Cooper, & Roter, 2003). Language barriers, misunderstanding in
cultural values, differences in cultural values regarding openness and disclosure in healthcare,
and screening of patients based on racial and cultural characteristics are variables that can
influence the patient-provider communication process (Cline & McKenzie, 1998). In a recent
study conducted with a nationally representative sample of 23,889 non-institutionalized U. S.
adults, Sentell & Halpin (2006) found "African Americans were 1.54 (95% confidence interval,
1.29 to 1.84) times more likely to have a work-impairing [health] condition than whites" (p.
864). When literacy level was considered, the effects of both African-American race and
education were no longer significant. Sentell & Halpin (2006) suggest, "Literacy inequity may
be an important factor in health disparities, and a powerful avenue for alleviation efforts, which
has been mistakenly attributed to other factors" (p. 862). Recent published findings involving
3,260 participants indicate, "Black individuals had worse self-reported health status and lower
influenza and pneumococcal vaccination rates" (Howard, Sentell, Gazmararian, 2006, p. 857).
Though currently unclear, it appears ethnicity and health literacy level are factors in health
related outcomes. These findings warrant further empirical investigation to determine the role of
ethnicity and health literacy level in health related outcomes.
Education. Though research has had mixed results on the effects of patient demographics,
one factor that is obviously linked to health communication and literacy skills and other
cognitive abilities that are utilized when interacting with a provider and engaging in a health
treatment is level of education. Current research results suggest that level of education is directly
related to a patient's ability to understand health information (Lukoschek, Fazzari, & Marantz,
2003). In a sample of 3,260, "individuals without a high school education had worse physical
and mental health and worse self-reported health status than those with a high school degree;
accounting for health literacy reduced these differences by 22% to 41%" (Howard, Sentell,
Gazmararian, 2006, p. 857). Patients with lower educational levels have decreased understanding
of health information compared to those with higher levels of education. Providers need to be
aware of patients' educational level to determine any difficulties that may be inherent in
comprehending medical information.
Gender. Gender influences patient health communication with providers and health
literacy behaviors, in that it aggravates an already asymmetrical process (Cline & McKenzie,
1998). Females are subordinate when interacting with male providers. In general, providers
spend more time with female patients; female patients talk more, ask more questions, and get
more information than men (Cline & McKenzie, 1998). However, disagreements and
interruptions by the provider suggest there are issues in the quality of communication between
females and their providers. Further, interactions with females and their providers result in more
communication, but often their concerns are not taken as seriously as those of male patients
(Thompson, 1998). Some reports claim there are gender differences in the way complaints are
elaborated (Thompson, 1998). Communication patterns between male patients and their
providers tend to be more similar; and males receive more legitimacy and acceptance than
females (Cline & McKenzie, 1998). During interactions with their providers women must remain
competent, while not challenging their provider with assertive behaviors. Due to the inconclusive
evidence, more information is needed to distinguish gender differences in reading, seeking,
exchanging, understanding, and using healthcare information.
Age. With the complexities of chronic health issues (many non-curable), vast experiences
and wisdom, and complicated treatments i.e. multiple medications, provider-patient interactions
with the elderly and their patient behaviors are significantly different from interactions with
younger patients. While the average skills of U. S. adults are inadequate for navigating the
healthcare system, the elderly are the most adversely affected subpopulation, which places them
at increased risk of negative outcomes when compared to other age groups (Schneider, 2006).
Elderly patients often want to be cared for and are less likely to challenge their provider (Cline &
McKenzie, 1998). Providers tend to take more control and use loud patronizing communication
and provide elderly patients with less information than they do for younger, more assertive
patients (Cline & McKenzie, 1998). The current literature indicates providers tend to be less
respectful, less supportive, more impatient, less engaged, and less attentive to older patients
(Cline & McKenzie, 1998). These factors indicate a need to better understand how to develop
interventions to address the variables that influence the quality of patient-provider interactions
and the functionality of patient health literacy skills.
Health Literacy and the Medical Encounter
Identifying the socio-demographical variables that influence the healthcare process is
important, but it is also imperative to understand what actually happens during a healthcare
encounter. To understand patient health literacy needs in the healthcare environment requires
knowledge about patient perceptions when engaging in healthcare experiences. This section will
review the current research regarding what is known about the patient experience and identify
gaps in the research needing further investigation.
Time restraints during healthcare visits are a primary concern when addressing the quality
of health communication between patients and their providers. On average, a general practitioner
conducts 120,000 to 160,000 patient interviews in a 40-year career (Center for the Advancement
of Health, 2003, para. 2). The average length of a patient-physician interaction is 20.4 minutes,
up from 16.3 minutes in 1989 (Mechanic, McAlpine, & Rosenthal, 2001). Researchers note that
although this amount of time is adequate, patients are often unable to disclose all their concerns,
ask important questions, and engage in meaningful dialogue because they are constantly
interrupted in an already limited time period (Thompson, 1998). Even if medical encounters
allow enough time for discussion, individuals who are not equipped with functional health
literacy skills may not be likely to assert themselves in discussions with their doctor.
Physicians also control interactions with patients by asking closed-ended questions which
serve to limit the amount of information they elicit and share; conveys a sense of urgency, even
in non-emergency environments; and asserts superiority. Physician's vocal intensity, lack of
fluency, perceived negative feedback, and abruptness may diminish the quality of patient-
physician communication (Thompson, 1998). In addition, sitting disrobed on an exam table in a
cold room during the encounter can be very intimidating for the patient (DiMatteo & Lepper,
1998). Individuals who lack health literacy skills are more likely to assume a submissive role in
the medical encounter and fail to properly represent themselves. They are also less likely to
obtain answers to questions they may have related to information they receive during their
medical encounters. Ninety % of physicians feel patients have trouble following directions and
identify many patients as having low levels of health literacy (Thompson, 1998). Identifying
individuals' health literacy needs prior to an appointment and intervening before there is a
problem can promote clear communication and the likelihood of increasing the quality of
Identifying Patients with Inadequate Health Literacy Skills
As practitioners and educators become more aware of the high levels of low health literacy
skills, there is increased pressure to identify individuals with low health literacy and provide
those individuals with assistance to ensure optimal health behaviors and health related outcomes.
First, practitioners and educators must realize that individuals with inadequate health literacy
often do not realize or acknowledge their lack of health literacy (Parker, Davis, & Williams,
1999). Due to the stigma associated with the inability to read, people with inadequate literacy
skills often feel ashamed when they have trouble reading. Thus, individuals will often hide their
inability to read and comprehend information (Parikh, Parker, Nurss, Baker, & Williams, 1996;
Baker et al., 1996). Often individuals with low literacy do not bring anyone to help them and
they do not ask for assistance (Parikh et al., 1996). This can make identifying individuals with
inadequate literacy difficult.
Health literacy skills can be evaluated using standardized assessments, but practitioners
and educators do not always have access or time for such measurements. However, there are
other informal means of identifying individuals with inadequate health literacy. One way to
evaluate an individual's ability to read is to ask him/her to read materials during the course of the
interaction, such as a label on a bottle. Often individuals who cannot read will say: they forgot
their glasses; they will read the information when they get home; or they will discuss it later with
family or friends (Parker, 2000). A practitioner should suggest reading it together to make sure
there are no misunderstandings or unanswered questions. A practitioner can also ask patients
about their education level. Individuals with an eighth grade education, or less, are likely to have
inadequate health literacy (Parker, 2000). Individuals with a ninth grade education or higher, are
harder to identify for health literacy problems and usually require formal assessment (Parker,
Practitioners and educators should also be aware of individuals who incorrectly fill out
forms or answer all items in an identical fashion. Practitioners and educators should always be
aware that anyone can have inadequate health literacy skills. Other indicators of poor readers are:
concrete thinking; literal interpretation of words and visuals; missing principal features; getting
lost in details; and inability to interpret perceptual information (Doak, Doak, & Root, 1996).
Individuals with low health literacy often avoid seeking printed materials and often look around
and have poor eye contact when receiving information (Doak et al., 1996). Healthcare clinicians
can benefit from taking the time to observe patients and assess indicators of low health literacy,
so remediation can be implemented from the initiation of care, rather than after a
misunderstanding has occurred.
The BRIEF screening tool proposed in this study takes one to two minutes to administer.
Therefore, it will take approximately 15 to 30 minutes of a clinician's work day (provided they
see 15 20 patients a day) to screen patients. Reasonably, many healthcare facilities cannot
afford to lose such valuable clinic time due to other patient needs. This dilemma creates an
opportunity to promote the alliance of health educators and healthcare provides in responding to
patient needs in the clinical care setting. Noland and Li (2006) suggest promoting the use of
health educators in the clinical setting as a reimbursable service is critical to advancing the field
of health education; gaining public trust; increasing professional visibility and accountability.
Further, health educators have the expertise to identify and appropriately respond to patients with
inadequate health literacy skills. Further as a reimbursable service, health educators can pay for
themselves while increasing the recognition of health education as a legitimate profession.
Currently, too often nurse educators attempt to fill the role as a clinical care provider and an
educator (particularly in nutrition and diabetes). This duality is certainly a contributor to the
excessive demands that lead to burn out in the nursing profession. Thus, if patient health
educators were employed as members of the healthcare team, they could provide reimbursable
patient health education services. Ideally, this role as members of the healthcare team would
advance the field of health education; contribute to increased quality of care; promote personal
health; and provide as a support resource for nursing professionals in the clinical setting.
Further, it takes additional time to respond to individuals with inadequate health literacy.
Detailed explanations about healthcare may need to be provided in a slower cadence, using non-
medical terms. Some patients may need to read and review written materials with their
healthcare provider or advocate. Though these adjustments would take additional time from the
clinician' s work day, these investments of preventative effort can result in better health outcomes
such as, but not limited to, fewer missed appointments; shorter periods of hospitalization;
increased adherence; and decreased use of emergency services (Andrus & Roth, 2002; Dewalt,
Berkman, Sheridan, Lohr, & Pignone, 2004; Pawlak, 2005). Reducing the adverse outcomes of
undetected inadequate health literacy can ultimately save time for clinicians and prevent the
unnecessary costs associated with inadequate health literacy. Thus the investment for employing
patient educators pays for itself and prevents poor patient outcomes.
In response to concerns related to required time for patient screening, providers should
recognize that individual screening for health literacy is only necessary once every five to ten
years, or after traumatic events. Further, if patients' health literacy level is posted in their
permanent patient record, all members of the healthcare team can be privileged to the
information and, based on the recorded health literacy assessment, can respond accordingly. The
benefits of screening and identifying patients' health literacy level clearly outweigh the costs in
time and effort. No longer should clinicians make assumptions about their patients' ability to
read, comprehend, apply, and provide health information.
Patients benefit from taking an active role in their medical care choices. Patients who are
active in their experience have greater satisfaction, greater alleviation of their symptoms, better
control of their chronic conditions, less distress about their illness, and have better response to
surgery and invasive diagnostic treatments (DiMatteo & Lepper, 1998). Patients who are
involved in their healthcare feel a greater sense of control over their health and their lives in
general, have more positive expectations for their health, and demonstrate better adherence to the
treatment, in which they helped in the decision making (DiMatteo & Lepper). Active patients are
less likely to blame providers for less than optimal outcomes, while less active patients give less
adequate histories, and tend to delay reporting symptoms. Lastly they are more likely to litigate
malpractice when they have less positive outcomes (DiMatteo & Lepper). Gaining knowledge
about what patients are currently doing when interacting with their providers and what behaviors
they are engaging in is key to gathering information and identifying the needs of patients to
increase their ability to read, seek, exchange, understand, and use health information.
Barriers to Patient Participation
There are several barriers that prevent patients from being active participants in their
healthcare. Foremost, patients find it difficult to increase their involvement; additionally provider
feedback can act as a barrier. For example, if the provider responds negatively, the patient is less
likely to be satisfied, than if the provider is positive. Cultural barriers also moderate the
likelihood of patient involvement (Thompson, 1998). Other barriers that affect a patient' s active
participation in his/her healthcare include being disrobed during an interaction led by a provider
who typically limits patient responses. Patients trying to behave and act as a "good patient" are
silent, in order to not take up the valuable time of a rushed provider. Another maj or barrier for
patients is their entry into the healthcare encounter while ill and with considerable anxiety,
thereby compounding an already complex interaction (DiMatteo & Lepper, 1998). Patients who
perceive communication barriers with their providers experience more pain and more difficulty
discussing their experience of pain with others, such as family members. Anxiety; coping style;
not wanting to distract the doctor; wanting to look like a good patient; and being concerned that
increased pain means progression of the disease are several factors that interfere with patients'
ability to communicate about pain (Thompson, 1998). Ultimately, many factors influence
whether a patient will participate actively in his/her healthcare. Understanding what prevents
patients from optimally interacting with healthcare providers and engaging in the healthcare
process is key to gathering information and identifying the needs of patients to increase their
ability to read, seek, exchange, understand, and use health information. Further, advocacy, social
support, and patient education resources and support can mediate the outcomes associated with
one's health literacy level.
Patient Education Programs
As researchers and health educators become more aware of the magnitude of inadequate
health literacy in the United States, large organizations and institutions of academics and health
join together to develop educational materials for dissemination. For example, the Florida Health
Literacy Study is a collaborative effort between the University of South Florida and Pfizer
Health Literacy Interventions to disseminate Pfizer health literacy educational materials to
patients to increase their ability to navigate through the healthcare process effectively (Florida
Health Literacy Study, 2003). Pfizer hosts another literacy program "Ask Me 3", under the
auspices of the National Partnership of Clear Health Communication. This Partnership of Clear
Health Communication has several large credible partnering agencies, such as the American
Medical Association. This program is a national communication campaign and provides a web
site, information and educational materials for patients and physicians (Partnership of Clear
Health Communication, n.d.). Bayer Institute for Healthcare Communication (n.d.) also
disseminates patient education using a coaching system and other educational resources to
P.R.E.P.A.R.E. patients to be active participants in the treatment process. Private organizations,
governmental agencies and other academic institutions have an array of health literacy programs
throughout the country for patients and physicians that provide instructional materials and
mentoring programs including an array of health literacy skills (Singleton, & Terrill, 2003).
Health Literacy Skills
Adequate health literacy includes the ability to evaluate information for credibility and
quality, analyze risks and benefits, determine dosages, understand test results, and locate health
information (Consumer Health Advisory Committee, 2000). To accomplish these tasks,
individuals need to be able to understand graphs and other visual information, obtain and apply
information, and calculate numbers. Language skills are needed to articulate their health
concerns and describe their symptoms (Consumer Health Advisory Committee, 2000). Patients
need to ask relevant questions, understand medical advice or medical instruction, and employ
decision-making skills. With increased accessibility to computers, patients can also benefit from
obtaining the ability to search the internet and evaluate web sites (Consumer Health Advisory
Reviews of patient training studies (Cegala & Broz, 2003; Cegala, Clinch, & Gade, 2003)
suggest the primary communication skills needed by patients are information seeking, expressing
concerns, assertiveness, information provision, and verifying information. Reflection,
summarizing, and processing knowledge are critical skills used in training patients to
communicate (Cegala & Broz, 2003). Patient skills, such as communication skills, have been
operationalized and empirically tested and have been shown to have a strong impact on the
information exchange between patients and their physicians (Cegala et al., 2003). Ultimately,
individuals need to be able to express their needs; articulate their symptoms; listen to their
healthcare providers; understand provided information and instructions; assimilate information
and use it to make decisions; participate in the decision making process; ask pertinent questions;
calculate and determine medication dosages; Eind quality information; verify information; and
follow their medical treatment plan. Educators and practitioners can benefit by including these
skills in their intervention development process to ensure program success.
In addition to obtaining skills, health educators and practitioners can create an
infrastructure for empowering patients. Health empowerment is described as the combined
function of self-efficacy and adequate health literacy skills (Hubley, 2002). Patient
empowerment is a central concept in models for improving healthcare outcomes and catalyzing
health reform efforts (Segal, 1998; Salmon & Hall, 2003). The premise of patient empowerment
suggests patients have a right to make decisions about their healthcare and can assume a
leadership role during the healthcare process. The concept of patient empowerment was founded
as a solution for the long-term care needs of chronically ill patients such as individuals with
diabetes and cancer (Segal, 1998). However, with healthcare reform initiatives underway it has
become apparent that patient empowerment is vital to help all individuals profieiently facilitate
the process and outcomes of their healthcare encounter.
If patients are equipped with functional health literacy skills and feel confident about
taking action and decision-making, this can translate into a belief that they can influence the
healthcare process and into higher levels of health empowerment. Patients who empower
themselves have improved health outcomes and consume fewer health care resources than
patients who do not (Office of the Public Advocate and Headway, 1992). Standard practice
should include providing all patients with supportive assistance, reinforcing patient participation
and mutually deciding what additional assistance or resources may be necessary. Efforts can be
made at the interpersonal and organizational level to promote an environment that supports
patient understanding and empowers patients to be active participants throughout the healthcare
process. The following section provides techniques for providing assistance and empowering
Creating a Shame-free Environment and Providing Resources
Providing a shame-free empowering environment will reinforce patients' ability to
effectively communicate and navigate through the healthcare system. Creating a supportive, non-
stigmatizing environment for low literacy patients can promote proactive behaviors rather than
perpetuate feelings of shame (Parker, 2000). These facilities can provide patients with surrogate
readers or advocates to assist them when getting information from practitioners, thereby
promoting a support system for successful healthcare outcomes (Potter & Martin, 2003). Patients
identified as having low levels of health literacy should be provided with support, verbal
explanations, written materials with only the necessary information presented in simple terms
and pictures, and alternative resources such as videotapes (Parker, 2000; Potter & Martin, 2003).
Implementing these practices will decrease the stigma associated with needing assistance and
emphasize the importance of understanding health materials in a shame-free environment. Health
education and care facilities can prepare educators and practitioners to be sensitive to low
literacy individuals to promote a shame-free environment (Parikh et al., 1996). Any efforts
healthcare and educational facilities take to empower patients with functional health literacy
skills and the self-efficacy to use them will promote optimal healthcare outcomes and an
economical and ethical practice.
This study utilized a multidimensional theoretical perspective for the purpose of placing
this research within a framework in which the information can be considered within a pragmatic
context (See Figure 2-1). The contributing theories and frameworks are discussed in the
The theoretical framework for this study is founded in Systems Theory which states open
systems, such as a healthcare system or a biological system, are made up of sub-systems. The
forces of other sub-units within the system affect all other sub-units, ultimately creating a ripple
effect throughout the entire system. The system as a whole works in a symbiotic fashion to create
a higher order state of equilibrium. Within complex systems, sub-systems interact with one
another to produce output (Giacomo & Weissmark, 1986). Systems Theory also assumes that
information must flow to create balance between systems and the environment. Systems rely on
input which is received and processed, resulting in an output which generates feedback which is
again received as input creating an ongoing cyclical reciprocal process. Reciprocal relations
occur with systems that provide feedback. The recursive nature of feedback results in
information transformation and can modify the output of the system thereby regenerating system
proficiency. When the flow of information is disrupted, the individual experiences imbalance
which can manifest in many forms. This process occurs with individuals, between individuals, in
organizations and beyond (Brown, Pryzwansky, & Schulte, 2001).
Provided these assumptions are true, then the following are true: (1) the actions or
inactions of a patient can affect his/her own healthcare process; (2) interaction between a patient
and his/her provider can affect the patient's healthcare process and outcomes; (3) the stated
affects ultimately affect the larger healthcare system in which these sub-systems interact. Thus
influencing factors associated with how patients interact with health information and healthcare
providers can influence the patient's health; their interactions with providers; health related
outcomes; and other systemic outcomes in general.
Illustrating systemic nature of healthcare encounters, Street (2003) applied an ecological
model to communication in medical encounters. The ecological approach, like systems theory,
evaluates the inter-relationships, the ecological approach focuses on how the organism (the
individual) interacts with his/her environment (medical encounter). Street (2003) proposes that
there is a dynamic series of interrelations occurring during medical encounters which impact the
quality of interaction including organizational; media; cultural; political-legal; and maybe most
importantly the interpersonal context. Within the interpersonal context, the ecological model
accounts for the multidimensional multivariable representation of the facets of the pre-disposing
and cognitive affective influences of the patient and the provider. The model provides a simple
yet dynamic understanding of the assumptions of the existing interactions happening within the
context of interaction between the patient and his/her healthcare provider. For example, the
model accounts for a patient's communication style; self-concept; linguistic resources; verbal
and nonverbal behavior; goals; perceptions; emotional state; and communicative strategies.
Within this context it seems logical to place the patient' s ability to exchange, comprehend, and
use health information. Such that the identification of one' s level of health literacy skills is a
primary factor when addressing the interpersonal context of medical encounters within the
ecological and systemic theoretical framework (research question #1). Thus, making these
assumptions this study proposed to identify a means of quickly assessing patients' health literacy
skills to support providers in adapting and responding to their patients' needs.
Further the ecological model accounts for the cultural context of medical encounters
suggesting that patient factors such as race, ethnicity, socio-economic status, education, gender,
etc. can influence the medical encounter and subsequent outcomes. This study seeks to address
these potentially influencing variables as they relate to one's level health literacy skills (research
question #2). These findings will contribute to identifying and understanding the influencing
variables associated with patients' ability to use health information and effectively interact with
Transtheoretical Model & Stages of Change
To evaluate participants' level of knowledge about health literacy (research question # 3)
and their readiness to engage in patient education (research question # 4) this study utilized the
Transtheoretical Model (TTM), specifically the Stages of Change. This theoretical model of
behavior change is founded on research conducted over the last fifty years (Prochaska, Norcross
& Diclemente, 1994). A list of the constructs associated with the Transtheoretical Model used in
this study can be found in Table 2-4.
The constructs of TTM interact over time to ideally produce behavior change such as
utilizing patient education resources or increasing awareness of health literacy. The five stages of
change relate to an individual's stage relative to engagement in a behavior; from contemplation
(not even considering the behavior) to maintenance (maintaining the behavior) (Prochaska,
Norcross & Diclemente, 1994). Self-efficacy serves as the most important predictor of
behavioral success. One's self-efficacy contributes to their success or failure at each stage of
change. This study also utilized the concept of consciousness raising. Based on the theoretical
construct of "consciousness raising," this study makes the assumption that if people are made
aware of health literacy and provided resources identified as patient health education, they will
become more aware of the constructs and utilize patient education resources. It is further
suggested that as the population increases their awareness of health literacy and patient health
education resources, they will increase their utilization of patient health education, thus
promoting optimal health related outcomes.
Ninety million individuals, have less than adequate health literacy skills (Kirsch,
Jungeblut, Jenkins, & Kolstad, 1993; Partnership for Clear Health Communication), which result
in approximately 73 billion dollars in unnecessary healthcare costs annually (Center on an Aging
Society at Georgetown University, 1999). The outcomes associated with inadequate health
literacy skills can affect anyone despite their age, race, education or income (Partnership for
Clear Health Communication, n.d.). Regardless of socio-cultural background the effects of
inadequate health literacy are detrimental to the individual and society as a whole. It is
imperative that researchers establish a time-efficient, economical, easy and valid assessment to
provide health professionals with a viable means of screening and intervening with patients. It is
also imperative that data is gathered about patients' awareness, readiness, confidence, and
utilization related to health literacy and patient health education. These data will provide
direction for health literacy screening development as well as assess the population's current
awareness of health literacy and awareness of, and utilization of, patient health education.
Table 2-1. General literacy and health literacy screening assessments.
4 years and
age and grade
22 (7 for short
and numeric test:
5-74 years Adults Only
grade level: 3rd
and below; 4ti'-
6th, 7t'-8ti', or 9"
0-8, O = failure at
first grade level,
8 = 8th grade
level or above
PIAT-R WRAT 0.88,
0.62-0.91 SORT-R 0.96,
Wide Range Achievement Test-Revised; REALM = Rapid Estimate of Adult Literacy in Medicine; MART = Medical
Terminology Achievement Reading Test; SORT-R = Slosson Oral Reading Test-Revised; PIAT-R = Peabody Individual Achievement
Test-Revised; IDL = Instrument for the Diagnosis of Reading; TOFHLA = Test of Functional Health Literacy in Adults. Table
adapted from Tables for Health Literacy: A Review (Andrus & Roth, 2002, p. 284-285) and published by Pawlak, 2005, in Economic
Considerations of Health Literacy, Nursing Economics.
Table 2-2. Scale for levels of TOFHLA/STOFHLA and general literacy skills.
TOFHLA TOFHLA Literacy Skills and Abilities
Level Score Score
Inadequate 0-16 0-225 Able to perform uncomplicated tasks involving brief
and uncomplicated texts and documents.
226 275 Able to locate information in text, make low level
Marginal 17-22 inferences using printed materials and integrate easily
identifiable pieces of information.
275 325 Able to integrate information from relatively long or
dense texts or documents, determine appropriate
arithmetic operations based on information contained
in the directive, and identify quantities needed to
Adequate 23-36 326 375 Adults at these levels demonstrate proficiencies
associated with long and complex documents and text
376 500 passages; able to determine and interpret qualitative
and quantitative data needed to perform an operation.
Table 2-3. Levels for grade equivalent scale of the Rapid Estimate of Adult Literacy in Medicine
REALM REALM Grade Skills and Abilities
Level Score Level
Limited 0-44 0 -6 Not able to read most low literacy health materials; will
need repeated oral instructions; materials should be
composed of illustrations or video tapes. Will need low
literacy materials; may not be able to read a
Marginal 45-60 7 -8 Will struggle with most patient education materials.
Adequate 61-66 9 and Will be able to read and comprehend most patient
above education materials.
Table 2-4. Description of Stages of Change, Self-efficacy & Consciousness Raising.
Stages of Change
Precontemplation Has no intention to take action in next 6 months
Contemplation Intends to take action in next 6 months
Preparation Intends to take action within the next thirty days and has taken
some behavioral steps in this direction
Action Has changed overt behavior for less than 6 months
Maintenance Has changed overt behavior for more than 6 months
Self-efficacy Confidence that one can engage in the healthy behavior across
different and challenging situations
Consciousness Raising Finding and learning new facts, ideas, and tips that support the
healthy behavior change
Organizational Media Interpersonal CutrlCnet Political/Legal
Context Context Context Context
Transtheoretical Patient Health
Model -- - Education
Stages of Change Intervention
Figure 2-1. Theoretical Framework for Research Study.
The purpose of this study was to (1) validate the BRIEF form health literacy screening
process with a VA ambulatory care population; (2) identify relationships between health literacy
level and patient variables; and (3) identify stage of behavior for VA patients related to health
literacy and patient health education. The research questions for this study include: (1) Does the
brief screening correlate with the validated assessments and identify those patients in an
ambulatory VA healthcare setting with inadequate health literacy? (2) What patient variables are
associated with health literacy level? (3) What is the patients' level of knowledge, readiness and
confidence related to health literacy? (4) What is the patients' level of confidence and readiness
related to receiving patient health education? The following sections describe the methodology
of the study.
The correlation analysis required 85 pairs of scores. Based on a power analysis table, a
minimum of 225 participants was necessary to conduct the regression analysis, with .30 effect
size (medium) and .10 accuracy. Ultimately 378 participants provided data for this study. Data
were collected in rural and non-rural settings. The target population for this proj ect was a
convenience sample of VA patients presenting for care in a VA ambulatory care setting.
Participants were patients in eight ambulatory clinics where trained volunteer data collectors
were on staff. Participants were chosen at random by volunteer recruiters; inclusion criteria
required the participants to speak English to verbally assent to participate.
A self-administered survey and three screening instruments were utilized to collect data for
this study. The participant survey instrument consisted of 19-items (See Appendix A). The brief
self-administered participant survey had seven demographical items: age, gender, marital status,
race/ethnicity, language, home ownership, and education. One item assessed English as a first
language. One Hyve-point Likert item (1 = excellent; 5 = poor) asked participants to rate their
reading ability. Three additional dichotomous items pertained to health status related to high
blood pressure; stroke; and diabetes. Finally, eight questions with Hyve-point Likert scale
response options (strongly agree to strongly disagree) gathered data about the participants'
ability to define health literacy; awareness of patient education resources; confidence in seeking
support related to health information and health literacy; received support and patient education
resources; and intention to seek patient education resources and support related to health
Two instruments were used to test the proposed BRIEF screening tool: (1) STOFHLA,
Short-Test of Functional Health Literacy in Adults (Parker et al., 2001); and (2) The REALM,
Rapid Estimate of Adult Literacy in Medicine (Davis et al., 1993).
The STOFHLA is a shortened version of the original TOFHLA (Parker, Baker, Williams,
& Nurss, 1995; Nurss, Parker, Williams, & Baker, 2001). STOFHLA is comprised of two prose
passages with thirty-six fill-in-the-blank response items worth one point each. The possible score
range is 0 36. The maximum time for administration is seven minutes. STOFHLA scores
divide health literacy skills into three criterion levels: (1) inadequate (0-16); (2) marginal (17-
22); and (3) adequate (23 -36) (See Appendix B). In a group of 21 1 patients, Cronbach's alpha for
the STOFHLA was 0.97. Spearman's correlation between the STOFHLA and the REALM was
0.80 (Baker, Williams, Parker, Gazmararian, & Nurss, 1999).
The REALM assesses health literacy through the use of three columns of twenty-two
words each. The words in each column are listed in ascending order of difficulty. The REALM
produces a summed score based on the number of correctly pronounced words in each column.
REALM scores range from 0 to 66 (See Appendix C) and are divided into three criterion levels,
namely: limited (0 44); marginal (45 60); and adequate (61 66). Davis and colleagues (1993)
tested the REALM with three other standardized reading tests to establish instrument validity
with a sample of 203 patients: the reading recognition section of the Peabody Individual
Achievement Test-Revised (PIAT-R), the Wide Range Achievement Test-Revised (WRAT-R),
and the Slosson Oral Reading Test-Revised (SORT-R). The REALM correlated well with the
three other tests. Correlation coefficients were 0.97 [PIAT-R], 0.88 [WRAT-R], and 0.96
[SORT-R], (p < .0001). To determine test-retest reliability 100 inmates at a state prison were
given the REALM twice, one week apart. Test-retest reliability was 0.99, p < .001). Intra-subj ect
reliability for REALM has been reported as 0.97 (Davis et al., 1993).
This study employed the BRIEF health literacy screening tool which is comprised of the
following four items (See Appendix D): (1) How often do you have someone help you read
hospital materials? (2) How confident are you filling out medical forms by yourself! (3) How
often do you have a problems learning about your medical condition because of difficulty
understanding written information? and (4) How often do you have a problem understanding
what is told to you about your medical condition? Response options were offered in a Hyve-point
Likert scale for each of the items [items 1, 3 & 4 (1 = always to 5 = never); and item 2 (1 = not at
all to 5 = extremely)].
Though the validity of the proposed tool has not been previously published, the individual
items have been evaluated to examine their ability to identify individuals with inadequate health
literacy. Chew and colleagues (2004) examined 16 screening questions. In a sample of 332
veteran participants; 15 (4.5%) had inadequate health literacy and 25 (7.5%) had marginal health
literacy on the STOHFLA. Three of the screening questions, "How often do you have someone
help you read hospital materials?" "How confident are you filling out medical forms by
yourself!" and "How often do you have problems learning about your medical condition because
of difficulty understanding written information?" were effective in detecting inadequate health
literacy; area under the receiver operating characteristic curve of 0.87, 0.80, and 0.76. Wallace
and colleagues (2006) continued Chew's (2004) work and tested the three individual items with
305 participants. They computed the area under the receiver operating characteristic curves
(AUROC) for each item, using REALM scores as a reference standard. In Wallace's (2006)
sample, 54 (17.7%) had limited and 52 (17.0%) had marginal health literacy skills. One
screening question, "How confident are you filling out medical forms by yourselfi" was accurate
in detecting limited (AUROC of 0.82; 95% confidence interval [CI] = 0.77 to 0.86) and
limited/marginal (AUROC of 0.79; 95% CI = 0.74 to 0.83) health literacy skills; the item had a
significantly greater AUROC than the other questions (P<.01).
Based on the findings of Chew, Wallace and their colleagues, three items were included in
the proposed BRIEF health literacy screening tool. However, because the Veteran Health System
serves a large portion of elderly patients, auditory health information can pose problems for
patients who have hearing problems. Thus, the item, "How often do you have a problem
understanding what is told to you about your medical condition?" was added in an attempt to
increase the construct validity of the tool by addressing the comprehension of oral health
information. The addition of this item completed the development of the 4-item BRIEF health
literacy screening tool to be validated against the previously published REALM and STOFHLA.
This study was approved as an expedited survey by the University of Florida Institutional
Review Board (UFIRB), proj ect number 629-2005. Therefore, this study did not require written
consent. Thus data collectors verbally recruited participants and participants verbally consented
to participate in the study. This study received cooperation and support from the staff of the
Veteran Health System. All data were collected on site at eight Veteran Health System
ambulatory care clinics and hospitals in North Florida/South Georgia between March 2006 and
August 2006 (See Table 3-1). This study used an action research method using volunteer
healthcare providers to collect data in the clinical setting. Volunteer data collectors were
recruited via an inter-office email generated by the research study co-investigator, a VA
registered nurse (See Appendix E). Data collectors attended a 30-minute training session on site
where they were provided with data packet administration instructions, practice opportunities,
and materials. Training was provided by the principal investigator.
Participants were recruited by twenty-one trained volunteer data collectors. The maj ority of
data collectors were nurses (n = 17), others included the principal investigator; a nutritionist; a
dental technician; and a nurse educator volunteer (See Table 3-2). The use of a diverse set of
providers enabled a diverse patient sample and demonstrated the use of the BRIEF health literacy
screening process in multiple ambulatory care settings. Healthcare providers followed one of two
procedures to recruit participants; they either invited participants and collected the data
themselves in their ambulatory care setting, or they invited the potential participant to participate
and then brought the patient to the investigator' s examination room (in the same ambulatory
setting). Who collected data from the participants (whether the investigator or the volunteer
recruiter/data collector), was determined by convenience for the volunteer recruiter/data
collector. Upon meeting the patient, data collectors invited the patient to participate and
answered any of the patient' s questions. Once the patient agreed to participate the data collector
provided a private exam room for the participant, along with three health literacy screening
assessments and one participant survey. All health literacy instruments and the survey required a
total of approximately twelve to fifteen minutes of the participant' s time. Administration times
for the individual instruments are illustrated in Table 3-3. Upon completion of the instruments,
the data collector thanked the participant for their information and offered to answer any
questions. The data collector then put the completed assessments and participant survey into a
manila envelope labeled with the clinician data collector name and site location. The envelopes
were then sealed and provided to the study administrator who kept the data stored in a locked
cabinet. To ensure confidentiality and anonymity, no identifying information was collected from
participants. The study had a response rate of approximately 90%; the reasons for refusal were
not assessed and thus cannot be determined. No incentives for participation were provided.
Data were analyzed using the SPSSC v. 12.0 software package. The level of significance
was set at a 95% level of confidence, with a p-value of .05. Univariate statistical analyses
included basic descriptive statistics including means, frequencies and proportions to provide
preliminary statistical information and assess data patterns (Noland, 2000). Frequency
distributions and descriptive statistics were calculated to obtain information about gender;
race/ethnicity; education; health status; knowledge; and health literacy level. The following
sections provide details of the analyses conducted to answer the four proposed research
Research Question #1
Does the BRIEF screening tool correlate with the STOFHLA & REALM to identify those
patients in an ambulatory VA healthcare setting with inadequate health literacy?
Bivariate Analyses. A Pearson Product Moment correlation coefficient was calculated in
order to determine the comparative validity of the screening tools. The Pearson Product Moment
correlation coefficient is commonly used to analyze the magnitude of association between two or
more variables. This study utilized two previously validated health literacy screening tools, the
Rapid Estimate of Adult Literacy in Medicine (REALM) and the Test of Functional Health
Literacy in Adults (STOFHLA) to test the validity of the proposed instrument, the four item
BRIEF. A normal distribution of data and a true linear relationship between variables was
Upon finding significant correlations between the BRIEF health literacy screening tool and
the validated tools, other analyses were run to determine the accuracy of the BRIEF when
compared to the REALM and the STOFHLA; and optimal cut-off points for determining varying
levels of health literacy. The accuracy of diagnostic tools is typically evaluated using measures
such as sensitivity; specificity; predictive values; and likelihood ratios (Espallardo, 2003).
Sensitivity is defined as the "proportion of patients who were positive for the test among all
patients with the disease" (Espallardo, 2003, p. 229). Specificity is defined as the "proportion of
patients who were negative for the test among all patients with the disease" (Espallardo, 2003, p.
229). Positive predictive value is defined as the "proportion of patients with the disease among
all patients who were positive for the test" (Espallardo, 2003, p. 230). Negative predictive value
is defined as the "proportion of patients who do not have the disease among those patients who
were negative for the test." (Espallardo, 2003, p. 230). "False negative" is defined as the
proportion of patients who had a negative test result, but were really positive for the disease.
"False positive" is defined as the proportion of patients who had a positive test result, but were
really negative for the disease. These diagnostic concepts are defined and illustrated in Tables 3-
4 & 3-5.
This study utilized the measures of sensitivity" as the primary indicator of accuracy of the
BRIEF health literacy screening tool. A perfect health literacy screening test would have 100%
sensitivity and 100% specificity. It would accurately identify all the true cases of
inadequate/marginal health literacy, and it would never mislabel someone with adequate health
literacy as someone with inadequate/marginal health literacy skills. However, this degree of
accuracy is rare.
Because most screening tests are imperfect, a balance between sensitivity and specifieity is
a necessary compromise. To establish the optimal balance, sensitivity and 1-specifieity can be
plotted on a graph, called a "ROC curve," meaning Receiver-Operator-Characteri stic (ROC). The
ROC can provide statistical data to determine the optimal cutoff point for demarcating data in
diagnostic tests, creating balance between sensitivity and specifieity. Pepe (2000) describes the
A statistical tool that is becoming popular for describing diagnostic accuracy is the receiver
operating characteristic (ROC) curve. To define an ROC curve, first consider diagnostic
tests with dichotomous outcomes, with positive outcomes suggesting presence of disease.
For dichotomous tests, there are two potential types of error. A false-positive error occurs
when a non-diseased individual has a positive test result, and conversely, a false-negative
error occurs when a diseased individual has a negative test result. The rates with which
these errors occur, termed the false-positive and false-negative rates, together constitute the
operating characteristics of the dichotomous diagnostic test...ROC curves generalize these
notions to non-binary tests in the following fashion: Let D be a binary indicator of true
disease status with D = 1 for diseased subj ects. Let X denote the test result with the
convention that larger values of X are more indicative of disease. For any chosen threshold
value c, one can define a dichotomous test by the positivity criterion X [greater than or
equal to] c and calculate the associated error rates. A plot of 1 minus the false-negative rate
(or true positive rate) versus the false-positive rate for all possible choices of c is the ROC
curve for X (p. 308).
Receiver operating characteristic (ROC) curves were calculated to plot the sensitivity
versus 1-specificity and areas under the ROC curve (AUROC) to test the performance of the
BRIEF health literacy screening items.
Finally, to test independence between the proposed score intervals, an analysis of variance
(ANOVA) was conducted for the health literacy screening tools. An ANOVA was also
conducted to test for group differences between the proposed BRIEF score levels. These analyses
were conducted to determine if the four item BRIEF assessment could improve the detection
capability of the previously recommended single health literacy screening items and to identify
score intervals for clinical interpretation (Chew, Bradley & Boyko, 2004; Wallace, Rogers,
Roskos, Holiday & Weiss, 2006).
An ANOVA is an inferential statistic; it can be used with more than one independent
variable, each of which must include at least two levels. The ANOVA procedure is appropriate
for experiments with more than one independent variable including more than two levels for
each independent variable (Weaver, 2003, para. 1; BBN Corporation, 1997, para 5). An ANOVA
compares the variance of scores within a set parameter. The variance of levels produces the sum
of squares within groups (SSws), and sum of squares between groups (SSbg). Variation in overall
mean values is due to factor or to variation due to random error. The ANOVA model presented
in Figure 3-1, provided, F= the overall mean, Z- the level effect, and E the random error
Multivariate Analyses. A Principal Component Analysis was conducted to determine if the
BRIEF in fact measures one distinct construct "health literacy." Principle component analysis is
an analysis of interdependence, which allows() the researcher to reorient the data so the first
few dimensions account for as much of the available information as possible" (Lattin, Carroll, &
Green, 2003, p. 83). A Principal Component Analysis informs us on how much of the variance
is accounted for by the identified principal components yielding in a given analysis (Lattin,
Carroll, & Green, 2003). Linear combinations of the original variables (i.e. BRIEF items) are
replaced with standardized variables. By function, the analysis produces weighted eigenvectors
associated with the covariance matrix of the standardized versions of the original variables. The
variance of the principal components is produced by the eigenvalues associated with the
eigenvectors. Principal components are typically identified using the calculated eigenvalues.
Kaiser' s rule suggests, "retaining only principal components with eigenvalues exceeding unity,"
suggesting only principal components with eigenvalues greater than one should be retained
(Lattin, Carroll, & Green, 2003, p. 113).
Research Question #2
What patient variables are associated with health literacy level in an ambulatory VA
Multivariate Analyses. A Stepwise multiple linear regression analysis was conducted, with
dummy coded variables, to determine predictors of the dependent variable health literacy,
including several independent variables: six socio-demographic variables (age, education level,
gender, ethnicity, perceived ability to read and perceived ability to define "health literacy") and
three health status variables (diabetes, stroke, high blood pressure). A multiple regression model
can be used with multiple explanatory variables and a continuous outcome variable, as in the
BRIEF health literacy screening tool.
Multiple regression is commonly used to predict and understand the relationship between
an outcome variable and several independent variables (StatSoft, 2006). In areas of inconclusive
research, such as health literacy, stepwise regression is a popular method to explore and
determine predicting independent variables. This analysis identifies the independent variables
(age, gender, socio-economic level, etc.) which account for variance in the dependent variable
(health literacy level), while controlling for other independent variables (Rosner, 2000).
Stepwise multiple regression analysis computes the ordinary least squares to determine the best
linear combination of independent variables which predict the dependent variable. The equation
for the multiple regression statistic is illustrated in Figure 3-2. Ultimately, Stepwise multiple
regression computes the best fit of the regression line.
Where Y is the dependent variable and "Sizi + 82Z2 63Z3") are the independent variables,
the "b's" are regression coefficients, which represent the change in the dependent variable (y)
when the independent variable changes one unit; the "c" is the constant. Error is assumed in
multiple regression; multivariate normality and relationship linearity is observed.
Research Question #3
What is the patients' level of knowledge, readiness & confidence related to health literacy
in an ambulatory VA healthcare setting?
Univariate statistical analyses calculated frequencies and proportions to provide statistical
information about the participants' reported level of knowledge and confidence related to health
literacy and health information. Univariate statistical analyses can be used to illustrate statistical
information and to identify data patterns (McDermott & Sarvela, 1999; Noland, 2000).
Research Question #4
What is the patients' level of confidence and readiness related to receiving patient health
education in an ambulatory VA healthcare setting?
As with Research Question # 3, to answer the fourth and Einal research question in this
study univariate statistical data analyses were conducted to calculate frequencies and proportions
to provide information about the participants' readiness & confidence related to patient health
education. As stated previously, univariate analyses are commonly used to illustrate statistical
data patterns (McDermott & Sarvela; Noland).
The instrument reliability was calculated for the proposed BRIEF health literacy screening
tool (See Appendix D) based on the study sample data using Cronbach's (1955) alpha measure
of internal consistency (Traub, 1994). Cronbach's alpha is commonly used to assess the
reliability of an instrument using an item analysis. To determine the benefit of adding the fourth
item (How often do you have a problem understanding what is told to you about your medical
condition?) a reliability analysis was conducted with three and four items resulting in the
following respective alpha levels: BRIEF with 3-items, a = .722 and BRIEF with 4-items, a =
.773. These statistical findings suggest that adding the fourth item to the BRIEF screening tool
increases the internal consistency of the construct measure.
Chapter three describes the methods used to answer the four research questions posed in
this study. The proposed methods examined the relationships between demographical variables
and health literacy and tested the proposed BRIEF health literacy screening tool against the
validated REALM and STOFHLA instruments. Chapter three includes a description of the
sample population, data collection, instruments, research design, research variables, and data
analysis procedures. Frequency rates were calculated to determine participant stage of readiness
related to patient education and health literacy. A Pearson' s Product Moment correlation was
conducted to test the association, and comparative validity of the screening tools. A Principal
Component Analysis was conducted to determine the number of distinct constructs measured by
the BRIEF health literacy screening tool. Receiver operating characteristic (ROC) curves were
calculated to plot the sensitivity versus 1-specificity and areas under the ROC curve (AUROC) to
test the performance of the BRIEF health literacy screening tool. An Analysis of Variance
(ANOVA) was also conducted to test group differences among the proposed score intervals.
Finally, a stepwise multiple linear regression analysis was conducted to determine predictors of
the dependent variable, health literacy. Analyses for the research questions were tested at a .05
significance level. The following chapter discusses results from these analyses.
Table 3-1. Frequency and percentage of participants recruited at the eight VA ambulatory care
Site # VA VISN 8 Sites Frequency Percent (%)
1 Gainesville, FL 244 64.6
2 Valdosta, GA 5 1.3
3 Lacanto, FL 3 .8
4 St. Augustine, FL 6 1.6
5 Leesburg, FL 18 4.8
6 Tallahassee, FL 55 14.6
7 Lake City, FL 37 9.8
8 The Villages, FL 10 2.6
Total 378 100.0
Table 3-2. Frequency and percentage of participant data collected by each of the volunteer data
Site # Data Collector Frequency Percent (%)
1 Principal Investigator 158 41.8
1Nurse Collector #1 5 1.3
1 Nurse Collector #2 (Co-PI) 39 10.3
1 Nurse Educator Volunteer 38 10.1
1Nurse Collector #3 3 .8
1 Nurse Collector #4 1 .3
2 Nurse Collector #5 5 1.3
3 Nurse Collector #6 3 .8
4 Nurse Collector #7 6 1.6
5 Nurse Collector #8 6 1.6
5 Nurse Collector #9 8 2.1
5 Nurse Collector #10 4 1.1
6 Nurse Collector #11 11 2.9
6 Nurse Collector #12 10 2.6
6 Nurse Collector #13 9 2.4
6 Nurse Collector #14 15 4.0
6 Nurse Collector #15 10 2.6
7 Nutritionist 9 2.4
7 Dental Technician 3 .8
7 Nurse Collector #16 25 6.6
8 Nurse Collector #17 10 2.6
Total 378 100.0
Table 3-3. Administration time for the REALM, STOFHLA, BRIEF, & Participant Survey.
Instrument Time to Administer (minutes)
Participant Survey 5
Table 3-4. Conceptual illustration of the sensitivity, specificity, false positive, false negative,
positive predictive value, and negative predictive value.
Health Literacy Status
Positive a b a+b
Test (true positive) (false positive)
Result Negative c d c+d
(false negative) (true negative)
Table 3-5. Definitions and equations of sensitivity, specificity, false positive, false negative,
positive predictive value, and negative predictive value.
proportion of patients who
were positive for the test
among all patients with the
proportion ofpatients who
were negative for the test
among all patients with the
proportion of patients who had
a negative test result, but were
really positive for the disease
proportion of patients who had
a positive test result, but were
really negative for the disease
Probability the patient has the
disease, when test results
indicate positive for disease
Probability the patient does not
have the disease, when test
results indicate negative for
c/(a+c) or (1 sensitivity)
b/(b+d) or (1 specificity)
Figure 3-1. Analysis of Variance (ANOVA) Equation Model.
Y = blxl+b2X2 ...+bnxn + c
Figure 3-2. Multiple Linear Regression Model.
The level of significance was set at a 95% confidence level, with an alpha of .05. Basic
descriptive statistics including frequencies and proportions are presented. The demographic
statistics are illustrated in Table 4-1. Participants were predominantly white (73.5%) males
(94.2%). The mean age was 61.5 (SD = 11.9) years. Mean age by ethnicity was: whites, 62.7
years (SD = 11.4); African Americans, 56.7 years (SD = 12. 1); Hispanics, 61.9 (SD = 12.6); and
Native Americans, 68.6 (SD = 11.4). The sample consisted of 278 (73.5%) white participants;
with 69 (18.3%) African Americans; 12 (3.2%) Hispanics; 12 (3.2%) Native Americans; 1 (.3%)
Asian American; and 1 (.3%) self-described as "other." Three (.8%) participants did not provide
their ethnicity. Though the veteran population is diversifying, this sample is representative of the
current veteran population receiving Veteran Health System services.
In addition to measuring health literacy level, participants rated their reading ability (five-
point scale). Nearly 86% report reading "good" to "very good" while 15% reported reading
"fair" or "poor" (See Table 4-2). Ninety-seven % (366) of participants report speaking English as
their first language. Seventy-seven % (290) of participants reported owning their home. Sixty-
four % (242) of respondents reported having high blood pressure; 35% (132) reported having
diabetes; and 11% reported having had at least one stroke.
Research Question # 1
Does the BRIEF screening tool correlate with the STOFHLA & REALM to identify
those patients in an ambulatory VA healthcare setting with inadequate health literacy?
To illustrate the levels of health literacy in the sample population participant health literacy
scores for each of the three tools were calculated and then categorized into three groups:
REALM: limited (0 -44); marginal (45 60); adequate (61 -66); STOFHLA: inadequate (0 -
16); marginal (17 22); adequate (23 36); and the proposed BRIEF: inadequate (4 12);
marginal (13 16); adequate (17 20). Data are provided in Table 4-3. The participants' scores
spanned the full range of possible scores: 4-Item Brief (5-20); REALM (0-66); and STOFHLA
Scores from the three screening tools (BRIEF, REALM, & STOFHLA) suggest between
7% and 20% of respondents had inadequate health literacy skills and between 8% and 37% had
marginal health literacy skills and between 43% and 83% had adequate health literacy skills. The
average scores for the three screening tools were as follows (standard deviation in parentheses):
BRIEF = 15.39 (SD = 3.67); REALM = 59.42 (SD = 8.96); and STOFHLA = 29.83 (SD = 8.03),
respectively (See Figure 4-1). These data suggest the average individual has marginal health
literacy skills as measured by the REALM and BRIEF; and adequate as measured by the
STOFHLA. Cross tabulation tables illustrate the BRIEF, REALM, & STOFHLA data in Tables
4-4 & 4-5.
A Pearson Product Moment correlation was calculated to determine the comparative
validity of the screening tools (See Table 4-6). Pearson correlation results were: r (378) = .40, p
<.01 for the BRIEF and REALM; r (378) = .42, p <.01 for the BRIEF and STOFHLA; and r
(378) = .61, p <.01 for the REALM and STOFHLA. The coefficients provide evidence that all
three of the health literacy screening tools are positively correlated. These correlations suggest
patients scoring high on one tool will also score high on the other two tools. The addition of the
fourth item to the BRIEF, "How often do you have a problem understanding what is told to you
about your medical condition?" did increase the internal validity of the screening measure as
indicated by the increased correlation coefficient (See Table 4-6). A general rule of thumb for
determining if there is a relationship between variables is a minimum correlation coefficient of at
least .30 (Hinkle, Wiersma, & Jurs, 1988). With this in mind, a significant correlation exists
between the BRIEF, the REALM and the STOFHLA. Franzblau (1958) provides a more precise
scale of correlation coefficients for determining the degree of a correlational relationship: .00 -
.20 = no or negligible correlation; .20 .40 = low degree of correlation; .40 .60 = moderate
degree of correlation; .60 .80 = marked degree of correlation; and .80 1.00 = high correlation.
According to Franzblau's (1958) standards the association between the BRIEF and the validated
tools is moderate. To further demonstrate the construct validity of the BRIEF, the Principal
Component Analysis findings suggest the BRIEF health literacy screening tool measures one
distinct construct "health literacy" (eigenvalue = 2.388) accounting for 60% of score variance.
The remaining eigenvalues were less than one, and thus were not retained. These findings
validate the BRIEF as a health literacy screening tool as compared to the REALM and
The area under the ROC for the individual screening items and the BRIEF were calculated
for detecting individuals with inadequate/marginal health literacy skills using the STOFHLA and
REALM as state variables. "State" variables are recognized as the "true" indicator of presence of
"disease" when comparing screening tools using the ROC method. AUROC curves, including
confidence intervals and sensitivity, were calculated to indicate the degree to which the BRIEF
would identify respondents with (1) inadequate health literacy skills; versus (2)
inadequate/marginal health literacy skills. The REALM and STOFHLA served as the state
variables. Using the REALM as the state variable, the BRIEF had a slight increase in accurately
identifying individuals with inadequate health literacy, .79 (95% CI = .70-.87), versus inadequate
and marginal, .69 (95% CI = .64-.75), respectively. Again, using the REALM as the state
variable, the BRIEF screening tool had a higher AUROC than any of the single BRIEF items,
indicating the BRIEF screening tool was a more sensitive screening tool for identifying
inadequate health literacy skills than the single items (See Table 4-7). With the STOFHLA as the
state variable, findings indicate the BRIEF items identify individuals with inadequate health
literacy skills, .76 (95% CI = .69-.83) with slightly more accuracy than individuals with
inadequate or marginal health literacy skills 74 (95% CI = .67-.80). All of the items had an
AUROC greater than 0.5 at 95% CI. The BRIEF had a higher AUROC than any of the single
items, indicating the BRIEF is a better screening tool than the single items (See Table 4-8).
Figures 4-2, 4-3, 4-4, and 4-5 illustrate the AUROC curve for the BRIEF in identifying
individuals with inadequate and inadequate/marginal health literacy using the STOFHLA as the
state variable. Figures 4-6, 4-7, 4-8, and 4-9 illustrate the AUROC curve for the BRIEF in
identifying individuals with inadequate and inadequate/marginal health literacy using the
REALM as the state variable. Sensitivities and 1-specificities for the BRIEF using the REALM
and STOFHLA as state variables are shown in Tables 4-9 & 4-10.
Based on previous research (Chew et al., 2004; Wallace et al., 2006) and the statistical
findings, three levels are recommended for interpreting BRIEF score intervals: 4-12 =
inadequate; 13-16 = marginal; and 17-20 = adequate. Additionally, an Analysis of Variance
(ANOVA) was conducted to determine if the group differences for the score intervals were
significant to further validate the proposed scoring intervals. Findings indicate the three
proposed BRIEF levels were significantly different from one another on the REALM (F = 28.63,
p <.000) and STOFHLA (F = 35.32, p <.000). Post Hoc Tukey analyses suggests all levels were
significantly different from one another on the REALM and STOFHLA at p <.00.
Research Question # 2
What patient variables are associated with health literacy level in an ambulatory VA
A step-wise multiple regression analysis was conducted, with dummy coded variables, to
examine the degree of association between the outcome variable (STOFHLA) and the
explanatory variables, including socio-demographic variables (age, education level, gender,
ethnicity, perceived reading ability and perceived ability to define "health literacy") and health
status variables (diabetes, stroke, high blood pressure). Gender and socio-economic status was
originally included however they were ultimately excluded from the final analysis. Because the
sample population was predominantly male (94.2%) gender was excluded from the regression
analysis. Further the socio-economic variable was deemed unsuitable due to the item, which
inquired about the participants' ownership of his/her home rather than income level. Further the
inclusion of these two variables in the preliminary analysis did not glean any significant findings.
Findings indicate the R2 Of .34 was statistically significant, F (7, 357) = 26.23, p < .000,
suggesting the explanatory variables: age, perceived reading ability, educational level, ethnicity,
perceived ability to define "health literacy," and reporting having Diabetes were j ointly
associated with 34% of the STOFHLA score variance. The interpretation of the unstandardized
regression coefficients of the statistically significant explanatory variables are as follows. For
each unit increase in STOFHLA score the participants' age decreased by 1.72 years (b = -1.72, t
(3 57) = -4.50, p =.000); perceived reading ability increased .21 units (b = .21, t (3 57) = 6.90, p
=.000); and educational level increased 1.2 units (b = 1.20, t (357) = 3.59, p =.000). Perceived
ability to define "health literacy" was associated with higher health literacy scores on the
STOFHLA (b = .99, t (357) = 2.48, p =.013). Being African American (b = -2.76, t (357) = -
3.01, p =.003) or Hispanic (b = -4.65, t (357) = -2.31, p =.022) was associated with lower health
literacy scores. Interestingly, African Americans were the youngest group (56.7 years) with
Native Americans being the eldest (68.6 years), white participants were the second eldest group
(62.7 years); the Hispanic participants had a mean age of 61.9 years. These Eindings should be
interpreted with caution due to the small sample sizes by ethnic orientation. Further, having
diabetes was associated with lower health literacy scores (b = 1.56, t (3 57) = 2. 16, p =.03 1).
Whether inadequate health literacy resulted in their diabetic condition cannot be determined.
Research Question # 3
What is the patients' level of knowledge, readiness & confidence related to health
literacy in an ambulatory VA healthcare setting?
To answer the third research question a univariate statistical analysis was conducted.
Frequencies and proportions were calculated to examine participant responses to survey items
related to their level of knowledge, readiness & confidence related to health literacy. The
maj ority (65.9%) of participants reported they were able to define "health literacy." However,
participants were not required to provide a definition or demonstrate their knowledge of health
literacy. Further, 21.5% of participants reported they were neutral in feeling as if they could
define health literacy and 12.7% strongly disagreed with the statement. The maj ority of
participants felt confident in their ability to seek support related to health literacy (86.9%); of the
remaining participants 9. 1% were neutral and 4. 1% disagreed. Eighty-four % of participants
reported receiving support with health information; while nearly 10% were neutral, and 6.1%
disagreed. Eighty-seven % intended to seek support related to health information in the future;
10.4% were neutral, and 2.1% disagreed. Frequencies and proportions are illustrated in Table 4-
Research Question # 4
What is the patients' level of confidence and readiness related to receiving patient
health education in an ambulatory VA healthcare setting?
Univariate statistical analysis was conducted to answer the final research question.
Frequencies and proportions were calculated to examine participant responses to survey items
related to their readiness & confidence related to patient health education. The maj ority (77.2%)
of participants is aware of health education resources that are available to them; 14.2% are
neutral, and 8.6% are not aware of available resources. Eighty-one % of participants report
having received health education resources in their lifetime; 11.55% were neutral, and 7.2% had
not received any resources. However participants did not report which resources they were aware
of or what health education resources they had received.
The maj ority of participants were confident in their ability to seek patient health education
(89.8%); 6.7% were neutral and 3.5% disagreed. Nearly 86% intend to seek patient health
education resources in the future; while 11.3% were neutral, and 3% do not intend to seek
resources. The meaning of the terms "ability to seek," "intention to seek," "patient heath
education resources" and "future" was subj ect to participants' interpretation, as definitions were
not provided. These findings suggest the maj ority of participants have received patient health
education and are confident in their ability to seek patient health education. Frequencies and
proportions are illustrated in Table 4-13.
This chapter reports findings from the participant survey and the health literacy screening
tools. A population profile was illustrated; most veteran participants were older white males.
Findings suggest between 7% to 20% of respondents self-report inadequate health literacy skills
and between 8% to 37% demonstrate marginal health literacy skills and between 43% to 83%
possess adequate health literacy skills. Bivariate analysis indicates a significant correlation
among all three screening tools. The BRIEF tool was able to identify individuals with inadequate
and marginal health literacy skills. However the tool has higher sensitivity (accuracy) when
identifying individuals with inadequate health literacy, rather than those with adequate health
literacy. Multivariate analysis indicated demographic variables were jointly associated with 34%
of the STOFHLA score variance. Finally, univariate statistics suggest the majority of
participants were knowledgeable about health literacy and health education; confident in their
ability to seek support; and had received support related to health literacy and patient health
education. Chapter 5 presents a discussion and recommendations from the study for future
research in the field of health literacy and patient education.
Table 4-1. Demographical distribution by age, gender, education, and ethnicity.
Demographical Variables N
Male 356 (94.2%)
Female 19 (5%)
Missing 3 (.8%)
Average 61.5 (SD=11.9)
Elementary School (Grades 1st 5th) 4 (1.1)
Junior High School (Grades 6th th) 11 (2.9)
Some High School (Grades 9th 12th) 56 (14.9)
High School/GED 98 (25.9)
Some College 126 (33.3)
College Degree 80 (21.2)
Trade School 1 (.3)
Missing 2 (.5)
African American 69 (18.3)
White 278 (73.5)
Hispanic 12 (3.2)
Native American 12 (3.2)
Asian American 1 (.3)
Other 3 (.8)
Missing 3 (.8)
Table 4-2. Frequency of participants' self reported reading ability.
Perceived Reading Ability (%)
Excellent 95 (25.1%)
Very Good 113 (29.9%)
Good 113 (29.9%)
Fair 42 (11.1%)
Poor 13 (3.4%)
Table 4-4. Cross tabulation table for BRIEF and STOFHLA levels for sample data.
STOFHLA Levels Inadequate Marginal Adequate Total
Inadequate 17 13 4 34
Marginal 10 14 5 29
Adequate 49 111 155 315
Total 76 138 164 378
Table 4-5. Cross tabulation table for BRIEF and REALM levels for sample data.
REALM levels Inadequate Marginal Adequate Total
Inadequate 15 9 1 25
Marginal 31 49 33 113
Adequate 30 80 130 240
Total 76 138 164 378
Table 4-6. Pearson Product Moment correlation coefficients between the REALM,
& BRIEF Item Scores.
N BRIEF BRIEF BRIEF BRIEF BRIEF
Item #1 Item #2 Item #3 Item #4 All Items
STOFHLA 378 .32 .42 .28 .28 .42
REALM 378 .34 .38 .28 .21 .40
Table 4-7. Areas under the receiver operating characteristic curve for BRIEF and BRIEF items
with REALM as the state variable.
Test Variable(s) Inadequate Inadequate & Marginal
BRIEF .79 (.70-.87) .69 (.64-.75)
BRIEF Item #1 .73 (.62-.84) .63 (.57-.69)
BRIEF Item #2 .71 (.59-.84) .68 (.63-.74)
BRIEF Item #3 .69 (.59-.79) .65 (.59-.70)
BRIEF Item #4 .68 (.58-.78) .59 (.53-.65)
Table 4-3. Participants' level of health literacy as measured by the BRIEF, REALM, &
Health Literacy Level
Table 4-8. Areas under the receiver operating characteristic curve for BRIEF and BRIEF items
with STOFHLA as the state variable.
Test Variable(s) Inadequate Inadequate & Marginal
BRIEF .76(.69-.83) .74 (.67-.80)
BRIEF Item #1 .66(.56-.77) .64 (.56-.72)
BRIEF Item #2 .75(.65-.84) .69 (.61-.77)
BRIEF Item #3 .65(.56-.74) .66 (.59-.73)
BRIEF Item #4 .68(.59-.78) .66 (.59-.74)
Table 4-9. Performance of BRIEF in detecting inadequate and marginal health literacy using
STOFHLA as the state variable.
Inadequate Inadequate and Marginal
Health Literacy Health Literacy
Table 4-10. Performance of BRIEF in detecting inadequate and marginal health literacy using
REALM as the state variable.
Inadequate Inadequate and Marginal
Health Literacy Health Literacy
Table 4-11. Stepwise Model: Unstandardized Coefficients, Standardized Regression
Coefficients, t-test Statistics, and Partial r-squares.
Variables b Std Error P t P
Intercept 43.75 2.89 15.14 .000
Age -1.72 .38 -.23 -4.50 .000
Perceived reading ability .21 .03 .32 6.90 .000
Educational level 1.20 .33 .17 3.59 .000
African American -2.76 .92 -.14 -3.01 .003
Perceived ability to define .99 .40 .12 2.48 .013
Hispanic -4.65 2.01 -.10 -2.31 .022
Diabetes 1.56 .72 .09 2.16 .031
Table 4-12. Frequencies and proportions of participant responses to items about health literacy
and heath information.
Agree Neutral Disagree
Participants reported they... N
...were able to define "health literacy." 245 80 47
(65.9) (21.5) (12.7)
...were confident in their ability to seek health literacy 324 34 15
support. (86.9) (9.1) (4.1)
...have received support regarding health information. 314 37 23
(84.0) (9.9) (6.1)
...intended to seek support related to health information 327 39 8
in the future. (87.4) (10.4) (2.1)
Table 4-13. Frequencies and proportions of participant responses to items about their knowledge,
readiness & confidence related to patient health education.
Agree Neutral Disagree
Participants reported they... N
...were aware of health education resources. 288 53 32
(77.2) (14.2) (8.6)
...were confident in their ability to seek health education 336 25 13
support. (89.8) (6.7) (3.5)
...have received health education resources. 303 43 27
(81.2) (11.5) (7.2)
...intended to seek patient health education resources in 319 42 11
the future. (85.8) (11.3) (3.0)
b 50 43
(u 40 37 REL
n- 30 2
Inadequate Marginal Adequate
Lev el of Health Lite racy Skills
Figure 4-1. Participants' level of health literacy as indicated by the BRIEF, REALM, &
3 0.2 0.4 0.6 0.8
Source of the Cuive
Diagonal segments ar-e produced by ties
Figure 4-2. The individual BRIEF items for identification of inadequate health literacy using the
Receiver Operating Characteristic (ROC) Curves with STOFHLA as the state
) 0.2 0.4 0.6 0
Diagonal segments aree produced by ties
Figure 4-3. The BRIEF health literacy assessment for identification of inadequate health literacy
using the Receiver Operating Characteristic (ROC) Curve with STOFHLA as the
3 02 04 06 08
Source of the Cuive
Diagonal segments are produced by ties
Figure 4-4. The individual BRIEF items for identification of inadequate and marginal health
literacy using the Receiver Operating Characteristic (ROC) Curves with STOFHLA
as the state variable.
02 04 06 0
Diagonal segments alee produced by ties
Figure 4-5. The BRIEF health literacy assessment for identification of inadequate and marginal
health literacy using the Receiver Operating Characteristic (ROC) Curve with
STOFHLA as the state variable.
00 02 04 06 08 10
Diagonal segments are produced by ties
Figure 4-6. The individual BRIEF items for identification of inadequate health literacy using the
Receiver Operating Characteristic (ROC) Curves with REALM as the state variable.
Source of the Curve
S02 04 06 0
Diagonal segments ale produced by ties
Figure 4-7. The BRIEF health literacy assessment for identification of inadequate health literacy
using the Receiver Operating Characteristic (ROC) Curve with REALM as the state
0 0.2 0.4 0.6 0.8
Source of the Cuive
Diagonal segments are produced by ties
Figure 4-8. The individual BRIEF items for identification of inadequate and marginal health
literacy using the Receiver Operating Characteristic (ROC) Curves with REALM as
the state variable.
02 04 06 0
Diagonal segments are produced by ties.
Figure 4-9. The BRIEF health literacy assessment for identification of inadequate and marginal
health literacy using the Receiver Operating Characteristic (ROC) Curve with
REALM as the state variable.
SUMMARY, DISCUSSION, AND RECOMMENDATIONS
The purpose of this study was to establish a means for clinicians and health educators to
quickly screen patients for inadequate/marginal health literacy and to determine if there were
explanatory variables associated with one's health literacy level. Data was also collected to
determine the participants' awareness of health literacy; their awareness of health education; and
their readiness to utilize health education. This study was conducted with a predominantly male
veteran population in an ambulatory clinical care setting within the Veteran Health System in
North Florida/South Georgia.
The protocol utilized two validated health literacy screening assessments: Rapid Estimate
of Adult Literacy in Medicine (REALM) and Test of Functional Health Literacy in Adults, short
form (STOFHLA) to investigate the proposed BRIEF screening tool. A self-administered survey
was also used to collect participant data. Findings indicate the BRIEF is significantly correlated
with the REALM and STOFHLA and thus can be utilized as a valid measure of health literacy.
Clinicians and health educators can use the BRIEF screening tool to quickly identify patients
with inadequate health literacy. Further, several demographic and personal variables were
significantly associated with health literacy level, including age, ethnicity, educational level, and
diabetic condition. These personal characteristics can be used as potential indicators of
inadequate health literacy. These findings also have implications for identifying and responding
to priority populations when developing intervention efforts. Findings indicate the majority of
participants were knowledgeable about health literacy and health education; confident in their
ability to seek support; and had received support related to health literacy and patient health
education. However, findings also indicate a significant portion of the population are not aware
of health literacy and are not aware of the patient health education resources that are available to
them. These findings, their implications and limitations are further discussed in the following
Health literacy is a national health issue, affecting one in three Americans and costing the
healthcare system billions of dollars annually. An individual's level of health literacy affects how
he/she navigates the healthcare process, communicates with healthcare providers, manages
health issues, and adheres to a healthcare regimen. The outcomes of inadequate health literacy
produce billions of dollars in unnecessary costs through prescription drug misuse; unnecessary
hospitalization; litigation; illness; and death. Research suggests there is a need to identify
patients with health literacy needs and respond to these needs throughout their healthcare process
(Safeer & Keenan, 2005; Baker, 2006).
Healthcare providers face the challenge of identifying the needs of their patients and
responding accordingly. This challenge becomes especially troubling when a providers' accuracy
in identifying individuals with low literacy is only 40% (Lindau et al., 2002). Subsequently,
providers are left in need of a quick and efficient means of identifying patients with inadequate
health literacy skills. Knowledge of a patient' s health literacy level, much like other patient data,
will allow providers to adjust their healthcare interactions to more effectively meet the needs of
their patients, thereby improving the quality of patient-provider interaction, promoting adherence
and thus resulting in better health outcomes. In response to this clinical need, this study offers
clinicians and educators an efficient health literacy screening tool; provides knowledge of socio-
demographic variables associated with health literacy; and identifies the population's awareness
of health literacy and readiness for patient health education. The following sections provide a
discussion of these findings.
This sample is representative of the current veteran population receiving Veteran Health
System services in North Florida. This sample consists primarily of older, white males. More
than 50% of participants report having high blood pressure; more than one third have diabetes;
and 10% report having had a stroke in their lifetime. Based on results of the three screening
tools, 7% to 20% of respondents had inadequate health literacy skills; 8% to 37% had marginal
health literacy skills; and 43% to 83% had adequate health literacy skills. As stated previously,
the refusal rate for this study was 10%. Though this percentage is not unusually high, one might
speculate that individuals who know they cannot read would not want to participate in a literacy
study. Thus individuals with inadequate health literacy skills may have avoided participation.
Further, if such is the case, the percentage of individuals with inadequate health literacy skills
may be under-represented by study findings.
Estimates of inadequate health literacy for the BRIEF (20. 1%) were higher than those of
the REALM (6.6%) or STOFHLA (9.0%). It should be noted the BRIEF estimates are more
representative of previous findings of larger studies suggesting 33% of individuals have
inadequate health literacy skills (Kirsch, Jungeblut, Jenkins, & Kolstad, 1993; Partnership for
Clear Health Communication, n.d.; Institute of Medicine, 2004). Due to the variance in
measurement among the three screening tools (REALM; STOFHLA; BRIEF), it cannot be
determined if the differences in measurement result from error in the validated measures
(REALM & STOFHLA) or the proposed screening tool (BRIEF). However, previous research
supports the findings of the BRIEF screening tool and thus, further supports the validity of the
BRIEF in the clinical ambulatory setting (Chew et al., 2004; Wallace et al., 2006).
The discrepancy in results from the three assessments (BRIEF, REALM, & STOFHLA) is
likely a function of the slight differences in the scope of each tools' measurement of the
construct of health literacy. However, it should be noted the Eindings of this study suggest the
BRIEF measures one distinct construct, "health literacy." Despite discrepancies between the
three screening measures, study findings provide evidence supporting the assertion that all three
of the health literacy screening tools are positively correlated; and the BRIEF screening tool will
accurately identify participants with inadequate and marginal health literacy skills. Findings also
suggest the addition of the fourth item, "How often do you have a problem understanding what is
told to you about your medical condition?" increased the internal validity of the BRIEF
screening measure. The results of the ROC (receiver operating characteristic) analysis provide
further evidence that the BRIEF is more sensitive when detecting inadequate and marginal health
literacy when compared to the single items.
However, data indicate that though the BRIEF is sensitive, it is not always accurate due to
false positives. This finding suggests the BRIEF will identify individuals as having inadequate
health literacy when they in fact have adequate health literacy skills. Findings also indicate the
BRIEF items identify individuals with inadequate health literacy skills with more accuracy than
individuals with marginal health literacy skills. Despite these minor issues, the BRIEF tool will
assist providers in identifying individuals with health literacy needs in the clinical setting. The
BRIEF will allow clinicians and educators to be sensitive to patients' needs; if it is discovered no
assistance with health literacy is needed, then it is simply not provided. In this case it is better to
be safe than sorry.
Further, it should be noted the BRIEF tool only requires one to two minutes to administer
while the administration times for the other tools are between four (REALM) and seven
(STOFHLA) minutes. This time may seem insignificant when considering one assessment.
However clinicians commonly see many patients a day, thus precious time can be spared using
the BRIEF screening measure. For example, if the average patient visit is 20 minutes and a
clinicians has twenty visits a day, if the physician administers the BRIEF it will take away less
than two minutes from each patient visit, about 20 to 30 minutes to screen all visiting patients
from the daily schedule; versus approximately 60 minutes (REALM) or 140 minutes
(STOFHLA) respectively for the other two instruments. The loss of approximately one to two
patient interactions per day is made acceptable through greater satisfaction and better patient
outcomes. Further, once patients have been screened, re-evaluation is only necessary every Hyve
to ten years, or after traumatic events (i.e., head injury).
Chew and colleagues (2004) recommended three items [How often do you have someone
help you read hospital materials; How confident are you filling out medical forms by yourself;
How often do you have problems learning about your medical condition because of difficulty
understanding written information?] and suggested using a Hyve point Likert scale, using the
response "sometimes" as the cut-off for inadequate health literacy. Thus, a response of
"sometimes" on any of the three items, which would be scored as a '3', indicated
inadequate/marginal health literacy. As with Wallace and colleagues (2006), who suggested
using the item, "How confident are you filling out medical forms by yourself!" who also
recommended a five-point Likert response scale, using the response "sometimes," as the cut-off
for inadequate/marginal health literacy skills. Likewise, the findings of this study recommend the
following scale for interpretation. A BRIEF score of: 4-12 = inadequate; 13-16 = marginal; and
17-20 = adequate health literacy skills.
Findings from this study should be interpreted with caution. Further research is needed to
validate the efficacy of the BRIEF health literacy screening tool to determine the rate of
inadequate health literacy in the general population and among priority sub-populations. Further,
though the three measures used in this study represent varying aspects of the concept of health
literacy, contributing to error in measurement, the REALM and STOFHLA are currently the
most studied and validated health literacy screening measures available. Thus, researchers and
practitioners are faced with the challenge to continue efforts in measuring health literacy in
diverse populations. These efforts remain imperative as researchers seek to understand the
interpersonal and cultural contexts of communication in medical encounters (Street, 2003).
In addition to the use of the BRIEF health literacy screening as a means for identifying
individuals with inadequate health literacy, findings of the step-wise multiple regression analysis
suggested there were several statistically significant explanatory variables which could help
predict an individual's health literacy skills. Perceived reading ability; age; educational level;
ethnicity; perceived ability to define health literacy; and having diabetes were j ointly associated
with 34% of the STOFHLA score variance.
One' s perception of his/her ability to read was the greatest predictor of health literacy
skills. These findings provide implications for the role of self-efficacy in determining one's
ability to read and understand health information. As with many other behaviors requiring skill,
an individual's self-efficacy is a significant predictor in determining if the individual can
demonstrate that particular skill. Therefore, in addition to providing patients with skills,
educators and clinicians should provide verbal support and empower patients to promote self
confidence, thus increasing one's sense of self-efficacy contributing to the acquisition and
successful utilization of health literacy skills without embarrassment or shame.
Further, people who report they have the ability to define health literacy are more likely to
have higher health literacy skills. This association is logical but likely involves a moderating
variable such as self-efficacy; awareness; or educational level. Further research is needed to
decipher if the relationship is direct or moderated by another variable. However, findings provide
implications for allocating resources for large scale health literacy public communication
campaigns to increase awareness and open communication channels with the general public.
Communication campaigns can increase the awareness of the general population and promote the
likelihood of individuals becoming aware of the topic of health literacy the associated outcomes
and thus seek knowledge and support when interacting with health information and the
Being older, less educated, and being a minority are variables related to lower levels of
health literacy; these findings replicate previously published findings (Kirsch, Jungeblut,
Jenkins, & Kolstad, 1993; Sudore et al., 2006). Though the Department of Health and Human
Services (2000) suggest anyone is likely to need assistance with health information, current
literature and the findings of this study suggest there are demographic variables associated with
and thus predictive of an individuals' health literacy level. Though all patients should receive
screening and assistance with health information, health educators and clinicians should be alert
to priority populations that are adversely affected by inadequate health literacy such as patients
who are poor, elderly, uneducated, and/or a minority.
Finally, in this sample of participants, having diabetes is predictive of having lower health
literacy scores. Whether having inadequate health literacy skills was a contributing variable in
the outcome of diabetes, cannot be determined. What is relevant is the role of inadequate health
literacy in the management of a condition such as diabetes. Previous research has identified
associations between management of diabetes and health literacy level (Schillinger et al., 2002;
Kutner, Greenberg & Baer, 2006). These findings provide implications for clinical practice. If
participants with diabetes are likely to have lower health literacy scores, providers need to be
aware of this risk factor and respond through clinical intervention. Further, these patients may
have difficulty managing their diabetic condition, therefore requiring individualized
modifications to their management regimen. Patient education may provide an option for
monitoring their condition and providing them with the necessary information and disease
Though anyone is likely to need assistance with health information regardless of
demographic status (U.S. Department of Health and Human Services, 2000), certain populations
are at increased need and offer support to previous research which suggests being elderly
(Schneider, 2006; Kirsch et al., 1993), a minority (Kirsch et al., 1993; Cooper, & Roter, 2003;
Sudore et al., 2006b), and having less education (Lukoschek, Fazzari, & Marantz, 2003; Howard,
Sentell, Gazmararian, 2006, p. 857) is associated with low and marginal health literacy skills.
Study results imply immediate need for increasing health literacy assistance to these groups. For
many reasons, initially identifying these groups for assistance with healthcare information can
begin to lessen existing health disparities among the general population. The findings of this
study also contribute supportive evidence for the association between socio-demographical
variables and health literacy skills, contributing knowledge to support the influence of the
cultural context within the framework of the ecological model of medical encounters (Street,
Based on the findings of this study it is recommended that health educators and clinicians
utilize the findings to: (1) equip themselves' with an efficient, inexpensive screening tool to
identify inadequate health literacy; (2) identify demographical characteristics that may put
individuals at risk for having inadequate health literacy skills; and (3) address health literacy
when exchanging health information with patients.
In making the connection between screening patients' health literacy skills and intervening
with individuals' with inadequate health literacy, this study attempted to identify the sample
population's level of awareness and readiness to engage in health literacy related initiatives set
forth by educators, clinicians, and the veteran healthcare system in general. The findings of this
study indicate the maj ority of participants were knowledgeable about health literacy; confident in
their ability to seek support; and had received support related to health information. The maj ority
of participants reported they were able to define "health literacy." However, they were not
required to define the concept. This limitation should be amended in future research; it is
recommended that future studies ask participants to define "health literacy" in an effort to
distinguish between perception of one' s ability to define the concept and their actual ability.
Further, more than 30% of respondents were neutral or strongly disagreed about their ability to
define "health literacy." Based on the results of these survey items, there is an identified need to
continue efforts in raising awareness and educating patients about the concept of health literacy
and the associated outcomes (increased risk of hospitalization; increased cost of healthcare; etc.).
Unfortunately, 16% of respondents reported not receiving support with health information;
12% did not report intention to seek support related to health information in the future; and
similarly, 13% did not report feeling confident in their ability to seek support related to health
literacy. These findings suggest that among this small portion of the population, information
related to health literacy is needed to enhance the level of knowledge possessed by consumers of
the VA health system. The higher the percentage of patients who know what health literacy is;
are aware of their need for assistance; and are confident enough to seek assistance, the better the
expected health outcomes for the general patient population. A health communication campaign
highlighting the concept of health literacy; identifying associated outcomes and interventions to
prevent adverse outcomes may be an effective way to inform and educate the entire population.
Increased awareness cannot amend the issue of health literacy. However increased awareness can
mobilize the target population into a stage of contemplation and promote moving them into the
action phase of obtaining information about health literacy and the help they need to understand
and use health information.
The maj ority of participants reported awareness of available health education resources.
However, participants were not required to list the resources; therefore there is no way to
determine which resources they were aware of, nor how many they could identify. Future
research should collect this data to better understand specific issues associated with awareness
and access to health education resources. More than 20% of this sample reported being unaware
of available patient health education resources. Based on these findings it is suggested further
efforts are needed to inform the population of the current availability of resources within the VA
health system. Because 18% of participants reported they had not received health education
resources in their lifetime; it may be possible the sample population is not aware when they are
receiving patient education or are unaware of services they are entitled to receive. If so, this
provides implications for patient health education practices within the VA. One way to address
this issue is to label and specifically identify patient education materials as such.
Fourteen % of the sample did not report intention to seek patient health education
resources in the future. This may be related to the 10% of participants who did not report feeling
confident in their ability to seek health education support. Either way, there is a significant
portion of the population who is unaware of resources; has not received health education
resources; does not intend to seek health education resources; and does not report having the
confidence to seek educational resources. These findings provide a clear indication for the need
to raise the populations' awareness of available resources, and provide access and support for
individuals who may need health education but may not be prepared to ask for help. This is a
common challenge for clinicians and practitioners. However providing information and shame-
free assistance can promote the likelihood of moving patients from pre-contemplation and
contemplation into the action phase of obtaining health education resources and participating in
their healthcare and adhering to the healthcare regimen.
Through the combined efforts of researchers, educators, clinicians, and administrators,
with institutional support and resources, patients can more effectively share in the responsibility
for their healthcare. Patients who have the skills to actively participant in their healthcare process
can more effectively function as a reliable knowledgeable resource, who can exchange and
comprehend health information and make functional medical and personal health decisions to
assume ownership of their healthcare process and produce better fiscal and health related
outcomes. When patients assume this role the professional members of the healthcare team also
benefit, because (1) the patient is assuming responsibility for himself as an influencing factor in
their health outcomes; and (2) health literate, functional patients are more likely to adhere to
healthcare regimens and manage health conditions resulting in reduced use of healthcare
services, which in turn helps to preserve the vitality of healthcare providers. In closing, screening
patients for inadequate health literacy; informing them about health literacy and the associated
outcomes; and providing health education and information resources is a win-win for patients,
clinicians and the larger healthcare system. In the case of health literacy clearly, as the adage
states, an ounce of prevention is worth a pound of cure.
Study limitations should be noted when interpreting results. Patient participants were
recruited from the North Florida region and therefore results cannot be generalized to other
populations or geographical regions of the United States. Further, because the sample was not
randomized, potentially confounding variables may have introduced error into measurement that
is not accounted for in the analysis. Thus, inferences should be made with caution. Further the
homogeneous nature of this small sample may have influenced findings. Self-reports on the
BRIEF and the participant survey also pose a limitation to inferences made from study findings.
It is incumbent upon the medical community to acknowledge the issue of health literacy as
a national health priority. Researchers have the responsibility to continue the development of
screening tools and strategies to ensure that patients receive assistance in overcoming barriers
that limit their ability to function adequately in the healthcare environment. Specialists in health
literacy suggest that past efforts have been effective in advancing the field of study toward the
goals of the Institute of Medicine; however, persistent efforts for the next several decades are
imperative for continued progress (Parker & Kindig, 2006). Saha (2006) suggests enhanced
health literacy levels are a fundamental component of reducing health disparities among
minorities and individuals with low socioeconomic status. Incremental and systemic change is
necessary to improve the health literacy of the general public, particularly priority populations
(Chew, Bradley, Boyko, 2004). Additionally, researchers, educators and providers need to find
innovative ways to identify individuals with inadequate levels of health literacy and intervene to
protect the integrity of their healthcare process. While policy makers and administrators need to
allocate funds and support programs that can serve as resources for patient health educators,
healthcare providers, and patients. There are several ways health educators, healthcare providers,
and other health professionals can collaborate to increase the quality of healthcare experience
and outcomes for individuals with low health literacy. Recommendations and implications for
the roles of educators; health professionals; and future research needs follow.