The purpose of this study was to determine if social support intervention would improve the support experienced by stroke survivors and if improvement would result in better psychosocial outcome. Subjects were drawn from a community-based sample that had received rehabilitation services in the hospital and at home. The study was a randomized trial (n = 48 experimental group, n = 40 control group), and measures were taken at entry into the study, immediately after intervention (or, at a comparable time for the control group), and again 3 months later. No significant differences were found between groups either on social support measures or psychosocial outcomes. Secondary analyses showed that significant changes in support had been experienced within the sample as a whole in relation to the subjects' gender, living situation, marital status, and employment status. Findings are discussed in terms of methodologic and programmatic concerns, and recommendations are made for further research with this potentially valuable intervention.