Modifiable Psychological Variables as Predictors of Return to Work: Temporal Patterns in a Longitudinal Observational Study
Abstract
Background: Common mental disorders (CMDs), including depression, anxiety, and adjustment disorders, are leading causes of long-term sickness absence (SA) and work disability in high-income countries. Despite the significant burden, evidence-based interventions to support return to work (RTW) remain limited. This study aimed to investigate the predictive role of selected modifiable psychological factors on RTW among individuals on sick leave for CMDs, with particular emphasis on temporal dynamics during the early phase of the RTW process.
Methods: A longitudinal observational study was conducted with bi-weekly self-reported questionnaire measurements for 3 months and quarterly follow-ups for 12 months. Outcomes were self-reported RTW and registry-based net SA days (NSAD). Analyses included robust linear mixed models for focal predictor trajectories, Bayesian logistic regression with a regularizing horseshoe prior for the probability of RTW, and ordered beta regression for NSAD since sick-days data have lower and upper bounds.
Results: Baseline measurements of RTW-SE and Work Ability Scores (WAS) were also associated with RTW at three months. Improvement in RTW self-efficacy (RTW-SE) during the first three months showed the strongest and most consistent association with increased probability of RTW, both cross-sectionally at 3, 6, and 12 months, and longitudinally.
Improvement regarding physical fatigue and WAS were associated with approximately 18% less NSAD at 3 and 6 months, while cognitive exhaustion showed similar impact but only at three months.
Related
- Bayesian additive regression trees for evaluating treatment benefit predictors using observational data
- Treatment persistence drives estimator performance in longitudinal causal inference based on observational data: A simulation study
- Doubly Robust Estimation and Inference for Longitudinal Win Probabilities with Ordinal Outcomes in Observational Studies
- Modifiable determinants of adherence to tyrosine kinase inhibitors in classical myeloproliferative neoplasms: A mixed-methods systematic review protocol
- The Effect of Omitted Variables on the Sign of Regression Coefficients
BibTeX
@misc{johansson2026modifiable,
title = {{Modifiable Psychological Variables as Predictors of Return to Work: Temporal Patterns in a Longitudinal Observational Study}},
author = {Magnus Johansson and Alice Giöstad and Aleksandra Bujacz and Anna Ek and Petter Gustavsson and Lars-Göran Öst and Anna Finnes},
howpublished = {PsyArXiv},
year = {2026},
month = oct,
doi = {10.31234/osf.io/89s2x_v3},
url = {https://osf.io/preprints/psyarxiv/89s2x_v3/},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Modifiable Psychological Variables as Predictors of Return to Work: Temporal Patterns in a Longitudinal Observational Study
AU - Magnus Johansson
AU - Alice Giöstad
AU - Aleksandra Bujacz
AU - Anna Ek
AU - Petter Gustavsson
AU - Lars-Göran Öst
AU - Anna Finnes
PY - 2026
DA - 2026/10/30
PB - PsyArXiv
DO - 10.31234/osf.io/89s2x_v3
UR - https://osf.io/preprints/psyarxiv/89s2x_v3/
AB - Background: Common mental disorders (CMDs), including depression, anxiety, and adjustment disorders, are leading causes of long-term sickness absence (SA) and work disability in high-income countries. Despite the significant burden, evidence-based interventions to support return to work (RTW) remain limited. This study aimed to investigate the predictive role of selected modifiable psychological factors on RTW among individuals on sick leave for CMDs, with particular emphasis on temporal dynamics during the early phase of the RTW process. Methods: A longitudinal observational study was conducted with bi-weekly self-reported questionnaire measurements for 3 months and quarterly follow-ups for 12 months. Outcomes were self-reported RTW and registry-based net SA days (NSAD). Analyses included robust linear mixed models for focal predictor trajectories, Bayesian logistic regression with a regularizing horseshoe prior for the probability of RTW, and ordered beta regression for NSAD since sick-days data have lower and upper bounds. Results: Baseline measurements of RTW-SE and Work Ability Scores (WAS) were also associated with RTW at three months. Improvement in RTW self-efficacy (RTW-SE) during the first three months showed the strongest and most consistent association with increased probability of RTW, both cross-sectionally at 3, 6, and 12 months, and longitudinally. Improvement regarding physical fatigue and WAS were associated with approximately 18% less NSAD at 3 and 6 months, while cognitive exhaustion showed similar impact but only at three months.
ER -