Introduction Aggression and violence are highly complex problems in acute psychiatry that often lead to the coercive interventions. The Safewards Model is an evidence-informed conflict-reduction strategy to prevent and reduce such incidents. The aim of this study was to evaluate the implementation of this model with regard to coercive interventions in inpatient care. Materials and Methods We evaluated outcomes of the implementation of the Safewards Model in two locked psychiatric wards in Germany. Frequency and duration of coercive interventions applied during a period of 11 weeks before and 11 weeks after the implementation period were assessed through routine data. Fidelity to the Safewards Model was assessed by the Organization Fidelity Checklist. Results Fidelity to the Safewards Model was high in both wards. The overall use of coercive measures differed significantly between wards [case-wise: χ 2 (1, n = 250) = 35.34, p ≤ 0.001; patient-wise: χ 2 (1, n = 103) = 21.45, p ≤ 0.001] and decreased post-implementation. In one ward, the number of patients exposed to coercive interventions in relation to the overall number of Patients decreased significantly [χ 2 (1, 281) = 6.40, p = 0.01]. Furthermore, the mean duration of coercive interventions overall declined significantly [ U (55,21) = −2.142, p = 0.032] with an effect size of Cohen’s d = −0.282 (95% CI: −0.787, 0.222) in that ward. Both aspects declined as well in the other ward, but not significantly. Discussion Results indicate that the implementation of the Safewards interventions according to the model in acute psychiatric care can reduce coercive measures. They also show the role of enabling factors as well as of obstacles for the implementation process.

Zusammenfassung Ziel Rund 80 % der Patienten in psychiatrischen Kliniken haben wesentliche arbeitsbezogene Teilhabebeeinträchtigungen und sind vom Arbeitsmarkt ausgeschlossen. Methode Stichtagsbefragung von 176 Patienten in (teil-)stationärer Behandlung bzgl. ihrer arbeitsbezogenen Teilhabebeeinträchtigungen, ihres Unterstützungsbedarfs und konkret erhaltener Hilfe. Ergebnisse Von 90 der 176 angefragten Patienten lagen Daten vor. 63 % der Patienten (n = 57) gaben Unterstützungsbedarf an; von diesen würden 84 % an einem Jobcoaching teilnehmen. 49 % der Patienten wurden auf das Thema Arbeit angesprochen. Unabhängig vom Unterstützungsbedarf erhielten nur 20 % der Patienten konkrete Hilfe. Starken Unterstützungsbedarf gaben junge Erwachsene (77 %) und Ersterkrankte (73 %) an. Schlussfolgerungen Es gibt einen hohen, aber unzureichend gedeckten Unterstützungsbedarf. Die routinemäßige Bedarfsermittlung und die Implementierung evidenzbasierter Methoden könnten arbeitsbezogene Teilhabechancen wesentlich verbessern.