In this paper, an attempt has been made to provide an overview of the major issues facing cancer researchers who are interested in the conceptualization and measurement of social support. Taken as a whole, the literature reflects a growing appreciation for the complexity of the social-support construct, and a need for greater conceptual and research specificity. The following issues were discussed: (1) the importance of measuring support in a way that will permit assessment of distinct types of support (e.g., emotional support, advice) by distinct providers (e.g., spouse, physician) since available evidence suggests that the impact of support is strongly affected by these factors; (2) the importance of using a multimethod approach to support measurement which includes both structural aspects and subjective assessments of the adequacy of the support network; and the problems inherent in assuming that structural variables alone are an indication of the adequacy of support; (3) the importance of considering negative as well as positive support, since there is evidence suggesting that such behaviors may be very prevalent in the interactions between cancer patients and those in their support network that they may have strong influence on subsequent health outcomes; (4) the need to consider the perspective of the provider as well as the recipient in obtaining judgments of the effectiveness of particular kinds of support, given the evidence that providers sometimes make attempts to be supportive that are regarded as unhelpful; (5) the importance of focusing on actual utilization as well as availability of the support network, since an understanding of how people mobilize their support networks has important theoretical and practical implications; and, finally, (6) the importance of studying more explicit, specific behaviors that occur between cancer patients and those in their support network, as well as more general judgments regarding whether support is adequate. Investigators in the social support area agree that it is time to move beyond demonstrations of a relationship between support and health outcomes to a more careful explication of the processes underlying support. This review highlights the importance of designing research that will shed light on the precise mechanisms through which support influences health outcomes. Hopefully, such research will clarify how social support interacts with other important psychosocial variables such as coping strategies and self-esteem.(ABSTRACT TRUNCATED AT 400 WORDS)