Effectiveness and cost-effectiveness of home treatment compared with inpatient care for patients with acute mental disorders in a rural catchment area in Germany

Abstract

Background and aimsHometreatment (HT) is considered as a safe and effective alternative to inpatient treatment (IT) for the treatment of patients with acute mental disorders. To date, there are only few studies on the effectiveness and no studies on the efficiency of HT vs IT in Germany. The aim of this study is to investigate the effectiveness and cost-effectiveness of HT in comparison with IT in a rural catchment area in Germany. MethodsIn a prospective observational trial, 60 patients with acute mental disorders treated by a HT team were compared with 58 patients who were eligible for HT but received IT. Treatment outcomes (change of psychotic symptoms [PANSS], depressive symptoms [HAMD] and overall clinical and functional impairment [HoNOS] from admission to discharge) were assessed. Treatment costs were assessed on the basis of reimbursement data. Effectiveness was estimated by means of mixed effects regression (MER) models. Cost-effectiveness was estimated by a net monetary benefit (NMB) regression model. Propensity score adjustment was applied for the control of selection bias in regression models. ResultsAs indicated by the results of the MER model, HT was more effective in comparison to IT with respect to HAMD reduction (b −4.11; p 0.004) and HoNOS total score decrease (b −4.43; p 0.021) but not as regards the PANSS total score (b −4.51; p 0.134). Unadjusted treatment costs did not differ between HT and IT, but after adjustment for propensity scores and for baseline values of the outcome measures HT was significantly less costly (b −7.151.10; p 0.028) than IT. NMB regression revealed a significant monetary benefit for a one unit change of the HAMD at a maximum willingness to pay (MWTP) of λ=0 € and λ=100 € and the HoNOS at λ=1000 €. ConclusionsHT is an effective and cost-effective alternative to IT for the treatment of people with acute mental disorders in the investigated catchment area. Results cannot be generalized for the whole of Germany. Further research is needed.