Health economic evaluation of a nurse-led care model from the nursing home perspective focusing on residents’ hospitalisations

Abstract

Background: Health economic evaluations of the implementation of evidence-based interventions (EBIs) into practice provide vital information but are rarely conducted. We evaluated the health economic impact associated with implementation and intervention of the INTERCARE model—an EBI to reduce hospitalisations of nursing home (NH) residents—compared to usual NH care. Methods: The INTERCARE model was conducted in 11 NHs in Switzerland. It was implemented as a hybrid type 2 effectiveness-implementation study with a multi-centre non-randomised stepped-wedge design. To isolate the implementation strategies' costs, time and other resources from the NHs’ perspective, we applied time-driven activity-based costing. To define its intervention costs, time and other resources, we considered intervention-relevant expenditures, particularly the work of the INTERCARE nurse—a core INTERCARE element. Further, the costs and revenues from the hotel and nursing services were analysed to calculate the NHs' losses and savings per resident hospitalisation. Finally, alongside our cost-effectiveness analysis (CEA), a sensitivity analysis focused on the intervention's effectiveness—i.e., regarding reduction of the hospitalisation rate—relative to the INTERCARE costs. All economic variables and CEA were assessed from the NHs' perspective. Results: Implementation strategy costs and time consumption per bed averaged 685CHF and 9.35 h respectively, with possibilities to adjust material and human resources to each NH’s needs. Average yearly intervention costs for the INTERCARE nurse salary per bed were 939CHF with an average of 1.4 INTERCARE nurses per 100 beds and an average employment rate of 76% of full-time equivalent per nurse. Resident hospitalisation represented a total average loss of 52% of NH revenues, but negligible cost savings. The incremental cost-effectiveness ratio of the INTERCARE model compared to usual care was 22′595CHF per avoided hospitalisation. As expected, the most influential sensitivity analysis variable regarding the CEA was the pre- to post-INTERCARE change in hospitalisation rate. Conclusions: As initial health-economic evidence, these results indicate that the INTERCARE model was more costly but also more effective compared to usual care in participating Swiss German NHs. Further implementation and evaluation of this model in randomised controlled studies are planned to build stronger evidential support for its clinical and economic effectiveness. Trial registration: clinicaltrials.gov (NCT03590470).