Process to Select Papers and Conduct Full-Text Review

All references from the initial search were imported in Zotero, a citation management software, and duplicates were removed. References from Zotero were exported to Rayyan, a screening software for systematic reviews, which was used for screening titles and abstracts (Ouzzani et al. 2016). Rayyan caught duplicates that were not initially flagged by Zotero, and the librarian resolved these. The first two authors independently screened the titles and abstracts in Rayyan. They jointly reviewed and discussed 33 percent of the references identified for inclusion and reached consensus. Next, they independently double-reviewed the remaining references and reached consensus by discussion for inclusion. Entries without abstracts were included in the pool of references if the title seemed to fit the inclusion criteria. During the title and abstract review, the first two authors identified eight abstracts containing only protocols, descriptions of future projects, or projected results. The public health librarian followed up with authors to obtain related publications. Only one author had published results, but the study was not included in the final results. The first two authors also identified twenty review articles; the librarian reviewed the reference lists and added studies that met the inclusion criteria.

After the title and abstract screening, the first two authors trained three coders to use a template to review and record information from the full-text articles (see Tables C-5, C-6, and C-7). The template included criteria from the roundtable's initial call for a review paper, the Centers for Disease Control and Prevention's (CDC's) “Seeking Best Practices: A Conceptual Framework for Planning and Improving Evidence-Based Practices” (Spencer et al. 2013), and the Guide to Community Preventive Services study design categories (HHS 2001). Based on the full-text information, both the coders and the first two authors excluded additional articles because they did not include anything about health literacy, a community, and an intervention in the body of the paper. Once coding was complete, the first two authors reviewed and synthesized information on the remaining studies.

TABLE C-5

Codes and Directions for Community-Based Health Literacy Intervention Review.

TABLE C-6

Centers for Disease Control and Prevention Best Practices Framework.

TABLE C-7

Drop-Down Lists for the Cells.

Magnitude of Intervention Effects

The roundtable asked the review team to report the magnitude of intervention effects when available. To assess magnitude, the team relied on reported effect sizes for statistically significant comparisons. One consequence of this methodological decision is that the results table (Table C-8) does not include information on the amount of impact suggested by qualitative analyses. Notwithstanding that decision, the team acknowledges the value of those qualitative analyses, especially when trying to reflect with fidelity the perspectives and experiences of community members. The results table adopts the convention of characterizing effect sizes as small, medium, large, or very large. These characterizations are based on conventional standards for Cohen's d effect sizes, for standardized regression weights (Cohen 1988; Sawilowsky 2009), and for odds ratios (Chen, Cohen, and Chen 2010). Of studies with quantitative outcomes that might have reported effect sizes, 84 percent (59) did not do so. Transparency about magnitude of effects is necessary for rational allocation of public health resources. The failure to calculate and disclose magnitude of effects is contrary to guidelines for reporting experiments (e.g., APA 2010).

Effect sizes that were reported were mainly small, or small to moderate, varying by outcome measure. Only two studies reported any effects that attained a large magnitude. One used fotonovelas with Latinas at a high risk of depression and found large effects of the intervention on knowledge of depression and efficacy to seek help for that condition (Hernandez and Organista 2013). Another achieved very large magnitude increases in breast cancer and cervical cancer screening among Korean Americans by conducting a single educational session led by community health workers (Han et al. 2017).

Health Literacy Skill and Capacity Building

Seven articles reported interventions to improve or increase a group's health literacy skills or an organization's capacity to serve its clients or constituents; an eighth related paper on Ethiopian immigrants in Israel is discussed below in the culturally sensitive interventions category. Most of these interventions occurred in countries other than those in North America and Western Europe. An intervention in Pakistan had a two-part purpose of increasing female college students' communication skills and sense of civic responsibility and increasing the iron-deficiency anemia knowledge of local women in a vocational education program (Ayub et al. 2015). The researchers described their program as a health literacy intervention because they aimed to create a trained, volunteer public health workforce to help meet the needs of low-literacy communities and build health knowledge. The researchers used pre- and posttests to identify statistically significant increases in students' perception of civic responsibility; students' perception of communication competency; and community women's knowledge of iron-deficiency anemia. Notably, the least educated community women showed the lowest knowledge gain.

(text continues on page 190)

The review identified several papers featuring the Ophelia (OPtimising Health LiterAcy and Access) approach developed by Richard Osborne and colleagues in Australia. A summary paper that reports intervention results from nine diverse sites in Australia met the inclusion criteria (Beauchamp et al. 2017). Readers can refer to the Ophelia protocol (Batterham et al. 2014), a Health Literacy Questionnaire validation study (Elsworth, Beauchamp, and Osborne 2016), and a detailed report of one of the nine sites (Goeman et al. 2016) (see results in Table C-8 for Goeman et al.) to learn more about the Ophelia model. The purpose of the nine interventions was to demonstrate “the successful application of the eight underlying principles to achieve development of health-literacy informed interventions with the potential to impact health and health outcomes” (Beauchamp et al. 2017, 5). Each site had a primary service organization that led the project and defined its own intervention outcomes based on their clients' responses to the Health Literacy Questionnaire, a survey instrument designed and validated by the research team. In general, the intended outcomes were improving clients' health literacy or identifying ways to make information or services more accessible. The authors report that of the six sites that used the Health Literacy Questionnaire to measure intervention outcomes, four found effects ranging mainly from small to moderate. Qualitative evaluation methods found a range of improvements that included self-efficacy, communication with providers, information seeking, and applying what was learned. Sites that involved volunteers reported they felt involved and useful in helping others.

A research team in the United Kingdom implemented an existing program called Self Care for People with the aim of increasing participants' self-care knowledge and skills and their confidence and intention to do self-care (White et al. 2012). The primary outcome was a change in the number of general practitioner consultations between baseline and followup over a 12-month period. The self-care program consisted of six three-hour sessions over a 3- to 6-week period. Participants received a handbook and exercises. The classes were held in small group sessions in workplaces and community-service locations for parents. Comparing the intervention groups to similar locations used as controls, the team did not find a significant difference in the use of general practitioner services after the self-care program. At the 6- and 12-month follow-ups, the researchers did find small but statistically significant improvements in intermediate outcomes, such as self-esteem and health literacy, defined as decisions about when to use services, intention to use services, and exposure to other self-care initiatives and resources.

In Shenzhen, China, a research team sampled communities by streets and residential buildings to be part of a health literacy intervention or control group (Zhuang et al. 2016). Both groups received basic health education through bulletin boards, posters, and health lectures; the intervention group also received text messages once a week for 1 year. The health content consisted of data, explanations, and recommended health behaviors encouraging action. Using a pre/post 22-item questionnaire with dimensions of health knowledge, behavior, and skill, researchers found that the intervention group had slightly higher health literacy scores after 1 year than the control group. The researchers also analyzed the acceptability of receiving messages and cost-effectiveness.

A research team in Georgia partnered with Meals on Wheels, a volunteer organization that delivers meals daily to homebound older adults. The team trained the volunteer drivers to deliver a health literacy skills intervention to the older adults they visited (Rubin et al. 2014). The intervention adapted a 2002 Meals on Wheels program used in southern Oregon as well as Questions Are the Answer (AHRQ) and the AskMe3 campaign (National Patient Safety Foundation). The volunteers scored near the top of the range on the Short Test of Functional Health Literacy for Adults, and their assignment was to foster interactive health literacy skills with the older adult clients so they could communicate more effectively with their health care providers. The paper reported the results of the training workshop and clients' recall of the coaching they received. The volunteers scored the workshops as effective, and client recall of the coaching interventions was high. Using a sample of clients who received the coaching and were contacted in the twelve months following, 93 percent recalled the coaching events; 83 percent said they thought about “good questions for good health”; and more than 75 percent could recall a specific coaching practice.

One study interested in people's self-monitoring skills used a group class on either sugar-sweetened beverage consumption or physical activity, behavioral diaries, and a teach-back call to assess people's accuracy in recalling messages and behaviors (Porter et al. 2016). The researchers measured participants' baseline health literacy with the Newest Vital Sign (Weiss et al. 2005). The researchers were interested in participants' ability to accurately complete food diaries and recall behavioral messages, as well as the amount of teach-back needed to confirm understanding. They found that overall, participants with the lowest levels of health literacy had the greatest difficulties with accuracy and recall and needed more teach-back. Compared to those who received the sugar-sweetened beverage content, the participants in the physical activity arm more accurately completed their diaries, but recalled fewer behavioral messages and needed more teach-back (Porter et al. 2016).

An academic–community partnership on Long Island, New York, used health literacy content in a miniature public health curriculum for community residents (Goodman, Dias, and Stafford 2010). The goal was to bridge cultural divides and increase residents' “science” and “research” literacy so that they could participate in a community health collaborative. Residents asked for the training so that they would be on “equal footing” with researchers. The curriculum consisted of eleven didactic sessions and four experiential workshops; sessions were held at public libraries because they were accessible and trusted local spaces. Adult literacy instructors helped to develop the curriculum. Outcome measures were knowledge change; participant satisfaction; and submission of a pilot grant proposal to the university partner. Pre- and posttests showed improved knowledge of public health research topics, and open-ended questions elicited positive responses to the curriculum with requests for more time spent on content and group activities. Ten of the thirteen community members in the sessions submitted pilot grant proposals.

Information-Seeking Interventions

Three projects focused on improving people's health information-seeking skills. In Australia, a team of clinicians, consumer representatives, and librarians led free public workshops to help adults improve their online skills in finding and using evidence-based health information (Gray, Elliott, and Wale 2013). Workshop participants were overwhelmingly positive about the workshops' effects on knowledge, skills, attitudes, and potential impact on seeking behaviors. Participants also reported that they intended to ask doctors new questions and change how they maintained their health and coped with chronic disease and pain. Higher income and being native born were associated with more positive reports about the workshops' impact. A similar intervention for working-class community members in the Washington, DC, area also found modest improvements in participants' ability to find and evaluate online information (Pomerantz et al. 2010). A computer skills and health information-seeking improvement intervention in suburban Maryland targeted older adults (Xie 2011). The research team offered public library-based classes using materials from the National Institute on Aging to help older adults learn how to use MedlinePlus.gov and NIHSeniorHealth.gov. Pre- and posttests showed significant improvements in computer and Web knowledge, along with decreased anxiety, increased efficacy, and increased participation in health care.

Culturally Sensitive Interventions to Build Knowledge and Skills

Five interventions aimed to increase not only health knowledge, but also behaviors and skills, such as communicating with others about health, seeking information, asking questions, and analyzing information, for distinct cultural groups; more interventions targeting cultural groups are discussed in the adult learners, chronic disease, mental health (other), and cancer sections. A project in a Midwestern U.S. city aimed to improve the functional, communicative, and critical health literacy skills of Iraqi immigrants for whom English is a second language (Hatamleh 2015). Using feedback from a community needs assessment, the researcher developed a three-hour educational session in Arabic that covered all types of health literacy skills and used a pre- and posttest to examine the effects. Functional health literacy (understanding information; familiarity with available resources; and ability to complete forms) did not improve as much as participants' self-reported ability to communicate with providers and critically assess information.

A 13-year program in Israel aimed to assist Ethiopian immigrants and improve their health by providing educational sessions and cultural liaisons and helping them communicate with health care workers (Levin-Zamir et al. 2011). The intervention also trained clinic staff to bridge cultural differences. The researchers reported improvements in clinic–patient relations; availability and accessibility of health services; improved navigation; and perceived well-being and self-management.

A health literacy model of behavior change was designed to help adults improve specific mental and physical health behaviors (Pleasant 2011). Facilitators varied in professional background and included physicians, nurses, nutritionists, physical therapists, social workers, and a pastor. Participants were recruited from a federally qualified community health center. Sessions were hosted in English and Spanish and held twice per week over a 6-week period. Interactive sessions employed lectures, activity, and goal setting. A few of the positive health changes include improved health values (cholesterol, blood pressure), increased exercise, improved nutrition, and increased physical ability. Additionally, participants reported changes in behaviors and attitudes, as well as gains in health literacy, such as reading and understanding food labels.

Two interventions for Indonesian refugees and asylum seekers in New Hampshire focused on explaining safe medicine use, empowering individuals to understand and access the health care system, and improving communication with health care providers (Mancuso 2011). One intervention was a community program on safe medicine use and provided a lecture, question-and-answer session, brochures, and wallet cards. A second intervention was a health fair with prevention information; body mass index, blood pressure, and cholesterol checks; instruction on how to read a food label; brown-bag medicine check-ups; and referrals to local care providers. Organizers received positive evaluations from community participants and health professionals.

Adult Learners and Health Literacy Interventions

A subset of health literacy knowledge and skill improvement interventions focused on adult learners. Several projects examined the effects of integrating health literacy content in adult education programs (Brown, Collie-Akers, and Fernandez-Ortega 2015; Chervin et al. 2012; Freedman 2011; Muscat et al. 2016; Soto Mas, Cordova, et al. 2015; Soto Mas, Ji, et al. 2015). As Soto Mas and two sets of colleagues (Soto Mas, Cordova, et al. 2015; Soto Mas, Ji, et al. 2015) observed, the National Action Plan to Improve Health Literacy (HHS 2010) calls for the integration of adult literacy curricula and health literacy content, but few examples exist of how to combine them. In one project, six adult education centers in the Northeast United States used a “study circles” approach to train adult education teachers to teach health literacy skills and integrate health literacy content into the curriculum (Rudd et al. 2005; cited in Chervin et al. 2012). The project goal was not only to improve adult learners' health literacy skills and increase their self-efficacy, but also to show how to teach health literacy as part of adult education. An evaluation found students' knowledge about health issues and self-efficacy increased significantly as a result of the health literacy instruction, and all six centers improved their capacity to teach health literacy. A knowledge and self-efficacy building intervention for primarily Latino adults attending adult community college or night school classes also found knowledge increases using either a fotonovela or brochure (Rodriguez 2015). A case study of a health literacy skills class in Atlanta, Georgia, described the effects of teaching functional health literacy skills to adult learners (Freedman 2011). The researcher was interested in how health literacy instruction could help adult learners to take greater control of their health and use what they learned in everyday life. She found that “careful attention to environmental factors and instructional strategies” can improve learning and help students carry lessons into everyday health behavior change (Freedman 2011, 81).

Five interventions conducted in English-speaking countries specifically considered the issues of health literacy for non-native speakers of English (Brown, Collie-Akers, and Fernandez-Ortega 2015; Muscat et al. 2016; Otilingam et al. 2015; Soto Mas, Cordova, et al. 2015; Soto Mas, Ji, et al. 2015). Soto Mas and colleagues implemented their Health Literacy and ESL (English as a second language) Curriculum with native Spanish speakers in two different U.S. communities to see if they could improve participants' health literacy as measured by the Test of Functional Health Literacy in Adults (TOFHLA) (Soto Mas, Cordova, et al. 2015; Soto Mas, Ji, et al. 2015). The Health Literacy and ESL Curriculum is a 12-unit package that includes traditional grammar, vocabulary, reading, writing, speaking, civic and life skills, math, and two health units. Certified ESL instructors teach the curriculum in 6 or 12 weeks. A 6-week implementation in New Mexico used a local elementary school, large hotel chain (employer), and community church as classroom sites. Using the Spanish version, the research team found that total TOFHLA, raw numeracy, and reading comprehension scores increased significantly; only two participants scored inadequate functional health literacy at posttest compared with thirteen at baseline (Soto Mas, Cordova, et al. 2015). Using a control group and the English-language TOFHLA to measure health literacy, a 6-week implementation in Texas significantly increased the intervention group's TOFHLA score compared with a control group (Soto Mas, Ji, et al. 2015).

Two interventions focused on nutrition topics and Spanish-speaking Latinas of Mexican origin (Brown, Collie-Akers, and Fernandez-Ortega 2015; Otilingam et al. 2015; see also Castañeda et al. 2016 in the cancer section). A nutrition literacy project used the SNAP-ED program, a federal program to reduce food insecurity and improve nutrition and wellness of low-resource groups, to reach Spanish-speaking women with low literacy skills (Brown, Collie-Akers, and Fernandez-Ortega 2015). The researchers tried the intervention with SNAP-ED clients as well as bilingual peer educators. The clients chose MyPyramid, reading Nutrition Facts labels, and using portion control for meals as the skills they wanted to learn. A qualitative evaluation found that participants rated the workshop and curriculum “very useful.” Otilingam and colleagues (2015) developed and tested Buenos Hábitos Alimenticios para Una Buena Salud, a program of two 2-hour nutrition workshops for Latinas. The workshops had culturally relevant nutrition information and skill-building activities, such as reading Nutrition Facts labels and preparing low-fat meals. The workshops were held in a community clinic familiar to workshop participants. Researchers used the Newest Vital Sign as a health literacy pre- and posttest and measured knowledge and self-reported health behaviors (Weiss et al. 2005). Compared with control groups, intervention groups showed health literacy, knowledge, and self-reported behavior improvements. Participants said the most important thing they learned was how to read nutrition labels.

An Australian project adapted a U.K. adult and health literacy program and conducted a feasibility study of an integrated adult and health literacy curriculum to improve functional, communicative, and critical health literacy skills (Muscat et al. 2016). The study mainly enrolled culturally and linguistically diverse students who spoke a language other than English. The students reported being satisfied and engaged with the content and said they were better prepared to communicate with health care professionals, including asking questions. However, the results of the evaluation questionnaires showed improvements in topic-specific health literacy, but no significant improvement in a composite measure of generic functional, communicative, and critical health literacy skills (Muscat et al. 2016).

Early Childhood Education and Health Literacy Interventions

The review includes three articles that involved Head Start programs, parents, or children (Burgette 2016; Fleary 2012; Stockwell et al. 2010). One project aimed to improve community members' oral health through a health literacy intervention. The project focused on training Early Head Start teachers and staff so they could educate parents and children less than three years old about good oral health and prevention, especially during pregnancy and for infants and young children (Burgette 2016). The educational intervention was Zero Out Early Childhood Caries. Baseline and 24-month reports indicated that children enrolled in this program were more likely to have had a dental visit and less likely to have negative quality of life results than children in the Medicaid control group. Parents' health literacy was not a significant factor for the Head Start children's results. A second intervention recruited mothers with children in a Texas Head Start program to participate in a 5-week educational intervention about healthy lifestyles, including diet/nutrition, physical activity, sleep hygiene, parenting skills, and mental wellness (Fleary 2012); the mothers' knowledge about all topics increased. In New York City, researchers partnered with adult literacy experts to create a culturally appropriate educational program in English and Spanish and care kit for upper respiratory infections, based on CDC campaign materials (Stockwell et al. 2010). The purpose was to educate low-income Latino parents with children in Early Head Start programs about appropriate care for upper respiratory infections. Researchers reported that parents' knowledge increased, and parents' attitudes about appropriate care changed.

Schools as Health Literacy Intervention Sites

This section reports on four studies that used schools as the setting for health literacy interventions; three additional school-based and one university-based study are reported in the mental health section (Fung et al. 2016; Hart, Jorm, and Paxton 2012; Noble, Hedmann, and Williams 2015; Pinto-Foltz, Logsdon, and Myers 2011), and one additional school-based study is reported in the medicine section (Chang et al. 2015). An intervention in Spain and one in Portugal tried to change student, school, and community engagement for health improvement. One project examined how schools in low-income, immigrant communities in Spain could use family-oriented health programs to reduce inequalities and foster social inclusion (Flecha, García, and Rudd 2011). Using a case study evaluation approach, the researchers found that the schools that paid close attention to a community's expressed needs, created a welcoming space for community members to gather, and included the community's “cultural intelligence” or rich knowledge about their own experience and needs were the most successful in addressing the community's health literacy needs and improving adults' and children's health literacy. Pais and colleagues (2014) worked with high school students to enhance health literacy through a community profiling project. Students assessed community needs and resources and took a leadership role in working with community members, professionals, and political representatives to address identified needs. The authors present two case studies from secondary schools in Porto, Portugal. Students negotiated topics with their teachers and workgroups and selected examination of health rights and health care service costs. Students participated in every phase of the research, created poster presentations, and discussed findings at a symposium at the University of Porto. Students said they valued the opportunity to share and listen to varied opinions during the poster symposium. Teachers observed students' increased ability to make critical judgments about past and present situations. The authors suggest the students' experience helped them exercise their citizenship, thus developing a foundation for critical health literacy in the future.

Diamond and colleagues (2011) used the Building Wellness Curriculum to design an after-school program for third through eighth grade students. Students came from urban communities, had high risk for chronic disease, and qualified for no-cost school lunches. Topics included food choices, healthy choices, body function, and skills about communicating with a health provider. The course was offered for one semester during each academic year over a span of 6 years and used discussion, in-class activities, reflection, and take-home activities. Researchers used the REALM Teen program to assess health literacy at years one, three, and six. Findings suggested consistent increases in knowledge and retention of information. Teachers' anecdotal reports indicate students demonstrated positive behavior changes and increased curiosity about health and the body. By contrast, a school-based project in Australia used general practice doctors and teachers to lead a health curriculum to increase high school students' health literacy, help-seeking behaviors, and access to local health services (Harrison, Ollis, and Savige 2016). Researchers found the students were receptive to the doctors and teachers as health resources, but the volume of information was overwhelming and not sufficiently in-depth on key topics. The evaluation found that even though students could recall information, there was little improvement in student knowledge, confidence, help-seeking behavior, or ability or desire to access health services.

Other Settings or Topics for Health Literacy Interventions

Son and colleagues (2016) worked with health care students to facilitate a three-day curriculum about “reproductive health literacy” at a juvenile detention center (see the cancer section for an additional study with incarcerated women) (Ramaswamy, Simmons, and Kelly 2015). The authors described “reproductive health literacy” as knowledge, self-efficacy, and communication skills. Participants were female and averaged age 16 years. Study findings were increased knowledge about sexually transmitted infections and increased self-efficacy in condom use. However, there was no significant improvement in sexual autonomy or contraception use.

Adult residents of the Richmond, Virginia, metropolitan area participated in a computer-based intervention to increase health literacy about preterm births (Vanderbilt et al. 2016). The majority (95 percent) of participants were African-American; about 20 percent of participants were male. Six modules included information about preterm births and related risk factors. The researchers said the modules used adult learning theory and plain language, among other techniques, to make the content accessible; the modules were reviewed by a range of experts, including health literacy experts. Women showed significant knowledge gains for all six modules. However, men did not show significant increases for the modules Let's Talk Patient and Provider Communication and It Takes a Village.

An intervention in Atlanta, Georgia, aimed to improve maternal and child health outcomes by building the knowledge of mothers who were homeless, had addiction problems, and lived in a rehabilitation facility (Oves 2013). Information for the mothers pertained to newborn care, preventive services, and social services. The mothers participated in a 6-hour class and received a manual. Using pre- and posttests, the researcher found increases in knowledge after the class and in a subsample at 2 to 4 months post-intervention.

Chronic Disease Risks and Management

Eight interventions focused on low health literacy adult groups with respect to chronic disease risk reduction and disease self-management. The interventions integrated health literacy techniques, such as simplified content and easy, action-oriented recommendations, to improve health literacy and sometimes change behavioral or physiological outcomes related to diabetes and cardiovascular risks or disease.

Chan's (2012) dissertation research in Hong Kong included two studies of older Chinese adults (average age 73 years) with one or more chronic diseases; the researcher chose older adults with chronic diseases because the literature review identified older adults as having low health literacy. To improve their health literacy as well as other outcomes, one study examined the effects of multiple-exposure, group self-management classes on self-management behaviors; self-efficacy; health status; and health care use (Chan 2012). Chan's interventions used simplified content, including plain Chinese and concrete examples. Using an intervention and control group, the researcher found statistically significant improvements in all self-management behaviors and self-efficacy outcomes and half of the health status measures. No change in health care use was found. In the second study (Chan 2012), the researcher proposed that well-designed diabetes self-management programs could improve adherence, health literacy, and health outcomes (Chan 2012, 16). The plain Chinese-language intervention included multiple-exposure, group diabetes educational sessions designed to improve a range of outcomes, including knowledge, quality of life, mental health, food consumption, and physiological measures. Using an intervention and control group, the researcher found statistically significant improvements in diabetes-related knowledge and quality of life, along with reductions in total energy and saturated fat intake. No changes in physiological measures, such as blood sugar and blood pressure levels, or other nutritional and quality-of-life measures were found.

Lam (2014) also conducted an intervention with Chinese adults (45 years and older) in Hong Kong, again because of their low health literacy and its relation to chronic disease management; the purpose was to increase their physical activity and improve their diabetes outcomes. Lam describes the intervention as culture, language, disease, and age appropriate, as well as theory based and “health literacy-oriented,” meaning the intervention was “a patient self-care empowerment program to nurture patients' ability to obtain, understand, comprehend, and analyze the health information and services that they needed [sic]” (Lam 2014, 18). A nurse and certified exercise trainer led the six-week classes followed by a telephone follow-up. Compared with the control group that received standard diabetes education, the intervention participants doubled their physical activity levels from sedentary to acceptable levels; no health literacy results were reported.

Two different interventions targeted Korean Americans with chronic conditions (Kim et al. 2014; Kim et al. 2015). One focused on older adults (average age 70 years) with high blood pressure who were selected because of low health literacy (Kim et al. 2014). A high blood pressure management intervention for Korean-American older adults added a health literacy component to educate participants about words and phrases relevant to blood pressure as well as communication skills with providers (Kim et al. 2014). The team used a high blood pressure health literacy scale they had validated in other research. The intervention was delivered in small groups at churches and senior centers over 6 weeks. In comparison to the control group, the intervention group had better medication adherence; lower blood pressure; better health literacy scores at the 12- and 18-month follow-ups (but not the 6-month follow-up); more blood pressure knowledge; and less depression. In the other study with first generation Korean-American immigrant adults (35 years or older), researchers selected the participants because of low health literacy and reported problems even using health materials in their native language (Kim et al. 2015). The health literacy component focused on building skills for tasks such as reading food labels and understanding medical terms; this was done through 2-hour group education sessions over 6 weeks. Although the researchers identified health literacy as a secondary outcome, it was not measured and reported separately in the results. The researchers found the intervention group improved on all outcomes of hemoglobin A1c levels, diabetes-related knowledge and self-efficacy, and quality of life in comparison with the control group (Kim et al. 2015).

An intervention on diabetes and blood pressure knowledge and self-efficacy involved older African Americans (average age 72 years) with low health literacy in senior affordable housing complexes in Washington, DC, and Maryland (Bertera 2014). The researcher designed and tested educational materials featuring community members as “talent”; combined the African American tradition of oral storytelling with the Hispanic medium of fotonovelas; and provided group educational sessions. Using intervention and control groups and pre- and posttests, the researcher showed a statistically significant increase in intervention participants' knowledge and self-efficacy. Using Chew et al.'s (2008) item set, the researcher found intervention participants with the lowest scores reported the greatest gains in self-efficacy.

Two interventions were implemented with Hispanic community members (see the culturally sensitive interventions, adult learners, mental health [other], and cancer sections for more interventions with Hispanic/Latino community members). Using data from a community health needs assessment and feedback from community leaders, a team in Shelbyville, Kentucky, created an educational program, “The Environment and Your Health” (Ramos et al. 2013). The curriculum placed special emphasis on cardiovascular disease and diabetes. Hispanic residents took two sets of health topic-specific classes 9 months apart, and pre- and posttests showed significant knowledge gains. In Boston, Massachusetts, researchers adapted a successful cardiovascular risk reduction program for African American women (Love Your Heart) and low-income Hispanic women (Romero et al. 2016). Study participants were recruited from two predominantly Hispanic housing projects and one local community health center. They attended weekly 90-minute sessions for 6 weeks. Based on pre- and posttest data, the culturally tailored Spanish-language cardiovascular disease prevention program showed knowledge improvements after the program (Romero et al. 2016).

Mental Health Literacy Interventions

Seventeen articles reported interventions to increase knowledge about mental health, increase confidence in identifying a person with a mental health problem, and reduce social stigma about mental illness. The majority of the mental health literacy studies focused on depression, psychosis, and stigma, though other topics were covered. Communities participating in these interventions varied by occupation (coaches, police officers, college students, farmers, community health workers, and foster care workers), age (older adults, school-aged youth), and race/ethnicity (Asian, Latino, African American). Intervention formats included traditional class settings, workshops for adult and student groups, and use of illustrated novels (fotonovelas).

The review identified seven papers using the Mental Health First Aid (MHFA) curriculum, a standard program in community mental health literacy research (Kitchener and Jorm 2008). The intervention focuses on improving knowledge, behavior, and attitudes toward people with mental health problems and encourages help-seeking and self-help strategies. The course is typically delivered in a group setting over multiple sessions for a total of twelve hours; however, two interventions used 4-hour courses (Armstrong et al. 2011; Hart, Jorm, and Paxton 2012). Most articles describing an MHFA program included a topical breakdown of content used in training, such as an introduction to mental health disorders, mental health first aid, and practice-based skills (Armstrong et al. 2011).

Several interventions used a questionnaire to assess mental health literacy, which included vignettes about depression and schizophrenia/psychosis. The questionnaire and vignettes can be translated into community members' preferred languages and tailored in duration and delivery. Only one article described qualitative findings on use of MHFA in diverse community groups, including social workers, religious leaders, and health professionals (Svensson, Hansson, and Stjernswärd 2015). Six focus groups were held to understand how the MHFA course influenced participants. Benefits of the course were described, including understanding and humility toward people with mental illness and useful advice on handling a mental health illness crisis.

Three interventions decreased stigmatizing attitudes about mental illness (Anderson and Pierce 2012; Armstrong et al. 2011; Lam, Jorm, and Wong 2010; Wong et al. 2017). In addition to reducing stigma in adult Chinese immigrants in Melbourne, Australia, Lam and colleagues increased the participants' ability to recognize symptoms for depression and schizophrenia as well as improved beliefs about treatment, though there was no change in beliefs about seeking professional services for treatment (Lam, Jorm, and Wong et al. 2010). In another study focused on Chinese adults living in Hong Kong, Wong et al. (2017) improved participants' ability to recognize symptoms of depression and schizophrenia and confidence in offering help to others. Similar to the Lam study, Wong and colleagues (2017) did not see a decrease in perceived dangerousness and dependency and suggested that traditional Chinese beliefs about mental health may require long-term strategies to change. Anderson and Pierce (2012) trained coaches and leaders of sports teams in rural settings in Australia to improve their understanding of mental health conditions, increase their confidence in assisting those with mental illness, and decrease stigma. Results indicated an increase in knowledge regarding their reported ability to identify depression and schizophrenia/psychosis. Half of the coaches demonstrated an increase in self-confidence about helping someone experiencing mental illness, and stigma was reduced for three of every four participants. Armstrong and colleagues (2011) trained community health workers within rural communities in India to reduce stigma, but there was only evidence of a minor reduction. However, ineffective but culture-typical pharmacological interventions (vitamins, herbal medicines, and appetite stimulants) to treat mental health conditions decreased, and three months after course completion, participants' ability to identify depression and psychosis remained statistically significant.

The Mental Health First Aid curriculum has been applied to other topic areas, such as preventing eating disorders (Hart, Jorm, and Paxton 2012) and raising awareness of drought-related mental health problems (Hart, Berry, and Tonna 2011). Hart and colleagues (2012) trained college students living in residence halls at the University of Melbourne to improve knowledge, behavior, and attitudes toward people with eating disorders. Though there was no change in knowledge like other studies, participants demonstrated improved ability to recognize an eating disorder and appropriate mental health strategies, such as knowledge of effective treatments and informal help seeking. Hart, Berry, and Tonna (2011) described a program for people living in rural communities of New South Wales who had experienced severe drought. The intervention aimed to raise awareness among community members and rural farmers about mental health literacy and the effects of the drought on mental health resilience through community events. The program generated a free mental health telephone crisis line that received more than 270 calls from local farmers; booklets about local mental health services; and multiple community events and forums to raise awareness about mental health issues.

In addition to the seven MHFA studies, an additional 10 papers in this section represented a wide variety of mental health interventions. Two studies focused on reducing mental illness stigma (Fung et al. 2016; Pinto-Foltz, Logsdon, and Myers 2011). Fung and colleagues recruited adolescents and adults in Hong Kong for five interactive 3.5-hour workshops. Findings demonstrated that both adolescent and adult intervention group participants increased mental health literacy and decreased stigma. Researchers observed that education is more effective for adolescents whereas contact is more important for adults. A 10-week program, called “In Our Own Voice,” used narrative storytelling and videos to reduce mental illness stigma and improve mental health literacy in a sample of 13-to 17-year-old girls attending two urban public high schools (Pinto-Foltz, Logsdon, and Myers 2011). Unlike the Fung intervention, stigma did not change post-intervention; however, mental health literacy scores increased at 4- and 8-week follow up.

Three interventions aimed to improve “dementia literacy” (Noble, Hedmann, and Williams 2015; Rodriguez 2015; Taylor et al. 2012). Noble and colleagues used a program called Old SCHOOL (Seniors Can Have Optimal Aging and Ongoing Longevity) Hip-Hop to train fourth and fifth grade students attending a school in Harlem, New York, using three 1-hour sessions with music, role-play skits, and short films to overcome cultural barriers and beliefs about Alzheimer's and dementia, familiarize students with symptoms, and develop an appropriate response to them. Knowledge of symptoms about Alzheimer's disease increased after the intervention and was retained at the three-month follow-up. Researchers developed a 16-minute DVD titled “Looking Out for Dementia” to educate three Aboriginal communities in Australia about dementia (Taylor et al. 2012). Findings reveal increased awareness about dementia as a health issue, appreciation by communities receiving information in their native language, and the need to describe dementia as a term that does not have a comparable term in some languages. A fotonovela educational intervention designed to increase knowledge of dementia and the benefits of early diagnosis was implemented in 13 community college and adult night school classes with English- or Spanish-speaking adults (Rodriguez 2015). Compared with a brochure with dementia content and Latino family photos and a brochure with no dementia content and Latino family photos, the fotonovela content was rated most highly by participants, although both the fotonovela and brochure with dementia content and culturally appropriate photos elicited knowledge increases.

Fotonovelas were also used in two additional studies (Hernandez and Organista 2013; Unger et al. 2013). Hernandez and Organista trained promotoras to deliver fotonovelas with depression and stigma information to Latinas 18 to 55 years old and at high risk for depression. This intervention increased depression literacy, decreased stigma, and increased help-seeking knowledge and behavior. Unger and colleagues used the Secret Feelings fotonovela in a Hispanic community to increase knowledge about depression. Knowledge gain was higher in the group receiving the fotonovela compared to those who received a text pamphlet. Stigma decreased significantly in the intervention group, though scores regressed to near baseline at follow-up. Participants who received the fotonovela were more likely to report they shared the fotonovela with more than one other person compared to the group receiving the text pamphlet.

Two papers described the effect of community health workers delivering mental health literacy programs in rural villages in India (Mindlis et al. 2015; Shidhaye et al. 2017). Mindlis and colleagues assigned villages to the intervention group if they had received CHW training to support their village needs, mental health education workshops, and community reintegration activities for people with mental health conditions. Intervention villages demonstrated increased mental health literacy about depression and lower levels of stigma; however, both intervention and control villages retained a fear of medications because of potential addictive effects. Findings suggest that tailored interventions localized by village and delivered by CHWs may positively influence mental health literacy. Shidhaye and colleagues implemented a community education program (VISHRAM) in 15 rural villages in Vidarbha, India, to address mental health and suicide. CHWs conducted small group meetings and home visits, established partnerships, and obtained buy-in from key stakeholders. The result was increased demand for mental health literacy services. Requests for treatment increased six-fold over the 18-month implementation and reached more than 1,400 participants. Another important finding was a decrease in the median cost of care from baseline, which was attributed to the likelihood that villagers no longer had to seek care in the private sector, but obtained specialist services locally and at no cost, and CHWs advocated for mental health support and care.

Dowrick and colleagues (2013) aimed to increase understanding of and access to mental health services and described the “output” as improved health literacy. They designed an intervention to work at the community, health care system, and policy levels. Their target community was older adults and South Asian and Somali adults in the United Kingdom. The researchers used community engagement, primary care quality improvement, and tailored psychosocial services to provide improved service access. Group interventions were offered over 8 to 10 weeks for 1 to 2 hours each and presented case examples of people experiencing anxiety and depression. Primary care practices also received training. Qualitative evaluation suggests that participants in the intervention group who received the well-being program improved in comparison to those receiving usual care.

Two studies trained participants to assist individuals with mental illness or to counsel children (Hansson and Markström 2014; Mosuro, Malcolm, and Guishard-Pine 2014). Swedish police officers were trained over 3 weeks for a total of 31 hours on mental health literacy (Hansson and Markström 2014). The training included a video about various types of mental health conditions and the opportunity for students to role-play potential situations. The intervention group demonstrated an improved attitude toward people with mental illness, confidence in assisting someone with mental illness in accessing help, belief that medication and psychotherapy can help, and belief that people with mental illness can fully recover. Mosuro, Malcolm, and Guishard-Pine (2014) enrolled foster caregivers (parents/caretakers) in Great Britain in a nationally accredited 30-hour course modified for foster caregivers that uses didactic teaching and role-play sessions. The course was designed to increase caregivers' confidence in their mental health knowledge and ability to help counsel the children in their care. The post-course assessment demonstrated increased mental health awareness, but no impact on coping skills.

Following a devastating earthquake in Japan, Tuerk and colleagues (2013) developed an intervention to assess mental health literacy in the adult population of Mito, Japan. Participants completed questionnaires about feelings and experiences of posttraumatic stress during the 2.5-hour training. Forty-one participants turned in their questionnaires after the training. Results showed that 27 percent met criteria for posttraumatic stress disorder and 15 percent reported the earthquake was the index event. Other findings from the surveys included problems of physical health in the past 7 days and a lost sense of community. The researchers suggested that interventions such as these can address citizen concerns, inform subsequent outreach and training efforts, and help community members learn how others respond to natural disasters.

Interventions on Medicines and Health Literacy

Six articles in the review pertain to medicine and health literacy. Outcomes of interest were knowledge, self-efficacy, and communication skills.

Two studies used games to deliver an intervention. The first recruited adults ages 18 to 64 from an urban, multiethnic community setting to assess how interactive, educational board games influenced participants' knowledge about medicines and communication skills with their pharmacist (Burghardt et al. 2013). Participants who played the games were significantly more likely to report intent to seek pharmacist medication advice in the future compared to control group participants. The second study to use games was a dissertation focused on older adults living independently in a retirement community (Creech 2014). Older adults participated in a single session about medicine safety as part of a longer health class. The author used the Newest Vital Sign tool to assess health literacy (Weiss et al. 2005). Results demonstrated an increase in knowledge, but not in self-efficacy.

Adherence was the focus of two studies in this review. The first describes a school-based program in Taiwan (Chang et al. 2015). Pharmacists and teachers developed a course to help primary and middle school students communicate with their doctors, read medicine labels and instructions, assume responsibility for taking their own medicines, and talk with their pharmacists. Researchers noted an increase in knowledge, self-efficacy, and skills in the intervention groups compared with the control groups. The second study on adherence focused on older adults attending an inner city day center (Martin, Kripalani, and DuRapau 2012). Researchers instructed participants on use of a personalized, illustrated daily medicine schedule, PictureRx. Participants also completed baseline measures, including one for health literacy using the Newest Vital Sign (Weiss et al. 2005). Twenty-five percent of participants had limited health literacy, and another 25 percent had a diagnosis of dementia. Six weeks later participants were reassessed. Findings demonstrated that medicine adherence and self-efficacy improved after the intervention. Following the study, changes in program protocol included using teach-back methods by nursing staff and revising medicine instructions to avoid acronyms, using plain text descriptions instead.

Two studies focused on increasing knowledge about antibiotics and complementary and alternative therapies. David and colleagues (2017) delivered a 60-minute seminar modeled after the Ask Me 3 communication campaign to educate adults 18 and older about antibiotic resistance. Researchers used the Newest Vital Sign to assess participants recruited through local churches and YMCAs (Weiss et al. 2005). The slide presentation described the differences between viruses and bacteria, provided examples of when antibiotic use is appropriate, explained how antibiotics work, defined antibiotic resistance and examples of what could cause antibiotic resistance, and outlined steps to decrease misuse. Knowledge scores increased, but the authors noted that health literacy affected knowledge gains particularly in those with limited health literacy. Adults over age 65 living in a rural community participated in a 7-week course about complementary and alternative therapies (Shreffler-Grant, Nichols, and Weinert 2017). The course used face-to-face classes and webinars, and provided take-home materials for participants. Researchers assessed health literacy using the Newest Vital Sign (Weiss et al. 2005), the Chew Health Literacy Screener (Chew et al. 2008), and a scale specific to complementary and alternative therapy literacy. Findings demonstrated that health literacy mean scores increased, but were not statistically significant.

Cancer and Health Literacy

Eight papers reported health literacy interventions about cancer focused primarily on increasing knowledge and self-efficacy. Most of the studies focused on women, and three of the studies addressed cost.

Culturally tailored and peer-led interventions were used to increase knowledge about breast and cervical cancer. Using promotoras to facilitate the six-week course, Nuestra Cocina: Mesa Buena, Vida Sana, researchers delivered a cooking class to Latinas over age 40 (Castañeda et al. 2016). Topics in the course addressed women's health and cancer basics, navigation of screening services, myths about breast cancer, communication with providers, and risk reduction strategies. Findings demonstrated an increase in knowledge and a greater likelihood of having a mammogram or colorectal cancer exam after the intervention. However, there was no significant increase in health literacy as measured by the Chew Health Literacy Screener, which assessed for confidence in filling out medical forms, asking for help reading medical material, and interpreting written medications (Chew et al. 2008). Korean-American women from 23 ethnic churches in an urban setting participated in an intervention about breast and cervical cancer screening. Community health workers received training and later trained participants in CHW homes, food courts, and ethnic cafes and grocery stores (Han et al. 2017). Workshops were designed to build health literacy skills regarding medical terminology about screening, medical instructions, and steps to navigate the health care system. Following the intervention, CHWs conducted monthly phone calls with participants and provided navigation support for up to six months. The researchers used an Assessment of Health Literacy in Cancer Screening (52 items), which assesses print literacy, numeracy, and cancer-specific terms. The intervention group had a higher likelihood of receiving mammograms and cervical cancer screening at a 6-month follow-up compared with control participants. Researchers assessed cost-effectiveness using an incremental cost effectiveness ratio. The cost per screening for someone in the intervention group to receive screening compared with someone in the comparison group was $236. More details about the cost-effectiveness study can be obtained in Schuster et al. (2015).

Two studies designed interventions on empowerment for adolescents and young adults. Researchers used a cervical cancer education program called the “Sexual Health Empowerment (SHE) Project” delivered to women over age 18 in a county jail in Kansas City, Kansas (Ramaswamy, Simmons, and Kelly 2015). Session topics focused on increasing knowledge, reducing perceived barriers, improving self-efficacy and ability to communicate with providers, and navigating the health care system. Results demonstrated increased scores in knowledge about cervical cancer, increased self-efficacy for screening, and increased confidence in navigating providers and health systems. In a qualitative evaluation of the After Cancer Care Ends, Survivorship Starts for Adolescent and Young Adults (ACCESS AYA), researchers explored how it improved health literacy among participants. The participants included community health providers, AYA survivors, caregivers, and cancer advocates. One important finding among AYA survivors was the desire to practice self and community advocacy, which may be tied to the program's emphasis on self-advocacy with medical professionals. AYA survivors also described increased knowledge and establishing community connections with other survivors. Cancer advocates reinforced the need for education and dissemination of materials as well as in-person group meetings for survivors to share information and experiences.

Environmental Health Risks, Communication, and Behaviors

Two papers reported interventions to clearly communicate environmental data and information in communities affected by health hazards (Paul et al. 2015; Ramirez-Andreotta et al. 2016). Both sets of researchers noted that communicating environmental risk data and recommendations in ways that motivate health-protective behaviors is especially challenging. One article used health literacy as the core concept (Paul et al. 2015), and the other used environmental health literacy, which they define as an “understanding of the connection between environmental exposures and human health” (Ramirez-Andreotta et al. 2016). A paper on well water contamination identified low health literacy in the community as a contributing factor to low levels of recommended water-testing behaviors (Paul et al. 2015). Community presentations on health risks and protective actions, an information campaign, community distribution of water kits, and test results produced a three-fold increase in water testing. In the other intervention, researchers did two home visits with residents living in a Superfund designated area (Ramirez-Andreotta et al. 2016). Researchers took biological samples from residents and environmental samples around the home. They also collected resident information with a questionnaire and asked residents to complete activity duration and dietary logs for the 4 days before the second home visit. Residents received individual and summary toxicity-level reports and phone calls to discuss results. The community hosted three meetings. An evaluation showed residents used the information to cope, change family household behaviors, and take steps to reduce exposures.

Policy/System Intervention

The review identified one paper reporting the effects of a national health policy change in Sweden that the authors stated was intended to increase population health literacy skills (Mahmud et al. 2010). The policy emphasized primary care, more individual responsibility for self-care, and population health. County councils were expected to provide local primary care services (health squares) and health promotion initiatives to strengthen health and health literacy in communities. Researchers were interested in how the policy was implemented at local levels and how well it supported the policy goals. They analyzed thirty “health squares” (primary care centers) and found that people's limited understanding of health promotion and empowerment concepts was a weakness, but a strength was a wide choice of health information materials and individualized counseling with staff. The authors suggest that the actors in the policy/community process should build a common understanding of health promotion and empowerment so the policy could be implemented as intended.

Excluded Papers of Interest for Community Health Literacy Interventions

Some excluded papers described interventions with interesting features relevant for community-based health literacy interventions. Although the interventions did not match the inclusion criteria, the team believes the papers can foster discussion about how health literacy insights can contribute to the broad domain of community-based interventions. Some excluded papers may also suggest potential research partnerships and intervention designs that can benefit from health literacy insights.

Authors of included papers discussed their challenges in recruiting large numbers of community members or entire communities to participate in research. Therefore, three excluded papers on health literacy issues for recruitment and retention of community members in research studies are of special interest (Nagler et al. 2013; Pelto et al. 2016; Rexroth and Friedland 2010). These papers were not included in the review because they did not describe interventions and outcomes, but they could inform dialogues on community engagement and “research adherence,” that is, a community's sustained interest and participation in research. Community members' interest, willingness, and continued participation in a study could be a valuable community intervention outcome with health literacy being a prime contributor.

Three interventions helped health profession students be more effective communicators by interacting with community-dwelling members outside health care visits (Grice et al. 2014; Hjertstedt, Barnes, and Sjostedt 2014; Milford et al. 2016). These papers were excluded because the interventions focused on educating the students to interact one-on-one with individual community residents. Nevertheless, these interventions showed opportunities to get health professionals “out of the classroom and into the community,” as the title of one paper put it (Milford et al. 2016). One intervention trained first and second year medical students to counsel Head Start children, parents, and staff on pediatric obesity prevention (Milford et al. 2016). An intervention for dental students sent them into older adults' homes to educate them about oral health and measure their oral health literacy and oral health status (Hjertstedt, Barnes, and Sjostedt 2014). Third-year pharmacy students were assigned to older adults living in senior housing (Grice et al. 2014). The students used health literacy techniques such as Ask Me 3 and teach-back to collect health histories, do safety checks, and assess medicine use, among other tasks.

One study used an interesting environmental intervention design to increase health literacy, but was excluded because people were recruited as individuals (Crim 2013). University faculty and staff were the target of an intervention to affect individually perceived health literacy and purchasing behaviors in on-campus eateries. The researcher designed and implemented a red, yellow, and green point-of-purchase labeling system to rate foods. Color-coded stickers, lighted menu boards, posters, emails, and table tents communicated the rating system. Exposure times were varied across the eateries: one eatery had the intervention for 6 weeks, the second for 4 weeks, and the third for 2 weeks. Comparing responses from a paper-and-pencil pretest and an online survey posttest, 42 percent of respondents said the color coding influenced their awareness, understanding, and food choices, but not intentions or knowledge. There was a significant increase in “green” food sales and a significant decrease in “red” food sales, regardless of length of intervention.

Two examples—one from Bangladesh and the other from Venezuela—hint at how community-level interventions could be designed with health literacy outcomes that extend beyond individual knowledge and behavior change. Since the mid-1990s, a research group in Bangladesh has helped build communities' capacity to address their own problems (Bhuiya, Hanifi, and Hoque 2016). Using a Participatory Rural Appraisal model, the researchers worked with rural, village-run organizations on a “self-help for health” program. The stated purpose was to build organizational capacity to address collective health problems, improve health literacy through individual awareness related to health improvements, and support participatory planning and monitoring in villages. Using intervention and comparison communities, surveys, project documents and workplans, and reports, researchers reported a number of improvements in the intervention community, including more organizational capacity as well as better health outcomes, such as lower infant mortality and increased vaccination. However, health literacy was mentioned only in the abstract, and villagers' literacy rates were mentioned only in passing in the body of the paper. The paper does not explain what the intervention did to improve literacy or health literacy and if there were any changes.

In a second example, researchers in Venezuela described an advocacy and social mobilization intervention over a 4-year period to increase health literacy about breast cancer in Venezuela (Eid and Nahon-Serfaty 2015). They used health literacy as the justification as well as the proposed outcome, but did not explain how the campaign incorporated or tried to affect health literacy. Nongovernmental organizations, physicians, scholars, journalists, public officials from state and municipal governments, and women who had experienced breast cancer created an action-oriented network to develop advocates and activists, enhance patient–provider communication, and promote social dialogue among stakeholders. The network organized multiple activities such as national campaigns, a cancer walk, educational materials, digital media, conferences, and public opinion forums. Annual workshops taught attendees how to create messages about breast health to promote attitudinal and behavioral changes. The focus of these efforts culminated in mobilizing multiple levels of local communities and national organizations for a national policy on breast cancer. However, the researchers did not report specific health literacy components.

Community Engagement in Planning and Implementing Interventions

The first theme pertains to the seriousness with which researchers are engaging with communities to design interventions. The many positive results may reflect, in part, that interventions frequently incorporated community perspectives and input even at formative stages, leading to interventions that participants often characterized as a good fit. High levels of community involvement at multiple stages in the intervention process are appropriate and admirable for health literacy improvement work. To the degree that health literacy and cultural competence are conceptually linked, community involvement ensures that interventions are relevant, understandable, and useful in helping communities to protect and promote health. Also, community specificity may partially account for the number of unique intervention designs and implementations, a low number of randomized trials, and frequent use of qualitative evaluations and self-report data in the final set of papers.

A few papers illustrated intervention failure when community perspectives were not appropriately considered. An evaluation found that school-based health programs for immigrant and culturally marginal school children and parents in Spain often ignored the expressed needs of the community and community members' cultural intelligence (italics in original) (Flecha, García, and Rudd 2011). The researchers concluded “the course [on how to keep children from being involved with drugs] failed because the abilities and knowledge that the participants had acquired from their experience with drugs in their families were ignored” (Flecha, García, and Rudd 2011, 213). The negative consequences for health literacy were that family members did not have sufficient trust in the school and teachers as information sources and were less interested in participating in the planned activities.

Community specificity may also help account for the low number of replicated interventions, with the Mental Health First Aid and Ophelia models being notable exceptions. The Australian adult literacy program and the U.S. Meals on Wheels health coaches adapted existing interventions for new communities (Beauchamp et al. 2017; Muscat et al. 2016; Rubin et al. 2014). Researchers were more likely to create their own materials and programs, and sometimes measurement tools, than to try an intervention effective with a different community, setting, or topic. Encouraging researchers to publicly share their instruments, materials, and program designs may motivate more adaptation or replication with appropriate cultural considerations. A few research teams have posted their materials online: Georgia Meals on Wheels (Rubin et al. 2014), the Australian Ophelia model (Beauchamp et al. 2017), and Buenos Hábitos Alimenticios para una Buena Salud (Otilingam et al. 2015).

Reporting back to communities characterizes community-engaged research, although report-back is not an element in the CDC Best Practices framework nor an activity the included papers typically addressed. Two excluded papers did not report interventions, but did discuss the health literacy and ethical issues as well as experiences and recommended practices for reporting biomonitoring and environmental exposures to individuals and communities (Brody et al. 2014; Ohayon et al. 2017). The papers indicate that reporting back clear and useful information to communities should be the norm. One set of authors concludes, “Thoughtful report-back can strengthen research experiences for investigators and participants and expand the translation of environmental health research in communities” (Brody et al. 2014, 40).

Health Literacy Components of Interventions

A second theme pertains to the way “health literacy” was used in interventions. The review team did not have preconceived ideas about when and how health literacy would appear in an intervention. Instead, through the review process, four ways emerged regarding how health literacy was included at one or more points in an intervention. The papers included described interventions that (a) identified or characterized the community of interest as “low health literate”; (b) used a health literacy instrument or measure to advance the analysis; (c) designed programs or materials using health literacy principles or techniques; or (d) assessed changes in knowledge, self-efficacy, skills, behaviors, or other outcomes that authors considered to be included in health literacy or related to health literacy. Table C-8 refers to these possible loci of health literacy in interventions as (a) community, (b) measurement, (c) content, or (d) outcome, respectively. On the one hand, the appearance of health literacy considerations at multiple points in an intervention is a positive sign that researchers take health literacy seriously as a pervasive factor in intervention effectiveness. On the other hand, the lack of consistency in using these four elements (or others the field may be interested in) is a sign that researchers do not share a common view of when health literacy should be considered in interventions.

Some examples illustrate problems created when the health literacy concept is inconsistently applied. During the full-text review, the review team excluded many articles that used “health literacy” in the title or abstract only and not in the body of the paper. The interventions were typically trying to increase knowledge and sometimes self-efficacy, empowerment, or other factors related to health attitudes and behaviors; and to change one or more health behaviors. The papers, however, did not address explicitly the four health literacy components. In a school-based health program to change family knowledge and behaviors for diabetes, researchers aimed to train healthy adolescents to become health coaches for family members with diabetes (Gefter et al. 2016). The researchers described a group—low-income ethnic minority students from families with diabetes—with a high likelihood of having low health literacy. However, the researchers do not report health literacy measures, techniques, or outcomes for the students or intervention; yet, the abstract conclusion says “this program can increase health knowledge and some psychosocial assets of at-risk youth and holds promise to empower these youth with health literacy [emphasis added] and encourage them to adopt healthy behaviors” (Gefter et al. 2016). A culturally adapted cardiovascular risk reduction program for African Americans manifests a similar approach (Brewer et al. 2016). The researchers describe many contextual factors that would suggest limited health literacy being a major consideration for both community members and the intervention implementation and results. However, health literacy was not evident in the intervention itself or the actual results.

At the very least, inconsistencies in the ways “health literacy” is invoked in an intervention can create problems when developing coherent reporting schemes for comparative reviews such as this one. From a scientific perspective, the lack of explicit and consistent use of “health literacy” to inform intervention design, implementation, and reporting may impede building the body of knowledge about what characterizes effective and ineffective community-based health literacy interventions. As evidenced by the examples above, interventions may also append “health literacy” without sufficient evidence to warrant the connection.

Health Literacy and Knowledge Outcomes

The review team expected to find many knowledge-building interventions, and the third discussion theme pertains to how to think about knowledge improvement as a community intervention outcome. Although increasing health knowledge is popular, it can also be a disparaged outcome when researchers, practitioners, and policy makers expect every intervention to result in a behavior or health status change. The review team found many community interventions aimed to increase participants' knowledge of a health topic or general health care. Because it is axiomatic that educational interventions nearly always increase short- if not medium-term knowledge, it would be easy to dismiss knowledge interventions and results as finding the obvious. From another perspective, however, the large number of knowledge-building interventions is also a sign of a fundamental and structural problem in many communities. In addition to the well-documented problem that many health information materials are a poor match for people's skills, the studies in this review also show large-scale problems with communities' limited information awareness and access. Even in countries such as the United States that have universal public education and an abundance of communication channels and information sources, low levels of health knowledge are an ongoing problem; moreover, prior research shows health knowledge and health literacy are related (Dennison et al. 2011; Hom et al. 2012; Williams et al. 1998). The studies included paint a picture of many communities cut off from the large amount and variety of health information that exists beyond their personal and collective experiences. As a team of Australian researchers observed about the adult education students in their intervention, “Students had limited or no formal experience learning about issues regarding their health” (Muscat et al. 2016, 5).

One excluded paper, a cardiac risk reduction intervention with African-American church-goers in Rochester, Minnesota (Brewer et al. 2016), provides information about the large and ongoing need to build knowledge in communities, as well as the limited role that health literacy insights have played in shaping many community interventions. The researchers for this intervention note an abundance of information about how to prevent cardiovascular disease and examples of successful cardiac risk reduction interventions for African Americans, and yet there is an extremely low level of ideal cardiac health among African Americans. But this information had not diffused in the African American community in Rochester. The intervention came about when three African American churches in Rochester asked local researchers for help with a health and wellness program. The researchers introduced church-goers to a culturally adapted version of Life's Simple 7, the American Heart Association's cardiac risk reduction program. Not surprisingly, researchers found they had to tackle low levels of cardiac health knowledge as well as other factors to show improvements in health outcomes. This study and many others in the review suggest that low levels of knowledge about reliable health information remain an ongoing problem that could negatively affect the development of higher level skills, such as effective communication and critical analysis. A continued focus on knowledge building as part of health literacy interventions may be necessary because communities often come to interventions isolated from the larger universe of science-based health information and lacking foundational knowledge that affects health literacy.

Challenges Applying an Evaluation Framework

The last discussion theme addresses the fit between evaluation frameworks for generally-used public health interventions and the community-based health literacy interventions in this review. The review team considered several evaluation models before selecting the CDC Best Practices framework (Spencer et al. 2013). The framework assesses the impact and quality of real-world programs to improve health and includes the general categories of Effectiveness, Reach, Feasibility, Sustainability, and Transferability. Each general category has several subcomponents. The framework includes a separate set of criteria about evidence quality, but the review team did not agree with definitions for “weak,” “moderate,” and “strong” evidence. The team specifically disagreed with the rating that all peer-reviewed studies are “strong” in quality. The team decided not to use the Best Practices evidence quality categories and instead reported on the study design categories from the Guide to Community Preventive Services (HHS 2001).

The review team applied the CDC Best Practices framework to individual papers, but the framework was not especially useful for identifying papers for inclusion. Few papers had the majority of the information in the framework, especially under Feasibility, Sustainability, and Transferability. The team included as much information about Effectiveness and Reach as papers reported. However, under the Effectiveness category, the majority of papers did not report the magnitude of the effects; the significance of the intervention and results; the benefits or risks of the intervention for the intended population/group; or how the intervention contributed to health equity. Under the Reach category, papers typically did not discuss how many people the researchers intended to reach or could have reached with the intervention. The review template is included as Tables C-5, C-6, and C-7 in the hope that it will stimulate discussion about future intervention design and reporting. For the current review, the Best Practices framework categories appear better suited to be guides to what community health literacy interventions should consider and report than helpful pointers to evaluate what has already been published.