Найдите свой следующий любимый книге

Станьте участником сегодня и читайте бесплатно в течение 30 дней
The Handbook of Health Behavior Change, Fifth Edition

The Handbook of Health Behavior Change, Fifth Edition

Читать отрывок

The Handbook of Health Behavior Change, Fifth Edition

1,417 pages
21 hours
Apr 28, 2018


This revised and updated fifth edition of the highly acclaimed “gold standard” textbook continues to provide a foundational review of health behavior change theories, research methodologies, and intervention strategies across a range of populations, age groups, and health conditions. It examines numerous, complex, and often co-occurring factors that can both positively and negatively influence people’s ability to change behaviors to enhance their health including intrapersonal, interpersonal, sociocultural, environmental, systems, and policy factors, in the context of leading theoretical frameworks. Beyond understanding predictors and barriers to achieving meaningful health behavior change, the Handbook provides an updated review of the evidence base for novel and well-supported behavioral interventions and offers recommendations for future research.

New content includes chapters on Sun Protection, Interventions With the Family System, and the Role of Technology in Behavior Change. Throughout the textbook, updated reviews emphasize mobile health technologies and electronic health data capture and transmission and a focus on implementation science. And the fifth edition, like the previous edition, provides learning objectives to facilitate use by course instructors in health psychology, behavioral medicine, and public health.

The Handbook of Health Behavior Change, Fifth Edition, is a valuable resource for students at the graduate and advanced undergraduate level in the fields of public or population health, medicine, behavioral science, health communications, medical sociology and anthropology, preventive medicine, and health psychology. It also is a great reference for clinical investigators, behavioral and social scientists, and healthcare practitioners who grapple with the challenges of supporting individuals, families, and systems when trying to make impactful health behavior change.


  • Revised and updated to encompass the most current research and empirical evidence in health behavior change
  • Includes new chapters on Sun Protection, Interventions With the Family System, and the Role of Technology in Behavior Change
  • Increased focus on innovations in technology in relation to health behavior change research and interventions


  • The most comprehensive review of behavior change interventions
  • Provides practical, empirically based information and tools for behavior change
  • Focuses on robust behavior theories, multiple contexts of health behaviors, and the role of technology in health behavior change
  • Applicable to a wide variety of courses including public health, behavior change, preventive medicine, and health psychology
  • Organized to facilitate curriculum development and includes tools to assist course instructors, including learning objectives for each chapter
Apr 28, 2018

Связано с The Handbook of Health Behavior Change, Fifth Edition

Похоже на «Книги»
Похожие статьи

Предварительный просмотр книги

The Handbook of Health Behavior Change, Fifth Edition - Springer Publishing Company


Theoretical Models of Health Behavior Change

Theories attempt to explain cause–effect relationships and help to provide a basis for understanding and predicting the occurrence of health-related behaviors, behavior change, and maintenance of change. Therefore, they also help to frame the development of strategies and interventions to effect behavior change and guide health behavior research. No single theory is all encompassing, often making it necessary to use multiple theories to understand how to promote change in specific behaviors.

Chapter 1, Individual Theories by Janevic and Connell provides a detailed review of six commonly used individual-level theories and models: Social Cognitive Theory, Self-Regulation Model, Health Belief Model, Theory of Planned Action, Transtheoretical Model, and Relapse Prevention Model. It concludes by asking researchers and clinicians to consider the need for more rigorous testing and subsequent theory modification to move the field forward and describes how we might do this. The authors call for taking theory testing and development in new directions to help ensure the relevance of individual-level theories in an environment where a strong evidence base and multilevel, multidisciplinary approaches are critical to addressing today’s biggest public health challenges.

Chapter 2, Understanding Population Health From Multilevel and Community-Based Models by Fitzgibbon and colleagues addresses how socioecological models that address the multiple levels of influence on behavior can be used for developing interventions that support healthy lifestyles. The authors note that we must pay attention to the communities in which people live and work, and they discuss the benefits and challenges of implementing community-based participatory research. They also note how the RE-AIM evaluation framework that identifies the importance of assessing reach, effectiveness, adoption, implementation, and maintenance when translating multilevel research to action can be applied to the evaluation of health behavior change trials.

The two chapters in this section emphasize that no one level of the socioecological model that affects behavior acts on its own; rather, each acts within the context of the others. Thus to be able to tackle our public health challenges described throughout this book, behavior change depends on a comprehensive approach for intervening on multiple levels.

Even though there are many theories and models available, there is still much we have not been able to explain about why and how people make changes in health-related behaviors and, even more challenging, how they maintain those changes. To fill in the black box of behavior change we must be able to move outside of it. Theories can be used to guide us but should not constrain our creativity. It is important for us to be open to integrating, expanding, and testing our theories to help us build on our base of knowledge and to apply what we learn from practice to modify and add to our theories. We also must be mindful of the importance of studying how to implement and disseminate our work using implementation science methodology and theories.


Individual Theories



•Describe how individual theories of health behavior change are used to guide both correlational and intervention-based research, and the prominent role that these theories have played in health behavior research over more than four decades.

•Discuss the major components and applications of the most commonly used individual theories, including Social Cognitive Theory (SCT), the Self-Regulation Model, the Health Belief Model (HBM), the Theory of Planned Behavior (TPB), the Transtheoretical Model (TTM), and the Relapse Prevention (RP) Model.

•Describe how individual theories have been applied to the burgeoning field of technology-enabled interventions and present opportunities for innovative approaches to theory development in this area.

•Comment on the current state of individual-level theory development, and describe proposals for making theories more valuable to researchers and practitioners.

The ultimate success of effective preventive and treatment regimens depends on individuals’ willingness to undertake and maintain required behaviors. Given low rates of screening, immunization, and adherence to prescribed medical regimens (Benner et al., 2002; DiMatteo, 2004; Fotheringham & Sawyer, 1995; Kimmel et al., 2007; Osterberg & Blaschke, 2005; Sackett & Snow, 1979), it is not surprising that behavioral scientists have devoted extensive effort to explain and predict individuals’ health-related decisions. This effort has resulted in the development and widespread application of numerous individual-level behavior change theories.

These theories tend to focus on cognitive variables, such as attitudes, beliefs, and expectations, and the factors that influence them. They are rational in that they assume that individuals wish to maximize positive health outcomes. In spite of increasing attention to multilevel frameworks and theories that account for broader influences on health behavior, individual-level theories remain widely used in the research literature (Noar & Mehrotra, 2011; Painter, Borba, Hynes, Mays, & Glanz, 2008). A close examination of how these theories are applied to empirical work reveals considerable variation. At one end of the spectrum, studies are informed, to some degree, by theory; at the other end, studies test, or attempt to build, theory (Michie & Prestwich, 2010; Painter et al., 2008). Recent guidelines for intervention development and evaluation have reiterated the importance of a strong theoretical base (Sniehotta, Presseau, & Araujo-Soares, 2015).

The ongoing popularity of individual theories can be seen in recent trends. First, 20th-century theories are being applied in new ways in the 21st century, as health behavior interventions are increasingly delivered through electronic means (Webb, Joseph, Yardley, & Michie, 2010). Second, a globally connected world and new communication channels are allowing researchers and practitioners to apply individual theories in populations and settings far from those in which they were developed. As lessons learned from this work are shared, we gain insight into how culture and setting affect the generalizability of these theories (King, 2015). Third, use of individual theories has extended well beyond the traditional health behaviors for which most were originally developed to study phenomena as diverse as antimicrobial stewardship (Heid, Knobloch, Schulz, & Safdar, 2016), safety behaviors among Boko Haram victims in Nigeria (Tade & Nwanosike, 2016), and older adults’ driving behavior (Kowalski, Jeznach, & Tuokko, 2014).

The appeal of individual theories lies in their potentially powerful applications to health behavior research and practice. For instance, they can form a blueprint for intervention development and evaluation by helping to identify the key determinants of health behavior to influence and measure. Theories also can be used to predict future health actions and, more generally, to help us sift through the complexity of human health-related behavior. Mitigating the potential value of these theories, however, is the reality of unclear evidence about whether explicit use of theories in intervention development actually enhances intervention efficacy, as well as the utility and aptness of each theory for a given health behavior, population, or setting. In addition, the considerable overlap in constructs or components across theories with varying terminology may make the existing menu of theories overwhelming to researchers and practitioners (Munro, Lewin, Swart, & Volmink, 2007; Noar & Zimmerman, 2005).

The overall goal of this chapter is to offer a snapshot of six commonly used individual health behavior theories: SCT; HBM; TPB; TTM (Stages of Change); and the RP Model. A summary of the origins and primary components is provided for each. We also describe their empirical application and the results of meta-analyses or reviews that lend insight into their value. We make note, where applicable, of critical attention that a theory has received, including noteworthy limitations or recommendations for future directions. In addition, we provide an overview of the growing application of individual theories to eHealth and mHealth interventions. We discuss overlap among theories and useful dimensions on which to compare individual theories. The chapter concludes with a discussion of recent observations on the state of health behavior theory research.


The central feature of SCT is the reciprocal nature of influences that produce behavior. Personal factors, existing behaviors, and the social and physical environments interact and, as a result, shape new behavior. In SCT, several key concepts explain and predict behavior and are based on how individuals learn: incentives, outcome expectations, and efficacy expectations. According to Bandura (1977), a person with given beliefs, information, attitudes, and needs, who is functioning in given social and physical environments, will engage in a behavior that will have a consequent outcome. In this way, behavior change and maintenance of behavior are largely a function of (a) expectations about the outcomes that will result from engaging in behavior and (b) expectations about one’s ability to engage in or execute the behavior. Thus, outcome expectations consist of beliefs about whether given behaviors will lead to given outcomes, whereas efficacy expectations consist of beliefs about how capable one is of performing the behavior that leads to those outcomes. The two constructs have been shown to be linked in predicting, for example, exercise behavior (Hallam & Petosa, 2004) and self-management of diabetes (Iannotti et al., 2006; Martin, Dutton, & Brantley, 2004; Shi, Ostwald, & Wang, 2010).

Both outcome and efficacy expectations reflect a person’s beliefs about capabilities and the connections between behavior and outcome. It is these perceptions, then, and not necessarily true capabilities, that influence behavior. In addition, it is important to understand that the concept of self-efficacy relates to beliefs about capabilities of performing specific behaviors in particular situations (Marks, Allegrante, & Lorig, 2005; Schunk & Carbonari, 1984); self-efficacy does not refer to a personality characteristic or a global trait that operates independently of contextual factors (Bandura, 1986, 1997). This means that individuals’ efficacy expectations will vary greatly depending on the particular task and context that confronts them.

Both outcome and efficacy expectations reflect a person’s beliefs about capabilities and the connections between behavior and outcome. It is these perceptions, then, and not necessarily true capabilities, that influence behavior. In addition, it is important to understand that the concept of self-efficacy relates to the beliefs about capabilities of performing specific behaviors in particular situations.

Bandura (1997, 2002) argued that perceived self-efficacy influences all aspects of behavior, including the acquisition of new behaviors (e.g., a young adult learning how to use a particular contraceptive device), inhibition of existing behaviors (e.g., decreasing or stopping cigarette smoking), and disinhibition of behaviors (e.g., resuming sexual activity after a myocardial infarction). A person’s self-efficacy also affects the amount of effort he or she will expend on a task, and the degree of persistence in the face of obstacles. Finally, self-efficacy affects people’s emotional reactions, such as anxiety and distress. Thus, individuals with low self-efficacy for a particular task may ruminate about their personal deficiencies, rather than thinking about accomplishing or attending to the task at hand; this, in turn, impedes successful performance of the task.

Efficacy expectations derive from four major sources (Bandura, 1986, 1997). The first, termed performance accomplishments, refers to learning through personal experience where one achieves mastery over a difficult task, which increases self-efficacy. Performance accomplishments attained through personal experience are the most potent source of efficacy expectations.

The second source is vicarious experience, which refers to learning that occurs through observation of events or other people, or models. To increase an observer’s self-efficacy, it is important that the model is viewed as having overcome difficulties through determined effort rather than with ease, and that the model is similar to the observer with regard to other characteristics (e.g., age, gender).

Verbal persuasion, or encouragement, constitutes the third source of efficacy expectations. Finally, one’s physiological state provides information that can influence efficacy expectations. For example, people who experience sweaty palms and a racing heartbeat before they are about to give a talk find that their self-efficacy plummets; to someone just beginning an exercise program, fatigue and mild aches and pains may be mistakenly interpreted as a sign that they should not be physically active.

The Hip Hop HEALS, a school-based intervention to improve children’s food choices, provides an example of how SCT constructs are typically operationalized in program design (Williams et al., 2016). In this program, children were given an opportunity for performance accomplishments (mastery) via a calorie balance video game; obtained vicarious experience in adopting healthy eating patterns via characters in animated cartoons and comic books; and received verbal persuasion in the form of encouragement from facilitators. Other recent interventions demonstrate SCT’s versatility; for example, SCT is the most frequently used theory informing the design of electronic games for children with diabetes (Lazem, Webster, Holmes, & Wolf, 2015) and the behavior change techniques used in popular commercially available activity monitors are commonly based on the goal-setting, self-monitoring, and feedback elements of SCT (Lyons, Lewis, Mayrsohn, & Rowland, 2014).

SCT has been most often applied to explaining or promoting lifestyle behaviors. In a review and meta-analysis of SCT-based diet and physical activity interventions for adult cancer survivors, Stacey and colleagues found a significant pooled intervention effect on physical activity, and that most studies were successful in changing at least one aspect of nutrition-related behavior (Stacey, James, Chapman, Courneya, & Lubans, 2015). However, in a caveat that is frequently found in reviews and meta-analyses of studies based on a given theory, the authors noted that SCT constructs were seldom well-described in the reviewed studies, and rarely measured or tested.

In a meta-analysis of observational studies of physical activity, 31% of the overall variance in outcomes was explained by SCT constructs (Young, Plotnikoff, Collins, Callister, & Morgan, 2014). They also found that self-efficacy and goals were more consistently associated with physical activity than outcome expectations or social–environmental factors. They point out that while self-efficacy is the SCT construct with the largest body of evidence supporting its links with behavior, goal setting via self-regulation appears to have similarly strong effects on behavior.


Given the size and complexity of SCT, it is not surprising that there have been theoretical offshoots. Self-regulation, drawn from SCT, has been developed as its own theoretical framework. As described by Zimmerman (1998), self-regulation refers to how individuals develop an understanding of their own behavior and outcome and efficacy expectations.

Leventhal and colleagues (Leventhal, Leventhal, & Contrada, 1998; Leventhal, Meyer, & Gutmann, 1980) were among the first to apply the self-regulation construct to health behavior. They contend that people function as a feedback system by establishing behavioral goals, generating plans to reach these goals, and establishing criteria for monitoring progress. This information is used to alter coping techniques, set new criteria for evaluating responses, and revise goals. The individual is, therefore, an information-processing system that regulates his or her relationship to the environment (Bandura, 1980). Individuals see and define their worlds, select coping strategies to manage threats, and change the way they represent problems when they obtain disconfirming feedback. The analogy provided is one of the people acting as scientists: formulating hypotheses about physiology and the effects of illness and creating a mental picture of his or her ability to take actions to prevent or cure illness.

Leventhal’s Self-Regulation Model suggests a process with four steps: (a) extract information from the environment; (b) generate a representation of the illness as dangerous to oneself; and (c) plan and act, which involve imagining response alternatives to deal with the problem and emotions it generates, and then taking selected actions to achieve specific effects. The feedback loop is achieved by the last step: (d) monitor or appraise how one’s coping affects the problem and oneself (Leventhal et al., 1980).

The work of Zimmerman and Leventhal has been highly influential, giving rise to important applications such as Clark and colleagues’ framework for conceptualizing chronic disease self- management (Clark, Gong, & Kaciroti, 2001). This framework has been applied to the design of self-management support interventions for asthma, heart disease, and other chronic illnesses. As part of these interventions, individuals are encouraged to select a goal related to disease management in an area that is personally important to them (e.g., increasing physical activity). After they observe their current behavior in this area and make judgments based on these observations, they set a behavioral change goal. As they make progress toward their goal, they experience cognitive reactions to their attempts at behavioral change, including changes to self-efficacy and outcome expectations. Interventions based on this framework have resulted in significant improvements in disease-management behaviors and health outcomes (Clark et al., 2000; Clark et al., 2004; Clark et al., 2010).


The HBM was developed in the early 1950s by social psychologists at the U.S. Public Health Service (Rosenstock, 1974) in an attempt to understand the widespread failure of people to accept disease preventives or screening tests for the early detection of asymptomatic disease. The HBM was later applied to patients’ responses to symptoms (Kirscht, 1974) and to adherence to prescribed medical regimens (Becker, 1974).

The basic components of the HBM are derived from a well-established body of psychological and behavioral theory. The antecedents are the same as those giving rise to SCT; in early iterations these antecedent concepts were called social learning theory (Miller & Dollard, 1941; Rosenstock, Strecher, & Becker, 1988). The HBM emphasizes personal beliefs and hypothesizes that behavior depends primarily on two variables: (a) the value placed by an individual on a particular goal and (b) the individual’s estimate of the likelihood that a given action will achieve that goal (Maiman & Becker, 1974). In the context of health-related behavior, these variables were conceptualized as: (a) the desire to avoid illness (or if ill, to get well) and (b) the belief that a specific health action will prevent (or ameliorate) illness (i.e., the individual’s estimate of the threat of illness and of the likelihood of being able, through personal action, to reduce that threat).

In the Health Belief Model context, it is understood that demographic, personal, social, and structural factors have the potential to influence health behaviors. However, these variables are believed to work through their effects on the individual’s health motivations and subjective perceptions, rather than functioning as direct causes of health action

Specifically, HBM consists of the following dimensions (Figure 1.1): perceived susceptibility, or one’s subjective perception of the risk of contracting a condition; perceived severity, or feelings concerning the seriousness of contracting an illness (or of leaving it untreated); while low perceptions of seriousness might provide insufficient motivation for behavior, very high perceived severity might inhibit action; perceived benefits, or beliefs regarding the effectiveness of the various actions available to reduce disease threat; perceived barriers, or the potential negative aspects of a particular health action that may act as impediments to undertaking the recommended behavior; and cues to action, or stimuli to trigger the decision-making process.

In the HBM context, it is understood that demographic, personal, social, and structural factors have the potential to influence health behaviors. However, these variables are believed to work through their effects on the individual’s health motivations and subjective perceptions, rather than functioning as direct causes of health action (Becker et al., 1977).

FIGURE 1.1 The Health Belief Model.

Source: From the National Institute of Neurological Disorders and Stroke. National Institutes of Health. Bethesda, MD 20892. https://stroke.nih.gov/resources/stroke_proceedings/Barch.htm

The HBM is well-represented in published research. Evidence supports the HBM’s ability to account for substantial amounts of variance in undertaking preventive health actions, seeking diagnoses, and following prescribed medical advice. Janz and Becker (1984) summarized findings from 46 HBM-related studies; 24 examined preventive health behaviors (PHBs); 19 explored sick-role behaviors (SRBs); and 3 addressed clinic utilization. A significance ratio was constructed that divides the number of positive, statistically significant findings for an HBM dimension by the total number of studies reporting significance levels for that dimension. In the preponderance of cases, each HBM dimension was found to be significantly associated with the health-related behaviors under study; overall, the significance ratio orderings (in descending order) were barriers (89%), susceptibility (81%), benefits (78%), and severity (65%).

Examples of HBM application across a range of preventive behaviors continue to proliferate. These include cancer screening (Janz, Wren, Schottenfeld, & Guire, 2003), condom use (Hounton, Carabin, & Henderson, 2005), vaccination (de Wit, Vet, Schutten, & van Steenbergen, 2005), and malaria chemoprophylaxis (Farquharson, Noble, Barker, & Behrens, 2004). The HBM has also been associated with heart disease management (George & Shalansky, 2006) and antihypertensive medication adherence (Yang et al., 2016). Although the HBM is often used to explain behavior, it has also served as a blueprint for intervention design. Jones, Smith, and Llewellyn (2014) conducted a systematic review of 18 HBM-based interventions targeted at improving adherence to a medical regimen. Perceived benefits and perceived susceptibility were included in virtually all studies; a cue to action was the model element least often addressed. About three quarters of reviewed studies reported a statistically significant effect of the intervention on improving adherence, but the authors found that the likelihood of intervention success did not appear to depend on which HBM constructs were addressed. Echoing a common criticism of the model, the authors observe that

there was no consistency within these studies in terms of the instruments used to measure health beliefs, the definition of the constructs, the number of items or response scales, whether items should be conditional or unconditional on taking preventive action or how the model was operationalized. (p. 265)

The authors argue that standardized tools are needed to measure HBM constructs, which could be adapted for given behaviors and diseases, to facilitate comparison across studies.

The HBM also has been criticized because it is limited to accounting for as much of the variance in individuals’ health-related behaviors as can be explained by their attitudes and beliefs. It is clear that other forces influence health actions; for example: (a) some behaviors (e.g., cigarette smoking and brushing teeth) have a substantial habitual component, obviating any ongoing psychosocial decision-making process; (b) many health-related behaviors are undertaken for what are ostensibly non–health-related reasons (e.g., stopping smoking to attain social approval); and (c) socioeconomic or environmental factors may prevent the individual from undertaking a preferred course of action (e.g., living in a city with high levels of air pollution). The model is based on the premise that health is a highly valued concern or goal for most individuals and that cues to action are widely prevalent; where these conditions are not satisfied, the model may not be useful in, or relevant to, predicting behavior.


The prediction of behavioral intentions, as the direct antecedent to behavior itself, is at the heart of Ajzen’s TPB (Ajzen, 1991, 2011) and its predecessor, the Theory of Reasoned Action (TRA; Ajzen & Fishbein, 1980). Predictors of behavioral intention, according to these models, are: (1) an individual’s attitudes toward the behavior, which are influenced, in turn, by his or her beliefs about the (a) likelihood and (b) desirability of outcomes resulting from the behavior; and (2) the subjective norm toward that behavior as perceived by the individual, determined by (a) what an individual perceives as the expectations of important others about the behavior, as well as (b) how motivated he or she is to comply with the beliefs of these significant others. The relative influence of these two components—attitude and subjective norms—on intention depends on the nature of the behavioral goal. For some behaviors, the attitudinal component will be the major determinant of intention, while for others, the more the individual believes that significant others are in favor of the behavior, the stronger will be his or her intention to perform it. A recent refinement of the model by its creator was the addition of descriptive norms as part of the normative beliefs component (Fishbein & Ajzen, 2010). Descriptive norms are what a person perceives as others’ actual performance of a particular behavior (such as the perceived extent to which peers practice safer sex); not just others’ views on the behavior.

The TPB expands on the TRA by including a component representing the perceived degree of control the person has over the behavior, a concept similar to Bandura’s construct of self-efficacy (Ajzen, 1991). These control beliefs take into account the presence and strength of personal and external factors that influence the behavior, such as having a workable plan, skills, social support, knowledge, time, money, willpower, and opportunity. By accounting for control perceptions, TPB increases its relevance to the large number of health-related behaviors over which people have incomplete volitional control, such as following nutrition guidelines.

FIGURE 1.2 Theory of Planned Behavior.

Source: Ajzen, I. (2013). Theory of Planned Behavior. Retrieved from http://people.umass.edu/aizen/tpb.html. Copyright 2006 Icek Ajzen. Reprinted with permission.

In sum, the TPB posits that when attitudes and subjective norms are favorable, and perceived control is high, strong intention to perform the behavior should result. The ultimate execution of the behavior is not solely explained by these predictors, as actual control plays an important part as well. According to the model (Figure 1.2), perceived behavioral control (PBC) can serve as a proxy for actual control over a behavior; as such, it may contribute to the prediction of actual behavior. In other words, successful performance of the behavior will be the end result if individuals have both intention and sufficient control over the internal and external factors that influence such performance.

Detailed guides are available to help researchers develop measures for each of the key TPB constructs and design TPB-based behavioral interventions; for example, selecting the specific theory components to attempt to influence for a given behavior and population (Ajzen, 2013). The TPB has been used extensively in health research and applied to a wide range of health-related behaviors (e.g., condom use, malaria prophylaxis, physical activity, cancer screening).

McEachan, Conner, Taylor, and Lawton (2011) performed a random-effects meta-analysis of 206 prospective studies of health behaviors using the TPB. Their analysis controlled for the effects of past behavior on future behavior, and assessed how behavioral type (e.g., health promoting vs. health risk), sample characteristics, and methodological factors (e.g., length of follow-up) moderated the ability of the TPB to predict future behavior. Results of this analysis indicated that the TPB explained 19% of the variance in behavior and 44% of the variance in intention across studies. Past behavior added 11% variance to the prediction of behavior and 5% to behavioral intention. In this analysis, the TPB more successfully predicted behaviors related to diet and physical activity than risk detection, safer sex, and abstinence from drugs. Behaviors in the shorter term were better predicted than behaviors in the longer term, and self-report measures of behavior were better predicted than objective measures. Notably, attitudes remained a strong predictor of intentions after controlling for past behavior, which, according to the authors, strengthens the case for trying to target attitudes as part of interventions.

The author of the TPB has responded to recurrent criticisms of this model (Ajzen, 2011). For example, Ajzen notes that the model has been said to ignore the affective component of decision making, and to rely on a rational-actor approach to behavioral choices. He argues that both emotion and irrationality can be subsumed in the behavioral, normative, and control beliefs components of TPB, as these beliefs are not always formed in a rational or even accurate manner and are subject to influence by emotional state. Ajzen also describes attempts made by researchers to improve the predictive validity of the TPB by adding components. For example, when predicting genetic testing behaviors, Wolff, Nordin, Brun, Berglund, and Kvale (2011) added anticipated affective outcomes to the model’s behavioral beliefs component; Kor and Mullan (2011) added perceived autonomy support to the prediction of sleep-hygiene behaviors. Ajzen cautions that for the sake of model parsimony, any new predictors proposed for addition to the model should meet strict criteria; for example, having a causal relationship with intention and action and conceptual independence from other components (Ajzen, 2011).

The debate over the utility of the TPB continues. A more recent round of criticism and counter-criticism was sparked by a commentary entitled, Time to retire the Theory of Planned Behavior (Sniehotta, Presseau, & Araujo-Soares, 2014), in which the authors assert that the TPB has well-documented limitations and that two aspects of the conceptual core of the TPB in particular are not supported: the notion that the attitude–behavior association is mediated by intention, and the assertion that TPB constructs as a group fully mediate all external influences on behavior. They also assert that the model is better at explaining intention than actual behavior, which limits its utility in practice. Ajzen responds to these criticisms by reiterating the specific way that the TPB should be used, including an extensive and specific formative research process (which Sniehotta and colleagues describe in a rebuttal as being so burdensome as to be impossible for most researchers) and argues that any other way of using the theory is not a fair test (Ajzen, 2015).

Questions about its overall value notwithstanding, Ajzen suggests several ways to further refine the TPB. First, additional work might focus on how beliefs are activated according to mood and setting (Ajzen, 2011). Other potential areas of inquiry are the role that past behavior and habit formation plays in determining behavioral intention and how stable background factors—for example, personality traits or demographic characteristics—might affect the relative importance of TPB components (Ajzen, 2011). Howland et al. (2016) report on an innovative extension of the TPB to a dyadic level in an assessment of interpersonal influences on physical activity intentions and behaviors. In this study, a partner’s PBC predicted a person’s intentions to be physically active over and above their own PBC and other model predictors. The authors suggest that such a dyadic framework might also be extended to other individual theories.


The TTM (Prochaska & Velicer, 1997), known more familiarly as the Stages of Change Model, is one of the most prominent in health behavior research. The TTM focuses on behavior change, in contrast to models like the TPB that could be used equally to explain ongoing behaviors (Noar, Chabot, & Zimmerman, 2008). The TTM was developed by Prochaska and colleagues in the 1970s to identify the common elements of leading theories of psychotherapy and behavior change. In subsequent empirical work, the authors observed that smokers used different processes of change at different points in their attempts to quit, leading to the formulation of the stage aspect of the TTM in which behavioral change is viewed not as an event but as a six-stage process.

These stages, which represent a continuum, and through which movement may take place backward as well as forward, are defined as (Prochaska & Velicer, 1997): precontemplation, where people are not considering a health behavior change in the near future (usually operationalized as 6 months); contemplation, where people are intending to change in the next 6 months and need to be motivated to do so; preparation, in which people intend to take action in the immediate future and need the skills to do so; action, where people are making a specific behavioral change and can be supported by intervention strategies and guidelines; maintenance, where a new behavior becomes habitual and requires less ongoing effort, but where relapse prevention is still important. The final stage, termination, occurs when a behavior is permanently ingrained.

The Transtheoretical Model, known more familiarly as the Stages of Change model, is one of the most prominent in health behavior research. The Transtheoretical Model focuses on behavior change, in contrast to models like the Theory of Planned Behavior that could be used equally to explain ongoing behaviors.

Ten processes of change, or activities that people use to progress through the stages, are specified by the TTM. Examples are counterconditioning (substituting healthy for less healthy behaviors), stimulus control (removing cues for unhealthy habits), and contingency management (e.g., a reward system). Other TTM constructs include decisional balance (an individual’s perceived pros and cons of changing), self-efficacy (as per SCT), and temptation (the intensity of urges to engage in a specific behavior when in challenging situations). According to the TTM, an intervention will be more effective if it is stage-matched—that is, using the processes and principles of change appropriate for a given stage and a given health behavior. Thus, one way to test the validity of the TTM is to do a match–mismatch experiment to assess whether stage matching enhances the intervention’s effectiveness. At least one such study of a smoking cessation intervention found that participants in the matched condition made greater forward movement in stages (Dijkstra, Conijn, & De Vries, 2006), but another found no advantage for the stage-matched condition (Aveyard, Massey, Parsons, Manaseki, & Griffin, 2009).

Much past empirical work with the TTM has been with smoking cessation—the health behavior that gave rise to the theory—but it also has been applied to a wide range of other health-related behaviors. For example, a 2008 meta-analysis examined 48 different health behaviors, which have been studied using TTM principles in 120 datasets (Hall & Rossi, 2008). This meta-analysis found consistent support across behaviors for Prochaska’s so-called strong and weak principles of change, which state that for an individual to move from precontemplation to action, an individual needs to experience approximately one standard deviation (SD) increase in his or her perceived pros of changing, and about one-half SD decrease in the cons of changing.

A number of reviews and meta-analyses have examined TTM-based interventions across various behaviors: smoking cessation (Cahill, Lancaster, & Green, 2010), physical activity (Hutchison, Breckon, & Johnston, 2009), dietary change for diabetes (Salmela, Poskiparta, Kasila, Vähäsarja, & Vanhala, 2008), and diet and physical activity for weight loss (Tuah et al., 2011). Across reviews, authors noted the inconsistent and often incomplete application of the model and other methodological weaknesses, making it difficult to draw conclusive results about the value of the TTM for behavior change interventions (Cahill et al., 2010). Based on a review of 35 trials, Cahill and colleagues (2010) concluded that stage-based self-help interventions for smoking cessation are no more effective than non–stage-based equivalents. Tuah et al. (2011) found that TTM-based interventions for weight loss have limited impact in the short term, with no strong evidence for bringing about sustained weight loss.

Despite—or perhaps because of—its popularity and the appealing prospect of stage-matched interventions having the potential to produce unprecedented impacts on entire at-risk populations (Prochaska & Velicer, 1997), the TTM has inspired a number of critical articles describing what are viewed as the model’s shortcomings. Indeed, one observer commented that the TTM has not just fame, but notoriety (Brug et al., 2005). Another notes that the model seems to be better accepted among practitioners than researchers (Munro et al., 2007). Examples of articles critical of the TTM are Sutton (2001), Littell and Girvin (2002), Adams and White (2005), and West (2005). For example, some argue that there is only weak empirical evidence for the efficacy of stage-based interventions, the existence of discrete stages of behavior change, and the sequencing of readiness stages. Others have pointed to the arbitrary length of stages and the use of staging algorithms that lack validity, reliability, and consistency across studies; the difficulty of applying the model to complex behaviors like physical activity or dietary change; and a lack of evidence that moving people forward in stages ultimately results in behavior change or other health-related outcomes. Munro et al. (2007) also noted that stage-tailored interventions may be resource intensive and might not be appropriate when rapid behavior change is desired.

Counterpoints to these criticisms also have been made. For example, Prochaska points to methodological shortcomings in studies that failed to offer support for the model (Prochaska, Velicer, Nigg, & Prochaska, 2008). It has also been noted that the TTM tends to be selectively operationalized (Hutchison et al., 2009), resulting in an alternative explanation for nonsignificant findings. Ideas for improving how TTM is applied include: increased precision of methods of staging individuals and expanding the use of stage-transition strategies to include constructs from other theories/models (Brug et al., 2005). Better documentation of the fidelity of TTM application to interventions also has been suggested to deepen understanding of how—and whether—the model is useful in fostering behavior change (Hutchison et al., 2009; Salmela et al., 2008). An informal review of the recent literature, however, suggests that such documentation continues to be largely absent from published reports of TTM-based interventions. It is important to note that this shortcoming is not unique to TTM but is shared across theories.


RP is a cognitive behavioral model that was developed more than three decades ago to help people change addictive behaviors and anticipate and cope with relapse (Hendershot, Marlatt, & George, 2009; Hendershot, Witkiewitz, George, & Marlatt, 2011; Marlatt & George, 1984). Rooted in SCT, with the ability to self-regulate playing a key role in an individual’s vulnerability to relapse (Bandura, 1997), the RP model has been highly influential in practice. In fact, relapse prevention has become essentially a generic term describing interventions that build coping skills for dealing with situations that place individuals at high risk of relapse (Hendershot, Marlatt, & George, 2009). Unlike the other theories and models discussed in this chapter, RP does not address initial behavior change, but rather deals with the ubiquitous challenge of maintaining the change over the long term. A major contribution of this model is its elaboration of relapse not as a failure, or end state, but rather as a transitional process which may or may not lead to previous levels of the undesired behavior. Moreover, RP provides individuals with skills and strategies to prevent a single, temporary lapse from becoming a more protracted relapse (Marlatt & George, 1984).

The RP model recognizes two primary categories of factors that trigger or contribute to relapse (as described in Larimer, Palmer, & Marlatt, 1999): immediate determinants of relapse and covert antecedents. Among the immediate determinants are high-risk situations, which pose a threat to a person’s sense of control over the behavior (including negative emotional states like depression, interpersonal conflict, social pressure, and external cues to engage in the behavior); coping (that is, how the person responds to those situations; individuals who have mastered effective coping strategies are less likely to experience relapse, and success in coping also leads to greater self-efficacy to do so in the future); outcome expectancies for the behavior (most relevant are positive expectations; for example, relapse is more likely if a person believes that the behavior will help with short-term stress reduction); and the abstinence violation effect (where a person who has an initial lapse feels guilt, a lack of control, and other negative emotions, which in turn make full-blown relapse more likely).

A second category of factors (Larimer et al., 1999) consists of the broader and more subtle covert antecedents to relapse. These can be thought of as lifestyle factors that influence the extent to which a person is confronted with high-risk situations. Successful lifestyle balance of obligations and pleasurable activities is essential for managing overall stress levels that can foster high-risk situations for relapse. The RP model specifies cognitive and behavioral intervention strategies to deal with both immediate determinants of relapse, such as skills development and cognitive restructuring, and also encourages people to achieve a healthier, more balanced lifestyle to address relapse’s covert antecedents. Importantly, a reformulated dynamic version of the RP model now recognizes both tonic (stable personal characteristics) and phasic (situational or transient) influences on relapse, which interact in complex ways to determine the likelihood of relapse (see Hendershot et al., 2011, for a detailed discussion of this reformulated model).

Most empirical applications of the RP model have been to alcohol use and smoking (Hendershot et al., 2011) with an occasional application to a nonaddictive behavior such as physical activity (e.g., Stetson et al., 2005). The efficacy of the RP model has been examined in systematic reviews, meta-analyses, and randomized controlled trials (RCTs), as well as in studies of specific model components such as self-efficacy and negative affect (Hendershot et al., 2011). It has been noted that while the overall clinical effectiveness of the RP model has received some empirical support, the diffuse application of RP approaches tends to complicate efforts to define RP-based treatments and evaluate their overall efficacy (Hendershot et al., 2011).

Both the RP model and the way it is applied to practice are continuing to evolve over time. For example, mindfulness-based RP refers to the addition of mindfulness techniques to the cognitive behavioral principles used in RP (Witkiewitz, Bowen, Douglas, & Hsu, 2013), with demonstrated promise in empirical studies (Witkiewitz et al., 2014). Other promising directions for this model (Hendershot et al., 2011) include: using nonlinear statistical approaches to examine the complex interactions proposed by the reformulated model; exploring genetic influences on relapse and relapse prevention; teasing out the mechanisms of effects in efficacious RP-based interventions; and use of functional MRI to explore the neural correlates of relapse.


While individual theories have changed little over the years, they are being applied in new ways, as intervention delivery methods keep pace with societal changes in communication and information exchange. For example, the last decade has seen a dramatic increase in theory-informed interventions delivered via the Internet (eHealth) or mobile devices (mHealth). mHealth includes short messaging service (SMS, or text messages) and apps on Internet-enabled smartphones, which may incorporate cameras, global positioning systems, and physiologic sensors into intervention design (Bull & Ezeanochie, 2016; Piette et al., 2015). A simple tabulation reveals that approximately one quarter of the behavioral intervention studies published between 2015 and 2016 in Health Education & Behavior incorporated technology-enabled components. Digital interventions have the potential to provide health behavior change support to a larger and more diverse audience than could be reached in traditional ways, given their low marginal costs and the fact that mobile phone use is now widespread across socioeconomic and demographic groups (Bull & Ezeanochie, 2016; Riley et al., 2011). Evidence is accumulating regarding the efficacy of many digital health behavior change interventions across diverse populations and behaviors (Hou, Charlery, & Roberson, 2014; Piette et al., 2015).

The degree to which classic theories of health behavior have been integrated into technology-enabled health interventions is the subject of a number of articles and reviews. For example, Winter and colleagues reviewed 240 technology-enabled interventions for cardiovascular disease prevention and treatment published between 2010 and 2015 and found that about half contained references to a behavioral theory or model—most commonly SCT, TTM, or TPB (Winter, Sheats, & King, 2016). Riley et al. (2011) reviewed mHealth interventions in the areas of smoking, weight loss, adherence, and disease management. The majority of smoking and weight loss studies invoked theory, and here, too, SCT was most frequently cited. In contrast, virtually no studies in this review targeting adherence or disease management cited a theoretical basis. The authors point out that even simple reminder interventions for adherence could potentially be made more powerful by addressing theory-informed mechanisms of behavior change (Riley et al., 2011). Bull and Ezeanochie (2016) found that only one tenth of electronically delivered interventions tested in RCTs between 2005 and 2014 included a reference to theory in their descriptions of the design, implementation, or analysis. In a review of 38 Internet-based interventions published between 2005 and 2015, Hou and colleagues (2014) found that most cited either SCT or TTM.

Overall, then, the use of theory in technology-enabled interventions appears to vary widely by intervention type and behavior of focus. Where theory is cited, it tends to be the same behavior change theories that are popular in nondigital interventions. SCT, in particular, is widely used, and seems to have ready application to the self-monitoring, individualized feedback, and goal-setting components of digital interventions. Reviews typically focus on research-tested interventions; use of theory may be markedly less common in commercially available apps for behavior change, and researchers have called for increased collaboration between health behavior change experts and commercial app developers (Zahry, Cheng, & Peng, 2016). Just as with nonelectronic interventions, there is general agreement that behavior change theory is useful for technology-enabled interventions but also mixed empirical evidence on this point. For example, in a review of text-messaging interventions on a range of health topics, the authors found that those that were theory-based were no more effective than those that were not (Head, Noar, Iannarino, & Grant Harrington, 2013).

Digital interventions have the potential to provide health behavior change support to a larger and more diverse audience than could be reached in traditional ways, given their low marginal costs and the fact that mobile phone use is now widespread across socioeconomic and demographic groups.

Other researchers have identified limitations to applying existing individual theories to the new types of interventions that are currently being developed, especially mHealth interventions. For example, they differ from traditional programs because they can incorporate frequent interaction with participants, up to multiple times per day, in the same context that health behaviors take place; for example, at home or at social events. Further, real-time inputs from participants—including self-reported symptoms, behavior, and information on current context—can be used to individualize interventions (Riley et al., 2011). Indeed, degree of tailoring and personalization of text–messaging-based health promotion interventions is thought to be positively associated with intervention efficacy (Head et al., 2013).

Ideally, then, health behavior theories used in these interventions would be able to accommodate a high level of interactivity and adaptivity. Traditional behavior change theories and models, however, are generally linear and static in nature (Riley et al., 2011). Martín and colleagues write: SCT and other related theories were developed primarily to explain differences between individuals, but explanatory theories of within-person behavioral variability are increasingly needed to support new technology-driven interventions that can adapt over time for each person (Martín et al., 2014). As an initial step to improving the fit of these theories to the more intensive, interactive interventions that are enabled by today’s technology, Riley and colleagues developed a dynamic computational model of SCT based on a control systems fluid analogy derived from engineering principles, which enables precise predictions of an individual’s behavior at a specific time based on specific changes in theory components (Riley et al., 2016).

Going forward, there are a number of innovative strategies for melding technology with theory to develop effective and appealing interventions for behavior change. Examples include: virtual reality (VR), gaming, alternate world reality gaming, and virtual coaches (Winter et al., 2016). One potentially powerful complement to traditional behavior change theories that can be incorporated into VR and gaming applications is story immersion—that is, an embedded narrative where participants feel part of the story (Lu, Baranowski, Thompson, & Buday, 2012). For example, a participant who takes on the part of a protagonist undergoing a behavior change in a vividly rendered world may be influenced by theoretical constructs like social norms and vicarious experiences woven into the story’s narrative. Finally, health communication and dissemination theories might inform mHealth interventions to maximize user engagement and take maximum advantage of the cost-efficient scalability of technology-enabled interventions (Bull & Ezeanochie, 2016).

Although technological advances provide exciting opportunities for health behavior scientists and practitioners, Winter et al. (2016) offer two notes of caution. First, they point out that while the pace of technology development is rapid, funding cycles for research are slow, potentially leaving investigators in a permanent state of catching up. Second, having more options for intervention delivery means more whiches to consider when applying behavioral theory: Which behavioral techniques delivered through which technology channels should be used for which targeted behaviors and outcomes and for which population segments? (p. 611).


Although presented here as distinct, the behavioral theories discussed in this chapter share many common elements; for instance, beliefs, self-efficacy, and social influences are all found across theories, though they may go by different names. While the original developers of these theories tended to focus on particular types of health behaviors (e.g., HBM for single-action preventive behaviors like screening; TTM for an addictive behavior, smoking), in practice all have been applied to a wide range of health-related behaviors. Noar and Zimmerman (2005) summarize the constructs that are common to the HBM, TRA, TPB, SCT, and TTM, showing how a given concept is labeled and described within each theory. A network analysis by Gainforth, West, and Michie (2015) provides empirical support for the close relationship and common origins of most behavior change theories.

Differences in individual theories also are easy to spot: for example, TRA/TPB is concerned with behavioral intention as the outcome; other theories attempt to predict actual behavior. Some theories, like TTM, are stage-based and describe processes and predictors of stage movement en route to behavior change; others are continuous and describe how predictors work together to influence the likelihood of action (Head & Noar, 2014). Theories may be developed with the ultimate goal of prediction, designing behavior change strategies, or both (Head & Noar, 2014), though the studies reviewed in this chapter suggest that this distinction is largely lost in the real world of theory application. Theories also differ in the extent to which they prescribe how to operationalize and measure theoretical constructs. As noted by Young et al. (2014), there is particularly wide variation in how elements of SCT are operationalized. In comparison, the developers of the TPB provide specific, modifiable items to measure theory constructs.

The overlap among popular individual theories can create confusion for practitioners and applied researchers who are trying to decide which option is best for a given application (Gainforth et al., 2015). Often their choice will be based on what is familiar or commonly used in a given field. Many of the reviews cited in this chapter have noted the tendency for study authors to cite multiple theories as influences on intervention design. Some research has suggested that multiple-theory interventions may be less effective than their single-theory counterparts (Gourlan et al., 2016; Prestwich et al., 2014), though the reasons for this are not clear.

Noar and Zimmerman (2005) describe two broad directions the field could take to address theory overlap. The first is to focus on integrations of similar theories, in which well-supported constructs from single theories are combined, tested across a range of behaviors, and then refined as needed. According to Head and Noar (2014), a first major attempt at such a unified theory was made in 1991 as a result of a workshop attended by the developers of major theories. The resulting Integrated Behavioral Model, however, was seldom used subsequently. Over a decade later this model was refined further by Fishbein and Ajzen, who recast it as the Reasoned Action Approach (RAA; Head & Noar, 2014). To date, few empirical published studies have applied the RAA under that name; illustrating, perhaps, that the uptake of theoretical innovations into practice is slow and uncertain. Another example of an attempt to make a composite, or best of, theory is Ryan’s (2009) Integrated Theory of Health Behavior Change, which blends concepts from multiple theories, selecting components based on the strength of empirical evidence for their efficacy. The second broad direction is for researchers to place more emphasis on head-to-head comparisons of theories, a point that we will return to in the following section. While such comparisons are still uncommon, several recent examples can be found in the literature (Gerend & Shepherd, 2012; Montanaro & Bryan, 2014; Murphy, Vernon, Diamond, & Tiro, 2014).


Judging from the extensive application of individual-level theories to current health behavior research—and especially the handful of popular theories described in this chapter—their value and utility would appear to be widely accepted. However, the actual link between theory use and intervention effectiveness appears to be weak, as suggested by a rigorous meta-analysis of nearly 200 physical activity and healthy eating interventions (Prestwich et al., 2014). Using a theory coding scheme to assess the use of theory to develop and evaluate interventions, these investigators found that very few reviewed studies used theory extensively. No meaningful associations were found between intervention effectiveness and type of theory use (e.g., linking behavior change techniques to a particular theoretical construct) or an overall theory score representing extent of theory use. Other commentators have repeatedly called attention to shortcomings in health behavior theory development, particularly in the realm of theory testing and of theory refinement.

Ideally, the field of health behavior theory would be dynamic, with ongoing development and modifications as new empirical evidence emerges (Munro et al., 2007; Noar & Head, 2014). Instead, however, some researchers refer to some popular theories as having a scientifically unhealthy immortality (Sniehotta et al., 2015); that is, they are never improved upon, nor retired. This state of stagnation may be due to a lack of rigorous testing and subsequent theory modification that would genuinely move the field forward (Head & Noar, 2014; Noar & Mehrotra, 2011; Weinstein, 2007; Weinstein & Rothman, 2005). Although it is beyond the scope of this chapter to describe these critiques in detail, several issues will be highlighted here, and the reader is referred to more nuanced discussions in the cited works.

The first is the means by which theories have traditionally been evaluated. The most common approach to theory testing is correlational; that is, the which theory explains the most variance in a particular behavior approach. However, whether data are cross-sectional or longitudinal, associations between cognitive variables and health behaviors are inflated because the relationships are likely bidirectional. Thus, any observed cognition/behavior association may be explained, in part, by the effect of the behavior on how one thinks about that behavior. This may be particularly true in the case of ongoing or habitual behaviors (Painter et al., 2008; Weinstein, 2007). For example, a person who has experienced the benefits of regular physical activity is likely to have more favorable attitudes toward exercise. In other words, correlational approaches assess whether given theory-based constructs are associated with a particular health behavior, when the true question of interest is whether changes in theory-based constructs lead to changes in health behavior (Noar & Mehrotra, 2011; Sniehotta et al., 2015). Another criticism of the correlational approach is that it offers little precision or insight about the criteria required for empirical data to support a given theory; for example, what does it mean if some, but not all, hypothesized relationships among theoretical constructs are supported by the data in a given study (Hagger, 2015)?

Health behaviors have a tendency to cluster and share common determinants at individual, interpersonal, and environmental levels. No existing individual-level theory explicitly considers changing multiple health behaviors simultaneously.

Sniehotta and colleagues argue that in spite of these issues, individual-level theory remains relevant, as scholars from multiple disciplines seek innovative strategies to motivate behavior change with the goal of improving population health and quality of life. As part of this process, insightful, in-depth theories that help us understand how this can be accomplished will remain in demand. It is therefore incumbent on health behavior scientists to improve methods of theory development in a way that mirrors recent improvements in methods for the development, reporting, and evaluation of interventions (Sniehotta et al., 2015, p. 186).

Head and Noar (2014) note the tension between the generalizability of a theory and its accuracy and utility in a given domain. The former is often of interest to theory developers and the latter to those applying the theory (Head & Noar, 2014). They advocate for living, behavior-specific versions of general theories that would incorporate knowledge gained from empirical work on that behavior and result in adding or removing theoretical constructs as indicated. Such behavior-specific theory could inform the design of future interventions and a more useful version of a general theory (Noar & Head, 2014).

Noar and Mehrotra (2011) have proposed a multimethodological theory-testing framework as a way to increase the rigor of theory testing and ultimately enhance the value of theory to practice. According to the authors, any single theory-testing approach has inherent limitations. For example, while theory-based interventions appear to lend support to that theory if they are proven effective, in cases where the control condition consists of minimal intervention or none at all, the specific role of the theory in intervention efficacy remains unclear. An alternative approach using sequential, complementary methodological tools has been suggested to test theory. These tools include: randomized lab experiments, randomized field experiments of theory-based interventions and analysis of mechanisms, and meta-analysis of both lab and field experiments. Brewer and Gilkey (2012) propose competitive hypothesis testing as another means of increasing the value of theory testing. In this approach, researchers identify specific components of two theories that give rise to competing predictions about the nature of the relationships among particular variables; empirical data are then used to determine which theory better supports the hypotheses. Noar and Zimmerman (2005) suggest eight important research questions for direct comparisons of theories; for example, Are certain theories or elements of theories better predictors of one-time behaviors as opposed to behaviors that must be maintained over time?

A second, broader issue regarding individual-level theories of behavior change is that they typically downplay the importance of both the biological underpinnings and consequences of behavior and the social context in which it occurs (Institute of Medicine Committee on Health and Behavior: Research, Practice, and Policy, 2001). Most of the theories discussed in this chapter incorporate interpersonal influences on attitudes, beliefs, or other antecedents of behavior; such variables are likely to be influenced indirectly by one’s environment (e.g., workplace restrictions on smoking may factor into one’s pros and cons of smoking cessation as per the TTM; living in an unsafe neighborhood may impede self-efficacy for daily physical activity).

However, Glass and McAtee (2006) write that much of public health continues to treat behaviors such as diet, smoking, violence, drug use and sex work as if they were voluntary decisions, without regard to social constraints, inducements, or pressures. These authors propose a topographic metaphor to illustrate the multilevel, interacting influences on behavior. Individual behaviors take place on the surface of a flowing stream, and are influenced by the biological levels below the water’s surface; above the surface, the social, built, and natural environments all shape individual behavior as well, with the life course represented by the horizontal flow of water. Risk regulators—factors like material conditions, social norms, and policies—facilitate or constrain a given behavior, and that affects biological systems, which also interact with behavior. This complex, multidimensional way of thinking about health behavior is a framework for a next-generation approach to the study of behavior and health, including generating theories and organizing research (Glass & McAtee, 2006). King (2015) has also advocated for behavioral theories and models that embrace complexity and serve to bridge constructs from behavior, and the micro, social/cultural, and physical environments.

Finally, application of individual theories to multiple health behavior change merits attention. Health behaviors have a tendency to cluster, and share common determinants at individual, interpersonal, and environmental levels (Klein et al., 2016). No existing individual-level theory explicitly considers changing multiple health behaviors simultaneously (Noar et al., 2008). Noar et al. (2008) suggest three possible approaches to using theory as it relates to multiple behaviors, each with differing implications for theory-based interventions: (a) a behavior change principles approach, wherein common principles of health behavior change can be taught to individuals, who can apply them to multiple health behaviors; (b) a global health/behavior category approach, which suggests using broader categories such as weight control or management of a particular illness to organize and appeal to the motives for change regarding a variety of health behaviors; and (c) a multiple behavioral approach, which involves intervening on multiple behaviors that may cluster together, like alcohol use and smoking. Empirical testing could reveal which of these theoretical approaches is optimal for given behaviors or circumstances. Klein et al. (2016) describe a variation of these approaches: repurposing evidence-based interventions from one related behavior to another. Finally, careful descriptive study of how multiple health behaviors cluster is also warranted.

Other related questions that merit future exploration have to do with sequencing—that is, is it best to address multiple behaviors simultaneously or sequentially? Is there a logical hierarchy to changing behaviors (such as gateway behaviors), and a maximum number

Вы достигли конца предварительного просмотра. Зарегистрируйтесь, чтобы узнать больше!
Страница 1 из 1


Что люди думают о The Handbook of Health Behavior Change, Fifth Edition

0 оценки / 0 Обзоры
Ваше мнение?
Рейтинг: 0 из 5 звезд

Отзывы читателей