O7D.3 Working conditions and health behaviour as causes of educational inequalities in self-rated health: an inverse odds weighting approach

Background This study aims to estimate to what extent working conditions and health behaviours mediate the increased risk of low educated workers to report a poor health. Methods Respondents of the longitudinal Survey of Health, Ageing, and Retirement in Europe (SHARE) in 18 European countries were selected aged between 50 years and 64 years, in paid employment at baseline and with information on education and self-rated health (n=15,126). Health behaviours and physical and psychosocial work characteristics were measured at baseline, while self-rated health was measured at 2 year follow up. We used loglinear regression models and Inverse Odds Weighting causal mediation analysis to estimate the total effect of low education on self-rated health and to decompose the effect into natural direct (NDE) and natural indirect effects (NIE). Results Lower educated workers were more likely to be in poor health compared to higher educated workers. The total effect of low education on self-rated health was RR=1.81 [95% CI 1.66–1.97]. For work conditions, having a physical demanding job was the strongest mediator, followed by lack of job control and lack of job rewards. NIE through working conditions was RR=1.16 [95% CI 1.06–1.25], explaining about 30% of educational inequalities in self-rated health. For health behaviour, body mass index and alcohol were the strongest mediators, followed by smoking. NIE though health behaviour was RR=1.14 [95% CI 1.07–1.20], explaining about 27% of educational inequalities in self-rated health. Conclusions Preventive interventions focusing on reducing physical work demands as well as improving health behaviour may contribute to reducing educational inequalities in self-rated health among workers in Europe.