[Prediction of Utilization of Psychological Support During Vocational Rehabilitation]

Comprehensive psychological support is a distinctive feature of vocational retraining centers, pivotal in stabilizing the health of the rehabilitants, reducing dropout rates and facilitating return to work. For optimal resource planning within the vocational retraining centers, and thereby ensuring the best possible psychological support during vocational rehabilitation, it is crucial to be able to estimate the number of appointments of the rehabilitants in advance. Even though routine data is available, predictions using regression analysis have not been undertaken previously.

To predict the number of appointments in the psychological service, a hierarchical linear regression analysis with 25 potential predictors was conducted using retrospective routine data from 404 rehabilitants. In specifying the regression model, data-driven methods and theory were combined. The Andersen healthcare utilization model served as the theoretical background.

The final regression model accounted for 39.5% of the variance in the utilization of the psychological service. Ten variables predicted a higher number of appointments: female gender, boarding status, low resilience, desire for psychological support, past psychological treatment, intake of psychotropic drugs, absence of two specific professional fields, Federal Employment Agency/Jobcenter as the paying institution, and non-participation in a specific preparatory course.

The routine data available at the beginning of vocational rehabilitation in a vocational retraining center is generally suitable to predict the number of appointments in the psychological service. As such, it can be utilized for resource planning. Various factors prove to be informative, suggesting that a focus on single factors (e. g., the presence of an ICD-10 F-diagnosis) seems disadvantageous. Limitations of the regression-analytic approach are highlighted.