Association of self-reported religiosity and mortality in industrial employees: the CORDIS study.

This study examined the association between self-reported religiosity and mortality in industrial employees, while controlling for workplace and socioeconomic factors. Subjects were 3638 Jewish Israeli males who participated in a 12-year follow-up study. During this period 253 deaths were recorded. The prevalence of negative workplace and sociodemographic factors: lower education, non-European origin, heavy physical work, blue-collar jobs and adverse job and environmental conditions, was highest among religious employees, and lower in traditional and nonreligious employees in descending order. Using Cox's proportionate hazard model an age by religiosity interaction on mortality was uncovered. In younger employees (age <55 years) religiosity was associated with lower adjusted mortality, after controlling for negative workplace and sociodemographic factors. Compared with nonreligious employees, the hazard ratios for the religious and traditional employees were: 0.64 (p=0.016) and 0.39 (p=0.118), respectively. In older employees (age >/=55 years), religiosity was associated with higher adjusted mortality. The corresponding hazards ratios were 1.69 (p=0.011) and 1.08 (p=0.004), even after controlling for the above possible confounding variables. It was concluded that religiosity had a protective effect on mortality in younger employees, but the reverse was true for older employees. This opposite trend could not be explained by negative sociodemographic and workplace conditions. The possibility of involvement of yet another potent factor of social isolation was discussed.