Work absence and return 2: Predictors of early retirement: a Dutch prospective cohort study. Oral presentations: Day 2: Thursday, September 8, 2011. 22nd International Conference on Epidemiology in Occupational Health EPICOH 2011 September 7-9, 2011, Oxford, UK

Objectives Despite the current debate on early retirement, relatively few longitudinal studies investigated determinants of (non-disability) early retirement. The aim of the present study was to identify (changes in) risk factors that predict early retirement. Methods Data of the Netherlands Working Conditions Cohort Study were used. Employees aged 55–63 years filled out questionnaires at baseline (n=2845), after 12 months (n=1829), and after 24 months of follow-up (n=1501). Generalised estimation equations (12-month time-lag model) were used to study predictors of retirement before the official retirement age of 65. In addition, logistic regression analysis was applied to study the influence of changes in independent variables during the first 12 months of follow-up on early retirement during the second 12 month period. Results In total, 200 employees retired early. Early retirement was predicted by older age and working in the same job for a longer period of time in the multivariate analysis, whereas employees with high mental demands or lower satisfaction with their salary were less likely to retire early. With respect to changes in risk factors, the occurrence of enterprise restructuring and a decrease in job demands predicted early retirement in the multivariate analysis. Employees that reported an increase in support from colleagues less often retired early. (Change in) health was not a significant predictor. Conclusions Work-related factors, and sudden changes in these factors, and financial factors predict early retirement in employees. A healthy social climate at work, challenging work, and financial incentives might support the prolongation of working life