44 Chapter 2 from work on physical activity (7). Among employees with physically demanding jobs, the loss of physical activity at work after exit from work was not compensated by leisure-time physical activity, because such compensation would require huge lifestyle adaptations (55, 56). This will likely result in an increased BMI, if no modifications are made in their eating patterns (53). Moreover, employees with a high SES often perceive time as a barrier for leisure-time physical activity during their working life (57). After exit from work, this barrier can dissolve as people with a high SES can more actively engage in leisure-time physical activity and thereby their BMI may decrease more than among people with a low SES. Methodological considerations A strength of this review is the inclusion of studies with a longitudinal study design, which enabled the investigation of health changes after exit from work across socioeconomic groups. Nevertheless, this is the first review investigating the effects of exit from work on health across socioeconomic groups and identified some important methodological considerations for future research. First, some studies using the same dataset were evaluated separately. Still, some overlap could exist between these studies and therefore, findings may have been multiplicated to some extent. Though, many aspects, i.e. study population, type of exit route, health outcome or SES indicator, varied between the studies using the same dataset, making it unlikely that findings are multiplicated. Second, the positive effects of exit from work among higher socioeconomic groups were mainly found after early/statutory retirement. Unfortunately, in the present review, we did not have sufficient studies to investigate the influence of unemployment or disability pension on health across socioeconomic groups. From previous literature, we know that the type of exit route plays a role in the course of health (7, 33). Hence, more research is needed to gain insight into the effects of unemployment or disability pension on health across socioeconomic groups. Third, all studies obtained a high score in the quality assessment. Still, the quality assessment did highlight a limitation. Many studies did not specify the exact timing of the work exit. The lack of specified information on the exact timing of this transition makes it difficult to relate changes in health to the transition itself. All studies measured health prior to exit from work to control for health at baseline, as health on itself can impact exit from work (i.e. endogeneity) (58). However, if the transition was measured over a period of a few years, health changes could either occur before the transition or as a consequence of the work exit. Thus, it remains unknown whether the change in health is caused by the work exit itself or whether the change in health already started before exit from work (58).
RkJQdWJsaXNoZXIy MTk4NDMw