Background: Depression is a major health problem for community-dwelling elderly adults. Since limited resources are available to decrease the high prevalence of depressive symptoms among the elderly adults, improved support for them can be provided if we can determine which intervention is superior in ridding depressive symptoms. Objective: To compare the effectiveness of the physical fitness exercise program and the cognitive behavior therapy program on primary (depressive symptoms) and secondary outcomes (6-min walk distance, quality of life, and social support) for community-dwelling elderly adults with depressive symptoms. Design and settings: A prospective randomized control trial was conducted in three communities in northern Taiwan. Participants: The elderly adults in the three communities were invited to participate by mail, phone calls, and posters. There were a total of 57 participants who had depressive symptoms and all without impaired cognition that participated in this trial. None of the participants withdrew during the 9 months of follow-up for this study. Methods: Fifty-seven participants were randomly assigned to one of the three groups: the physical fitness exercise program group, the cognitive behavior therapy (CBT) group, or the control group. The primary (Geriatric Depression Scale-15, GDS-15), and secondary outcomes (6-min walk distance, SF-36, and Inventory of Socially Supportive Behaviors scales, ISSB) were collected immediately (T2), at 3 months (T3), and at 6 months after the interventions (T4). Results: After the interventions, the CBT group participants demonstrated significantly lower symptoms of depression (p= 0.009) at T2 and perceived more social support from those around them (p< 0.001, <0.001 and =0.004, respectively) at three time-point comparisons than the control group. Moreover, after intervention, participants in the physical fitness exercise program group had decreased GDS-15 scores at three time-point comparisons (p= 0.003, 0.012 and 0.037, respectively), had a substantially greater 6-min walk distance (p= 0.023), a better quality of life (p< 0.001), and a better perceived social support at T2 (p< 0.001). Conclusions: Immediately after a 12-week intervention, there were significant decreases in depressive symptoms and more perceived social support amongst those in the CBT group. When considering the effectiveness in the decrease of depressive symptoms longer term, the increase in the 6-min walk distance and raising the patients' quality of life, physical fitness exercise program may be a better intervention for elderly adults with depressive symptoms.