Multilevel, risk group-oriented strategies to decrease sickness absence in the public sector: evaluation of interventions in two regions in Sweden

Abstract

Purpose: Sickness absence has been identified as needing to be addressed through multilevel interventions, but knowledge regarding optimal design and implementation of such interventions is scarce. The aim of this study was to evaluate the implementation and effects of a large-scale multilevel intervention in the public sector in Sweden. Methods: The overall effect of the intervention was assessed using mixed-effect models. Sickness absence data (before, and 6 or 12 months after the intervention) for 90 intervention groups and 378 reference groups was retrieved from administrative personnel systems from the two participating regional councils. The implementation processes were evaluated using qualitative content analysis of qualitative interviews conducted at two timepoints. Results: The results show that the vast majority of implemented measures were on an individual level and the integration of the intervention differed between the two regions. The reception and perception of the intervention activities seem to have been influenced by the implementation process, and how well the interventions were communicated and integrated, both regarding the integration of the different interventions and their integration into the discrete context and existing routines. No short-term overall effects on sickness absence were found. Conclusions: The results point to the many challenges in implementing complex interventions, especially where organizational measures are involved—including adequate participation by, and communication between, the involved actors, as well as sufficient resources. The results indicate potential learning effects regarding the awareness of organizational factors in sick leave, after implementing and integrating multilevel strategies.