The authors assessed the effect of a control-relevant psychosocial intervention in 31 nursing home residents with either major depressive episode or minor depression. An initial group of 22 residents were randomized to either active treatment or waiting list. Four of 11 residents randomized to active treatment were deemed Responders, compared with 0 of 11 on the waiting list (P < 0.05). Of the total of 31 residents who participated in the intervention, 14 (45%) were deemed Responders during the intervention period. For these Responders, the Hamilton Rating Scale for Depression (Ham-D) and Geriatric Depression Scale scores improved significantly during the intervention. The improvement in the Ham-D was not sustained 2 months after intervention was terminated. These findings suggest that a psychosocial intervention enhancing socialization according to each resident's choice had a positive therapeutic impact on almost half of the nursing home residents with major or minor depression. However this effect could not be sustained by the residents without the support of the structured program.