Health Literacy

Burden of Low Literacy and Low Health Literacy

In 2003, the US Department of Education conducted a survey entitled “National Assessment of Adult Literacy” (NAAL). The most comprehensive examination of adult literacy to date, the NAAL surveyed more than 19,000 adults age 16 and older and included items intended to measure health literacy directly. More than one-third of respondents (36 percent) taking the NAAL scored in the lowest two (“basic” and “below basic”) out of four categories on health literacy items, suggesting that approximately 80 million adults in the United States have limited health literacy, including related prose, document, and quantitative skills.¹² These adults may have difficulty with even simple tasks such as reading and understanding the instructions on a prescription bottle or filling out an insurance form. Although the NAAL did not independently report on prose, document, or quantitative health literacy, its predecessor, the National Adult Literacy Survey (NALS), reported similar proportions of individuals scoring in the lowest proficiency levels across these domains.^11,13 More recent (although not nationally representative) data suggest that many adults may have higher print literacy than quantitative literacy.¹⁴

Although a significant proportion of the general population has low health literacy, certain groups have an even higher prevalence of the problem. Such groups include the elderly, minorities, individuals who have not completed high school, adults who spoke a language other than English before starting school, and people living in poverty.¹² For instance, the NAAL demonstrated a higher prevalence of poor health literacy among the elderly. Compared with the 36 percent of all adults who scored in the bottom two categories on the NAAL survey, 59 percent of adults age 65 and older scored in the “below basic” and “basic” range.¹² This association between age and health literacy has proven consistent in other studies of literacy in health care settings. However, the majority of these studies are cross-sectional, making it difficult to determine whether the higher prevalence of poor health literacy in the elderly population results from a cohort effect (e.g., fewer educational opportunities; higher prevalence of a native language other than English) or whether literacy declines with age or cognitive function.¹⁵ Both factors likely play a contributing role.

The NAAL also reported a strong relationship between health literacy and race or ethnicity. White respondents scored better on the survey than any of the other racial or ethnic groups evaluated. Only 9 percent of white respondents scored in the lowest (“below basic”) category on the NAAL survey, but 24 percent of black, 41 percent of Hispanic, 13 percent of Asian, and 25 percent of American Indian and Native Alaskan respondents scored in the “below basic” range.¹² Differences in the quality of education received by disadvantaged members of nonwhite populations may, at least partially, explain this finding. Further, issues of language and acculturation likely play a significant role. The association between health literacy and race and ethnicity raises the question of whether health literacy serves as a mediator of racial and ethnic disparities in health. If literacy is related to health outcomes, disparate health literacy levels among different groups could contribute to differential health outcomes.

In addition to age, race, and ethnicity, educational attainment plays a predictably strong role in health literacy. In the NAAL study, more than three-quarters (76 percent) of respondents who had not completed high school scored in the “below basic” or “basic” range of health literacy, compared with only 13 percent of individuals with 4-year college degrees.¹² Although one's literacy level is related to one's educational status, the correlation between years of education and literacy is imperfect. People often score reading grade levels that are several grades lower than the last year of school they completed.¹⁶ In addition to the ability to read, the ability to complete 12 years of education may draw on several factors, including social support, community resources, motivation, and family expectations.

Using statistical modeling and demographics, such as those above, the National Center for Education Statistics and others^17-20 have provided estimates of local and regional literacy and health literacy prevalence. As might be expected, these estimates suggest variation across states and counties,^18,20 which might affect health outcomes in important ways. To assist clinicians and policymakers in estimating the health literacy prevalence in their own environments, calculators based on such work are now available online.¹⁹

Measuring Health Literacy

To date, instruments for measuring health literacy skill levels have focused primarily on the ability to read and, in some cases, to use numbers. A variety of measures focusing on these skills are available and have been applied in the health setting (see Tables 1 and 2). Currently, no instruments are widely available to measure oral health literacy or a comprehensive set of skills that have been conceptualized as the components of health literacy.

Table 1

Measures of health literacy.

Table 2

Measures of numeracy.

Commonly used measures of health literacy. The instruments most commonly used in the health literature to measure health literacy are the Rapid Estimate of Adult Literacy in Medicine (REALM)²¹ and the Test of Functional Health Literacy in Adults (TOFHLA).²² The REALM is a word recognition test that assesses whether a person can correctly pronounce a series of health-related words listed in order of increasing difficulty. The REALM has been validated as an instrument of reading ability and is highly correlated with traditional reading assessments in the educational literature (correlation with the Wide Range Achievement Test [WRAT]: r = 0.88).²¹

The TOFHLA employs a different approach and assesses both reading skills and numeracy. It assesses reading skills using a modified cloze procedure. In this procedure, subjects read health-related passages in which every fifth to seventh word has been deleted; they then fill in the blanks by selecting the correct word from four choices.²² The TOFHLA assesses numeracy by asking a subject to respond to health-related prompts, such as pill bottle instructions and appointment slips. While developing and validating the TOFHLA, the authors found that the reading comprehension subtest and quantitative or “numeracy” subtest were highly correlated (r = 0.79). The TOFHLA has also been noted to be highly correlated with the REALM (r = 0.84) and the WRAT (r = 0.74).²² A short version (S-TOFHLA)²³ is available and has also been widely applied in the literature.

The most common instruments used to measure numeracy in the health literature are the Schwartz and Woloshin Numeracy Test and the WRAT math subtest. Neither of these focuses specifically on the health context. The Schwartz and Woloshin Numeracy Test consists of three items that assess individuals' understanding of probability and their ability to convert between percentages and proportions.²⁴ The WRAT math subtest assesses individuals' ability to count, read numerical symbols, and perform simple arithmetic operations.²⁵ A growing number of newer tools (e.g., Diabetes Numeracy Test) measure numerical skills in the health context, but have not been widely employed to assess the relationship between numeracy and health outcomes.

No gold-standard instrument is currently available to assess adequately the more global concept of health literacy, including the interactions of reading ability, numeracy, and oral literacy. However, as recommended by policymakers, work to define and measure a wider set of skills that might more adequately reflect health literacy has begun.²⁶

Relationship Between Health Literacy and Outcomes

In the past 15 years, researchers have demonstrated that low literacy can have far-reaching consequences for an individual's health. In our 2004 systematic review and related articles,^49,50 we identified 44 articles describing results that addressed the relationship between literacy and use of health care services, health outcomes, costs of health care, and disparities. The report found that low or inadequate literacy (compared to adequate literacy) was strongly associated with poorer knowledge or comprehension of health care services and health outcomes.^49,50 Limited literacy was also associated with higher probability of hospitalization, higher prevalence and severity for some chronic diseases, poorer global measures of health, and lower utilization of screening and preventive services.^49,50 In many cases, however, the evidence was mixed; both outcomes assessed and analytic methods differed across studies.^49,50 Although literacy was often related to health outcomes in bivariate associations, the relationship sometimes weakened and became statistically nonsignificant after the investigators adjusted results for covariates such as age, education, socioeconomic status, health care access, or experience in the health care setting, calling into question whether low literacy was truly an independent problem or merely a marker of other social problems. Outcome differences were rare between a middle literacy group (marginal) and the adequate group. Only one study that was reviewed examined differences in costs and one study examined differences between race or ethnicity groups, resulting in insufficient data to reach conclusions concerning these issues.

Based on these findings, the 2004 review recommended that future research: (1) examine more closely and include in analytic models factors that may be confounding the relationship between literacy and health outcomes (e.g., age, income, or health insurance status); (2) consider other factors, referred to as mediators, that may be in the causal pathway between health literacy and health outcomes (e.g., self-efficacy, self-care, trust, and satisfaction); (3) consider prospective cohort studies to examine the relationship between literacy, age, and changes in health outcomes such as health status; (4) stratify outcomes by numeracy level to gain a greater understanding of how these skills may uniquely affect health outcomes and under what conditions numeracy would be a useful indicator for targeting individuals for interventions; and (5) examine the effect of literacy on costs and on racial, ethnic, and age-related disparities.

Effects of Interventions To Reduce Burden of Low Health Literacy

In our prior review,^49,51 we identified 29 articles describing interventions to mitigate the effects of low literacy on health outcomes. Of the 29 articles, 20 measured literacy in individual participants and were performed in developed countries. These 20 studies tested a wide range of interventions for improving health outcomes in patients with poor literacy. Most of the interventions occurred in a single session and attempted to make health information more readily available to patients with limited literacy. Some studies compared standard handouts with materials that were written in simpler, easier-to-read prose. Others compared standard materials with pictographs, booklets, videotapes, or CD-ROMs specially designed for low-literacy audiences. A few interventions used multiple methods.

In aggregate, these studies suggested that interventions may reduce the adverse health effects associated with low literacy.^49,51 However, few studies examined each type of intervention; few examined the interventions' effects in literacy subgroups; a minority examined outcomes other than knowledge; and many had methodological flaws limiting conclusions.

Based on observations from our 2004 review, we recommended that (1) additional studies of interventions be pursued, (2) any new investigations measure the interventions' effects by literacy subgroup, and (3) investigations examine a broader range of outcomes.

Need for Update of the Earlier Review

Given the ongoing concern about an association between health literacy level and poor health outcomes and the potential to reduce these outcomes with novel interventions, the US Department of Health and Human Services (HHS) has released a National Action Plan to Improve Health Literacy.⁵² Additionally, several national organizations, including the IOM,⁵³ the American Medical Association (AMA),⁵ the National Institutes of Health (NIH), and HHS (Healthy People 2010),³ have promoted health literacy as a research priority. With such attention, the research community in this field has responded with considerable new work since 2004. Additionally, AHRQ has released a Health Literacy Universal Precautions Toolkit based on evidence and best practices.⁵⁴

To synthesize the increasing volume of literature on health literacy and further the larger goal of improvements in health literacy, AHRQ commissioned the RTI–UNC EPC to update its 2004 systematic review to examine the effects of health literacy on health outcomes and interventions to improve those outcomes. In this updated report, we focus on the same key questions as the original report, but we expand our conception of literacy to health literacy and consider—separately and in combination—print literacy, numeracy, and oral health literacy skills. In the results chapters of this report (Chapters 3 and 4), we include only studies that have been published since our last review; we did not systematically reabstract studies from our earlier review or reassess their quality. We did, however, reorganize data about intervention studies from our first review to highlight features of the interventions reviewed earlier and allow interpretation of these features in light of current evidence. Additionally, we compared all findings from the current review to findings from our 2004 review to allow for comprehensive conclusions.

Further, following our review of information available through publications and our review of the quality of the studies based on that information, we queried intervention authors from both the first review and this updated review about key features of the interventions that they had not reported in published articles. This additional information is included in Appendix A.

Production of This Report

Technical Expert Panel

We identified technical experts in the field of health literacy to provide assistance throughout the project. The TEP was expected to contribute to AHRQ's broader goals of (1) creating and maintaining science partnerships as well as public-private partnerships and (2) meeting the needs of an array of potential customers and users of its products. Thus, the TEP was both an additional resource and a sounding board during the project. The TEP included eight members: five technical/clinical experts; one member whose expertise and mission concerns the interests and perspectives of patients and consumers; one potential user of the final evidence report; and an AHRQ health literacy expert (see Acknowledgments, page iv).

To ensure robust, scientifically relevant work, the TEP was called on to provide advice on substantive issues or possibly overlooked areas of research. TEP members participated in conference calls and discussions through e-mail to refine the scope of this update (including inclusion/exclusion criteria) and discuss our preliminary assessment of the literature. Because of their extensive knowledge of the literature on health literacy, including numerous articles authored by TEP members themselves, and their active involvement in professional societies and as practitioners in the field, we also asked some TEP members to participate in the external peer review of the draft report.

Use of This Updated Systematic Review

This updated report addresses the key questions outlined in Chapter 2 through a systematic review of published literature. We anticipate that the report will be of value to the AMA for its various efforts to inform and educate physicians. This report can also inform practitioners about the current state of evidence and provide an assessment of the quality of studies that aim to improve health for people with low health literacy. Researchers can obtain a concise analysis of the current state of knowledge in this field and will be poised to pursue further investigations that are needed to improve health for low-health-literacy populations. Health educators can also use this report to guide future interventions to improve health communication. Finally, policymakers can use this report to inform new strategies and the allocation of resources toward future research and initiatives that are likely to be successful.