Background: Much evidence has accumulated over the last three decades that low social support is related to both mental and physical health. Despite this large and convincing literature, reviewers have noted that there exists remarkably little evidence that social support can be increased by an appropriate intervention. This study reports on the development and evaluation of a new intervention for social support which takes account of the stress-buffering and direct effect models. Method: Eighty-one individuals scoring low on social support were randomly allocated to the intervention or a waiting-list control condition. Treatment consisted of 10 weekly sessions administered in a group format, and 49 participants (nine males) completed assessments at the beginning and end of a 10-week period, and at 10-week follow-up (intervention condition only). Results: The intervention proved to be successful at increasing functional support but not structural support. The intervention was also successful in increasing the social skill of self-disclosure, and decreasing depression. Gains made between pre- and post-treatment were maintained at 10-week follow-up. Conclusions: Based on published analyses of the effects of social support on health, the results imply that the intervention would be useful for stress-buffering purposes, but not for the general health-promoting effects that are associated with good social integration.