The overall structure of the ECHC framework (figure 1) is drawn from social ecology, which posits that factors outside the individual influence his or her behavior (Bronfenbrenner 1979). That is, individuals are nested within groups, which are embedded in the broader community. We selected this multilevel structure because consumer engagement interventions can take group or population approaches, and contextual factors influence individual behavior and the success of interventions. The ECHC's structure of concentric circles shows that individuals, groups, and communities may influence one another at any point in time, but not necessarily in a linear fashion or with equal importance.
The first level is concerned with individuals’ engagement in their health and health care and is the most detailed because, as we noted earlier, consumer engagement is fundamentally about individual behaviors. Although groups and communities can have unique identities and act collectively, most consumer engagement is ultimately acted out by individuals (Hibbard 2009a). The surrounding group level focuses on individuals’ interactions with and influences on one another (i.e., interpersonal relationships and behaviors). Thus, in the ECHC framework, the term “group” refers to relationships among two or more individuals whose relationship may be based, for example, on friendship, family, religion, ethnicity, exchange of services, or employment.1 The community level represents the larger geographic space, consisting of the physical, economic, sociocultural, and political structures in which individuals (and the groups with which they identify) navigate (Buttimer 1969; Hunter 1975). Although some literature uses the term “community” to describe a large group of people with a common attribute (e.g., the breast cancer community), the ECHC framework considers these collections of people as groups. Finally, at the bottom of the framework we list those characteristics identified in the literature as likely influencing consumer engagement within and across levels.
Next we examine more closely each of the framework's three levels and the relationships within and among them. Our goal was to develop a framework for classifying consumer engagement interventions, specifically interventions sponsored by or put in place by a community's multistakeholder entities, to facilitate dialogue about their logic and role. Here, however, we cannot present an exhaustive set of specific hypotheses about how any or all of a level's characteristics could impact an individual's engagement.
At the center of the ECHC framework is an adapted version of the transtheoretical model of individual behavior change (Prochaska 2008; Prochaska et al. 1994). The transtheoretical model is not, however, a theory but a model that reflects a common, but not universal (e.g., see Herzog 2010), conceptualization of behavior change as an individual's progression through a series of stages and processes that can be targeted by interventions (Prochaska 2008; Prochaska et al. 1994). For example, in the precontemplation and contemplation stages, a key process may be becoming increasingly aware of how a particular behavior affects an individual's health and well-being (Prochaska et al. 1994). Because of its widespread use in the literature (Glanz, Rimer, and Viswanath 2008; Painter et al. 2008) and its intuitive, staged approach (Munro et al. 2007; Prochaska, Redding, and Evers 2008), we chose this model as a starting point, modifying it to incorporate the role of activation and to emphasize the dynamic nature and varied time horizon of change.
The original transtheoretical model consists of five stages of change: (1) precontemplation (no change intended within six months), (2) contemplation (change intended within next six months), (3) preparation (change intended within a month), (4) action (already has made changes), and (5) maintenance (sustaining change beyond six months) (Prochaska et al. 1994); the time frames reflect the original focus on using the model to design interventions to guide the planned behavior change. The ECHC framework specifies activation as a distinct characteristic that influences an individual's decision-making process and actions taken to change his or her behavior. This decision-making process has four stages: (1) recognizing one's role in health and health care, (2) gathering information, (3) weighing the options, and (4) engaging in healthy, self-management, shopping, or health care encounter behaviors. In figure 1, the two core conceptual dimensions of consumer engagement discussed earlier in this article—activation and one or more of the four behavior types—are denoted in boldface. In addition, we eliminated any reference to specific time frames and added bidirectional links to better reflect the dynamic nature of decision making and behaviors over time. For various reasons, individuals may implicitly or explicitly (re)consider their behaviors.
To improve activation and the performance of engaged behaviors, individuals must be aware of, acknowledge, and embrace the idea that there is room to improve their own health and that they have an important, active role in doing so. This step primarily concerns individuals’ self-perception of their role and ability to be engaged in their health and health care (Bandura 1986, 2004; Prochaska et al. 1994; Swann and Bosson 2010; Terry 2005). Recognizing their role provides motivation to pursue behavior change, and how individuals perceive this role affects their motivation to pursue behavior change, including the scope and types of information sought. Individuals’ effort in looking for information and where they seek it can vary greatly and include searching the Internet, asking friends or family for advice, reading pamphlets in their provider's office or at the pharmacy, and talking to their provider about their health and health care. The information gathered and how it is interpreted can further shape how an individual perceives his or her role.
The next step is weighing the options, or understanding, assessing, and synthesizing information to evaluate the direct or indirect benefits and costs of engagement alternatives. In economics, this process is known as estimating trade-offs to maximize utility, or assessing the “opportunity costs” when choosing one's behavior (Culyer and Newhouse 2000). In the transtheoretical model of behavioral change, this concept is referred to as individuals’ “decisional balance sheet” of pros and cons (Prochaska et al. 1994). Weighing the options may be executed as an intrinsic, unconscious process or as one that is more explicit and deliberate. If individuals conclude that the benefits of change sufficiently outweigh the costs, they will be expected to perform and maintain the engaged behaviors. Individuals’ choosing to take (or not to take) action may affect their health and well-being over time. Finally, the framework indicates that an individual's degree of activation is associated with the behavior change and that his or her degree of activation can change over time (Fowles et al. 2009; Greene and Hibbard 2011; Harvey et al. 2012; Hibbard 2009b; Hibbard et al. 2007a; Mosen et al. 2007; Remmers et al. 2009).
The ECHC framework assumes that individuals may progress (or regress) among these stages at different rates. Individuals have different capacities, and different constraints on their capacities, that affect motivation and movement. For example, a diagnosis of diabetes may prompt individuals to rethink their role in managing their own health and instigate information gathering about diabetes-specific management recommendations. However, these same people may not view themselves as active “shoppers” for high-quality providers and may not look for information about their provider's record on diabetes care. Or they may not have “real” choices because of insurance restrictions or because other providers are not taking new patients. Since we designed our framework to accommodate these variations, defined time periods are not assigned to a stage.
In addition, although we presented our framework linearly, we recognize that the consumer engagement process is dynamic (Hibbard and Mahoney 2010; Prochaska 2008). The bidirectional arrows indicate the potential to progress or regress in linear and nonlinear ways over time. For example, an individual who exercises regularly may move from taking action to gathering information to understanding better why he or she has not seen a change in weight or decrease in blood pressure. In another case, an individual may believe that the costs associated with using performance reports and switching physicians are prohibitively high but may revisit this calculus when he or she moves to another town or his or her physician retires.
While the process is dynamic, individuals may pause at any point. For instance, individuals may gather information about one of their providers or possible treatments but stall and do nothing because they find the information hard to understand and believe that the additional effort to comprehend the information (e.g., time, potential embarrassment in seeking help, likely relevance) outweighs expected benefits. Or an individual who did not lose weight despite exercising may not return to gather more information but move back to role recognition because this feedback altered his or her self-conception as a confident, apt manager of his or her own health. One factor affecting progress is the source of an individual's motivation to change. Individual change that is more intrinsically than extrinsically motivated is thought to be more powerful, especially when people are experiencing stress, because the intention and motivation is grounded in their own values and interests, not in others’ (Anderson and Funnell 2010; Ryan and Deci 2000).
As indicated by our examples, static traits like age, gender, personality, cognitive abilities, race, and ethnicity and more variable states such as health status, income, socioeconomic status, self-efficacy, emotions, experiences, self-conception, degree of self-regulation, knowledge, awareness of choices, skills, and beliefs and values can influence the engagement process and degree of engagement (Hibbard and Mahoney 2010; IOM 2001; Protheroe, Nutbeam, and Rowlands 2009; Rosen, Anell, and Hjortsberg 2001; Schumaker, Ockene, and Riekert 2009; Swann and Bosson 2010).
Group membership affects access to potentially beneficial information, social support, and material resources (Berkman and Kawachi 2000; Granovetter 1973; Hermann-Pillath 2010). Most individuals are members of social networks and psychologically identify with one or more groups, each of which has its own identity and culture (Charles et al. 2006; Leary 2010; Tajfel and Turner 1979). Group culture, which is made up of the values, attitudes, and beliefs that are learned and transmitted among group members, is expressed through implicit and explicit rules that regulate and reinforce appropriate, socially sanctioned behavior and punish unacceptable displays of values and beliefs (Kreuter and Haughton 2006; Leary 2010; Link and Phalen 2001). This ongoing socialization can impede or enhance consumer engagement, depending on how consumer engagement in health and health care is defined and valued (Swann and Bosson 2010).
For example, some groups of consumers might not trust specific sources of information or might believe it is disrespectful to question providers about treatment (Cutilli 2010; Hesse et al. 2005; Kleinman 1980). Members of these groups also may disregard institutionally sponsored campaign materials encouraging people to ask their providers specific questions about their health care or prevent a family member from asking questions during a visit with a provider. Alternatively, groups that strongly value being “informed” or “not pushed around” may be positively affected by the same campaign efforts, if the campaign affirms these core elements of their identity. Instead of automatically discarding these materials, members of these groups may bring them to their next office visit for easy reference. For their part, providers may adhere to and reinforce nonactivated consumer roles in health care encounters, for example, by repeatedly interrupting patients and outwardly dismissing their concerns, or they may affirm activated roles by soliciting patients’ input in a collaborative manner and making them aware of any options (Dy and Purnell 2012; Emanuel and Emanuel 1992).
A group's influence on a consumer's engagement is likely to be greater if a member strongly identifies with, is attached to, and is embedded in the group; the group has a strong identity; and it has a high degree of cohesiveness and interdependence (Berkman and Kawachi 2000; Friedkin 2004; Lau 1989; Swann and Bosson 2010). For example, Ahern and colleagues (2009) found that the same high degree of collective efficacy was associated with more smoking in neighborhoods having permissive smoking norms and with less smoking in neighborhoods having strong antismoking norms. The forces that prevail may depend on the issue and situation. For example, if a minority woman with diabetes identifies with her ethnic group more strongly than she does as a diabetic, she may hesitate to discuss alternative medicine treatments with her mainstream medical providers (Graham et al. 2005; Shelley et al. 2009), even if others in her diabetes support group encourage this discussion.
Other group-level characteristics that can shape consumer engagement are sociodemographic makeup, integration into mainstream culture, and economic, political, and social2 resources (Fiske 2010; Link and Phalen 2001; Reitz and Sklar 1997; Yzerbyt and Demoulin 2010).
A community's economic, political, social, and physical infrastructures influence (1) the relevance and role of consumer engagement in improving the community's health and health care; (2) the will and ability to take collective action with respect to consumer engagement; and (3) the legitimate, acceptable options for consumer engagement initiatives in the community. Although scholars and practitioners disagree on the relative influence of these factors (Helman 2001; Lynch et al. 2002; McLeroy et al. 1988; Roblin and Becker 2009), powerful individuals and entities can have tremendous influence over community agendas, including how resources are distributed (Fiske 2010; Gradstein and Schiff 2006). Context is important to understanding the opportunities for and characteristics of community consumer engagement initiatives (Hibbard 2009a, 2009b; Larson et al. 2009).
A community's economic infrastructure affects both the total amount of resources available for improving a community's health and the locus of power. For example, if a few large purchasers have significant economic power in a community's health care sector, they may use this power to compel providers to supply performance data for their employees to use in choosing providers or for the employer to use in deciding payments. The purchasers may also pursue this performance-reporting agenda, regardless of the providers’ readiness. In turn, providers may use their social power to persuade patients to maintain their existing provider relationships, despite any purported quality differences.
Likewise, a community's “mainstream culture” represents historically powerful groups’ values and beliefs, such as the role of collectivism versus individualism in community life or the role of folk medicine versus Western medicine. The importance of the mainstream culture is its power to set norms that influence behavior, as observed in the evolution of smoking behavior and the treatment of depression (Kleinman and Hall-Clifford 2009; Stuber, Meyer, and Link 2008). Its influence may vary with the matter being considered and how embedded a group is in the mainstream (Swann and Bosson 2010; Wolff et al. 2010). Finally, geography and physical infrastructure affect consumer engagement by shaping the connections among people, groups, and resources (Sampson, Morenoff, and Gannon-Rowley 2002; Williams and Collins 2001). Geography affects access to resources that can reduce the costs of engagement like social support, healthy food, safe places to exercise, and the accessibility and availability of high-quality providers (Cohen, Inagami, and Finch 2008; Larson, Story, and Nelson 2009; Nemet and Bailey 2000; Stanley, Cantor, and Guarnaccia 2008).