Research on the effects of multiple roles on women's health has in the past been conducted within the context of two competing hypotheses: the scarcity hypothesis and the expansion hypothesis. Empirical evidence is more supportive of the expansion than the scarcity hypothesis, i.e., women who occupy several roles are healthier than those with few. However, this generalization obscures important health differentials related to types of roles occupied and attributes of those roles. Research on multiple roles is now shifting from examining numbers of roles to analyzing the effects of specific role combinations, patterns, and characteristics. Further research is needed to identify ways in which rewards and stresses within each role interact to produce health outcomes.