CURISSUE1

Cutillietal.2018.pdf

Home >Psychology homework help >CURISSUE1

1.5

ANCC Contact Hours

Introduction Older adults in our country are at risk for suboptimal health due to their high rate of low health literacy. Fifty- nine percent of older adults have low health literacy. Research in the older adult population has shown that low health literacy is associated with increased mortal- ity, worse physical functioning and mental health, fair/ poor rating of health, heart failure health outcomes, less preventive care and health-promoting behaviors ( Baker et al., 2007 ; Bostock & Steptoe, 2012 ; Chen, Hsu, Tung, & Pan, 2013 ; Kobayashi, Wardle, Wolf, & von Wagner, 2015 ; Mahnoush et al., 2015 ; Mottus et al., 2014 ; Smith

BACKGROUND: Low health literacy in older adults has been associated with poor health outcomes (i.e., mortality, decreased physical and cognitive functioning, and less pre- ventive care utilization). Many factors associated with low health literacy are also associated with health disparities. Interaction with healthcare providers and sources of health information are infl uenced by an individual’s health literacy and can impact health outcomes. PURPOSE: This study examined the relationships between health literacy, sources of health information, and demographic/ background characteristics in older adults (aged 65 years and older) related to health literacy and disparities. METHODS: This descriptive, correlational study is a sec- ondary analysis of the 2003 National Assessment of Adult Literacy, a large-scale national assessment. RESULTS: Older adults with lower health literacy have less income and education, rate their health as poor or fair, have visual or auditory diffi culties, need help fi lling out forms, reading newspaper, or writing notes, and use each source of health information less (print and nonprint). Many of these characteristics and skills are predictive of health literacy and associated with health disparities. CONCLUSION: The results expand our knowledge of characteristics associated with health literacy and sources of health information used by older adults. Interventions to improve health outcomes including health disparities can focus on recognizing and meeting the health literacy demands of older adults.

Health Literacy, Health Disparities, and Sources of Health Information in U.S. Older Adults

Carolyn Crane Cutilli ▼ Lynn C. Simko ▼ Alison M. Colbert ▼ Ian M. Bennett

Carolyn Crane Cutilli, PhD, RN, Patient and Family Education Specialist, Hospital of the University of Pennsylvania, Philadelphia; and Adjunct Professor, Master’s Degree Program, American International College, Maple Glen, PA.

Lynn C. Simko, PhD, RN, CCRN, Clinical Associate Professor, School of Nursing, Duquesne University, Pittsburgh, PA.

Alison M. Colbert, PhD, PHCNS-BC, Associate Professor and Associate Dean for Academic Affairs, School of Nursing, Duquesne University, Pittsburgh, PA.

Ian M. Bennett, MD, PhD, Department of Family Medicine, Family Medicine Research Section, University of Washington School of Medicine, Seattle .

No confl icts of interest or sources of funding for the authors .

DOI: 10.1097/NOR.0000000000000418

et al., 2015 ; White, 2008 ), more diffi culty with activities of daily living and activity limitations ( Wolf, Gazmararian, & Baker, 2005 ), and more rapid decline in executive function ( Sequeira et al., 2013 ). Health liter- acy has been found to be a mediator for health outcomes in older adults with heart failure ( Wu, Moser, DeWalt, Rayens, & Dracup, 2016 ).

Many public and private organizations have made health literacy a priority and invested resources to help educate healthcare providers, including the American Medical Association (n.d.) , Institute of Medicine ( Nielsen-Bohlman, Panzer, & Kindig, 2004 ), The Joint Commission (n.d.) , Pfi zer (2015) , and Agency for Healthcare Research and Quality (2010) . The Centers for Disease Control and Prevention (2014) emphasizes the importance of older adults having adequate health literacy to address health concerns often associated with aging. Health literacy as a national priority is also dem- onstrated by its inclusion in Healthy People 2020 and the 2003 National Assessment of Adult Literacy (NAAL).

The NAAL is a national assessment that examined the relationship between literacy/health literacy and background characteristics in a representable sample of the U.S. population. Several characteristics examined in the NAAL have been identifi ed as variables associated with health disparities (e.g., disability, racial/ethnic group, geography, socioeconomic status) ( Healthy People 2020, n.d. ). The NAAL provides the opportunity to use a large-scale national assessment to examine

Andrea Mustafa

possible relationships between health literacy and disparities in older adults. For the overall adult popula- tion, lower income and education are associated with lower health literacy ( Kutner, Greenberg, Jin, & Paulsen, 2006 ; White, 2008 ).

Although insight into disparities related to health lit- eracy is critical, it is equally important to understand how those relationships then translate into specifi c health behaviors, such as health information seeking and interactions with providers. Reaching patients with information they deem useful is an important founda- tional step in self-management. The literature notes that older individuals with inadequate health literacy and chronic illnesses (asthma, diabetes, and congestive heart failure) have lower mean knowledge scores about their chronic condition compared with those with adequate health literacy ( Gazmararian, Williams, Peel, & Baker, 2003 ). Thus, we need to fi nd a way to support knowledge acquisition in this group. A fi rst step is understanding where older adults access health information.

As expected, the NAAL showed that the largest per- centage of all adults with below basic health literacy used each source of printed health information less (newspaper, magazine, books, Internet) compared with others with higher health literacy. Although participants with low health literacy used each nonprint source (radio/TV, family/friends/coworkers, and doctor/health- care provider) more than print sources, they still had the highest percentage of adults who did not use each non- print source ( Kutner et al., 2006 ). It is interesting to note that the percentage of those who use doctor/healthcare providers “a lot” decreases as health literacy decreases.

Current research on sources of health information (health information-seeking behavior) is heavily fo- cused on Internet usage. The Program for the International Assessment of Adult Competencies (PIAAC) showed that adults with a high school diploma used more text-based sources compared with those without a diploma ( Feinberg et al., 2016 ). In the NAAL, the average health literacy scores were highest for adults who sought health information from the Internet ( White, 2008 ). The literature suggests that older indi- viduals use the Internet less than younger persons but the gap is closing ( Kontos, Blake, Chou, & Prestin, 2014 ; Levy, Janke, & Langa, 2014 ; Pew Research Center, n.d. ; Tennant et al., 2015 ). Other studies note that for older adults, the physician is a main source of health informa- tion ( Campbell & Nolfi , 2005 ; Hall, Bernhardt, & Dodd, 2015 ; Morey, 2007 ). The NAAL data provide the oppor- tunity to examine the sources of health information used by older adults on a national level for strategy and policy development to impact health outcomes. Healthcare providers need to provide education where older adults are seeking it, especially those with low health literacy and at risk for health disparities.

This descriptive, correlational study is a secondary analysis using data from the 2003 NAAL to examine the relationships between health literacy, demographic/ background characteristics of older adults (aged 65 years and older), and sources of health information. The conceptual model for this study, “Causal Pathways Between Limited Health Literacy and Health Outcomes” ( Paasche-Orlow & Wolf, 2007 ), directed the selection of

variables and allowed for further exploration of the relationships posited. Specifi cally, the model suggests that an individual’s health outcome is affected by access and utilization of healthcare, provider–patient interac- tion, and self-care, all of which are infl uenced by health literacy level and key demographic variables. Self-care is conceptualized in the model as patients’ knowledge/ skills and extrinsic factor such as health education. For this study, self-care is operationalized within the con- text of health information and seeking: knowledge/skills (use of the Internet and e-mail, understanding medica- tion dosing, help needed with completing tasks such as writing notes, mathematics, reading newspaper) and health education (newspapers, magazines, books, TV/ Radio, family, friends and coworkers, and healthcare professionals). Using the model as a guide, this study sought to further explore how health literacy level and key demographic variables directly tied to health dis- parities are related to information-seeking self-care.

The research questions for this study are as follows:

1. What sociodemographic (educational attain- ment, income, race, region of the country, gen- der, marital status, U.S. citizenship, country of birth), background (computer usage, cognitive tasks, language, health status, vision, hearing, disabilities), and extent of health information use (sources of health information) variables are associated with older adults’ health literacy?

2. What variables related to health literacy, health disparities, and sources of health information are related to the health literacy levels of older adults?

Methods The 2003 NAAL was implemented through the U.S. Department of Education, Institute of Educational Sciences, National Center for Education Statistics (NCES). This large-scale national assessment was de- signed to provide an estimation of literacy and health literacy for U.S. populations such as older adults. To de- crease the burden on individual participants, a fraction, rather than all, of the assessment items were adminis- tered to each participant, resulting in no accurate indi- vidual score. The NCES used marginal maximum likeli- hood models to estimate health literacy scores for populations. AM software was developed to provide these estimates ( NCES, n.d.-b )

The health literacy measurement included 28 health literacy questions embedded in the literacy tasks ( White, 2008 ). The NAAL examined health literacy in relationship to various background demographics/ characteristics. Specifi cally, the questions focus on functional health literacy tasks centered on the follow- ing domains: clinical (medications, diagnosis, and treat- ment), preventative (self-care, preventing disease), and health system navigation (informed consent, health in- surance coverage) ( Kutner et al., 2006 ). The popula- tion’s health literacy (prose, document, and quantitative tasks) was categorized on the basis of mean health lit- eracy score: below basic (0–184), basic (185–225), inter- mediate (226–309), and profi cient (310–500) ( White &

Dillow, 2005 ). These categories were developed by a committee using the Bookmark method and “quasi- contrasting groups approach” ( Baldi, 2009 ).

SAMPLE This study included 2,668 nonincarcerated older adults (aged 65 years and older) who were part of the 18,000-person household sample from the NAAL study. The household sample was determined through a four- stage, stratifi ed area sample: primary sampling units of counties or groups of contiguous counties, secondary sampling units (segments) of area blocks, housing units with households, and fi nally eligible persons in house- holds. This sample was weighted to represent the total U.S. population. The participant’s assessment was con- sidered complete and included if the background ques- tionnaire and at least one task from each of the three scales were answered.

For individuals who completed the questionnaire but failed to answer any literacy tasks, regression-based im- putation methods were used ( Greenberg & Jin, 2007 ). The imputation procedure was instituted to avoid non- random unknown biases due to refusal. The analysis concluded that nonresponsive bias was negligible at the screening and background questionnaire stages ( Kutner et al., 2006 ). Participation in the NAAL was strictly voluntary.

DATA COLLECTION The assessment was administered on a one-on-one situ- ation using a computer-assisted personal interviewing system. Participants utilized everyday aids and other tools such as eyeglasses, magnifying glasses, rulers, and calculators when completing tasks. The assessment began with the 35-minute questionnaire on background information, followed by seven core literacy tasks ( Kutner et al., 2006 ). Topics in the background ques- tionnaire include political and social participation, labor force participation, literacy practices, job training and skills, family literacy, and areas described in the re- search questions.

The ability of the subjects to participate in the main assessment was determined by completion of seven ini- tial tasks. Those who struggled with these tasks were given an alternate assessment designed to present easier tasks fi rst and move onto highly contextualized material usually found at home or in the community. The NAAL consisted of 152 tasks divided into 13 blocks, with ap- proximately 11 questions per block. Each participant was given a booklet with three blocks of questions. Health literacy assessment questions were embedded in the assessment ( Kutner et al., 2006 ).

For this secondary analysis, the data were accessed through the public-use fi le “NAAL_2003_Health.am” ( NCES, n.d.-b ) located on the NCES’ NAAL website ( NCES, n.d.-a ). In this large database, missing data were managed during the data collection process. To compen- sate for missing data and avoid bias from the participant’s refusal to answer, this study used imputed answers. Imputed answers are based on the answers given by par- ticipants with the same background characteristics.

VARIABLES Thirty-two variables from the 2003 NAAL were chosen for this study. In this study, health literacy was the de- pendent variable and the sociodemographic and back- ground variables were independent variables. Variables were selected on the basis of their identifi cation in the literature on health literacy, health disparities, and sources of health information (health information- seeking behavior). The variables examined include gen- der, educational attainment, race, income, marital sta- tus, region, U.S. citizenship, country of birth, vision, hearing, language, disabilities, help with tasks, and sources of health information ( Cutilli, 2010 ; Healthy People 2020, n.d. ; Kutner et al., 2006 ; Paasche-Orlow & Wolf, 2007 ). Because the study was a secondary analy- sis, some variables did not produce usable data due to erroneous outcomes or error messages. These variables are citizenship, country of birth, language, and needing help with mathematics.

STATISTICAL ANALYSES The data were opened in the AM software for data edit- ing and statistical analyses. A data fi lter was set for the age 65 years or older. Data editing such as removing nonapplicable values and collapsing categories was completed as needed. Descriptive statistics include fre- quencies and measures of central tendency. To answer the fi rst research question, bivariate analyses using in- dependent t tests with Bonferroni adjustments as needed were conducted to explore the relationships of health literacy with sociodemographic and background characteristics associated with health literacy, health disparities risk factors, and sources of health information.

To assist with the clinical interpretation of the differ- ences, effects sizes were reported for the difference in health literacy for each variable. An effect size is consid- ered to be the smallest immediate difference that is clin- ically meaningful in the target population for the out- come of interest (i.e., health literacy in this study). Reporting the differences between the groups using an effect size index provides a more accurate interpreta- tion of the clinical signifi cance of results. As per Cohen (1992) , the difference between two group mean scores falls under the index Cohen’s d . Accordingly, an effect size of 0.20 is considered a small effect, 0.50 is consid- ered a medium effect, and 0.80 is considered a large ef- fect. Medium and large effects are considered substan- tial and of practical importance.

Finally, to answer the second research question, a simultaneous multiple linear regression was conducted to determine the predictive relationships of study vari- ables on health literacy. The selection of variables for regression was based on outcomes of the bivariate anal- yses. For inclusion, the variable had to have at least half of the categories in the bivariate analyses demonstrate signifi cance. The AM software allowed for the testing of the overall model but did not provide a measure of the robustness of the model through the quantifi cation of the variance explained. In addition, although the AM software is able to test the contribution of each predictor

to the overall model, it is unable to quantify the differ- ences within each level of the predictor.

Results PARTICIPANTS The 2,668 participants of this study represent older adults in the United States. Table 1 describes demo- graphic characteristics of the sample population. The majority of participants were female, white, married, having some high school education or were high school graduate, and earned above $40,000 per year. The mean health literacy score for older adults was 214 (translat- ing to basic level), with 59% of the population having below basic or basic health literacy.

SOCIODEMOGRAPHIC CHARACTERISTICS There was no difference between genders, with the mean health literacy score in both males (214, 2.3) and females (214, 3.4) at the basic level. Mean health literacy scores were at the basic level (217, 2.2) for White/ Hispanic category and below basic (182, 6.25) for Black and Other (including multiracial) category, with a sig- nifi cant difference demonstrating medium effect size.

The variables related to U.S. citizenship, country of birth, and language usage could not be used because of erroneous data or error messages.

The variables in Table 2 demonstrated statistically signifi cant differences between mean health literacy scores of categories within the variable. For income, as anticipated, the mean health literacy score increased as the income increased; however, the mean score re- mained relatively constant for income greater than $60,000. The signifi cant differences in health literacy occurred most frequently between the lowest income level and all others, with medium to large effect sizes for almost all. Health literacy increased with educational attainment, demonstrating statistical differences be- tween most categories with medium to large effect sizes.

BACKGROUND CHARACTERISTICS

Knowledge/Skills Table 2 shows the results for help with forms, reading a newspaper, and writing notes. The health literacy level declined as the need for help increased. The most differ- ences between mean health literacy scores were statisti- cally different with medium to large effect sizes. The older adult population that needed the most help had an average score in the below basic range (162–181). Even those who did not need help had basic health literacy (221–225). Erroneous data or error messages prevented the use of variables related to understanding medica- tion dosing and obtaining help with mathematics.

Health Variables related to health such as vision, hearing, and overall health are given in Table 2 . The mean health lit- eracy scores for those who answered “yes” to vision (190) and hearing (203) diffi culties were at the basic level and statistically different from the scores of those who did not ( p < .05). The effect size was medium and small, respectively. For self-reported overall health, the mean health literacy score increased with improving health. Signifi cance differences were found for fair/poor health compared with good to excellent health, with ef- fect sizes ranging from small to large.

Health Education (Sources of Health Information) The results of the association between health literacy and health education (sources of health information) are shown in Tables 3 and 4 . Table 3 presents mean health literacy scores/standard errors associated with frequency (a lot, some, a little, and none) for each source of health information. This includes results of bivariate comparisons with indication of statistical signifi cance and effect size. Table 4 shows the percentage of older adults associated with frequency, source of health infor- mation, and health literacy level.

Tables 3 and 4 show that overall health literacy de- creases as the use of each sources of health information decreases. In Table 3 , approximately half of the scores were basic (192–226) and the lowest mean health liter- acy scores were associated with utilizing each source of health literacy “none.” Use of the Internet at all

TABLE 1. SOCIODEMOGRAPHIC CHARACTERISTICS OF THE PARTICIPANTS

Variable Categories %

Gender Female 55

Race/ethnicity

White 85

Black 7

Hispanic 5

Other 3

Region

South 37

Midwest 24

Northeast 20

West 19

Marital status

Married/living as married 59

Separated, divorced, or widowed 37

Never married 4

Education

Some high school/high school graduate/GED

Vocational school or some college

College graduate/graduate school

Household income

Above $40,000 69

Below $40,000 31

Diffi culty hearing 25

Learning disability 1

Health literacy level

Below basic 29

Basic health 30

Intermediate 38

Profi cient 3

Andrea Mustafa

TABLE 2. COMPARISON OF HEALTH LITERACY MEAN SCORES BY EDUCATION, INCOME, HELP WITH TASKS, AND HEALTH (VISION, HEARING, AND OVERALL) EFFECT SIZES REPORTED ONLY FOR SIGNIFICANT DIFFERENCES

Variable Health

Literacy Mean SE Effect Sizes for Pairwise Comparisons a

Household income 2 3 4 5 6 7

1. $0–$14,999 183.0 4.1 0.4 0.5 0.7 0.8 1.3 1.1

2. $15,000–$19,999 201.6 4.1 NS NS 0.6 1.2 1.0

3. $20,000–$29,999 212.3 4.7 NS NS 0.7 NS

4. $30,000–$39,999 219.2 5.2 NS 0.8 NS

5. $40,000–$59,999 227.3 5.0 NS NS

6. $60,000–$99,999 252.6 6.7 NS

7. $100,000 + 241.3 11.9

Bonferroni adjusted α = .00244

Education 2 3 4 5

1. Still in high school/less than/ some high school

167.2 4.8 0.5 1.0 1.1 1.5

2. GED/high school equivalency 194.5 7.2 NS 0.7 1.2

3. High school graduate 216.1 2.8 NS 0.8

4. Vocational/some college/ associate degree

224.5 3.7 0.6

5. College + 250.7 4.8

Bonferroni adjusted α = .005

Overall health 2 3 4 5

1. Excellent 231.2 6.4 NS NS 0.8 0.9

2. Very good 231.7 4.0 0.4 0.8 1.0

3. Good 212.9 3.2 0.4 0.7

4. Fair 191.6 3.8 NS

5. Poor 179.0 8.7

Bonferroni adjusted α = .005

Get help with forms 2 3 4

1. A lot 162.6 5.9 0.7 1.0 1.3

2. Some 203.4 6.2 NS 0.4

3. A little 216.5 4.1 NS

4. None 225.3 2.4

Bonferroni adjusted α = .008

Help with writing 2 3

1. A lot/some 175.2 6.6 0.4 0.9

2. A little 201.0 6.2 0.4

3. None 220.8 2.2

Bonferroni adjusted α = .017

Help with newspaper

1. A lot, some, a little 181.3 4.8 0.8

2. None 223.1 2.0

Diffi cultly seeing

1. Yes 189.8 5.2 0.5

2. No 218.6 2.1

Diffi culty hearing

1. Yes 203.3 4.3 0.3

2. No 217.1 2.2

Note . NS = not signifi cant. a Effect sizes reported only for comparisons signifi cant at Bonferroni adjusted α or p < .05 for t -test results.

TABLE 3. SOURCES OF HEALTH INFORMATION VARIABLES WITH MEAN HEALTH LITERACY SCORES, COMPARISONS, AND EFFECT SIZES

Variable Mean SE Effect Sizes for Pairwise Comparisons a

Receive health issue information from newspapers

2 3 4

1. A lot 227.2 4.9 NS 0.4 0.8

2. Some 221.8 2.7 0.3 0.7

3. A little 207.8 3.8 0.4

4. None 182.4 5.7

Bonferroni adjusted α = .008

Receive health issue information from magazines 2 3 4

1. A lot 223.0 3.6 NS NS 1.1

2. Some 226.6 2.6 NS 1.1

3. A little 212.6 4.6 1.1

4. None 171.5 4.5

Bonferroni adjusted α = .008

Receive health issue information from the Internet 2 3 4

1. A lot 235.4 6.1 NS NS 0.6

2. Some 249.9 5.1 NS 0.9

3. A little 249.1 7.5 0.9

4. None 203.1 2.2

Bonferroni adjusted α = .008

Receive health issue information from radio/TV 2 3 4

1. A lot 204.3 3.5 0.3 0.3 NS

2. Some 219.1 2.7 NS 0.6

3. A little 222.9 4.7 0.5

4. None 191.5 7.0

Bonferroni adjusted α = .008

Receive health issue information from books 2 3 4

1. A lot 228.3 4.3 NS NS 1.1

2. Some 222.9 2.3 NS 1.1

3. A little 216.6 4.0 0.8

4. None 171.5 5.1

Bonferroni adjusted α = .008

Receive health issue information from family members or friends or coworkers

2 3 4

1. A lot 209.2 5.1 NS NS NS

2. Some 218.8 3.0 NS 0.4

3. A little 221.5 4.5 0.5

4. None 197.0 3.4

Bonferroni adjusted α = .008

Receive health information from doctors/ healthcare providers

2 3 4

1. A lot 216.1 2.3 NS NS 0.7

2. Some 218.0 3.0 NS 0.8

3. A little 212.8 5.8 0.6

4. None 177.6 6.6

Bonferroni adjusted α = .008

Note . NS = not signifi cant. a Effect sizes reported only for comparisons signifi cant at the Bonferroni adjusted α .

frequency levels was associated with intermediate health literacy. Also, the use of print materials was associated with higher mean health literacy levels. Most signifi cant differences exist between print and nonprint sources, as well as those who use sources “none” compared with other levels of frequency. The effect sizes ranged from small to large (0.3 to 1.1), with some of the greatest ef- fect sizes noted for print sources.

Table 4 shows that a larger percentage of older adults with lower health literacy use each nonprint source less than those with higher health literacy. A very large per- centage (41%–93%) of older adults do not use the Internet as a source of health information. The health- care provider is used “a lot” by 41%–47% of older adults regardless of health literacy level. Besides the health- care provider, the TV/radio is the next most frequent

source of health information used by most older adults (85%–93%).

REGRESSION From the original 32 variables, 15 were placed into the regression analysis to determine which are statistically signifi cant ( p < .05) and have the greatest impact on the mean health literacy scores (see Table 5 ). The results re- veal that the overall regression model was a statistically signifi cant estimator of health literacy and impact of a specifi c variable on health literacy through the unstand- ardized beta coeffi cient ( β ). The coeffi cient can be posi- tive or negative depending on the assigned value of cat- egories within the variables. The following variables were signifi cant (estimate in the parentheses): house- hold income (4.284); educational attainment (9.249);

TABLE 4. PARTICIPANTS’ (PERCENTAGE) USE OF SOURCES OF HEALTH INFORMATION BY HEALTH LITERACY LEVEL

Sources of Health Information A Lot (%) Some (%) A Little (%) None (%)

Below b asic (0–184)

Newspaper 12 30 26 32

Magazine 9 27 21 42

Books 10 28 21 41

Internet 2 2 3 93

Radio/TV 30 35 20 15

Family/friends/coworkers 17 34 22 27

Doctor/healthcare Providers 41 29 18 12

Basic (185–225)

Newspaper 22 39 23 16

Magazine 9 30 21 42

Books 17 45 21 17

Internet 5 8 6 81

Radio/TV 27 46 20 7

Family/friends/coworkers 14 39 26 21

Doctor/healthcare providers 43 38 14 5

Intermediate (226–309)

Newspaper 26 43 20 11

Magazine 21 50 20 9

Books 20 48 23 9

Internet 8 17 10 65

Radio/TV 21 47 25 7

Family/friends/coworkers 15 41 30 14

Doctor/healthcare providers 43 38 15 4

Profi cient ( ≥ 310)

Newspaper 23 43 23 11

Magazine 7 59 29 5

Books 27 34 34 5

Internet 4 26 29 41

Radio/TV 14 33 46 7

Family/friends/coworkers 18 42 35 5

Doctor/healthcare providers 47 22 29 2

get help fi lling out forms (6.213) and reading newspaper articles (13.623); overall health ( − 5.698); and receive health information from doctors/healthcare providers ( − 5.228), books ( − 5.982), Internet ( − 5.231), and maga- zines ( − 5.552).

Discussion This secondary analysis of the NAAL was a unique op- portunity to examine relationships between health lit- eracy and demographic/background characteristics in the U.S. older adult population. This study has three key fi ndings: (1) Results support most relationships de- scribed in the conceptual model and literature associ- ated with health literacy and help identify potential ways to impact health disparities through health liter- acy interventions. (2) There are similarities and differ- ence in the utilization of health information sources based on health literacy level. (3) The results support established health literacy predictive relationships and identify variables (knowledge/skills, health education) that make the model more robust.

For the fi rst research question, the results demon- strate that most relationships in the conceptual model/ literature between background variables and health lit- eracy exist in the older adult population. Furthermore, the results show that several, although not all, of the variables related to health literacy and health disparities may have potential to impact health outcomes in older adults. In the health disparities literature, gender is identifi ed as a variable that impacts health conditions/ outcomes. However, the results did not demonstrate a

difference between mean health literacy scores and gen- der in this study. Thus, health disparities related to gen- der will most likely not be resolved through health lit- eracy interventions.

In contrast, the literature on health literacy and health disparities repeatedly demonstrates that those from lower income brackets and lower educational at- tainment have lower health literacy and experience health disparities. This study showed that there were signifi cant differences in the mean health literacy score between the lowest income level/lowest educational level and most other income/educational levels. Interventions such as additional support for health management in communities with lower income and education may help older adults take care of their health and potentially prevent health disparities.

A strong relationship between health and health lit- eracy was demonstrated through self-reported overall health. This is consistent with the literature noting that older adults with lower health literacy had worse health status ( Baker et al., 2002 ; Baker, Gazmararian, Sudano, & Patterson, 2000 ; Mottus et al., 2014 ; Sudore et al., 2006 ; Wolf et al., 2005 ). Although vision and hearing defi cits were not signifi cant in predicting health literacy when compared with the other variables in the regres- sion model, the signifi cant differences noted between the mean health literacy score of those with these disa- bilities are important to consider when developing in- terventions to decrease disparities in this population.

The fi rst research question is also answered by exam- ining sources of health information and health literacy. The sources used by older adults vary on the basis of

TABLE 5. REGRESSION ANALYSIS SHOWING CONTRIBUTION OF INDIVIDUAL PREDICTORS TO THE OVERALL PREDICTION MODEL

Predictors β SE t Statistic p

Constant 219.522 17.479 12.559 .001 *

Race/ethnicity − 8.247 4.137 − 1.994 .051

Approximate household income (eight categories) 4.284 1.004 4.266 .001 *

Educational attainment (six categories) 9.249 1.531 6.039 .001 *

Diffi cultly seeing words and letters in newspapers even with glass/lenses

2.709 4.038 0.671 .505

Diffi culty hearing in normal conversation even with hearing aid − 2.496 4.316 − 0.578 .565

Get help from family/friends fi lling out forms 6.213 2.098 2.962 .004 *

Get help from family/friends to read newspaper articles 13.623 4.561 2.987 .004 *

Get help from family/friends to write notes 5.387 2.855 1.887 .064

Overall health − 5.698 1.694 − 3.364 .001 *

Receive health information from doctors/healthcare providers − 5.228 1.942 − 2.691 .009 *

Receive health issue information from books − 5.982 2.32 − 2.578 .012 *

Receive health issue information from the Internet − 5.231 2.272 − 2.303 .025 *

Receive health issue information from magazines − 5.552 2.177 − 2.55 .013 *

Receive health issue information from newspapers − 3.663 2.093 − 1.751 .085

Receive health issue information from radio/TV 2.21 1.924 1.149 .255

Root MSE 41.922 1.731 – –

* p < .05.

health literacy level, although there are some similari- ties. Most recent studies on sources of health informa- tion or health information-seeking behaviors have fo- cused on using the Internet. Older adults using the Internet have higher education, incomes, and health literacy and make better healthcare decisions ( James, Boyle, Yu, & Bennett, 2013 ; Kobayashi, Wardle, & von Wagner, 2015 ; Pew Research Center, n.d. ). Studies show that older adults did not rely on the Internet for infor- mation and have healthcare providers as the main source of information ( Gollop, 1997 ; Kutner et al., 2006 ; Morey, 2007 ; Tian & Robinson, 2008 ). The results of this study support fi ndings in the literature. The implication for healthcare providers is to understand that older adults do not use the Internet to the same extent as other segments of the population and it may not be the preferred source of health information.

For the remaining sources of health information, the results of this study support the literature stating that the percentage of adults using each source increases with increasing health literacy ( Kutner et al., 2006 ). However, the extent to which the source is used varies when comparing the general adult population and older adults. For example, the percentage of older adults who use doctor/healthcare provider “a lot” increases with in- creasing health literacy whereas for the general adult population, the percentage decreases with increasing health literacy. Thus, older adults use healthcare provid- ers differently than do other adults and interventions should be tailored to address this difference. It is im- perative that providers make health education a priority and be prepared to be the main source of health infor- mation for older adults. Providers need to adjust their strategy to educate older adults on the basis of health literacy level, reaching out to those with the lowest health literacy (because they seek health information the least) and being prepared to direct those with higher health literacy to various reputable sources.

As the fee-for-service model of healthcare fi nance is replaced by fee based on quality, providers have an op- portunity to change their approach to patient and fam- ily education. Providers can try methods that engage patients and provide education in a way that meets the patients’ health literacy needs. This process is actually less diffi cult if providers follow the principles of Universal Health Literacy Precautions ( Agency for Healthcare Research and Quality, 2010 ), educating all older adults with using simple, everyday language. Because lower health literacy has been associated with decreasing cognitive ability, providing education that decreases the load on cognition is essential ( Gakumo, Enah, Vance, Sahinoglu, & Raper, 2015 ; Kobayashi, Wardle, & von Wagner, 2015 ; Kobayashi, Wardle, Wolf et al., 2015 ; Mottus et al., 2014 ; O’Conor et al., 2015 ; Tennant et al., 2015 ).

TV/radio is the next most used source of health infor- mation for older adults of all health literacy levels. The popularity of shows such as Dr. Oz , the inclusion of the health reports on local and national news broadcasts, and the popularity of health stations on satellite radio demonstrate the desire to receive health information via this medium. Healthcare providers must push for more

health information to be made available through TV and radio and be available to provide information when needed for these sources. New electronic sources (Internet, cable), which provide on-demand content to various devices such as laptops and tablets, are another method to present information in a format similar to TV and radio. New technology will require healthcare pro- viders to partner with older adults and technology spe- cialists to develop access to these devices while decreas- ing the load on cognition.

On the whole, print health sources were used by a lower percentage of older adults when compared with nonprint sources with one notable exception. Families/ friends/coworkers were used less often than some forms of print resources for older adults with basic and inter- mediate health literacy. More research to understand the context in which family/friends/coworkers are used by older adults would help identify potential strategies. Also to meet the educational needs of older adults with lower health literacy, the following interventions should be examined: Increasing provider time for more exten- sive education and/or having health educators working with the provider answer questions and supply addi- tional education during the offi ce visit. Because re- sources are fi nite, prioritizing interventions using the most common nonprint sources is very important.

The second research question is addressed by the ex- amination of variables in the regression analysis. On the basis of the conceptual model, two (income and educa- tion) of the nine signifi cant predictors of health literacy have been identifi ed as impacting health literacy. The other signifi cant predictors have not been discussed in the literature. This study identifi ed self-care variables related to patients’ knowledge/skills (help with forms and reading newspaper) and health education (sources of health information) as impacting health literacy. The literature on predicting health literacy using patients’ skills has focused on using single questions related to ability or confi dence to complete tasks. These studies demonstrated that the use of screening questions is as effective as other more lengthy health literacy assess- ments such as the Rapid Estimate of Adult Literacy in Medicine (REALM) or the Test of Functional Health Literacy in Adults (TOFHLA) ( Wallston et al., 2014 ). For example, Chew, Bradley, and Boyko (2004 , p. 588) used the following questions to determine adequacy of health literacy: “How often do you have someone help you read hospital materials?” Like this question, the NAAL ques- tions about knowledge/skills could provide the founda- tional work to pursue additional single-question screen- ing tools used to identify those with low health literacy.

Where patients seek health information has not been cited as a variable that predicts health literacy. In this study, four sources of information (doctors/healthcare providers, books, Internet, and magazines) were found to be statistically signifi cant in the regression analysis. Thus, use (or lack of use) of these sources may have po- tential to predict health literacy. When educating older adults, providers should ask about the sources of health information used to help the provider understand pa- tients’ health literacy levels and direct patients to relia- ble information they are more likely to use.

Limitations This study is limited by being a secondary analysis. The principal investigator did not have control over the orig- inal research questions and data collection. As a result, the analysis of data was limited by an insuffi cient num- ber of observations and thus several variables (i.e., lan- guage, citizenship) could not be examined. To increase observations, categories of variables were collapsed (i.e., race, help with reading newspaper). By collapsing White and Hispanic into one category, any discussion about race in the U.S. older adult population is limited by the lack of meaningful analysis. This may have also contributed to race not being signifi cant in the regres- sion analysis. The variable “help with reading” was also collapsed into “help” versus “no help.” This may have impacted the estimate in the regression analysis be- cause it was substantially higher than other variables’ estimates. Although the NAAL is the only large-scale na- tional study of health literacy in the United States, the data from the study are more than 10 years old. Thus, the data do not refl ect changes in the population such as technology skills and the increase or decline in the use of certain sources of health information such as the Internet and newspapers.

Conclusion This secondary analysis examined health literacy and multiple background questions in the U.S. older adult population using the NAAL data. The results support most relationships described in the conceptual model and literature associated with health literacy and as a result have potential to impact health disparities through health literacy interventions. These results also demonstrate the similarities and difference in the utili- zation of health information sources based on health literacy level, support established health literacy predic- tive relationships, and identify variables (skills, knowl- edge) that make the model more robust.

The results confi rm relationships already established in the literature between sociodemographic variables and health literacy. Lower health literacy in older adults is associated with income less than $15,000 (in 2003), high school graduate or less, vision and hearing defi cits, and fair/poor overall health. Income and education were the strongest predictors of health literacy when compared with other variables. Because these charac- teristics are also associated with health disparities, the potential role of health literacy in the decreasing dis- parities needs to be examined. Interventions (e.g., com- munity health workers, postdischarge phone calls) fo- cused on older adults with these characteristics may impact their ability to manage their health and could potentially lessen disparities. Rubin et al. (2014) devel- oped a program to train volunteers for Meals on Wheels as health literacy coaches for older adults.

Knowledge/skills and health education variables under self-care (needing help with skills such as fi lling out forms, reading newspaper, and writing notes; and sources of health information) have an interactive rela- tionship with health literacy. Knowledge/skills/health education variables not only were utilized by older

adults but can also be indicators of health literacy. In this study, the fi rst two knowledge/skills listed earlier were identifi ed as predictive and thus have the founda- tional potential to be used as single-item screener ques- tions and make the conceptual model more robust.

Sources of health information results demonstrated some similarities and differences by health literacy level. They showed that the opportunity to educate older adult patients, regardless of health literacy level, is dur- ing the interaction with the doctor/healthcare provider and the Internet is not the preferred source. Results also show that as health literacy decreases, the percentage of older adults using each source of health information de- creases. Thus, those with lower health literacy may need the healthcare system to reach out and engage them in learning about health concerns rather than assuming they will use nonprint sources. Future research should be focused on interventions that engage older adults while supplying education in formats most commonly used such as the healthcare provider and TV/radio. Some sources of health information (print and nonprint materials) used by older adults are predictive of health literacy. They may provide the foundation for determin- ing additional single-item screeners for identifying those with low health literacy and make the conceptual model more robust.

REFERENCES Agency for Healthcare Research and Quality . ( 2010 ). Health

literacy universal precautions toolkit . Retrieved from http://www.ahrq.gov/qual/literacy/

American Medical Association . ( n.d. ). Health literacy kit . Retrieved from http://www.ama-assn.org/ama/pub/ about-ama/ama-foundation/our-programs/public- health/health-literacy-program/health-literacy-kit. page?

Baker , D. , Gazmararian , J. A. , Sudano , J. , & Patterson , M. ( 2000 ). The association between age and health liter- acy among elderly persons . Journals of Gerontology Series B-Psychological Sciences & Social Sciences , 55 ( 6 ), S368 – S374 .

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

Baker , D. , Wolf , M. S. , Feinglass , J. , Thompson , J. A. , Gazmararian , J. A. , & Huang , J. ( 2007 ). Health literacy and mortality among elderly persons . Archives of Internal Medicine , 167 ( 14 ), 1503 – 1509 .

Baldi , S. ( 2009 ). Technical report and data fi le user’s manual for the 2003 National Assessment of Adult Literacy (No. NCES 2009-476). Washington, DC : U.S. Government Printing Offi ce .

Bostock , S. , & Steptoe , A. ( 2012 ). Association between low functional health literacy and mortality in older adults . BMJ , 344 , e1602 . doi:10.1136/bmj.e1602

Campbell , R. J. , & Nolfi , D. A. ( 2005 ). Teaching elderly adults to use the Internet to access health care infor- mation: Before–after study . Journal of Medical Internet Research , 7 ( 2 ), e19 . doi:102196/jmir.7.2.e19

Centers for Disease Control and Prevention . ( 2014 ). Health literacy . Retrieved from http://www.cdc.gov/health literacy/

Chen , J. Z. , Hsu , H. C. , Tung , H. J. , & Pan , L. Y. ( 2013 ). Effects of health literacy to self-effi cacy and preventing

http://www.ama-assn.org/ama/pub/about-ama/ama-foundation/our-programs/publichealth/health-literacy-program/health-literacy-kit.page

http://www.cdc.gov/healthliteracy/

Andrea Mustafa

care utilization among older adults . Geriatrics and Gerontology International , 13 ( 1 ), 70 – 76 . Advance on- line publication. doi:1111/j.1447-0594.2012.00862.x

Chew , L. , Bradley , K. , & Boyko , E. J. ( 2004 ). Brief questions to identify patients with inadequate health literacy . Family Medicine , 36 ( 8 ), 588 – 594 .

Cohen , J. ( 1992 ). A power primer . Psychological Bulletin , 112 ( 1 ), 155 – 159 .

Cutilli , C. C. ( 2010 ). Patient education corner. Seeking health information: What sources do your patients use? Orthopaedic Nursing , 29 ( 3 ), 214 – 219 .

Feinberg , I. , Frijters , J. , Johnson-Lawrence , V. , Greenberg , D. , Nightingale , E. , & Moodie , C. ( 2016 ). Examining as- sociations between health information seeking behavior and adult education status in the U.S.: An analysis of the 2012 PIAAC data . PLoS One . Retrieved September 14, 2106, from http://journals.plos.org/plosone/ article?id = 10.1371/journal.pone.0148751#authcontrib

Gakumo , C. A. , Enah , C. C. , Vance , D. E. , Sahinoglu , E. , & Raper , J. L. ( 2015 ). “Keep it simple”: Older African Americans’ preferences for a health literacy interven- tion in HIV management . Patient Preference and Adherence , 9 , 217 – 223 .

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

Gollop , C. J. ( 1997 ). Health information-seeking behavior and older African American women . Bulletin of the Medical Library Association , 85 ( 2 ), 141 – 146 .

Greenberg , E. , & Jin , Y. ( 2007 ). 2003 National Assessment of Adult Literacy: Public-use data fi le user’s guide (No. NCES 2007-464). Washington, DC : U.S. Government Printing Offi ce .

Hall , A. K. , Bernhardt , J. M. , & Dodd , V. ( 2015 ). Older adults’ use of online and offl ine sources of health in- formation and constructs of reliance and self-effi cacy for medical decision making . Journal of Health Communication , 20 ( 7 ), 751 – 758 .

Healthy People 2020 . ( n.d. ). Disparities . Retrieved from https://www.healthypeople.gov/2020/about/foundation- health-measures/Disparities

James , B. D. , Boyle , P. A. , Yu , L. , & Bennett , D. A. ( 2013 ). Internet use and decision making in community-based older adults . Frontiers in Psychology , 4 , 1 – 10 . doi:10.3389

Kobayashi , L. C. , Wardle , J. , & von Wagner , C. ( 2015 ). Internet use, social engagement and health literacy de- cline during ageing in a longitudinal cohort of older English adults . Journal of Epidemiology & Community Health , 69 ( 3 ), 278 – 283 . doi:10.1136/iech-2014-204733

Kobayashi , L. C. , Wardle , J. , Wolf , M. S. , & von Wagner , C. ( 2015 ). Cognitive function and health literacy decline in a cohort of aging English adults . Journal of Internal Medicine , 30 ( 7 ), 958 – 964 . doi:10.1007/s11606-015- 3206-9

Kontos , E. , Blake , K. D. , Chou , W. Y. S. , & Prestin , A. ( 2014 ). Predictors of eHealth usage: Insights on the digital di- vide from the Health Information National Trends Survey 2012 . Journal of Medical Internet Research , 16 ( 7 ), e172 .

Kutner , M. , Greenberg , E. , Jin , Y. , & Paulsen , C. ( 2006 ). The health literacy of America’s Adults: Results from the 2003 National Assessment of Adult Literacy (No. NCES 2006-483). Washington, DC : U.S. Government Printing Offi ce .

Levy , H. , Janke , A. T. , & Langa , K. M. ( 2014 ). Health liter- acy and digital divide among older Americans . Journal of General Internal Medicine , 30 ( 3 ), 284 – 289 . doi:10.1007/s11606-014-3069-5

Mahnoush , R. , Javadzade , S. H. , Heydarabadi , A. B. , Mostafavi , F. , Tavassoli , E. , & Sharifi rad , G. ( 2015 ). The relationship between functional health literacy and health promoting behaviors among older adults . Journal of Education and Health Promotion , 3 , 119 . doi:10.4103/2277-9531.145925

Morey , O. ( 2007 ). Health information ties: Preliminary fi nd- ings on the health information seeking behaviour of an African-American community . Information Research , 12 ( 2 ), paper 297. Available at http://InformationR.net/ ir/12-2/paper297.html.

Mottus , R. , Johnson , W. , Murray , C. , Wolf , M. S. , Starr , J. M. , & Deary , I. J. ( 2014 ). Towards understanding the links between health literacy and physical health . Health Psychology , 33 ( 2 ), 164 - 173 . doi:10.1037/a0031439

National Center for Education Statistics . ( n.d.-a ). National Assessment of Adult Literacy . Retrieved from http:// nces.ed.gov/naal/

National Center for Education Statistics . ( n.d.-b ). Data fi les from the 2003 National Assessment of Adult Literacy . Retrieved from http://nces.ed.gov/naal/datafi les.asp

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

O’Conor , R. , Wolf , M. S. , Smith , S. G. , Martynenko , M. , Vicencio , D. P. , Sano , M. , … Federman , A. D. ( 2015 ). Health literacy, cognitive function, proper use and ad- herence to inhaled asthma controller medications among older Adults with asthma . Chest , 147 (5), 1307–1315. doi:10.1378/chest.14-0914

Paasche-Orlow , M. K. , & Wolf , M. S. ( 2007 ). The causal path- ways linking health literacy to health outcomes . American Journal of Health Behavior , 31 ( Suppl. 1 ), 819 – 826 .

Pew Research Center . ( n.d. ). Older adults and technology use . Retrieved from http://www.pewinternet. org/2014/04/03/older-adults-and-technology-use/

Pfi zer . ( 2015 ). Health literacy . Retrieved from http://www. pfi zer.com/health/literacy

Rubin , D. L. , Freimuth , V. S. , Johnson , S. D. , Kaley , T. , & Parmer , J. ( 2014 ). Training meals on wheels volunteers as health literacy coaches for older adults . Health Promotion Practice , 15 ( 3 ), 448 – 454 . doi:10.1177/ 1524839913494786

Sequeira , S. S. , Eggermont , L. H. , Silliman , R. A. , Bickmore , T. W. , Henault , L. E. , Winter , M. R. , … Orlow , M. K. ( 2013 ). Limited health literacy and decline in execu- tive functioning in older adults . Journal of Health Communication , 18 ( Suppl. 1 ), 143 – 157 . doi:10.1080/ 10810730.2013.825673

Smith , S. G. , O’Conor , R. , Curtis , L. M. , Waite , K. , Deary , I. J. , Paasche-Orlow , M. , & Wolf , M. S. ( 2015 ). Low health literacy predicts decline in physical function among older adults: Findings from the LitCog cohort study . Journal of Epidemiology & Community Health , 69 ( 5 ): 474 – 480 . doi:10.1136/jech-2014-204915

Sudore , R. , Mehta , K. M. , Simonsick , E. M. , Harris , T. B. , Newman , A. B. , Satterfi eld , S. , … Yaffe , K. ( 2006 ). Limited literacy in older people and disparities in health and healthcare access . Journal of the American Geriatrics Society , 54 ( 5 ), 770 – 776 .

Tennant , B. , Stellefson , M. , Dodd , V. , Chaney , D. , Paige , S. , & Alber , J. ( 2015 ). eHealth literacy and web 2.0 health information seeking behaviors among baby boomers and older adults . Journal of Medical Internet Research , 17 ( 3 ), e70 . doi:10.2196/jmir.3992

The Joint Commission . ( n.d. ). “What did the doctor say?” Improving health literacy to protect patient safety . Retrieved from http://www.jointcommission.org/ What_Did_the_Doctor_Say/

http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0148751#authcontrib

http://InformationR.net/ir/12-2/paper297.html

http://www.pewinternet.org/2014/04/03/older-adults-and-technology-use/

http://www.pfizer.com/health/literacy

http://www.jointcommission.org/What_Did_the_Doctor_Say/

Tian , Y. , & Robinson , J. D. ( 2008 ). Media use and health information seeking: An empirical test of complemen- tarity theory . Health Communication , 23 ( 2 ), 184 – 190 .

Wallston , K. A. , Cawthorn , C. , McNaughton , C. D. , Rothman , R. L. , Osborn , C. Y. , & Kripalani , S. ( 2014 ). Psychometric properties of the brief health literacy screen in clinical practice . Journal of General Internal Medicine , 29 ( 1 ), 119 – 126 . doi:10.1007/s11606-013- 2568-0

White , S. ( 2008 ). Assessing the nation’s health literacy: Key concepts and fi ndings of the National Assessment of Adult Literacy (NAAL) (No. OP423908). Chicago, IL : American Medical Association Foundation .

White , S. , & Dillow , S. ( 2005 ). Key concepts and features of the 2003 National Assessment of Adult Literacy (No. NCES 2006-471). Washington, DC : U.S. Department of Education, National Center for Education Statistics .

Wolf , M. S. , Gazmararian , J. A. , & Baker , D. ( 2005 ). Health literacy and functional health status among older adults . Archives of Internal Medicine , 165 ( 17 ), 1946 – 1952 . doi:10.1001/archinte.165.17.1946

Wu , J. R. , Moser , D. K. , DeWalt , D. A. , Rayens , M. K. , & Dracup , K. ( 2016 ). Health literacy mediates the relation- ship between age and health outcomes in patients with heart failure . Circulation: Heart Failure , 9 ( 1 ), e002250 . doi:10.1161/CIRCHEARTFAILURE.115.002250

For additional continuing nursing education activities on orthopaedic nursing topics, go to nursingcenter.com/ce.