The International Electronic Journal of Health Education

[Reprint (PDF) Official Layout of this Article]

IEJHE, 2000; 3(Special), Copyright © 2000

Health Behavior Models

Colleen A. Redding, PhD1; Joseph S. Rossi, PhD2; Susan R. Rossi, PhD3; Wayne F. Velicer, PhD4; James O. Prochaska, PhD5

1Dr. Redding is an Assistant Research Professor at the Cancer Prevention Research Center (CPRC) at the University of Rhode Island. She received her doctorate in 1993 at the University of Rhode Island in clinical psychology. Her clinical internship was in Health Psychology at UCLA Medical Center and her post-doctoral fellowship was at the CPRC. She has fifteen years of clinical and research experience in many medical and reproductive health care settings, and is an investigator on many federally funded research projects studying various health behavior changes. She has ten years of experience in the development and application of the Transtheoretical Model and stage-matched interventions to many problem behaviors.
2Dr. Joseph Rossi is a Professor in the Department of Psychology and Director of Research at the Cancer Prevention Research Center at the University of Rhode Island. He received his PhD in experimental psychology from the University of Rhode Island in 1984. He has published in a wide range of areas including research methodology and measurement, skin cancer prevention, smoking cessation, exercise adoption, weight control, diabetes self_management, HIV risk reduction, and expert system development. A recent study conducted by the Institute for Scientific Information and the American Psychological Society listed him among the 10 most influential authors in psychology.
3Dr. Susan Rossi is Diabetes Research Coordinator for ProChange Behavior Systems, Adjunct Assistant Professor at the College of Nursing and Coordinator of Adherence & Education Programs for the Rhode Public Health Partnership in Emerging Infectious Disease at the University of Rhode Island. She received her BS in nursing at Duke University in 1974, and an MSN in 1982 and PhD in nursing in 1994 at the University of Rhode Island. She has 13 years of research experience as a behavioral scientist focusing on applying the Transtheoretical Model to the dietary area, work which has been nationally and internationally recognized.
4Dr. Velicer is a Principal in ProChange, Co_Director of the Cancer Prevention Research Center, and is Professor of Psychology, University of Rhode Island. He received his PhD in psychology from Purdue University in 1972. He has published more than 120 papers on a variety of topics, including behavior change for health promotion/disease prevention, factor and component analysis, time series analysis, and measurement development. He was identified as one of the highest impact authors in psychology by studies in 1992 and 1996. He is an original developer of the Transtheoretical Model and a pioneer in the development of expert systems. He has been PI or co_PI on more than $40 million dollars in federal funding.
5Dr. Prochaska is President of ProChange, Director of the Cancer Prevention Research Center, and Professor of Psychology, University of Rhode Island. He received his PhD in clinical psychology in 1969 from Wayne State University. He has published more than 100 papers and 3 books on the Transtheoretical Model of behavior change for health promotion and disease prevention. A recent study conducted by the Institute for Scientific Information and the American Psychological Society listed him among the 10 most influential authors in psychology. One of the originators of the Transtheoretical Model, he has been PI on over $40 million in research grants on prevention of cancer and other chronic diseases.

Corresponding author: Colleen A. Redding, Assistant Professor, Cancer Prevention Research Center, University of Rhode Island, 2 Chafee Road, Kingston, RI 02881-0808; phone: 401.874.2830; email: e-mail: CREDDING@URI.EDU.

Introduction Health Belief Model Theory of Reasoned Actions Social Cognitive Theory Self-Efficacy Transtheoretical Model References

Since they define what to measure, models and theories of health behavior change are inherently linked to the measurement of health behavior. This chapter reviews different models of health behavior change as an overview, not an in-depth comparison. Furthermore, while we aim to present other models fairly, the authors are biased toward the Transtheoretical model -- we present best that with which we are most familiar.

We review here only the four most commonly used models of individual health behavior change, relying primarily on the criteria of Glantz and colleagues.1 Their review of articles published between 1992-1994 in health education, medicine, and behavioral science that use any theoretical framework (only 45% used a theory)1 revealed that the most used models were the Health Belief Model, Theory of Reasoned Action/Planned Behavior, Social Cognitive Theory and The Transtheoretical Model. There are many more theories and models relevant for health behavior covering such important topics as: community organization, communication, diffusion of innovations, social marketing, information processing, stress and coping, relapse prevention, and empowerment. Additional materials are cited in the text and interested readers can pursue these ideas further through these resources. An entire book is devoted to the topics covered in this chapter and is highly recommended for those who want more details.1

Theoretical models fundamentally guide both our current and future understanding of health behavior, as well as providing direction for our research and intervention development. As a metaphor, each model or theory provides a different roadmap of the health behavior territory. Of course, it is important to point out that the map is not the territory itself, and different maps (theories) describe the same territory differently. Even so, when we enter new territory, we still need a map. Even a roughly drawn or poorly scaled map is much better than none at all. The map points out the relevant landmarks (constructs) and how they are connected, and, perhaps, how far it is from one landmark to another. As different maps of the same territory evolve over time and are compared, the territory becomes clearer, thus allowing better maps to evolve, perhaps integrating the clearest features of different maps. So it goes with theoretical development as well. There is no final or true map, only a map or theory that best meets our needs right now. Thus, as we evaluate these different theories, we should ask comparable questions of them.

Evaluation and comparison of the different theories reveals that they are not so different in terms of their differential predictions. Most differences really amount to emphasis on one construct over another. Cummings and colleagues conclude that theories which integrate ideas from other competing theories provide more explanatory power.3 Similarly, Fishbein and colleagues went through an unprecedented consensus conference among theorists to outline the most important variables to study in relation to reducing HIV risk.3 Many constructs from each theory are actually fairly similar. One excellent research proposal for conducting critical tests of different theoretical predictions is outlined by Weinstein.4 In fact, the National Institutes of Health recently issued a cross-agency request for funding applications to systematically test different theoretical predictions.

Health Belief Model
Introduction Health Belief Model Theory of Reasoned Actions Social Cognitive Theory Self-Efficacy Transtheoretical Model References

The Health Belief Model (HBM) has the longest history of all the theories reviewed. It was originally conceived by social psychologists in the public health arena as a way of predicting who would utilize screening tests and/or vaccinations.3,5-9 According to the HBM, the likelihood that someone will take action to prevent illness depends upon the individual's perception that:

• they are personally vulnerable to the condition;
• the consequences of the condition would be serious;
• the precautionary behavior effectively prevents the condition; and
• the benefits of reducing the threat of the condition exceed the costs of taking action.9

These four factors, which are influenced by mediating variables, indirectly influence the probability of performing protective health behaviors by influencing the perceived threat of the illness and expectations about outcome.

The HBM has been used for intervening with health screening, illness, sick role, and precautionary behaviors.5,8-12 The model has undergone some modifications since its original formulation. Table 1 shows the four-construct model that is the most commonly described form of the HBM. The model's four key components are conceptualized as perceived: 1) susceptibility, 2) severity, 3) effectiveness, and 4) cost. 

Table 1. Health Belief Model Constructs



Perceived Susceptibility

One’s evaluation of chances of getting a condition

Perceived Severity

One’s evaluation of how serious a condition, its treatment, and its conseuqences would be

Perceived Benefits

One’s evaluation of how well an advised action will reduce risk or moderate the impact of the condition

Perceived Barriers

One’s evaluation of how difficult an advised action will be or how much it will cost, both psychologically and otherwise

Cues to Action

Events or strategies that increase one’s motivation


Confidence in one’s ability to take action

Perceived susceptibility refers to the probability that an individual assigns to personal vulnerability in developing the condition. The concept of perceived susceptibility has been found to be predictive of a number of health-protective behaviors. From an HBM perspective, the likelihood individuals will engage in precautionary behaviors to prevent cancer (e.g., quit smoking, eat a diet low in fat and high in fiber, exercise, get a mammogram or prostate exam) depends on how much they believe they are vulnerable to or at risk for cancer. In general, it has been found that people tend to underestimate their own susceptibility to disease.

Perceived severity refers to how serious the individual believes the consequences of developing the condition are. An individual is more likely to take action to prevent cancer if s/he believes that possible negative physical, psychological, and/or social effects resulting from developing the disease pose serious consequences (e.g., altered social relationships, reduced independence, pain, suffering, disability, or even death). Models of Health Belief frequently refer to perceived health threats. The combination of perceived susceptibility and perceived severity constitute a threat.

Perceived effectiveness refers to the benefits of engaging in the protective behavior. Motivation to take action to change a behavior requires the belief that the precautionary behavior effectively prevents the condition. For example, individuals who are not convinced that there is a causal relationship between smoking and cancer are unlikely to quit smoking because they believe that quitting will not protect against the disease.

Perceived cost refers to the barriers or losses that interfere with health behavior change. The combination of perceived effectiveness and perceived costs constitute the notion of outcome expectation. Belief alone is not enough to motivate an individual to act. Taking action involves cognitively weighing the personal costs associated with the behavior against the benefits expected as a result of engaging in the behavior. Benefits have to outweigh the costs involved.

Cues to action involve stimuli that motivate an individual to engage in the health behavior.9 The stimulus that triggers action may be internal or external. For example, angina may act as an internal cue to initiate action. External cues such as a spouse's illness or the death of a parent may also trigger health behavior changes in an individual who was not otherwise considering them. HBM factors also interact to trigger action. For example, when perceptions of susceptibility and severity are high, a very minor stimulus may be all that is needed to initiate action. However, more intense stimuli may be needed to initiate action if perceived susceptibility and severity are low.

More recent formulations of the HBM have included self-efficacy as a key factor. Self-efficacy is influenced by mediating variables and in turn influences expectations. In addition, some forms of the HBM refer to general susceptibility to illness as a key factor in the model. However, substitution of the general case over specific consequences is only appropriate if the intention of the precautionary behavior is to improve health in general.4 The value of health, another variable which is sometimes included, refers to interest in and concerns about general health,9 the extent to which an individual values health.6 According to this view of HBM, individuals concerned about being healthy in general are more likely to exercise regularly than individuals who place little value on health. Although both cues to action and the value of health have been included in some forms of HBM, their importance in predicting health behavior is unclear since nei ther variable has been systematically studied.9

Mediating factors (demographic, structural, and social variables) have also been explored in applying the HBM. Mediating variables (e.g., educational level) are believed to indirectly affect behavior by influencing an individual's perceptions of susceptibility, severity, benefits, and barriers.9 Becker and Maiman added the concept of motivation to the HBM.6 This has also been interpreted as readiness to change behavior.12

Theory of Reasoned Action/Planned Behavior
Introduction Health Belief Model Theory of Reasoned Actions Social Cognitive Theory Self-Efficacy Transtheoretical Model References

The theory of reasoned action (TRA) is a widely used behavioral prediction theory which represents a social-psychological approach to understanding and predicting the determinants of health-behavior.14-16 Over the years, TRA has been applied to many diverse health-related behaviors including: weight loss, smoking, alcohol abuse, HIV risk behaviors, and mammography screening. The theory of reasoned action states that the intention to perform a particular behavior is strongly related to the actual performance of that behavior. Two basic assumptions that underlie the TRA are: 1) behavior is under volitional control, and 2) people are rational beings. From the perspective of TRA, we behave in a certain way because we choose to do so and we use a rational decision-making process in choosing and planning our actions. The TRA was designed to predict behavior from intention, and proposes quasi-mathematical relationships between beliefs, attitudes, intentions, and behavior. A modified version of TRA includes the addition of perceived control over the behavior and is referred to as the Theory of Planned Behavior.3,7 Table 2 describes the main constructs used in TRA and the Theory of Planned Behavior (TPB).

Table 2. Theory of Reasoned Action/Planned Behavior



Behavioral Intention

Perceived likelihood of performing the behavior


The product of the behavioral belief multiplied by the evaluation of it

Behavioral Belief

Evaluation of the likelihood that performance of the behavior is associated with certain outcomes

Evaluation of B.B.

How good or how bad those outcomes would be

Subjective Norm

The product of the normative belief multiplied by the motivation to comply

Normative Belief

Perception of how much each personal contact approves or disapproves of the behavior

Motivation to Comply

Motivation to do what each personal contact person wants

Perceived Behavioral Control

The product of the control belief multiplied by the perceived power

Control Belief

Perceived likelihood of each facilitating or constraining condition occurring

Perceived Power

Perceived effect of each condition in making the performance of the behavior easier or more difficult

Predicting behavior is the ultimate goal of the TRA. According to the TRA, behavior is influenced by the intention to perform the behavior. Intention is influenced by three major variables: subjective norms, attitudes, and self-efficacy. Subjective norms involve an individual's perception of what significant others believe about his or her ability to perform the behavior. For example, whether or not someone intends to cut down on dietary fat by giving up bacon and red meat could be partly determined by what that person believes his or her spouse's opinion would be if s/he did. Attitudes can be conceptualized in terms of values. That is, an individual develops particular values about behaviors. For example, one attitude might be: eating a healthy diet is a good way to prevent heart disease and/or cancer. Self-efficacy is the confidence an individual feels that s/he can successfully perform the behavior of eating a healthy diet.

Two of the variables that influence intention, subjective norms, and attitudes are in turn influenced by beliefs. Two general types of beliefs are considered in TRA: normative and behavioral beliefs. Normative beliefs are situationally based social expectations, which are considered the rule. Normative beliefs influence subjective norms while beliefs about the behavior influence attitudes. An individual's attitudes toward a behavior are determined by his/her expectations about the outcome of performing the behavior, and the extent to which s/he values the outcome. Thus, from a TRA perspective, the likelihood that an individual will engage in health risk reduction depends upon how much s/he is convinced that healthy behaviors will prevent risk, and the degree to which s/he perceives the benefits will outweigh the costs.

The majority of TRA research has focused on the prediction of behavioral intention rather than on the behavior itself.13 Unfortunately, because the correlation between behavior and intention is not particularly impressive, research on attitudes and behaviors is often dismissed.14 Despite this shortcoming, Sonstroem has suggested that TRA can still be a use ful perspective as long as situation-specific attitude and intention measures are employed that specify congruent action, target, context, and time, and that the interactions between personal determinants and situations are emphasized.17

Social Cognitive Theory
Introduction Health Belief Model Theory of Reasoned Actions Social Cognitive Theory Self-Efficacy Transtheoretical Model References

This theory goes well beyond individual factors in health behavior change to include environmental and social factors. In fact, this theory may be the most comprehensive model of human behavior yet proposed. Bandura's Social Cognitive Theory (SCT),18 also referred to as Social Learning Theory, is a

Table 3. Social Cognitive Theory Constructs




Factors outside the person


One’s perception of the environment

Behavioral Capability

One’s knowledge and skills to perform a behavior


One’s anticipation of the outcomes of a behavior


How good or bad one evaluates the outcomes to be


Regulation of one’s own behavior

Observational Learning

Acquiring a new behavior by watching someone else perform it and observing the outcomes–a.k.a. modeling


Responses to a person’s behavior that affect how likely it is that the behavior will reoccur


One’s confidence in one’s own ability to perform a behavior

Emotional Coping Responses

Strategies used by someone to deal with emotionally challenging thoughts, events, or experiences

Reciprocal Determinism

Dynamic interaction of the person, the behavior, and his/her environment

behavioral prediction theory that represents a clinical approach to health behavior change.3,7 This theory has been widely applied to health behavior with respect to prevention, health promotion, and modification of unhealthy lifestyles for many different risk behaviors. SCT emphasizes what people think and its effect on their behavior.19,20  SCT proposes that behavior can be explained in terms of triadic reciprocity between three key concepts which operate as determinants of each other. Reciprocal determinism forms the basic organizing principle of SCT. This important concept states that there is a continuous, dynamic interaction between the individual, the environment, and behavior. Thus, a change in one of these factors impacts on the other two. SCT involves numerous key concepts, which have been associated with each of the three main constructs for the purpose of describing the SCT. Table 3 describes all the key constructs employed by SCT.

Bandura conceptualized influences on behavior that involved the concept of person in terms of basic human capacities that are cognitive in nature.18 Key concepts associated with the person include: personal characteristics, emotional arousal/coping, behavioral capacity, self-efficacy, expectation, expectancies, self-regulation, observational/experiential learning, and reinforcement.19,20



Personal characteristics have been operationalized as multiple, interacting determinants such as demographics (e.g., gender, race/ethnicity, education), personality, cognitive factors (e.g., thoughts, attitudes, beliefs, knowledge), motivation, and skills.
Emotional arousal/coping can interfere with learning and thus influence behavior. This refers to an individual's ability to respond to emotional stimuli with various techniques, strategies, and activities that help one to deal with arousing situations (e.g., fear, anxiety).
Behavioral capacity refers to the individual's possession of both the knowledge and skills necessary to perform a behavior.
Self-efficacy refers to an individual's confidence in his or her ability to perform a behavior in various situations. Self-efficacy has been recognized as an important mediating variable between knowledge, attitudes, skills, and behavior.13
Expectations are beliefs associated with the outcome of a behavior. Expectancies are the value an individual attributes to the anticipated outcome of performing a behavior.
Self-regulation refers to the individual's ability to manage or control behavior. Individuals use goal setting, self-monitoring, and self-reinforcement to regulate performance of a behavior.
Observational/experiential learning refers to the acquisition of a behavior through observation and experience. Learning can occur either through observation of another's performance of a behavior (modeling), or through personal experience, i.e., trial and error.
Reinforcement refers to the consequences that affect the probability a behavior will be tried again. Individuals are motivated to perform behaviors through rewards and incentives.20

In SCT, the relationship between behavior, person, and environment is interactive. The stereotypic picture of the relatively young executive who develops high blood pressure provides an illustration of how variables associated with person, (e.g., personal characteristics), interact with the environment and behavior. Consider a male in his early 40s who is obsessed with achievement, advancement, and recognition. This individual is a highly competitive workaholic who is driven to get things done quickly. Such individuals are sometimes described as being hostile and might be found operating in a highly stressful environment. Although simplistic and stereotypical, this picture represents a classic example of a "Type A personality." From an SCT perspective, this individual's predominant personality type negatively influences his behavior. Thus this individual is less likely to take the time to acquire the cognitive and behavioral skills necessary to successfully perform any risk reduction behavior (smoking cessation, stress management, etc.).

Influences on behavior which involve the environment can be physical, social, cultural, economic, political in nature,21 or situational in nature.20 In SCT, the person's perceptions of the environment are referred to as situations; this key variable can facilitate or inhibit behavior. In this reciprocal, interactive scheme, in which multiple determinants of behavior are assumed, behavior also exerts an influence on both the environment and the person. The environment and past experience with a particular behavior can also provide reinforcement for acting in a particular way. For example, as Americans have demanded the availability of healthier, lower fat, higher fiber choices in their environment, more and more eating establishments have changed their food preparation procedures and menus to reduce dietary fat. There are a wider variety of "heart healthy" menus available now. To reduce their risk, consumers have begun to take advantage of greater environmental choices by: purchasing more fruits and vegetables, substituting available lower fat products instead of high fat ones, changing their food preparation methods to broiling and baking instead of frying, and ordering lower fat food choices offered by restaurants.

Interactions are also assumed to occur between problem behaviors (e.g., eating high fat foods, lack of exercise, smoking) and physiological factors (e.g., nicotine, caffeine addiction).21 An individual's performance of associated behaviors can have an important impact on disease prevention. Engaging in exercise can trigger hunger, stimulating the desire for high fat food. Finishing a meal can act as a cue that triggers the desire for a cigarette. An individual may use smoking to relax in a high stress environment. To effectively prevent disease, an individual needs to engage in multiple healthy behaviors like exercise adoption, low fat/high fiber eating habits, mammography screening, wearing seatbelts, etc. SCT assumes that most behaviors are learned responses and can be modified. Thus, learning through observing the behavior of others (i.e., modeling) is important from a SCT perspective. SCT also places heavy emphasis on learning both cognitive and behavioral skills for coping with situations and making changes in health behavior. Thus, an individual who wants to quit smoking but lacks the cognitive and behavioral skills to effectively cope with stressful situations without cigarettes is less likely to successfully change smoking behavior in spite motivation to do so.

Introduction Health Belief Model Theory of Reasoned Actions Social Cognitive Theory Self-Efficacy Transtheoretical Model References

The concept of self-efficacy is recognized as one of Bandura's most important contributions to psychology and the field of health behavior change in general.22 Self-efficacy refers to the confidence an individual has in his or her own ability to successfully carry out a behavior. The importance of self-efficacy for behavior change has been widely recognized across multiple behaviors relevant to health risk reduction.23 Furthermore, its incorporation into almost all major theories of behavior change is further evidence of its important role in the behavior change process.

Bandura proposed that the actual performance of a particular behavior is highly related to an individual's belief in his/her ability to perform that behavior in specific situations. An individual with low self-efficacy is likely to have lower expectations of successfully performing the behavior and be more affected by situational temptations that are counterproductive to promoting and maintaining behavior change. In contrast, an individual who has high self-efficacy not only expects to succeed but is actually more likely to do so. For example, the likelihood that an individual will successfully perform a behavior like exercise is strongly dependent upon how confident that individual is that s/he can actually do activities, such as walking, jogging, swimming, or doing aerobics on a regular basis.22

Several factors influence an individual's self-efficacy, including persuasion by others, observing others' behavior (modeling), previous experience with performing the behavior, and direct physiological feedback.18 For example, individuals are more likely to attempt to quit smoking if: 1) a physician recommends that they do so, thus persuading them that quitting is a good idea, 2) they have observed others who have been able to quit and/or are coping well with trying to quit, 3) they have had past experience with quit attempts, and/or 4) they have been able to cope with the physical symptoms of nicotine withdrawal. Self-efficacy exerts such a strong influence on behavior change that confidence has been found to outperform past performance in predicting future behavior.Several factors influence an individual's self-efficacy, including persuasion by others, observing others' behavior (modeling), previous experience with performing the behavior, and direct physiological feedback.18 For example, individuals are more likely to attempt to quit smoking if: 1) a physician recommends that they do so, thus persuading them that quitting is a good idea, 2) they have observed others who have been able to quit and/or are coping well with trying to quit, 3) they have had past experience with quit attempts, and/or 4) they have been able to cope with the physical symptoms of nicotine withdrawal. Self-efficacy exerts such a strong influence on behavior change that confidence has been found to outperform past performance in predicting future behavior.24

Transtheoretical Model
Introduction Health Belief Model Theory of Reasoned Actions Social Cognitive Theory Self-Efficacy Transtheoretical Model References

The past 20 years of Transtheoretical Model-based research has found some common principles of behavior change which have applied to a wide range of health behaviors. These behaviors include: smoking cessation, exercise adoption, sun protection, dietary fat reduction, condom use, adherence to mammography screening, medication adherence, stress management, and substance abuse cessation, to name just a few.25-29 These problem behaviors are important from both a clinical and a public health standpoint because they are strongly associated with increased morbidity, mortality, and with decreased quality of life. The Transtheoretical Model (TTM) is a model of intentional behavior change that has produced a large volume of research and service across a wide range of problem behaviors and populations.25-29 This model describes the relationships among: stages of change; processes of change; decisional balance, or the pros and cons of change; situational confidence, or self-efficacy in the behavior change; and situational temptations to relapse. Table 4 describes all the constructs that collectively comprise the TTM.

This model has several advantages over other models. First, it describes behavior change as a process, as opposed to an event. Then, by breaking the change process down into stages and studying which variables are most strongly associated with progress through the stages, this model provides important tools for both research and intervention development. Secondly, its explicit focus on measurement of constructs has provided a strong foundation for the model. Across different problem behaviors and populations, different variables have been associated with stage movement for each stage of change.30

Table 4. Transtheoretical Model Constructs



Stages of Change


No intention to take action within the next 6 months


Intends to take action within the next 6 months


Intends to take action within the next 30 days and has taken some behavioral steps in this direction


Has changed overt behavior for less than 6 months


Has changed overt behavior for more than 6 months

Decisional Balance


The benefits of changing


The costs of changing



Confidence that one can engage in the healthy behavior across different challenging situations


Temptation to engage in the unhealthy behavior across different challenging situations

Process of Change

Consciousness Raising

Finding and learning new facts, ideas, and tips that support the healthy behavior change

Dramatical Relief

Experiencing the negative emotions (fear, anxiety, worry) that go along with unhealthy behavioral risks


Realizing that the behavior change is an important part of one’s identity as a person

Environmental Reevaluation

Realizing the negative impact of the unhealthy behavior, or the positive impact of the healthy behavior, on one’s proximal social and/or physical environment


Making a firm commitment to change

Helping Relationships

Seeking and using social support for the healthy behavior change


Substitution of the healthier alternative behaviors and/or cognitions for the unhealthy behavior

Reinforcement Management

Increasing the rewards for the positive behavior change and/or decreasing the rewards of the unhealthy behavior

Stimulus Control

Removing reminders or cues to engage in the unhealthy behavior and/or adding cues or reminders to engage in the healthy behavior

Social Liberation

Realizing that social norms are changing in the direction of supporting the healthy behavior change

These TTM findings inform the design of individualized, stage-matched, expert system interventions (see below) that target those variables most predictive of progress for individuals at each stage of change. One aspect of this model that often goes unrecognized is that it is the processes of change that drive movement through the stages of change.31 Thus, although commonly referred to as the "Stages of Change Model" since "stage" is the core construct around which other model constructs are organized, this is a misnomer since it focuses attention on only one construct from this multidimensional model. Naturally, model-based interventions are multidimensional as well. TTM research has found remarkable similarities across different kinds of behavior changes. We have found repeatedly that the stages of change have predictable relationships with the pros and cons of behavior change, confidence in behavior change, temptation to relapse, and the processes of change.

Stages of Change

Individuals do not change their behavior all at once; they change it incrementally or stepwise in stages of change. The stages most commonly used across research areas include: Precontemplation, Contemplation, Preparation, Action, and Maintenance. Individuals do not typically move linearly from stage to stage, but often progress and then recycle back to a previous stage before moving forward again. This change process is conceptualized most meaningfully as a spiral, which illustrates that even when individuals do recycle to a stage they've been in before, they may still have learned from their previous experiences.

Precontemplation describes individuals who for many reasons do not intend to change within the next six months. Some of these individuals may want to change at some future time, but just not within the next six months. Others may not want to change at all and, in fact, may be very committed to their problem behavior (e.g., a lifelong smoker or someone who regularly cultivates a deep tan).

Contemplation describes individuals who are thinking about changing their problem behavior within the next six months. They are more open to feedback and information about the problem behavior than their counterparts in Precontemplation.

Individuals in the Preparation stage are committed to changing their problem behavior soon, usually within the next 30 days. These people have often tried to change in the past and/or have been practicing change efforts in small steps to help them get ready for their actual change attempt.

The Action stage includes individuals who have changed their problem behavior within the past six months. The change is still quite new and their risk for relapse is high, requiring their constant attention and vigilance.

Maintenance stage individuals have changed their problem behavior for at least six months. Their change has become more of a habit, and their risk for relapse is lower, but relapse prevention still requires some attention, although somewhat less than for individuals in Action.

Processes of Change

The processes of change describe the ten cognitive, emotional, behavioral, and interpersonal strategies and techniques that individuals and/or change agents (therapists, counselors) use to change problem behaviors.25,26 Research has demonstrated that successful behavior change depends upon the use of specific processes at specific stages.32-35 TTM-based research has consistently found that different processes are used to progress to different stages. Thus, the processes mediate the transitions from stage to stage and can represent important intermediate outcomes of interventions. The processes of change are also ideal tools for process-to-outcome research and in many ways provide the foundation for TTM expert system interventions (see below). The processes of change are consistent with many SCT constructs and are quite similar to most conceptions of coping behaviors as well.36

Many studies across problem behaviors35,37 have found that the ten most used processes of change are organized into two higher order clusters of processes: the experiential processes -- Consciousness Raising, Dramatic Relief, Self-Reevaluation, Environmental Reevaluation, and Social Liberation; and the behavioral processes -- Helping Relationships, Counterconditioning, Reinforcement Management, Stimulus Control, and Self Liberation. The experiential set of processes are most often emphasized in earlier stages (Precontemplation, Contemplation, and Preparation) to increase intention and motivation; and the behavioral set of processes are most often utilized in later stages (Preparation, Action, and Maintenance) as observable behavior change efforts get underway and need to be maintained.

Decisional Balance

Decisional Balance, or the pros and cons of behavior change, describes the importance or weight of an individual's reasons for changing or not changing. The pros and cons relate strongly and predictably to the stages of change.38,39 These are the decision-making components of the TTM and also serve as two important intervening, or intermediate outcome variables. Individuals' decisions to move from one stage of change to the next are based on the relative weight given to the pros and cons of adopting the healthy behavior. The pros are the positive aspects of changing behavior, or the benefits of change (reasons to change). In contrast, the cons include the negative aspects of changing behavior, or barriers to change (reasons not to change). These two dimensions have been consistently supported by studies across many different problem behaviors in TTM-based research.39 Characteristically, the pros of healthy behavior are low in the early stages and increase across the stages of change, and the cons of the healthy behavior are high in the early stages and decrease across the stages of change.

Due to the recent popularization of the interactional stress approach, measures which assess variables that influence individual appraisal and characteristics of the invoked response, have a relatively short history in comparison to stressor and response type measures.26 One of the most promising interactional measures is the Derogatis Stress Profile (DSP).38 The DSP consists of 70 items designed to interactively measure environmental, affective, and personality characteristics of the stress equation. The DSP has moderate to high test-retest reliability and high internal consistency and predictive validity.39 Other interactional stress measures include the Jenkins Activity Survey (JAS), which assesses Type A behavioral patterns, and the Ways of Coping Checklist (WACC), which measures methods individuals use to cope with stressful demands.26

The pros and cons are particularly useful when intervening with individuals in early stages of change. Decisional balance is an excellent indicator of an individual's decision to move out of the precontemplation stage. The relationship between the stages of change and decisional balance has been shown to be remarkably consistent across at least 12 different problem behaviors.39 Not only has the relationship between stage and the pros and cons been replicated across problem behaviors, but the magnitude of the change across the stages of change has been replicated as well. Based on these data, the strong and weak principles of behavior change were formulated.38 The strong principle states that in progressing from precontemplation to action, the pros of change generally increase by about one standard deviation, whereas the weak principle states that correspondingly, the cons of change tend to decrease by about one-half of a standard deviation.

The TTM pros and cons constructs are quite similar to those also proposed by both the HBM (benefits/barriers) and the TRA/TPB (benefits/costs); and the evidence presented by Prochaska and colleagues39 across 12 problem behaviors does provide some support for all three models. However, only the TTM proposes the specific relationship between these constructs and the stages of change. Also, importantly, the TTM has gone beyond mere specification of components to deductively hypothesize the degree of change in the pros and cons that occurs from Precontemplation to Action across problem behaviors.38 This is an important, innovative step for the TTM and for the development of the science of behavior change in general.

Situational Confidence and Temptations

The self-efficacy construct utilized in the TTM40 integrates the models of self-efficacy proposed by Bandura,18,22 and the coping models of relapse and maintenance described by Shiffman.41 These variables have undergone considerable elaboration over time, with situational temptation to engage in the unhealthy behavior often viewed as an equally important companion construct to the more commonly used situational confidence measures. Confidence and temptation function inversely across the stages, and temptation predicts relapse better. Research has demonstrated that both the confidence and temptation constructs can be conceptualized psychometrically as unifactorial and/or multifactorial. Structural modeling analyses have repeatedly revealed a global higher order construct (confidence or temptations) which is comprised of several lower order situationally determined components.40,42,43 The lower order situational factors depend more strongly upon the problem behavior than the higher order construct. A global score is often useful as a general screening tool, while the situational subscale scores provide useful information for targeting intervention feedback to individuals at different stages of change.

Confidence and temptation both vary across the stages of change, with confidence rising and temptation decreasing across longitudinal profiles of smokers.30 A moderate, reciprocal relationship (r = -.60) has been found between temptation and confidence for both smoking cessation and safer sex behaviors.40,43 Confidence is typically lowest in the Precontemplation stage, since individuals have little performance feedback and/or little interest in change. Confidence is higher during Contemplation, outperforming demographic variables in its ability to predict movement into Preparation and Action stages.44 Even in the Maintenance stage where subjects have successfully altered the problem behavior for at least six months, temptation is one of the best predictors of relapse and recycling to earlier stages of change.45

Expert Systems

An expert system computer program mimics or codifies the reasoning of human experts. The program uses standardized decision rules or algorithms for assessment and providing feedback and applies those algorithms consistently. A TTM expert system is an integrated assessment and intervention delivery computer program.46-49 Expert systems have been used with many different populations and have been found to be effective for smoking cessation,33,50 sun protection,51,52 dietary fat reduction53 and mammography screening.54 The more recent development of multimedia expert systems provides nearly immediate feedback to respondents, who sit at the computer and completes a series of questions followed by feedback. Participants respond to several series of questions interspersed with feedback on different TTM constructs.47,48

Measurement and Research Foundations

What is not obvious from most descriptions of the TTM is the careful attention to measurement development and validation that is taught and practiced by TTM-Model contributors and originators.54,55 Construction of measures based on the TTM have typically employed the sequential methods of scale development described by Jackson56,57 and Comrey.58 Initial item pools are generated based on theoretical construct definitions. Many of the items are adapted from existing instruments, but item content is then modified to more closely reflect the problem behavior and the language used by the population being studied. This is necessary for intervention development, as well as being an important foundation for any research project. There is a large and increasing volume of research using this model.

Summary/Future Directions

We have presented each of four models of behavior change, with a clear emphasis on the TTM. The TTM has an explanatory advantage since it was conceived later than the other models. The TTM was clearly not conceived in a vacuum. As part of their "Transtheoretical" strategy, model originators consciously incorporated and built upon the strengths of their predecessors. As put so well by Isaac Newton in his letter to Robert Hooke (February 5, 1676), "If I have seen further Éit is by standing on the shoulders of giants." Also, importantly, the TTM is not a fixed entity. It must grow and develop over time incorporating and responding to new promising ideas and new challenging data. It is a sign of our times that other theories are now utilizing this same strategy and drawing upon TTM variables, especially stage of change, to integrate within their own framework. Others are also now using various eclectic theoretical frameworks to develop tailored feedback systems.59-62 We have made our biases clear. It is now up to the reader to investigate the strength of the evidence further and to keep these questions in mind as he or she evaluates and compares competing theories:

• How well does this model describe health behavior change?
• How parsimonious is this theory?
• How much variance is accounted for in studies applying this model?
• How much intervention development guidance is provided by this theory?
• How useful is this model in practice?
• How effective in practice are interventions based on this theory?
• How well measured or how clear are theoretically defined constructs?
• How well specified and tested are theoretically defined mediating mechanisms?

These are the questions that the science of behavior change will ultimately use to decide which models and which model-based components provide the most useful description of the health behavior territory.

Introduction Health Belief Model Theory of Reasoned Actions Social Cognitive Theory Self-Efficacy Transtheoretical Model References
1. Glantz K, Lewis FM, Rimer BK, eds. Health Behavior and Health Education: Theory, Research, and Practice. 2nd ed. San Francisco, CA: Jossey-Bass, Inc.; 1997.
2. Cummings KM, Becker MH, Maile MC. Bringing the models together: an empirical approach to combining variables used to explain health actions. Journal of Be havioral Medicine. 1980;3:123-45.
3. Fishbein M, Bandura A, Triandis HC, Kanfer FH, Becker MH, Middlestadt SE. Factors influencing behavior and behavior change. In: Baum A, Reveson T, and Singer J, eds. Handbook of Health Psychology. Hillsdale, NJ: Lawrence, Erlbaum, and Associates; in press.
4. Weinstein ND. Testing four competing theories of health-protective behavior. Health Psychology. 1993;12:324-33.
5. Becker MH. The health belief model and personal health behavior. Health Education Monographs. 1974;2:324-473.
6. Becker MH, Maiman LA. Sociobehavioral determinants of compliance with health and medical care recommendations. Medical Care. 1975;13:10-24.
7. Fishbein M, Bandura A, Triandis HC, Kanfer FH, Becker MH, Middlestadt SE. Factors influencing behavior and behavior change. Final Report of the Theorist's Workshop, Washington, DC; 1991.
8. Janz NK & Becker MH. The health belief model: a decade later. Health Education Quarterly. 1984;11:1-47.
9. Rosenstock IM. The health belief model: explaining health behavior through expectancies. In: Glanz K, Lewis FM, Rimer BK, eds. Health Behavior and Health Education: Theory, Research, and Practice. San Francisco, CA: Jossey-Bass; 1990:39-62.
10. Rosenstock IM. The health belief model and preventative health behavior. Health Education Monographs. 1974;2:354-86.
11. Kirscht JP. The health belief model and illness behavior. Health Education Quarterly. 1974;11: 1-47
12. Strecher VJ, Rosenstock IM. The health belief model. In: Glantz K, Lewis FM, Rimer BK, eds. Health Behavior and Health Education: Theory, Research, and Practice.,/i> 2nd ed. San Francisco, CA: Jossey-Bass; 1997:41-59.
13. Baranowski T. Beliefs as motivational influences at stages in behavior change. International Quarterly of Community Health Education. 1992;13:3-29.
14. Ajzen I, Fishbein M. Understanding Attitudes and Predicting Social Behavior. Englewood Cliffs, NJ: Prentice-Hall; 1980.
15. Fishbein M, Ajzen I. Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research. Reading, MA: Addison-Wesley; 1975.
16. Montaño DE, Kasprzyk D, Taplin SH. The theory of reasoned action and theory of planned behavior. In: Glantz K, Lewis FM, Rimer BK, eds. Health Behavior and Health Education: Theory, Research, and Practice. 2nd ed. San Francisco, CA: Jossey-Bass, Inc; 1997: 85-112.
17. Sonstroem RJ. Psychological models. In: Dishman R, ed. Exercise Adherence. Champagne, IL: Human Kinetics Publishing; 1988: 125-54.
18. Bandura A. Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ: Prentice-Hall; 1986.
19. Perry C, Baranowski T, Parcel G. How individual's, environments, and health interact: Social Learning Theory. In: Glanz K, Lewis FM, Rimer BK, eds. Health Behavior and Health Education: Theory, Research, and Practice. San Francisco: Jossey-Bass; 1990: 161-86.
20. Baranowski T, Perry CL, Parcel GS. How individuals, environments, and health behavior interact: Social Cognitive Theory. In: Glantz K, Lewis FM, Rimer BK, eds. Health Behavior and Health Education: Theory, Research, and Practice. 2nd ed. San Francisco, CA: Jossey-Bass, Inc; 1997: 153-78.
21. Ockene IS, Ockene JK. Helping patients to reduce their risk of coronary heart disease: an overview. In: Ockene IS and Ockene JK, eds. Prevention of Coronary Heart Disease. Boston, MA: Little, Brown and Company; 1992.
22. Bandura A. Self-efficacy: toward a unifying theory of behavior change. Psychological Review. 1977;84:191-215.
23. Strecher VJ, DeVellis BM, Becker MH, Rosenstock IM. The role of self-efficacy in achieving health behavior change. Health Education Quarterly. 1986;13:73-91.
24. DiClemente CC. Self-efficacy and the addictive behaviors. J of So and Clin Psychol. 1986;4:302-15.
25. Prochaska JO, DiClemente CC. Stages and processes of self-change in smoking: towards an integrative model of change. Journal of Consulting and Clinical Psychology. 1983;51:390-395.
26. Prochaska JO, DiClemente CC. Common processes of change for smoking, weight control, and psychological distress. In: Shiffman S, Wills T, eds. Coping and Substance Abuse. New York: Academic Press; 1985:345-64.
27. Prochaska JO, Norcross JC, Fowler JL, Follick MJ, Abrams DB. Attendance and outcome in a worksite weight control program: processes and stages of change as process and predictor variables. Addict Behav. 1992;17:35-45.
28. Prochaska JO, Redding CA, Harlow LL, Rossi JS, Velicer WF. The Transtheoretical Model and HIV prevention: a review. Health Education Quarterly. 1994;21:471-86.
29. Prochaska JO, Velicer WF. The Transtheoretical Model of health behavior change. Am J of Health Promotion. 1997;12:38-48.
30. Prochaska JO, Velicer WF, Guadagnoli E, Rossi JS, DiClemente CC. Patterns of change: dynamic typology applied to smoking cessation. Multivariate Behav Res. 1991;26:83-107.
31. Prochaska JO, DiClemente CC. The Transtheoretical Approach: Crossing the Traditional Boundaries of Therapy. Homewood, IL: Dow Jones/Irwin; 1984.
32. Marcus BH, Rossi JS, Selby VC, Niaura RS, Abrams DB.The stages and processes of exercise adoption and maintenance in a worksite sample. Health Psychology. 1992;11:386-95.
33. Prochaska JO, DiClemente CC, Velicer WF, & Rossi JS. Standardized, individualized, interactive and personalized self-help programs for smoking cessation. Health Psychology. 1993;12:399-405.
34. Redding CA, Rossi JS. The processes of safer sex adoption. Annals of Behavioral Medicine. 1993;15:S106. (Abstract)
35. Rossi JS. Common Processes of Change Across Nine Problem Behaviors. Paper presentation at the 100th annual convention of the American Psychological Association. Washington, DC; August, 1992.
36. Rimer BK. Perspectives on intrapersonal theories of health behavior. In: Glantz K, Lewis FM, Rimer BK, eds. Health Behavior and Health Education: Theory, Research, and Practice, 2nd ed. San Francisco, CA: Jossey-Bass, Inc; 1997:139-47.
37. Prochaska JO, Velicer WF, DiClemente CC, Fava J. Measuring processes of change: applications to the cessation of smoking. J Consult and Clin Psychol. 1988;56:520-8.
38. Prochaska JO. Strong and weak principles for progressing from precontemplation to action based on twelve problem behaviors. Health Psychology. 1994;13:47-51.
39. Prochaska JO, Velicer WF, Rossi JS, et al. Stages of change and decisional balance for twelve problem behaviors. Health Psychology. 1994;13:39-46.
40. Velicer WF, DiClemente CC, Rossi JS, Prochaska JO. Relapse situations and self-efficacy. Addict Behav. 1990;15:271-83.
41. Shiffman S. A cluster analytic classification of smoking relapse episodes. Addict Behav. 1986;11:295-307.
42. Clark MM, Abrams DB, Niaura RS, Eaton CA, Rossi JS. Self-efficacy in weight management. J of Consult and Clin Psychol. 1991;59:739-44.
43. Redding CA, Rossi JS. Testing a model of situational self-efficacy for safer sex among college students: stage and gender-based differences. Psychology & Health. In press.
44. DiClemente CC, Prochaska JO, Fairhurst S, Velicer WF, Velasquez M, Rossi JS. The process of smoking cessation: an analysis of precontemplation, contemplation and preparation stages of change. J of Consult and Clin Psychol. 1991;59:295-304.
45. Redding CA, Rossi JS, Fava JL, et al. Dynamic factors in the maintenance of smoking cessation: a naturalistic study. Proceedings of the Society of Behavioral Medicine. 1989;10;170 (Abstract)
46. Pallonen UE, Velicer WF, Prochaska JO, Rossi JS, Bellis JM, Tsoh JY, Migneault JP, Smith NF, & Prokhorov AV. Computer-based smoking cessation interventions in adolescents: description, feasibility, and six-month follow-up findings. Substance Use & Misuse. 1998;33:935-65.
47. Redding CA, Prochaska JO, Pallonen UE, et al. Transtheoretical individualized multimedia expert systems targeting adolescents' health behaviors. Cognitive and Behav Practice. In press.
48. Velicer WF, Prochaska JO, Bellis JM, et al. An expert system intervention for smoking cessation. Addict Behav. 1993;18:269-90.
49. Prochaska JO, Redding CA, Evers K. The transtheoretical model and stages of change. In: Glantz K, Lewis FM, Rimer BK, eds. Health Behavior and Health Education: Theory, Research, and Practice. 2nd ed. San Francisco, CA: Jossey-Bass, Inc. 1997:60-84.
50. Velicer WF, Prochaska JO, Fava JL, Laforge RG, Rossi, JS. Interactive versus non-interactive interventions and dose-response relationships for stage-matched smoking cessation programs in a managed care setting. Health Psychology. In press.
51. Weinstock MA, Rossi JS. The Rhode Island Sun Smart Project: a scientific approach to skin cancer prevention. Clinics in Dermatology. In press.
52. Weinstock MA, Rossi JS, Redding CA, Maddock JE. Randomized trial of intervention for sun protection among beachgoers. Journal of Investigative Dermatology. 1998; 110:589. (Abstract) and Journal of Dermatological Science. In press.
53. Greene GW, Rossi SR, Rossi JS, et al. Efficacy of an expert system intervention for dietary fat reduction. Manuscript in review.
54. Velicer WF, Prochaska JO, Fava JL, Norman GJ, Redding CA. Smoking and stress: applications of the Transtheoretical Model of Behavior Change. Homeostasis. 1998;38: 216-33.
55. Rossi JS, Rossi SR, Velicer WF, Prochaska JO. Motivational readiness to control weight. In: Allison DB, ed. Handbook of Assessment Methods for Eating Behaviors and Weight-related Problems: Measures, Theory, and Research. Thousand Oaks, CA: Sage; 1995:387-430.
56. Jackson DN. The dynamics of structured personality tests. Psychol Review. 1971;78:229-48.
57. Jackson DN. A sequential system for personality scale development. In: Spielberger CD, ed. Current Topics in Clinical and Community Psychology. New York, NY: Academic Press; 1970;61-96.
58. Comrey AL. Factor-analytic methods of scale development in personality and clinical psychology. J of Consult and Clin Psychol. 1988;56:754-61.
59. Brug J, Glantz K,Van Assema P, Kok G, van Breukelen GJP. The impact of computer-tailored feedback and iterative feedback on fat, fruit, and vegetable intake. Health Education & Behavior. 1998;25:517-31.
60. Campbell MK, DeVellis BM, Strecher VJ, Ammerman AS, DeVellis RF, Sandler RS. Improving dietary behavior: the effectiveness of tailored messages in primary care settings. Am J of Public Health. 1994;84:783-7.
61. Skinner CS, Strecher VJ, Hospers HJ. Physician recommendations for mammography: do tailored messages make a difference? Am J of Public Health. 1994;84:43-9.
62. Strecher VJ, Kreuter M, Boer DJ, Kobrin S, Hospers HJ, Skinner CS. The effects of computer tailored smoking cessation messages in family practice settings. The Journal of Family Practice. 1994;39:262-70.

[Reprint (PDF) Official Layout of this article]

Copyright © 2000 by IEJHE.