Related Experiment Video
Updated: Jul 10, 2026
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
[SIMBO: a Screening Instrument for Identification of Work-Related Disabilities--analyses of construct and prognostic
M Streibelt1, H Gerwinn, T Hansmeier
1Charité--Universitätsmedizin, Abteilung für Versorgungssystemforschung und Qualitätssicherung in der medizinischen Rehabilitation, Berlin. marco.streibelt@charite.de
Background:
For a number of years, work-related interventions in medical rehabilitation (MBO) have been developed. Basically, these interventions concentrate on vocational problems of rehabilitees whose health disorders are strongly associated with contextual factors of the environment as well as personal factors. Previous studies showed a close relationship between the success of an intervention and identification of a specific demand. In fact there are several clinical concepts regarding specific demand. But there still is a lack of appropriate instruments for use in identification of occupational challenges. Therefore SIMBO (Screening Instrument for Identification of a Demand for Medical-Vocational Oriented Rehabilitation) has been developed recently. By using a scale for the intensity of work-related problems as well as a cut-off point, SIMBO is able to identify patients with and without a demand for work-related interventions.
Method:
Analyses relative to construct validity and predictive validity were carried out on two different samples--a multi-clinic sample (patients with musculoskeletal disorders) and a sample from the German statutory pension insurance agency DRV Westfalen (successful applications for medical rehabilitation). In this context the cut-off level discussion is very important.
Results:
By means of the multi-clinic sample--irrespective of cut-off definition--the SIMBO-decision and the clinical identification of MBO-demand were found to agree in 74-78% of the cases. This corresponds to a maximum adjusted correlation of r=0.59 (phi coefficient). Compared to the external ratings of vocational problems given by DRV staff in handling the applications, however, only little agreement is found (64%, r=0.25). In fact, SIMBO had in 77% (r=0.50) of the cases been able to correctly predict work-related problems to be expected. So the result obtained using this instrument is far better than prediction of these problems in the external ratings by DRV staff (54%, r=0,21). Also, return to work (RTW) in good health after six months can be predicted correctly by SIMBO in 77% of the cases. This means that the probability of RTW in good health is reduced by 90% (Odds Ratio=0.1) if work-related problems had been identified by SIMBO.
Discussion And Conclusion:
Concerning its clinical as well as predictive quality, the validity of SIMBO-based ratings of work-related problems has been proven. Further, it has become obvious that SIMBO is suitable as an easy-to-handle tool for identification of a need for vocationally-focused interventions for use by the social insurance agencies which finance rehabilitation. Further interesting questions arise relative to application in different indications as well as potential uses as an outcome instrument.
