The health belief model (HBM), prodigiously researched, has enjoyed sustained popularity amid evolving social norms, theories and models, and the recent developments of advanced technology influencing health behavior change. Developed by US Public Health Service social psychologists in the 1950s, the HBM was conceptualized to model the failure of individuals to engage in disease prevention or detection programs. Around 1974, and later in the 1980s, additional components (knowledge about the disease, self-efficacy or confidence in one’s ability to perform the action of interest, cues to action), such as those examining responses to diagnosed illnesses and symptoms and adherence to medical regimens, were added to the model (Janz et al. 2003).
The HBM has been studied both as separate components and in its entirety, with multiple illnesses and preventive health behaviors, using various multivariate analytic techniques. The four basic components of the HBM are: (1) perceived risk or susceptibility (defined as one’s perception of vulnerability to developing a disease); (2) perceived severity (seriousness of the illness or consequences of not completing the behavior of interest); (3) perceived benefits (perceived efficacy of the action or behavior change of interest) and perceived barriers (tangible and psychological obstacles to completing the action or behavior change of interest); and (4) cues to action (strategies to activate behavior change) (Janz et al. 2003). In later formulations of the model, self-efficacy, the confidence in one’s ability to take action or complete the steps of behavior change, was adapted from social cognitive theory (Bandura 1986). Socio-demographics and knowledge about the desired action or illness to be prevented were also added, and hypothesized to affect perceptions and subsequently indirectly influence individual behavior as a result.
Theoretical Foundation And Measurement
Rooted in cognitive and stimulus–response theories, the HBM can be understood as a value expectancy theory . A value expectancy theory assumes that individual people are goal-oriented, and that the behaviors they perform in response to their beliefs and values are undertaken to achieve some end. Value expectancy concepts, revised to fit within the paradigm of health behaviors, resulted in the present interpretation of the model. For example, examined within the context of breast cancer and screening using mammography, when individuals recognize breast cancer as serious (perceived severity), understand themselves to be at risk of developing breast cancer (perceived susceptibility), frame mammography as efficacious in detecting breast cancer (perceived benefits), identify few obstacles to having a mammogram (perceived barriers) and are confident in their ability to complete all the steps necessary to have a mammogram (perceived self-efficacy), they will have a mammogram.
Descriptive, cross-sectional research on health behavior change usually assesses perceptions and knowledge about a disease (e.g., cancer, diabetes, heart disease) and the behaviors that manage illness. Intervention studies investigate the effects of education about health promotion behavior using these key components of the HBM. For example, in a diabetes control or insulin adherence regimen study, education may be based on individuals’ pre-education perceptions and knowledge. By changing perceptions and knowledge through a prescribed communication program, researchers expect to see posteducation differences in perceptions and/or knowledge, and, correspondingly, an increase in adherence to insulin regimen.
There are many advantages to using the HBM as the guiding framework for behavior change research. First, the conceptualization of the various components of the model has been well established by application to a variety of diseases. Studies as varied as cancer screening, diabetes, exercise behavior, and smoking cessation have all been conducted using the HBM as the guiding theoretical framework. Most studies attempt to effect health behavior by changing cognitions about relevant actions. Variations of the model, including the addition of culturally oriented variables, health-care access, and health service factors, have been published. Results have been mixed, with less than half of the variance in health behavior explained by the HBM framework. For example, the HBM variables together explain less than 50 percent of the variance in mammography screening behavior, leaving almost 50 percent of the variance unexplained. Unfortunately, the majority of the studies assess only linear relationships between HBM variables and outcomes, leaving unanswered questions about possible mediating or moderating relationships between variables.
A second advantage of the HBM is that it can be adapted for use with other behavior change theories and models. For example, the key components of the transtheoretical model of change (TTM) and the HBM can be merged to predict or affect health behavior change. According to the TTM, behavior change is not a dichotomous event; an individual progresses through a series of stages while changing behavior (from not thinking about it, to thinking about it, to taking action, then maintaining the change; Glanz et al. 2003). At each stage, research indicates that perceptions (such as those from the HBM) may differ significantly, allowing communication to be tailored or targeted to each stage. In mammography screening, for example, a woman in pre-contemplation (not thinking about having a mammogram) may report low perceived susceptibility and low perceived benefits; a woman in contemplation (thinking about having a mammogram) may indicate she has higher barriers than those in pre-contemplation (Champion et al. 2007). Given this information, researchers may focus on changing different perceptions at different stages to encourage forward movement.
Another example of the HBM being integrated with other theories is with the precede– proceed model (PPM). The PPM is a nine-phase framework for guiding community-based health programs (Glanz et al. 2003). Various phases deal with the individual, the community, organizations, evaluation, and policy. At the individual level, predisposing factors are defined as antecedents that provide a rationale for behavior change. The variables under this factor are not specified but left to the discretion of the model user. Researchers have inserted perceptions from the HBM under predisposing factors; further, a cue to action may be seen as an enabling factor – defined as external resources that facilitate behavior change (at the individual level) (Menon et al. 2003). Health communication may then be aimed at affecting change in perceptions (predisposing factors) and providing a cue to action (enabling factor).
The HBM, however, also has limitations, especially in regard to accounting for cultural differences among the audience. As boundaries between countries blur and advanced technology makes cross-over even easier, it is important to consider health in the larger contexts of culture and globalization. The HBM makes no allowances for cultural variations in beliefs. The underlying assumption of the model is that individuals will engage in preventive health behavior given the right set of beliefs. However, in cultures where preventive health orientation is lacking, fatalistic views of life are prevalent, or religion influences beliefs about health and death, the HBM would predict little or no health behavior.
Furthermore, the HBM assumes that health promotion technology or health-care is available to individuals. However, in economies where preventive tests, such as mammograms and glucometers, are not available to people, regardless of perceptions, health behavior change cannot occur. Additionally, even in societies where health insurance exists and covers costs of health promotion, many individuals are underinsured or uninsured, and without access to health-care. In countries where free national health plans provide for all citizens, long waits and plan-stipulated delays may incur. In all such instances, cognitive perceptions may end up having little to do with a woman’s decision to have a timely mammogram or adhere to an insulin regimen.
Given the long and widespread use of the HBM, the operationalization of concepts within the model is widely variable, even when studying the same health behavior. Thus, comparison of the predictive value of the HBM across studies remains difficult and findings have been rather inconsistent. For example, perceived susceptibility to breast cancer has been reported to have had a positive, a negative, and no effect on mammography screening. The operational definitions of susceptibility were different in these reports.
A final limitation is that the model has rarely been extended to examine moderating or mediating effects. The majority of multivariate analyses have investigated HBM variables in linear regression models. There remains the possibility that the HBM variables may moderate or mediate the effect of each other or other variables from other theories.
The answer to such limitations lies in further adapting the HBM to accommodate cultural and system variables such as fatalism and health-care access, among others. Planned in-depth analysis of curvilinear relationships within the HBM may further add to its utility as the basis for health communication. While some HBM constructs may not directly influence behavior, they may still act as mediators or moderators of the relationships between beliefs and the outcome. For example, while fatalism may not directly influence mammography use, it may act as a mediator of the relationship of barriers or benefits to mammography use. Inconsistent results when using the HBM underscore the importance of standardization of operational definitions of HBM variables within various disease foci. Despite these constraints however, the HBM remains a popular model, guiding behavior change and health communications for over six decades.
- Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice Hall.
- Champion, V., Skinner, C. S., Hui, S., et al. (2007). The effect of telephone versus print tailoring for mammography adherence. Patient Education and Counseling, 65(3), 416 – 423.
- Giger, J. N., Davidhizar, R. E., & Fordham, P. (2006). Multi-cultural and multi-ethnic considerations and advanced directives: Developing cultural competency. Journal of Cultural Diversity, 13(1), 3 – 9.
- Glanz, K., Lewis, F. M., & Rimer, B. K. (2003). Health behavior and health education: Theory, research, and practice. San Francisco, CA: Jossey-Bass.
- Janz, N. K., Champion, V. L., & Strecher, V. J. (2003). The health belief model. In K. Glanz, F. M. Lewis, & B. K. Rimer (eds.), Health behavior and health education: Theory, research, and practice. San Francisco, CA: Jossey-Bass, pp. 45 – 66.
- Menon, U., Champion, V. L., Larkin, G. N., Zollinger, T. W., Gerde, P. M., & Vernon, S. W. (2003). Beliefs associated with fecal occult blood test and colonoscopy use at a worksite colon cancer screening program. Journal of Occupational and Environmental Medicine, 45(8), 891– 898.
- Menon, U., Champion, V., Monahan, P. O., Daggy, J., Hui, S., & Skinner, C. S. (2007). Health belief model variables as predictors of progression in stage of mammography adoption. American Journal of Health Promotion, 21(4), 255 –261.
- Murray, E., Burns, J., See, T. S., Lai, R., & Nazareth, I. (2005). Interactive health communication applications for people with chronic disease. Cochrane Database Systematic Review, October 19, (4), CD004274.