Depression is a major cause of suicide among the elderly. Few previous community-based interventions against depression have reduced the suicide rate. This study aims to evaluate outcomes of a community-based program to prevent suicide among the elderly using a quasi-experimental design with a neighboring reference group. The program, including depression screening with follow-up and health education through primary care and public health nursing, was implemented for 10 years in Matsudai town, a rural area of Japan (population 6,015; suicide rate per 10(5) [65-year-olds] for males 290.6, and for females 361.3). Changes in the suicide risk were estimated by the incidence rate ratio (IRR). The female risk of completing suicide in the intervention area was reduced by 70% (age-adjusted IRR: 0.30; 95% CI: 0.14-0.67), while there was no change in the risk for males in the intervention area nor for males or females in the reference area (Kawanishi town: population 9,425; elderly suicide rate for males 212.2, females 151.9). A ratio of the female IRR in the intervention area to that in its prefecture was also estimated at 0.45 (95% CI: 0.19-0.97), showing that the reduction of suicide risk in the intervention area was greater than the historical trend. A community intervention against suicide using management of depression with nonpsychiatric, primary health care would be effective for elderly females, but not males.