Perceived social support among depressed elderly, middle-aged, and young-adult samples: cross-sectional and longitudinal analyses.

BACKGROUND: A number of studies have concluded that the perceived quality of support is more strongly associated with mental health than with the actual structure of personal networks. This study examined clinical, historical, and phenomenological variables associated cross-sectionally and longitudinally with perceived social support.
METHODS: Participants included elderly, middle-aged, and young-adult depressed samples derived from the Duke Clinical Research Center for the Study of Depression in Late Life.
RESULTS: Cross-sectional multivariate analyses revealed that perceived social support was: (1) for the elderly associated with pessimistic thinking, being divorced, having strange ideas, the degree of social interaction, and instrumental support; (2) for middle-age associated with dysthymia, divorce, pessimistic thoughts, social interaction, and instrumental support; and (3) among young adults with instrumental support only. Longitudinal multivariate analyses indicated that only perceived social support at Time 1 predicted perceived social support 1 year later among elderly and middle-aged subjects, whereas only instrumental support predicted perceived social support 1 year later among the young-adult sample. LIMITATIONS: The small number of subjects among the young-adult sample limit conclusions regarding this group. In addition, only patients provided data. Future studies should consider using multiple informants to enhance the accuracy of reported social support.
CONCLUSIONS: Our findings indicate that in addition to whatever else they do for depressed patients, clinicians must endeavor to address relationship or social support difficulties, especially in the elderly.