Abstract
This paper examines an important area of personal health behavior in which the possible link between stress and health status is not primarily a biological one but a behavioral one. The review of empirical literature is organized around four topics: 1) the limited payoff from the search for stable sociodemographic and personality correlates of adherence; 2) the inadequacy of our current theoretical formulations; 3) the role of the doctor; and 4) the doctor-patient interaction, which emphasizes mutual expectations rather than a one-way transfer of information. The concluding remarks address themselves to the need for a firmer linkage with the conceptual and empirical literature on stress and coping.