Assessment of the abbreviated Duke Social Support Index in a cohort of older Australian women

Objectives: To assess the acceptability, reliability and validity of the 11‐item Duke Social Support Index (DSSI) in community‐dwelling older Australian women, and to describe its relationship with the women's sociodemographic and health characteristics. Methods: Women aged 70–75 years were randomly selected from the national Medicare database, with over‐sampling of rural and remote areas. The mailed survey included items about social support, Medical Outcomes Study Short Form Health Survey (SF‐36), health service use, recent life events and sociodemographics. Results: All DSSI items were completed by 94% of the 12 939 participants. Internal reliability was reasonable for 10 of the 11 DSSI items and its factors, social interaction (four items) and satisfaction with social support (six items; Cronbach's alpha of 0.8, 0.6, 0.8). The factor structure was consistent for subgroups of women: urban/non‐urban; English speaking/non‐English speaking background; married/widowed. Summed scores were highly correlated with factor scores and showed good construct validity. Higher social support was associated with better physical and mental health, being Australian born, more educated and better able to manage on income. Conclusion: Ten of the 11 DSSI items provided an acceptable, brief and valid measure of social support for use in mailed surveys to community‐dwelling older women.