Résumé :
|
[BDSP. Notice produite par INIST-CNRS vajY2R0x. Diffusion soumise à autorisation]. Introduction : Psychosocial factors have been evidenced worldwide to influence mortality. But there are less studies that have shown long-term prediction of these psychosocial dimensions. Comprehensive multidimensional models (1) have been conceptualised and tested for health inequalities, mortality and morbidity but seldom for long-term mortality. Methods : Data from three Belgian large-scale studies were analysed, namely the "Belgian Heart Disease Prevention Project" (BHDPP), which started in 1972, the "Physical Fitness Study" (PFS) in 1977 and the "Belgian Interuniversity Research on Nutrition and Health" (BIRNH) which started in 1982 but was restricted to working males aged between 35-59 years old. Predictions by various baseline psychosocial data of All-causes and Cause-specific Mortality, respectively after 18 (BHDPP), 17 (PFS) and 10 (BIRNH) mean years of follow-up, were assessed by means of multivariate survival analyses, using Cox proportional hazard models. Results : After adjustment for classical risk factors, social factors showed to have a predictive power for mortality : marital status for all-causes mortality, occupation for all-causes mortality, and cardiovascular mortality. Conclusions : Other psychosocial dimensions still need to be included, and evidently a psycho-socio-biomedical model should be used in order to be able to prevent disease and efficiently promote health and well-being in the working world.
|