[Preventive effect of social support on mortality among elderly population: a longitudinal study]

The main objective of this study is to analyze the relationship between social support and the mortality among a non-institutionalized population cohort age 60 or above residing in Barcelona studied throughout the 1996-1999 monitoring period for the purpose of analyzing the relationship between the social support and the mortality of this cohort.

In 1996, a telephone survey was conducted of 755 individuals (316 males and 439 females) who had been surveyed in the 1992 Barcelona health survey and who, were age 60 or above that year. The questionnaire included socio-demographic, morbility and health condition, social support and social network-related variables. The deaths during the 1996-1999 period were recorded. Logic regression was used for analyzing the social support-mortality relationship.

The deaths for the period in question totaled 55 individuals (5.9% of the females and 9.2% of the males). Among the variables reflecting social support, for the males solely the situation of living with someone was found, in the bivariate analysis, to have a significant relationship to the mortality, which was higher for those males who lived with other family members, but not with their wives or female partner (OR = 3.7; IC 95% 1.4-9.6). Among the females, the existence of support from neighbors, the size of the family network, the number of contacts with the community network and the situation of living with someone else were related to the mortality in the bivariate analysis, although in the multivariate logic regression, solely the support of neighbors (OR = 3.6; IC: 1.1-11.1) was found to have a significant relationship.

The results are consistent with prior studies revealing a relationship between the social support variables and the mortality. The relationship between the different social support variables and the possible prevention mechanisms should be taken up in future studies.