The Health Literacy Skills Framework

Abstract

Although there are a variety of models and frameworks that describe factors that are associated with health literacy skills, few illustrate the full pathway from development and moderators of health literacy skills, their application, and the outcomes that result all in one framework or model. This article introduces the Health Literacy Skills conceptual framework that does encompass this full continuum. To develop the framework, the authors reviewed and built upon existing health literacy frameworks. The Health Literacy Skills framework hypothesizes the relations between health literacy and health-related outcomes and depicts how health literacy functions at the level of the individual. The framework also reflects how factors external to the individual (e.g., family, setting, community, culture, and media) influence the constructs and relations represented in the framework. The framework is organized into 4 primary components: (a) factors that influence the development and use of health literacy skills; (b) health-related stimuli; (c) health literacy skills needed to comprehend the stimulus and perform the task; and (d) mediators between health literacy and health outcomes. Previous theoretical frameworks lend support to the proposed causal pathways it illustrates. The authors hope this conceptual framework can serve as a springboard for further discussion and advancement in operationalizing this complex construct. The Health Literacy Skills framework could also be used to guide the development of interventions to improve health literacy. Future research should be conducted to fully test the relations in the framework.

Health literacy is on the public health agenda. The goal to “improve the health literacy of the population” was included as an objective in Healthy People 2010 and 2020 Objectives. In 2004, the Institute of Medicine released Health Literacy: A Prescription to End Confusion, which recommended that “the Department of Health and Human Services and other government and private funders should support research leading to the development of causal models explaining the relationship among health literacy, the education system, the health system, and relevant social and cultural systems” (Nielsen-Bohlman, Panzer & Kindig, 2004, p. 55). The U.S. Department of Health and Human Services’ 2010 National Action Plan to Improve Health Literacy reinforces the need for conceptual advances in the field by calling for the development and implementation of health literacy interventions on the basis of theories and models, drawing from such related disciplines as communication, education, cognitive science, and medical sociology (p. 44).

Pleasant, McKinney, and Rickard (2011) indicated that there is a lack of true theoretical frameworks that explain health literacy. Theory should be the foundation for developing reliable and valid measures of health literacy, which will allow the field to better study and understand the role of health literacy in health behavior change. This lack of theory has caused researchers to define health literacy in many different ways (Peerson & Saunders, 2009) and, thus, studies vary significantly depending on the definition and measures used (DeWalt, Berkman, Sheridan, Lohr, & Pignone, 2004; Peerson & Saunders, 2009; Pignone, DeWalt, Sheridan, Berkman, & Lohr, 2005).

The absence of a common definition and understanding of health literacy may have slowed the field's progress in developing measures and conducting solid methodological research. There are a variety of models and frameworks that describe factors associated with health literacy skills (Baker, 2006; Mancuso, 2008; Nutbeam, 2000; Paasche-Orlow & Wolf, 2007; von Wagner, Steptoe, Wolf & Wardle, 2009). A comprehensive health literacy theory or framework may spur more professional discussions to help lay the foundation for a new era of theory-driven research.

The majority of these frameworks illustrate the effects that health literacy has on health-related outcomes (Lee, Arozullah, & Cho, 2004; Manganello, 2008; Nutbeam, 2000; Paasche-Orlow & Wolf, 2007; Rootman et al., 2002; Schillenger, 2001; von Wagner et al., 2009). However, few illustrate the full continuum of relations among predictors, moderators, mediators, and outcomes of health literacy all in one theory or framework (Pleasant, 2011).

Purpose

The purpose of this article is to introduce a framework for conceptualizing health literacy that builds on existing theoretical frameworks. We undertake three activities in the article:

1.	First, we review and synthesize the several existing theoretical frameworks for health literacy and describe their strengths and weaknesses.
2.	Second, we describe how we have incorporated key concepts and constructs from existing frameworks to develop the Health Literacy Skills (HLS) conceptual framework.
3.	Third, we suggest how the HLS conceptual framework could be further tested and used to guide future research, evaluation, and intervention development efforts.

Identifying Key Constructs: A Review of Theories on Health Literacy Skills

The Institute of Medicine (Nielsen-Bohlman et al., 2004) definition is pointed to most consistently and considers health literacy to represent a “constellation of skills” necessary to function effectively in the health care environment and act on health care information. These skills comprehensively include the ability to interpret documents and read and write prose (print literacy), use quantitative information (numeracy or quantitative literacy), and speak and listen effectively (oral literacy) (Berkman, Davis, & McCormack, 2010).

Not all theoretical frameworks of health literacy embrace the Institute of Medicine's definition of health literacy as a skill or set of skills. For example, Sørensen and colleagues (2012) recently conducted a review of 17 definitions of health literacy and developed a new definition that “captures the essence” of these definitions found in the literature. Sørensen and colleagues’ definition states the following:

Health literacy is linked to literacy and entails people's knowledge, motivation, and competence to access, understand, appraise and apply health information in order to make judgments and take decisions in everyday life concerning healthcare, disease prevention, and health promotion to maintain or improve quality of life during the life course (Sørensen et al., 2012, p. 3).

Although their definition includes skills, it also includes concepts such as knowledge and motivation, which some researchers consider to be separate constructs from healthy literacy (Baker, 2006; Lee et al., 2004; Paasche-Orlow & Wolf, 2007; von Wagner et al., 2008). Because of this variation in conceptualizations and definitions of health literacy, we have limited our review of health literacy frameworks to those which focus on health literacy as a skill or set of skills.

The purpose of our review was to identify key constructs that were common across theoretical frameworks, determine which constructs were identified as influencing the development of health literacy skills (moderators), and which constructs were depicted as explaining why and how health literacy affected outcomes (mediators) (Kraemer, Wilson, Fairburn, & Agras, 2002). Moderators are variables that affect the direction and/or the strength of the relation between an independent variable and a dependent variable (Baron & Kenny, 1986). Mediators are variables that explain why (e.g., the mechanism through which) specific outcomes or effects occur (Baron & Kenny, 1986; Kraemer et al., 2002). Figure 1 depicts a general framework of health literacy skills. We use this general framework to describe where constructs related to the acquisition and utilization of health literacy skills are located in the health literacy theories and models we reviewed.

Figure 1  Framework for identifying constructs and their influence.

We found that researchers tended to either identify and detail constructs in the first part of the framework (i.e., constructs that moderate the development of health literacy skills) or in the second part of the framework (i.e., constructs that mediate the effects of health literacy on health outcomes). The theoretical frameworks rarely addressed the full continuum. For example, Baker (2006) and Mancuso (2008) place more focus on factors that influence the development of health literacy. In contrast, Paasche-Orlow and Wolf (2007) and von Wagner, Steptoe, Wolf, and Wardle (2009) focus on pathways between health literacy and health outcomes. Sørensen and colleagues’ (2012) recent review resulted in their development of a conceptual framework that acknowledges some general categories of determinants of health literacy (e.g., society, environmental, personal, and situational) and general pathways between health literacy and population level outcomes; however, it does not specify concepts that mediate the relation between health literacy and outcomes.

Specific Concepts Presented in Theories on Health Literacy Skills

In Table 1, we summarize and compare key characteristics and constructs from 10 different theories about health literacy skills and identify concepts from each that have informed the development of the HLS. Table 1 also indicates if the framework treats health literacy skills as dynamic or static, the definition of health literacy used by the authors and key contributions, strengths, and limitations of the framework. Next, we describe the key concepts found in our review of the different theoretical frameworks. Several concepts can be found in multiple theoretical frameworks and are subsequently described.

Table 1. Individual-level models of health literacy: A summary and comparison of key concepts from health literacy models, identifying the key concepts that have informed the development of the RTI Health Literacy Skills Conceptual Model (RTI HLSCM)

Framework Reference Model level Health literacy (HL): Dynamic/Static	How model defines health literacy	Key concepts included	Key contributions and strengths/limitations of model	Key concepts retained in RTI HLSCM
1. Baker (2006)  Individual-level modelDynamic	Presents Institute of Medicine's (IOM) and Healthy People's 2010 definition: “The degree to which individuals have the capacity to obtain, process, and understand basic health information and services needed to make appropriate health decisions.”States that this definition is static and presents framework that is dynamic (individual and health care encounter).Components of HL:▪ Print (ability to understand written health information).▪ Oral (ability to orally communicate about health).	Individual capacity (defined as reading fluency).Prior knowledge (includes vocabulary and conceptual knowledge of health and health care) affects capacity and health literacy.Potential mediators/moderators:culture and norms, barriers to change (new knowledge, positive attitudes, greater self-efficacy, behavior change).Outcomes: improved health outcomes.	Views underlying prior knowledge as a resource that affects HL, not as part of the definition.Strengths:▪ Identifies mediators/ moderators that may influence outcomes.▪ Emphasizes the role of prior knowledge and recognizes conceptual knowledge as well as vocabulary.Limitations:▪ Does not identify specific health outcomes.	▪ Prior knowledge.▪ Communication as component of HL.▪ Culture and social norms (from family, community, media, etc.) as factors that influence the relationship between health literacy and health outcomes.▪ Individual-level mediators: attitude, self-efficacy.
2 Lee, Arozullah, & Cho (2004)Individual-level modelDynamic	IOM definition: “the capacity of individuals to obtain, process, and understand basic health information and services needed to make appropriate health decisions.”	Theoretical framework of how HL affects outcome—through intermediate factors: depicts “net” effects of HL and intermediate variables.Mechanisms or intermediate factors linking HL to health outcomes (considered intercorrelated):▪ Disease & self-care knowledge.▪ Health risk behavior.▪ Preventive care & physician visits.▪ Compliance with medications. Outcomes:▪ Health status.▪ Emergency care.▪ Hospitalization. Moderators/control variables:▪ SES.▪ Gender.▪ Ethnicity.▪ Health insurance coverage.▪ Disease severity.▪ Income discrepancy. Ethnic composition of community	Strengths:Presents testable relationships based on framework.Limitations:▪ Does not directly consider individual's motivation, self-efficacy, or attitude.▪ Does not consider provider-level, system-level, or societal-level factors.▪ While acknowledging its importance, does not include social support as a potential mediator of the relationship between HL and health outcomes.	Differences between Lee et al. framework and RTI framework:Lee: knowledge of health and disease is a mediator between HL on health outcomes;RTI: knowledge contributes to one's HL level and is a result of applying HL skills to stimuli.Health outcomes also influence control variables.Describes environmental variables as influencing health literacy skills development, mediators, and health outcomes.
3. Mancuso (2008) Individual-level modelDynamic	Reviews and critiques various definitions of HL but does not totally adopt any.	Six dimensions of competence that are antecedents of HL:1. Operational.2. Interactive.3. Autonomous.4. Informational.5. Contextual.6. Cultural.Attributes of HL:▪ Capacity: individual skills in information processing, oral language, social skills, and others.▪ Comprehension: understanding information.▪ Communication: reading, writing, speaking, understanding, listening, and observing.Consequences of poor HL: increased costs, less knowledge of diseases and treatments, fewer self-management skills, poorer compliance, more errors, poor ability to negotiate and access the health care system, poorer health outcomes.	Uses concept/dimensional analysis to conceptualize the antecedents, attributes, and consequences of HL. Positions HL within the context of individual and society and identifies the interaction between the six competencies and the three attributes.Limitations: does not identify the pathway between antecedents/attributes and outcomes; does not distinguish between long term and short term outcomes; does not identify potential mediators between HL and health outcomes.	▪ Communication as a component of health literacy.
4. Manganello (2008)Individual-level modelStatic	Refers to IOM definition: “the capacity of individuals to obtain, process, and understand basic health information and services needed to make appropriate health decisions.”Framework specifies skills related to HL:Levels from Nutbeam (2000):▪ Functional.▪ Interactive.▪ Critical.Adds:▪ Media literacy.	Conceptual framework of adolescent HL.Individual traits (affect HL):▪ Age, race, gender, language, culture, education.▪ Social skills.▪ Cognitive skills.▪ Physical abilities.▪ Media use.HL (affect health outcomes):▪ Functional.▪ Interactive.▪ Critical.▪ Media literacy.Mediators:▪ Family and peer influences.▪ Mass media.▪ Education system.▪ Health system.Health outcomes▪ Health behavior.▪ Health costs.▪ Health-service use.	Strengths:▪ Considers HL based on the ecological model: incorporates both individual and environmental influences on HL and health outcomes.▪ Explains relationship between HL and health outcomes in specific subpopulation (adolescents), and recognizes that the structure of a framework explaining the relationship may vary by population.Limitations:▪ Does not address motivation, physician-patient interaction, and other psychological considerations.	Individual traits:▪ Culture.▪ Cognitive skills.▪ Physical abilities. Mediators:▪ Family.▪ Mass media.▪ Health care system.The influence of societal-level variables on all sections of the model.
5. Nutbeam (2000) Individual-level model, but identifies social and political goalsDynamic: HL can be improved through educational programs	World Health Organization's definition (Nutbeam, 1998): “The personal, cognitive and social skills which determine the ability of individuals to gain access to, understand, and use information to promote and maintain good health.”	Identifies 3 progressive levels of HL (from Freebody & Luke, 1990):1. Basic/Functional: sufficient basic skills in reading and writing to be able to function effectively in everyday situations. Interventions should focus on the educational goal of communicating information.2. Communicative/Interactive: more advanced cognitive, literacy, and social skills used to actively participate in everyday activities, to extract information and derive meaning from different forms of communication, and to apply new information to changing circumstances. Interventions should focus on the educational goal of developing personal skills.3. Critical literacy: more advanced cognitive skills which, together with social skills, can be applied to critically analyze information, and to use this information to exert greater control over life events and situations. Interventions should focus on the educational goal of personal and community empowerment.Outcome model categorization (model includes measures for each category):Health promotion actions: education, social mobilization, advocacyHealth promotion outcomes (intervention impact measures): health literacy, social action and influence, healthy public policy and organizational practiceIntermediate health outcomes (modifiable determinant of health): healthy lifestyle, effective health services, healthy environmentHealth and social outcomes	Strengths:▪ Creates a multidimensional conceptualization of HL, goes beyond functional literacy to Integrate concepts of interactive and critical literacy into HL.▪ Places health education and communication into the wider context of health promotion, and highlights HL as a key outcome from health education.▪ Recognizes how social context and environment influence health behaviors, which in turn, affect health outcomes.▪ Identifies both individual and community/social benefit outcomes from each of the three HL levels.Limitations:▪ Does not cleanly and separately distinguish concepts of knowledge, skills, motivation, and access (empowerment).▪ Limited consideration of other individual-level factors.	▪ Influence of environmental factors on health outcomes.▪ Health behaviors related to lifestyle, but considered an outcome rather than a mediator in RTI's framework.▪ Morbidity and mortality as health outcomes.
6. Paasche-Orlow & Wolf (2007)Individual-level model and identifies system-level factorsStatic in model, but authors discuss HL as dynamic (changes over time, depends on context: complexity of tasks, attributes of health care system)	IOM definition but adds the emphasis that HL must be examined in the context of the specific tasks that need to be accomplished (context specific.)	Influences on individual's HL: sociodemographic variables (race, ethnicity, education, age, occupation, employment, income, social support, culture, language), capabilities (vision, hearing, verbal ability, memory, reasoning).Mediator/moderators presented as falling into three domains:▪ Access and utilization of health care (patient factors including navigation skills, self-efficacy, perceived barriers & system factors (complexity, acute care orientation, tiered delivery model).▪ Provider-patient interaction (patient factors including knowledge, beliefs, participation in decision making & provider factors including communication skills, teaching ability, time, and patient-centered care).▪ Self care (patient factors including motivation, problem solving, self-efficacy, knowledge/skills, & extrinsic factors (support technologies, mass media, health education, resources).Health outcome: no specific outcomes specified	Considers not only patient-level characteristics, but also characteristics of the health care system as component-cause mechanisms of the relationship between HL and health outcomes.Strengths:▪ Causal model focusing on pathways between HL and health outcomes.▪ Presents factors that could explain the association between HL and health outcomes.Limitations:▪ Model does not address inappropriate use of services such as overuse of the emergency room, level of patient activation, patient health behaviors.▪ Framework is unidirectional.	▪ Sociodemographic variables.▪ Individual capabilities.▪ Navigation skills (but included as a component of HL skills in RTI framework, rather than a mediator of relationship between HL and health outcomes).▪ Mediators of the relationship between HL and health outcomes:– Individual/ patient mediators (motivation, self-efficacy).– Health care provider.– Health care system. – Media.
7. Rootman et al. (2002)Individual-level modelStatic (focus is generally on literacy)	No definition offered.	Conceptual model.Actions (including policy, community development) and determinants (living & working conditions, socioeconomic status, education, personal assets) affect literacy.Literacy:▪ General literacy (reading ability, numeracy, judgment, critical thinking, interpretation of evidence, communication, and negotiation skills).▪ Health literacy (knowledge of health, ability to seek health info, ability to interpret health info, knowledge and ability to seek health care, ability to understand and give consent, and ability to understand “risk.”▪ Other literacy (political, economic, etc.).Effects of Literacy:▪ Direct: medication use, compliance, etc. → health status → quality of life.▪ Indirect: use of services, lifestyles, income, safety practices, work environment, access to health info, stress level → health status → quality of life.	Strength: comprehensively and separately considers literacy and HL skills, but includes them both in their model.	Health information-seeking as a HL skill.
8. Schillenger (2001)	“Functional” HL: measure of a person's capacity to function in the health care setting as determined by literacy (comprehension of written health care materials) and numeracy (ability to understand and act on numerical health care instructions).Person may also have difficulties processing oral communication.	Framework for association between functional HL and chronic disease outcomes.Being a patient with low functional HL is related to:Ineffective visit-based clinician-patient communication:▪ Poor understanding of disease process.▪ Poor recall/comprehension of advice and instructions.▪ Passive communication.▪ Nondisclosure of functional HL problem.▪ Unorthodox health beliefsAndIneffective home-based monitoring and disease management support:▪ Poor adherence.▪ Inadequate self-care.▪ Poor self-management.▪ Poor problem-solving skills.▪ Low self-efficacy.Outcomes:▪ Worse clinical outcomes.▪ Worse functional outcomes.▪ Higher utilization of services.	Model conceptualizes how worse outcomes among those with chronic diseases are because of visit-based and/or home-based care mediators such as self-efficacy, lack of comprehension or ability to perform self-care, or inability to correctly interpret or act on results.Limitations:▪ Does not consider social support.▪ Does not integrate into health outcomes model, health system, or physician/provider factors. The latter are included in a separate model.	▪ Key role of patient-provider communication in model, but included as an outcome in RTI framework, rather than as a mediator. ▪ Self-efficacy as a mediator of the relationship between health literacy and health outcomes.
9. Sørensen et al. (2012)Model incorporates individual- and population-level componentsDynamic	Health literacy is linked to literacy and entails people's knowledge, motivation and competences to access, understand, appraise, and apply health information in order to make judgments and take decisions in everyday life concerning healthcare, disease prevention and health promotion to maintain or improve quality of life during the life course.	Antecedents of health literacy include societal, environmental, situational, and personal determinants.Dimensions of health literacy include the abilities to access, understand, appraise, and apply health information, which are affected by knowledge, competence, and motivation. These competencies allow for the navigation of three health-related domains: the healthcare setting, disease prevention, and community health promotion efforts.The consequences of health literacy, defined at the population level, include health service use and health costs, health behavior and health outcomes, participation and empowerment, as well as equity and sustainability.	Strengths:▪ Offers a conceptual model based on a comprehensive review of existing models of health literacy▪ Provides a broad perspective of health literacy across health-related domainsLimitations:▪ Does not specify pathways at the individual level▪ Does not consider ecological or environmental influences on the health literacy process▪ Does not include individual-level mediators of the effects of health literacy on health-related outcomes	▪ The use and application of health literacy skills across a range of health contexts▪ The influence of situational and individual determinants on the development and use of health literacy skills
10. von Wagner, Steptoe, Wolf, & Wardle (2008)Individual-level modelDynamic	Presents IOM and Healthy People 2010 definition”: “The degree to which individuals have the capacity to obtain, process, and understand basic health information and services needed to make appropriate health decisions”Authors’ definition: HL is a combination of cognitive skills, knowledge, and experience attained throughout the life span. The relationship between HL and health action is mediated by at least 2 processes: motivation based and skill based.	Theoretical basis for model: Unlikely that HL has direct effects on most health outcomes; likely to depend on a range of mediating factors, called health actions (actions to promote health, prevent disease, comply with diagnosis and treatment)Uses constructs from social cognition models of health to integrate HL into a wider framework of health actions.Epidemiological or structural determinants (of reading and math skills and resulting HL):▪ Individual influences: Cognitive abilities, Age-related cognitive decline; and Knowledge▪ External influences: Environmental influences; Formal educational opportunities; Experiential learningSociocognitive or psychological determinants (mediators of HL, including motivational constructs that affect the performance of health outcomes):▪ Motivational phase: knowledge and understanding; affecting beliefs and attitudes.▪ System factors: health care costs; accessibility of health information.▪ Volitional phase or action control: implementation skills, including task-specific skills.Actions based on sociocognitive or psychological determinants:▪ Access and use of health care.▪ Patient-provider interaction.▪ Management of health and illness.	▪ Presents theory for role of HL on health outcomes based on social cognition models of processing.▪ Builds on framework by Paasche-Orlow & Wolf (2007), adding additional explanation.▪ Framework described as having been tested retrospectively (i.e., applied to earlier studies), but not prospectively.▪ Applied to shared decision-making (consent comprehension), access, and use of primary prevention services (recommendations for screening) and adherence to mediation (management of chronic disease).Limitations:▪ Does not include cultural or media influences.	▪ Quality of the patient-provider interaction as an outcome.▪ Knowledge as a moderator of health literacy skills.▪ Psychological determinants, such as beliefs, attitudes, knowledge and decision making as mediators.

Communication

Communication is included within different parts of health literacy frameworks. Some include communication as a component of health literacy (Baker, 2006; Mancuso, 2008; Manganello, 2008; Nutbeam, 2000). The Institute of Medicine includes oral literacy—listening and speaking skills—as a component of the general literacy skills that are necessary for health literacy and is essentially the same as communication skills. Rootman and colleagues (2002) include communication as a factor that influences the development of health literacy, and as a component of general literacy. Paasche-Orlow and Wolf (2007) include communication within the patient-provider interaction as a mediator of the relation between health literacy and health outcomes. The quality and effectiveness of patient-clinician communication is also an important mediator of the relation between health literacy and health outcomes in Schillinger's (2001) framework.

Knowledge

Health-related knowledge is another concept that appears in various locations within health literacy frameworks. The Institute of Medicine includes cultural and conceptual knowledge (e.g., “an understanding of health and illness and a conceptualization of risks and benefits”) as components of literacy and health literacy (Nielsen-Bohlman et al., 2004, p. 37). Nutbeam (2000), Rootman and colleagues (2002), and Sørensen and colleagues (2012) also include knowledge as a component of health literacy, yet still consider health literacy to be a set of skills. However, other frameworks consider knowledge as a factor that influences the development of health literacy skills. Baker (2006) uses the term “prior knowledge” to indicate that it is a factor that contributes to the development of health literacy and suggests that prior knowledge consists of vocabulary and conceptual knowledge of health and health care. In their framework, von Wagner and colleagues (2009) include knowledge in two places: as a factor that influences health literacy and as a mediator of the relation between health literacy and health actions (which affect outcomes). Paasche-Orlow and Wolf (2007) include a patient's knowledge as a factor that affects the patient-provider interaction, which in turn influences health outcomes. Lee and colleagues (2004) include knowledge of disease and self-care as a mediator of the relation between health literacy and health outcomes.

Health Outcomes

How the frameworks specify health outcomes is another key difference among them. Some frameworks identify no specific health outcomes (Baker, 2006; Paasche-Orlow & Wolf, 2007), while others include multiple levels of health outcomes. Whereas some frameworks include morbidity and mortality as outcomes (Nutbeam, 2000), others only consider mediators such as health behaviors and use of health services as outcomes (Sørensen et al., 2012; von Wagner et al., 2009). However, some frameworks include both types of outcomes (Lee et al., 2004; Manganello, 2008; Rootman et al., 2002; Schillinger, 2001; Sørensen et al., 2012).

Societal Influences

In addition to individual-level factors, societal influences (e.g., culture, community, and media) are also included in different places in the frameworks. Sometimes they are included as moderators (e.g., Sørensen et al., 2012; von Wagner et al., 2009) and other times they are presented as mediators (e.g., Baker, 2006). Most frameworks demonstrate the effect of societal influences in both parts of the model (Lee et al., 2004; Paasche-Orlow & Wolf, 2007; Mancuso, 2008; Manganello, 2008; Nutbeam, 2000; Rootman et al., 2002; Sørensen et al., 2012). Societal factors likely exert an influence in both areas as they not only affect the extent and development of health literacy skills, but they also influence how they are applied in health care systems and interactions with health care providers. This presents a challenge to disentangling effects and confirming direct causal pathways.

Strengths and Limitations of Existing Frameworks

Each framework provides a unique contribution to developing a theoretical base for the development and influence of health literacy skills on health-related outcomes. While there are constructs that are consistently included in the frameworks we reviewed, how each relates to health literacy and health outcomes varies. The frameworks also vary in terms of what they are trying to explain (e.g., how skills are acquired versus how the application of skills affects health behaviors and other outcomes) and in terms of context as some are void of context and present health literacy skills and health outcomes very generally (Baker, 2006; Lee et al., 2004; Nutbeam, 2000; Rootman et al., 2002; Sørensen et al., 2012), while others present the application of health literacy skills specific to interactions within health care settings (e.g., Paasche-Orlow & Wolf, 2007). To date, most of the health literacy frameworks discussed do not appear to have been tested empirically to determine if the proposed relations, moderators, and mediators are accurate (Sørensen et al., 2012). As some frameworks do not define outcomes, testing these frameworks is difficult. In addition, many frameworks do not clarify how included constructs, such as mass media or patient-centered care, would be operationalized to test the framework.

Development of the HLS Conceptual Framework

Even though a number of frameworks for health literacy exist, there is no widely agreed upon framework for health literacy (Nielsen-Bohlman et al., 2004. We used findings from our review of existing frameworks (Baker, 2006; Mancuso, 2008; Nutbeam, 2000; Paasche-Orlow & Wolf, 2007; Schillinger, 2001; Zarcodoolas et al., 2005) to inform the development of a conceptual framework explaining how individuals acquire and apply health literacy skills and how health behaviors and outcomes are affected by health literacy skills. To create our conceptual framework, we took Paasche-Orlow and Wolf's (2007) invitation to build on their framework and on other earlier frameworks, and we also employed a socioecological perspective (Stokoles, 1992) under the assumption that health literacy is a social and dynamic construction (Pleasant et al., 2009). This perspective recognizes external factors that influence an individual's exposure to and cognitive processing and understanding of health-related information. In addition, this framework describes the relation between health literacy, comprehension of health information, health behaviors, and outcomes and incorporates concepts from health behavior theories by explicitly positioning a set of mediators between comprehension and health behavior.

Sørensen and colleagues (2012) recently reviewed health literacy definitions and models and developed an integrated conceptual model of health literacy that strives to bridge the gap between “medical” and “public health” literacy models. Their model is useful in providing a macro-level view of the domains and contexts in which health literacy operates and both supports and complements the HLS conceptual framework. The HLS framework depicts in more detail how an individual may respond to health-related stimuli by identifying clear pathways that can be empirically tested.

Three frameworks had the greatest influence on the development of the HLS conceptual framework presented in this article. Namely, Paasche-Orlow and Wolf (2007), Manganello (2008), and Baker (2006) each presented frameworks of health literacy that served as a basis for the development of a more comprehensive representation of the constructs that are related to the acquisition and utilization of health literacy skills. Our framework sought to simplify a multitude of complex relations, while also elucidating the nature of causation, with some variables operating as mediators and others as moderators. We hope this framework can serve as a springboard for further discussion and advancement in operationalizing this complex construct.

We developed the HLS conceptual framework to describe factors that influence an individual's development of health literacy skills, how health literacy skills influence comprehension, and how comprehension and a variety of other influencing agents (e.g., community, health care system, media, and family) affect variables that are associated with health-related outcomes. In addition, the framework presents a variety of different types of variables that have been shown to affect health-related outcomes as mediators to health literacy. To establish theoretical linkages between these constructs, we drew from extant literature that lends support to the proposed causal pathways.

The HLS conceptual framework (see Figure 1) hypothesizes the relations between health literacy and health-related outcomes and illustrates how health literacy functions at the level of the individual, while acknowledging that factors external to the individual (e.g., family, setting, community, culture, and media) influence all relations represented in the framework. It is organized into four primary components: (a) factors that influence the development and use of health literacy skills, (b) health-related stimulus, (c) health literacy skills needed to comprehend the stimulus and perform the task, and (d) mediators between health literacy and health outcomes. The underlying assumptions of the framework are as follows:

1.	Health literacy is a multidimensional, dynamic construct that we define as “the degree to which individuals can obtain, process, understand, and communicate about health-related information needed to make informed health decisions” (McCormack et al., 2010). Importantly, this definition is not context dependent—as is the case with many health literacy definitions—thus allowing it to be used across all of the different contexts in which individuals make health-related decisions. Moreover, the four dimensions of health literacy (print literacy, numeracy, communication, and information seeking skills) are defined as separate skills that can be developed, enhanced, refined, and even lost over the course of a lifetime.
2.	Demographics, individual resources, capabilities, and prior knowledge are considered interrelated background factors that affect the degree to which an individual acquires health literacy skills.
3.	In accordance with an ecological perspective, our framework assumes that health-related behaviors and outcomes have multiple levels of influences, including individual-level, system-level, and social-level factors, and that these influences interact across the different levels.

Next, we describe the components of the framework (Figure 2) moving from the left side of the framework to the right. The HLS conceptual framework includes concepts or domains (e.g., demographics, mediators, and behaviors) that will allow researchers to include more specific constructs that can be measured and tested in studies.

Figure 2  The health literacy skills conceptual framework. (Color figure available online.).

Factors That Influence the Development and Use of Health Literacy Skills

Our framework acknowledges the interdependence of various background factors and posits that these factors influence health-related behaviors and outcomes directly and indirectly. We hypothesize that demographic characteristics (e.g., age, race and ethnicity, income, and gender), prior knowledge (e.g., disease and illness experiences, conceptual knowledge of health and health care, and familiarity with health care vocabulary), resources (e.g., employment/occupation, social support, culture, language, literacy, and education), and capabilities (e.g., vision, hearing, verbal ability, memory, and cognitive functioning) influence the degree to which an individual develops, refines, and uses health literacy skills.

As described by Baker (2006), we believe that prior knowledge of the health topic influences the degree to which health literacy skills need to be used to understand a stimulus. For example, someone with more conceptual knowledge of health (e.g., how the body works, how bacteria can cause infection) will find it easier to understand a stimulus that references their current knowledge base. Consequently, prior knowledge influences an individual's ability to develop and utilize their health literacy skills to encode, store, and retrieve information (Lang, 2006).

Health Literacy Skills Needed to Comprehend the Stimuli and Perform the Task

To be able to obtain, process, understand, and communicate about health information, individuals must use a variety of skills. We specify print literacy, numeracy, communication, and information-seeking skills in the framework. Being able to navigate search engines and websites has become increasingly important as 74% of Americans use the Internet and, in 2008, 61% of these Internet users looked for health or medical information online (Fox & Jones, 2009). However, individuals not only need to be able to navigate websites, but to navigate through other stimuli such as print materials (e.g., brochures, fact sheets, and booklets). Sometimes referred to as reading fluency, print literacy is the ability to process written materials and includes the ability to read and understand text (prose literacy) and the ability to locate and use information in documents (document literacy; Baker, 2006). Numeracy skills are defined as the ability to apply arithmetic operations and the use of numerical information in printed materials; it is sometimes referred to as quantitative literacy (Baker, 2006; Rothman, Montori, Cherrington, & Pignone, 2008). As health information is often conveyed orally, especially during medical visits, communication skills are important in obtaining and sharing health information and include the ability to listen, speak, and negotiate. These dimensions of health literacy contribute independently to the overarching construct of health literacy skills.

Demand of Health-Related Stimulus

The HLS conceptual framework incorporates health-related stimuli that people receive in their daily life (e.g., a brochure, a prescription label, and a conversation with a doctor). When individuals encounter stimuli, they select the important parts of the messages to encode. Health literacy skills interact with characteristics of stimuli to influence how well they are encoded, stored, and retrieved, which affects comprehension of the messages, their assimilation into an individual's current knowledge base use, and their use in decision making (Lang, 2006).

Health literacy demand is defined as the complexity and difficulty of a stimulus. According to our framework, after exposure to a stimulus, the health literacy demand of the stimulus interacts with a person's health literacy skills to influence comprehension of the message.

In addition to the sheer number and frequency of stimuli exposure, there are three important message characteristics that need to be considered when determining the health literacy demand of a message: communication channel, message content, and message source (Zarcadoolas, 2010). In terms of the communication channel, health-related stimuli can be transmitted through interpersonal or mediated channels. Within mediated channels, there are also a number of formats to convey health information, such as print materials, audio recordings, video presentations, radio announcements, and so forth.

The content of the message can also vary in terms of language (e.g., plain language versus jargon and complex versus simple messages) and orientation (e.g., health promoting versus disease prevention, costs versus benefits, use of fear appeals, and narrative versus non-narrative). Decisions regarding the orientation of the message and the language used have important implications for comprehension (Rothman, Mano, Bedell, Detweiler, & Salovey, 1999).

Last, the source of the message, or the messenger, also plays a major role in influencing an individual's ability to process health information. In terms of interpersonal communication (e.g., between a doctor and patient) the communication skills of the messenger are critical to an individual's skill in interpreting the message being delivered. In addition, the relation of messenger to recipient can also influence comprehension of the stimuli (e.g., family member, friend, or doctor). Credibility of the messenger has been cited as being crucial in the public's trust of health information and messages (Hesse et al., 2005; Nelson, Hesse & Croyle, 2009).

Comprehension of the Stimuli

Comprehension involves learning what to do as well as how to do it (i.e., skill acquisition (McGuire, 2001)). We conceptualize comprehension of the stimuli or health information as a primary indicator of health literacy that is moderated by the health literacy demand of the stimulus. This framework supports the tenet that although comprehension is essential, it is often not sufficient to affect health outcomes (e.g., behaviors such as exercise, healthy eating, and drug use) and health status (e.g., morbidity, mortality, and health, or well-being) directly. While health literacy skills and message characteristics determine what someone will learn from a stimulus, mediating factors affect whether they will accept or adopt what they learn. Here, the basic assumption is that the effects of a stimulus depend on two factors: comprehension of the message and acceptance of what is learned (McGuire, 1968).

Other Mediating Factors Between Health Literacy and Health Outcomes

Mediating factors affect whether people retain, retrieve, and decide to use the information in the stimulus when making health-related decisions. While research has found that there is a direct relation between an individual's health literacy skills and his or her health outcomes (Berkman et al., 2010), many factors may also mediate this relation including health status, attitudes, emotions, motivation, and self-efficacy, which are further affected by ecological influences (e.g., culture, social support, community health care resources, the media, and access to health care resources including health insurance). Depending on the specific health behavior and outcome of interest, numerous mediators could potentially influence the relation between comprehension of a stimuli and health behaviors (e.g., motivation, attitude toward/perceived effectiveness of the behavior, fatalism, and decision-making skills).

Although it is tempting to want to identify someone as being “health literate” if they demonstrate a desired behavior (e.g., quitting smoking), there are far too many mediating factors that influence whether or not an individual engages in a behavior or has a positive health outcome to be able make this direct association. For example, an individual may understand that smoking can cause lung cancer, but may not have the motivation, health care, or social support to quit. Thus, a smoker may very well comprehend the health message but is not be able to act on it for other reasons. Other types of mediators include social support, decision-making skills, access to health care, trust in information/source/stimulus, fatalism, and perceived relevance of the message.

Health-Related Behaviors and Outcomes

The HLS conceptual framework includes two large categories that represent the application of health literacy skills: health-related behaviors (e.g., smoking, exercise, or medication adherence) and health status or outcomes (e.g., morbidity/mortality, disease state, health care service utilization, quality of life, or health/well-being).

Ecological Moderators

Our framework posits there are a variety of ecological influences (e.g., culture, community resources, family, media, health care system, and health care providers) that can moderate components of the conceptual framework in various ways. Health system-level moderators are generally believed to influence the relation between health literacy and health outcomes, but can also affect health literacy development (e.g., see Manganello, 2008; Paasche-Orlow & Wolf, 2007). Last, societal-level moderators include the environment, cultural differences, and access to different types of media, and can influence health literacy and health outcomes (e.g., see Mancuso, 2008; Manganello, 2008; Nutbeam, 2000).The dotted arrow in Figure 1 represents a feedback loop and shows that health behaviors and health status influence mediators, health literacy skills, and knowledge. This feedback supports the idea that health literacy skills are dynamic, and that as individuals interact with health-related stimuli, skills are learned or unlearned, reinforced, or degraded. In addition, the framework represents how health literacy skills can both develop and deteriorate as individuals’ capabilities such as vision, hearing, and cognitive capacity change over time.

Discussion

Theoretical frameworks that advance our understanding of health literacy and how it relates to health-service use and health outcomes are clearly needed (Pleasant et al., 2011). Theoretical frameworks provide the necessary basis for reliable measurement and the development of interventions to improve health literacy. Yet, current frameworks are limited because they often present only part of the causal pathway (moderators to health literacy skills, or health literacy skills to outcomes) rather than reflect the full continuum that includes both the acquisition and application of health literacy skills. The HLS conceptual framework offers another perspective about conceptualizing the complex interrelations between the myriad of factors that influence and are influenced by health literacy. It is distinct from other models because it illustrates how health literacy skills influence comprehension of a health-related stimuli and that health literacy's effect on health-related behaviors and outcomes is mediated by a variety of constructs (e.g., emotions, perceived norms, motivation, self-efficacy, access to health care, and perceived relevance of the message).

We propose the HLS conceptual framework as a springboard for further exploration of the relations between the various elements of health literacy and to encourage additional research such as

	empirically testing the framework;
	identifying the degree to which demographic variables, resources, capabilities, and prior knowledge affect individuals’ health literacy skills;
	investigating the degree to which health literacy skills influence different mediators and outcomes; and
	determining the pathway(s) through which health literacy skills affect different health-related behaviors directly or indirectly (through certain mediators).

Future studies could apply the conceptual framework to a single health behavior (e.g., diabetes management) or more globally (e.g., health-lifestyle behaviors) to help assess its utility in different contexts. In addition, investigating how experiences with health and health care can influence development, enhancement, and application of health literacy skills is needed.

The conceptual framework can be used to guide the development of interventions to improve the health literacy of individuals. The framework identifies barriers to acquiring health literacy skills (e.g., individual resources such as social support, language, and general literacy; capabilities such as memory, cognitive functioning, vision, and hearing), which can provide insight for health care systems interested in self-assessment. The HLS framework can inform the development of strategies to improve access to health information and navigation of the health care system. For example, if a health care system designed strategies for overcoming certain barriers, the effect of these strategies could be assessed by measuring health literacy skills, mediators, and relevant outcomes before and after the strategies have been implemented.

While the HLS conceptual framework advances current theoretical frameworks of health literacy by including both the determinants and outcomes of health literacy, it is not without limitations. To make the HLS conceptual framework applicable across health topics and behaviors, we did not include the universe of the potential mediators and moderators to health literacy and outcomes. Rather, we included these as conceptual categories that will allow researchers to include the specific mediators, moderators, and outcomes that are relevant to their particular research question. However, this lack of specificity may in fact limit the utility of the framework for some researchers.

In addition, the framework does not address key measurement issues that could affect the assessment of the underlying dimensions of health literacy skills (see McCormack et al., 2010). Using valid and reliable measures of health literacy skills will be vital to testing the conceptual framework. McCormack and colleagues (2010) developed the Health Literacy Skills Instrument, which assesses three domains of health literacy: print literacy, oral health literacy (listening skills only), and Internet information-seeking skills. In terms of content domains, it reflects health-related issues across the life course for health promotion and disease prevention, health care maintenance and treatment, and health system navigation. The Health Literacy Skills Instrument can be self-administered via a computer, which can reduce data collection costs and minimize potential discomfort or embarrassment among participants. The instrument can be accessed at http://www.rti.org/page.cfm?objectid=66F893E4-5056-B100-OC834F234F368198. Measuring prior knowledge will also be important in testing the HLS conceptual framework. For example, if a study or intervention focuses on cardiovascular disease, specific knowledge of this health issue should be assessed. To fully test the HLS conceptual framework in the general population, individuals’ broader conceptual knowledge of health and science should be assessed.

Research to more thoroughly investigate how age and culture influence the acquisition and application of health literacy skills will also advance the field by helping to identify where interventions can be most effective. For example, studies can be designed to determine whether community-based or system-based interventions that address issues related to the acquisition of health literacy skills are more effective than interventions that focus on enhancing individuals’ use of health literacy skills (e.g., interventions that focus on enhancing self-efficacy). Last, learning how to develop and/or adapt educational materials and messages so that those with different levels of health literacy skills can use the health information to make informed health decisions is paramount to addressing health disparities in our nation. The HLS conceptual framework is a first step toward developing a framework that can guide these efforts.