This page was translated automatically from Turkish. Read the original

Bu sayfa, 26. Dünya Psikiyatri Kongresi'nde sunulan çalışmanın yapay zekâ desteğiyle hazırlanmış Türkçe özetidir. Çalışmanın sahibi sunan yazarlardır; atıf için orijinal kaynağa başvurun.

EV1331

Co-design of a population-based risk calculator for adverse outcomes following discharge from involuntary psychiatric care

A POPULATION-BASED RISK CALCULATOR FOR ADVERSE OUTCOMES AFTER DISCHARGE FROM INVOLUNTARY PSYCHIATRIC CARE

A risk calculator intended for use in discharge planning following involuntary psychiatric care was found to be usable, but views on its clinical utility remained mixed unless the predicted risk was linked to a concrete action.

The aim is to develop the first risk calculator based on national registries to predict the risks of readmission, self-harm, and suicide among patients discharged from involuntary psychiatric care (IPC), and to report the co-creation phase in this study. Eight focus group sessions, four in-depth interviews, and surveys involving 25 physicians and two patient representatives were conducted at four clinics in the Stockholm region; interview data were transcribed verbatim and evaluated using thematic analysis, while survey data were evaluated using descriptive statistics. In the preliminary modelling phase, linked national registries covering all IPC episodes in Sweden between 2010 and 2020 (N=134,514) were analysed using Cox regression. The pooled mean System Usability Scale score was 76.1/100; the UTAUT performance expectancy score was 2.55/7, and the attitude towards using technology score was 3.13/7. Participants valued the calculator primarily for making risk assessment systematic and transparent; there was consensus that the tool should support, but not replace, clinical judgement, be used alongside dynamic clinical information, and not serve as the sole justification for decisions on detention or discharge. Feedback directly changed the calculator’s outputs, with involuntary readmission and suicide attempt replacing acute readmission and self-harm, and the interface was also redesigned. It was emphasised that the ratings concerned the prototype rather than the concept itself, varied considerably across clinics, and that the predicted risk had not yet been linked to an action at discharge; the next steps are completion of the statistical modelling, end-user training, and a cluster-randomised pilot study.

Tüm WCP 2026 özetleri