Improving mental health and well-being of hospital staff: mixed-methods process evaluation of the SEEGEN trial

Health workers experience high levels of occupational stress. There is a lack of knowledge about moderators and active ingredients of interventions aiming to reduce occupational stress in health workers. This paper presents results of the mixed-methods process evaluation of the SEEGEN (“SEElische GEsundheit am Arbeitsplatz KrankeNhaus) randomised controlled trial. Employees of three hospitals in Germany were recruited for this trial between 10/2019 and 01/2020. Participants allocated to the intervention group could choose from five modules according to their needs and interests. Outcome and process measures were collected with established scales, including emotional and cognitive strain (Irritation Scale), subjective well-being, psychosocial safety climate, and self-efficacy. Uptake of the intervention and fidelity was assessed using the SEEGEN Fidelity Scale. Multivariate general least squares ANOVA with allocation as interaction effect was used for analysis of moderators. Mediator analysis was done using ordinary least square regression. Qualitative data was collected via semi-structured interviews from a sub-sample, focusing on organisation, outcomes and practical applicability of the intervention. Interviews were analysed using qualitative content analysis. 415 hospital employees participated in this study. Average dose of the intervention across all modules was a close to two days. The effect of the intervention was moderated by some staff- and work-related variables, and mediated by fidelity to the intervention. Qualitative results of 23 participants confirmed and supplemented quantitative findings. The participants’ adjustments in terms of stress reduction related primarily to the individual level, in particular to changes in behaviour and the altered perception of work situations. Cross-study (data type) synthesis enabled a deeper understanding of the moderating variables of the intervention with regard to age of participants and the relevance of contextual conditions and confirmed the range of requirements of the participants depending on their respective work situation. The high complexity of the intervention with different modules and target groups hindered implementation and effectiveness. A step backwards seems indicated, i.e. the further evaluation of smaller components of the intervention, added by theory development and refinement. DKRS, DRKS00017249. Registered 08/10/2019, https://www.drks.de/DRKS00017249 .