Wilkins Kathryn
Health Statistics Division, Statistics Canada, Ottawa, Ontario, K1A 0T6.
Health Rep. 2003 May;14(3):21-34.
This article investigates the effect of social support on mortality among Canadian seniors.
The analysis is based on longitudinal household data from the National Population Health Survey (NPHS) for 2,422 people aged 65 or older in 1994/95. Vital status and date of death were established using data collected in 2000/01.
Multivariate proportional hazards models were used to study associations between four indicators of social support (marital status; social contacts; participation in organizations; and perceived emotional support) in 1994/95 and death by 2000/01. Separate analyses were performed for men and women.
When the influence of age, socio-economic status, stress, health-related behaviours and physical/mental health status was taken into account, no association between social support and mortality emerged for women, but such a relationship was evident for men. Married men had a 40% lower hazard of death, compared with their non-married counterparts. Participation in organizations also conferred a reduced likelihood of dying for men.
