[Zurück zur Trefferliste]

SWE - Stufenweise Wiedereingliederung - Analyse der Dauer der Arbeitsunfähigkeit und der Situation nach Rückkehr in die Arbeit


Projektkennung VfD_SWE_12_003399
Laufzeit von 06/2012, fortlaufend
Status des Projekts laufend



Fragestellung(en) Gelingt es durch Wiedereingliederungsprogramme, Arbeitnehmer schneller in ihre gewohnte Arbeitsumgebung einzubinden?
Reduziert eine stufenweise Wiedereingliederung die Belastungen nach Wiederaufnahme der Arbeit soweit, dass keine Unterschiede zu Nicht-Teilnehmern bestehen?

Hintergrund / Ziele Graded return-to-work programs exist in Germany since 1971 (so called Hamburger Modell). Based on a return plan by the physician and insured that is agreed upon by employer and sickness fund, participants increase their working hours slowly over a specified period of time. During program participation sick leave benefits are paid as participants are still classified as incapable of working. At present, no empirical analysis of the program effects in the SHI system exists. There is evidence from the German pension insurance scheme that program success after rehabilitation is highest for patients with mental disorders.

The study aims at analyzing the graded return-to-work program with respect to sickness duration using claims data of the Techniker Krankenkasse (TK). Research questions focus on identifying participants and analyzing the program effects. Thereby, we account for socio-economic factors, insurance-based characteristics and medical information. Hence, the outcome is not only viewed with respect to the length of inability to work or the sick leave payments but is linked to diagnosed illnesses and demand for medical services during sick leave.

As a second task the follow-up year of those on long-term sick leave after returning to work is analyzed. Research questions focus on two aspects: First, to match program participants and non-participants and second, to assess sickness spells, demand for medical services and resulting costs one year after the return to work. In comparing both groups, we account for socio-economic factors, insurance-based characteristics and medical information such as ICD diagnoses for the initial sickness spell.
Methodik We use data on sick leave from claims data of the Techniker Krankenkasse. Beginning in October 2010, information on the return-to-work program is included. We consider sick-leave spells starting from October 2010 to January 2011 to allow for tracing participation in the return-to-work program, termination of sick leave and new sickness episodes. As the selection process into the program is not observable, we apply a weighting based on propensity scores. By doing so, we analyze determinants of length of the sickness absence and resulting sickness benefits by applying models for survival analysis (Cox proportional hazard model).
In a second step, we consider sick-leave spells with a successful return to work prior to a length of 517 days. We apply a propensity score matching (Greedy matching) between participants and non-participants to further analyze differences in sickness spells, medical demand and treatment costs in a follow-up year.
Datenbasis Sekundärdaten
   Krankenkassen - GKV  (Stichprobengröße: 28.000)
Studiendesign Querschnittstudie
Untersuchte Geschlechter weiblich und männlich
Untersuchte Altersgruppen von 18 bis 99 Jahre
Ergebnisse In the considered time period, about 28 000 cases of sick leave longer than six weeks are analyzed. Among these cases, about 7 000 individuals participate in the return-to-work program. The most frequent diagnoses are musculoskeletal disorders, mental health problems and injuries. Participation probability was higher for women and increases with income. The results of the survival analysis show a positive program effect for participants being more than 120 days on sick leave. For those with mental health problems, we find a positive effect from sickness periods with more than 90 days.
For the follow-up period, we analyze 6434 insured-pairs and find only small differences in sickness spells and medical costs between treatment and control group. In detail, rtw participants show slightly lower expenditures on hospitals but higher for ambulatory services and pharmaceuticals. Moreover, differences in expenditure are related to diagnoses of the initial sickness spell. Furthermore, former RTW participants are more likely to take part in a rtw program in the follow-up year.

Forschende und kooperierende Einrichtungen






Stand: 24.03.2014