A problem-solving intervention for employees on sickness absence due to common mental disorders : effects, ethics and process

Abstract

Purpose: Employees with common mental disorders (CMDs) which include depression, anxiety-, and adjustment disorder, are more often on sickness absence and have longer sickness absence episodes than employees with other diagnoses. To support these employees in their return-to-work process, problemsolving in combination with a work-directed intervention have previously been evaluated with positive results on earlier return-to-work and reduced sickness absence when provided in an occupational health service context. However, this combination has not been tested in primary health care in Sweden. A problemsolving intervention with workplace involvement (PSI-WPI) on top of care as usual (CAU) was provided in primary health care to employees on sickness absence due to common mental disorders with the aim of decreasing sickness absence. The overall aim of this thesis was to evaluate the effectiveness of the PSI-WPI on top of CAU for employees on sickness absence due to CMDs when compared to CAU alone, and to examine related contextual factors, ethical aspects, and process outcomes.

Methods: A cluster-randomised controlled trial was conducted including 19 coordinators (PSI-WPI=9 and CAU=10) working in the primary health care delivering the intervention and 197 employees (PSI-WPI=85 and CAU=112) receiving the intervention. Data were collected from the Swedish Social Insurance Agency register on sickness absence, semi-structured interviews with coordinators, employees, and managers, and questionnaires to coordinators and employees. To evaluate the effectiveness of the intervention, generalized estimating equations were used to estimate the difference in mean registered net sickness absence days per month between PSI-WPI and CAU during an 18-month follow-up. To explore experiences, content analysis was used to analyse interview data which was subsequently categorised according to the consolidated framework for implementation research. To explore ethical aspects, the theoretical framework for systematic identification of ethical aspects of health care technologies was used to build a coding scheme and thematic analysis was used to analyse interview data. Process evaluation outcomes evaluated were reach, dose delivered, dose received, fidelity and dose response.

Results: There was no statistically significant effect on sickness absence days for employees receiving the PSI-WPI compared to CAU during the 18-month followup. However, sickness absence days decreased in both groups over time. Rehabilitation coordinators, employees, and managers reported that the structured process of the PSI-WPI was facilitating. As well, learning to identify problems and finding solutions, and early involvement of the manager in the return-to-work process was considered to be beneficial. Barriers to PSI-WPI included the time, and number of face-to-face meetings required as well as symptom severity. Ethical challenges of the PSI-WPI included workplace and health care differences in identified goals, values and norms. Further ethical challenges identified of the PSI-WPI were an increased sharing of information, unclear roles for coordinators and managers and juggling of the patient and employee roles. The process evaluation showed that rehabilitation coordinators delivering PSI-WPI agreed that their training and resources to deliver the PSI-WPI were sufficient. Out of 85 employees receiving PSI-WPI, 35 (41%) received the three sessions covering all five steps of the PSI-WPI.

Conclusions: In primary health care in Sweden, receiving the PSI-WPI on top of CAU did not result in a statistically significant effect on SA days compared to CAU alone. Even so, the PSI-WPI group received more sessions with the rehabilitation coordinators, three-part meetings and follow-up sessions compared to employees receiving CAU. Employees found it facilitating to learn about problems and solutions. The three-part meetings enabled a dialogue between the coordinator, employee and manager and a common view on how to move forward in the return-to-work process. Since time investment was reported as a barrier for the PSI-WPI it would be valuable to identify the employees that would benefit the most from the PSI-WPI. Ethical challenges were identified in the collaboration between the workplace and healthcare, and this collaboration in the Swedish setting needs further investigation. At this stage we cannot recommend the use of PSI-WPI in primary health care in Sweden. More research is needed on which employees require support from a PSI-WPI in returning to work and how much of the PSI-WPI that is needed to see effects on return-to-work outcomes.