Citation

Berkman LF, Melchior M, Chastang JF, Niedhammer I, Leclerc A, Goldberg M. Am. J. Epidemiol. 2004; 159(2): 167-174.

Affiliation

Department of Society, Human Development, and Health, Harvard School of Public Health, Boston, MA 02115, USA. lberkman@hsph.harvard.edu

Copyright

(Copyright © 2004, Oxford University Press)

DOI

unavailable

PMID

14718219

Abstract

The authors investigated associations between social integration and all-cause and cause-specific mortality among French employees of Electricity of France-Gas of France. A total of 12,347 men aged 40-50 years in 1989 and 4,352 women aged 35-50 years in 1989 comprised the sample. In age-adjusted survival analyses for all causes of death, men who were least socially integrated were 4.42 times as likely to die during follow-up (1993-1999) as those with the highest level of integration (p < 0.0001). After adjustment for age, occupation, smoking, alcohol consumption, body mass index, self-reported health, depressive symptoms, and region of France, relative risks for men ranging from the least socially integrated to the most socially integrated were 2.70 (95% confidence interval (CI): 1.17, 6.23), 1.95 (95% CI: 1.25, 3.04), and 1.37 (95% CI: 0.92, 2.04) in comparison with the most integrated men. In multivariate cause-specific analyses, isolated men had elevated risks of dying from cancer (relative risk = 3.60) and from accidents and suicide (relative risk = 3.54). Among women, in multivariate analyses, the relative risk was 3.64 (95% CI: 0.72, 18.58). The small number of deaths among women (n = 29) limited statistical power and prohibited cause-specific analyses. These results suggest that in this employed cohort of middle-aged men and women, social integration is an important predictor of mortality.

Language: en