Assessment of a short scale to measure social support among older people

Background: Social support strongly influences health, but research has bee n limited by the paucity of brief, inexpensive, psychometrically sound scal es suitable for use with older people. Method: The validity and reliability of the 11-item Duke Social Support Ind ex (DSSI) were tested among a random sample of 565 community dwelling peopl e aged 70 years and over. Results: A response rate of 76% was obtained. Participants had a mean age o f 76 years. Internal consistency using Cronbachs alpha for the overall inde x was 0.77. Test-retest reliability (n=117) scores ranged from 0.70 to 0.81 . Concurrent validity is supported by the DSSI's moderate to strong correla tions with the interview Schedule for Social Interaction. Construct validit y of the DSSI was supported by moderate correlations with health, quality o f life and loneliness. To further test construct validity a multiple regres sion model was used and 25.6% of the variance in social support was explain ed using three expected variables; self rated health, quality of life, and living arrangements. Conclusion: Strong evidence for reliability and validity of the 11-item Duk e Social Support Index support its use in aged care research and health pro motion strategies. Implications: The DSSI provides researchers with the opportunity to use a b rief measure of social support which has been assessed specifically with ol der Australians instead of single-item measures or scales which have limite d psychometric evidence.