Social ties and survival among the elderly in Tampere, Finland.

The association between social ties and survival was assessed using a stratified sample of 1060 elderly aged 60-89 years in the city of Tampere, Finland. During the 6.5-year follow-up, 240 men and 153 women died. Compared with married men, the death rate ratio among unmarried men was 1.7 and among widowers 1.2. The respective ratios for women were 1.2 and 1.2. Vicinity of children, living alone, loneliness, social contacts, and social participation were used as indicators of social ties. Men and women were analysed separately. None of the indicators were significant predictors of survival in proportional hazard analyses, after adjusting for age, perceived health, functional ability and occurrence of a disabling disease at baseline. The relative hazards ranged between 0.85 and 1.38, and all 95% confidence intervals included unity. However, when social participation was entered into the models as a continuous variable, it was strongly associated with increased survival in both sexes. Social participation is probably not protective as such, but it may rather reflect way-of-life, which is characterized by social competence or 'control of destiny'.