High job exposures linked to lower return-to-work after rehabilitation using job-exposure matrices

Background

Administrative data from social insurance provide large samples and long follow-up periods but lack individual-level information on job exposures. However, work-related factors are crucial for stable return-to-work (RTW) among working-age populations. Job-exposure matrices (JEM) allow the assignment of occupation-specific exposures and may help bridge this gap in health services research.

Methods

We conducted a retrospective nationwide cohort study based on the Scientific Use File “Completed rehabilitation 2014–2021” of the German Pension Insurance. The study included 621,695 individuals aged 18–63 years with recorded employment prior to rehabilitation. Job exposures were operationalised using the Overall Job Exposure Index (OJI), derived from the German Employment Survey, and categorised into low, medium, and high exposure. Stable RTW was defined as at least four consecutive months of employment within 24 months after rehabilitation. Associations were analysed using Cox regression models (unadjusted and stepwise adjusted).

Results

Stable employment at 24 months was observed in 74.4%, 68.4%, and 66.2% of individuals with low, medium, and high exposure, respectively. Higher exposure levels were consistently associated with a lower probability of RTW (Model 1). These associations remained after adjustment for sociodemographic factors (Model 2) and additional employment- and health-related variables (Model 3). Differences in RTW emerged immediately after rehabilitation and persisted throughout the follow-up period.

Discussion

Our findings suggest that job exposures can be approximated in administrative data using JEM-based indices. This approach provides a feasible method to incorporate work-related factors into routine data analyses. It enables more differentiated assessments of rehabilitation outcomes and supports the identification of target groups for interventions. Potential exposure misclassification at the individual level should be considered.

Key messages

• Job-exposure matrices enable the integration of occupational exposures into routine data, improving the analysis of return-to-work after rehabilitation.

• Higher occupational exposure is associated with lower return-to-work probability, highlighting the need for exposure-sensitive rehabilitation strategies.