or inhibit health behavior development and/or change and maintenance. Social support, peer pressure, and family environment are powerful reinforcers of behavior throughout the life cycle. Such sociocultural influences must be considered potential agents of change as we come to better understand and define their parameters.
Environmental factors, including characteristics of the physical setting as well as legal and policy issues, are also potent determiners of health behaviors. Access to exercise facilities, health assessment procedures, cigarette machines, the availability of "prudent diet" menus and information at school or workplace cafeterias, ambient air quality, noise level, lighting at work and home, the "corporate culture," the statutory regulations concerning smoking, seat belts, and bicycle helmets all influence health behavior and presumably health outcomes ( 16).
It has been argued, in fact, that in the absence of environmental changes, the likelihood of maintaining individual health behavior change is quite low ( 17). Thus, programs that focus on the intrapersonal and interpersonal levels without considering concomitant environmental modification are unlikely to achieve the desired long-term effects.
If this were not sufficiently complex, our model must also differentiate between the processes of health behavior development (e.g., smoking prevention, development of prudent dietary habits), health behavior change (shifting from healthdestructive to health-protective behaviors), and health behavior maintenance (adherence, preventing relapse). The technical approaches may be quite different in achieving the objectives in each area. Differences in populations and unique strategies based on the special needs of these populations may also apply.
Our experiences in the fields of learning and behavior modification at the clinical and small-group levels have also entered the public health domain through concepts of social marketing ( 18, 19), communication ( 20, 21), community organization ( 22), and social learning ( 23). These theoretical models designed to influence the lifestyle behaviors of populations (e.g., communities) are presently being tested in several community risk-reduction programs throughout Europe and the United States ( 24- 26).
In summary, there are exciting developments taking place in behavioral medicine-psychologists, physicians, nurses, health educators, epidemiologists, and many others have much to offer each other in seeking biobehavioral solutions to problems of health and illness. Using Neal Miller's classic statement on the subject as our watchword -- "we must be bold in what we try, but cautious in what we claim" ( 27) -- we can be sure of a long and viable future for this field.