The need for treating milder forms of depression has recently been of increased interest. This was a randomized, controlled study to evaluate the effects of telephone-based problem-solving therapy for mild depression. Comparison groups were a treatment-as-usual group and another group receiving stress-management training by telephone. From 1,742 family practice patients screened for depression, 54 with mild depression entered the study. Treatment was provided by experienced family practice nurses, trained and supervised in the treatments. The Hamilton Rating Scale for Depression was administered before and after the intervention period, and the Beck Depression Inventory and Duke Health Profile were administered at the end of the intervention period. Of the 36 subjects assigned to the problem-solving and stress-management groups, half dropped out early in the study. Five from the treatment-as-usual group were lost to follow-up. In the remaining subjects, there was a significant decrease in depression scores. There were no significant differences in the amount of decrease between the groups on any scores. The small sample and high dropout rate limit the interpretation of the findings. However, since all subjects tended to improve, regardless of treatment received, mild levels of depression may generally remit even without focal intervention, and watchful waiting may be a reasonable alternative for management.