Presentation M O30

M O30 (Oral Presentation):
Work participation after inpatient, occupational rehabilitation

Presented by: Irene Øyeflaten


Øyeflaten I1, 2, Lie SA3, Ihlebaek C4, Midtgarden IJ1, Maeland S2,5, Magnussen LH2,5,6, Eriksen HR2

  1. The National Centre for Occupational Rehabilitation, Rauland, Norway
  2. Uni Health, Uni Research, Bergen
  3. Department of Clinical Dentistry, UiB, Bergen
  4. Norwegian University of Life Sciences, NMBU, Ås
  5. Department of Occupational Therapy, Physiotherapy and Radiography, Bergen University College
  6. Department of Global Public Health and Primary Care, University of Bergen, Bergen



The aim of the presentation is to discuss different prognostic factors for return to work (RTW) after work rehabilitation (WR), illustrated by three different samples and methods using official register data and self-reports.


All three samples include rehabilitation clients on long-term sickness absence (mean 9-10 mo) mainly due to musculoskeletal and mental diagnoses. Follow-up time in sample 1 was 3 years and 4 months (n=1155). Structural equation modelling was conducted to examine the direct and indirect effect of exogenous variables on days on sickness benefits after WR, and if fear avoidance beliefs for work (FABW) mediated these effects. The follow-up period in sample 2 was 4 years (n=584). Proportional hazards regression models were used to analyze prognostic factors for the probability of being on a specific benefit or at work and the intensity of transitions between any of the following seven states during follow-up; working, partial sick leave, full sick leave, medical rehabilitation, vocational rehabilitation, partial disability pension (DP), and full DP. In the third sample (n=338), the relationships between health complaints, functional ability, coping and self-reported work status three years after WR, were analyzed using logistic regression analysis.


In sample 1, length of sickness absence before WR and FABW had a strong direct effect on days on sickness benefits after WR, and FABW mediated the effects of education and functional ability. These mechanisms were more important for RTW than the health complaints. In sample 2 the probabilities for working and for receiving sickness benefits and DP were associated with gender, diagnoses, type of work and length of sickness absence. In sample 3, poor functional ability and low coping were the variables most strongly associated with not having returned to work three years after WR.


The prognosis for RTW after WR depends on the length of previous sickness absence, being female, blue-collar work and low education. High FABW at referral to WR may mediate these effects, and seem to be a more important predictor for working during follow-up than reported health and functioning. However for those not working at follow-up, poor functioning and low coping appear to be essential.

Schedule Details

Monday September 29
15:45 - 17:15 Afternoon Concurrent Sessions (M O25 - M O37 and Seminar 3)
Session: Long-term work disability
Room A