Therapeutic leave (TL) or unaccompanied home visits before discharge, are well-established procedure during inpatient psychiatric treatment. At the beginning of 2020, reimbursable TL in Switzerland was limited to a maximum of 24 hours, without any available empirical data on the therapeutic relevance of this procedure. The present study examines the practice from 2015 to 2019 in a large psychiatric care facility in a rural setting. For this purpose, routine data of 2727 inpatient hospitalisations of 2028 patients were analysed descriptively. The relationship of TL and parameters of treatment outcome was investigated in multivariate regression models. In addition, it was examined how the new practice affected the duration and frequency of TL. Further, the attending physicians and psychologists of the facility were asked to assess the current procedure of TL (summer 2021). The data shows, that TL was a widely used and established instrument across all diagnostic groups. Regression analyses showed a significant correlation between the use of TL and better symptomatic treatment outcome as well as reduced rehospitalisation tendency before rationing. As expected, the change in practice led to a significant reduction in the duration of TLs. At the time of the survey after rationing, all 16 participating experts considered TL to be useful and 9 out of 16 advocated for more frequent or longer use of the instrument. In summary, the study suggests a positive impact of TL on treatment outcome parameters, including potentially reduced risk of relapse and urges caution in further reducing the established instrument of therapeutic leave before additional research results are available.