Patient-related characteristics or local tradition: what predicts the admission to a locked ward or the use of coercive measures in psychiatric inpatient treatment?

Prior research shows that locked doors and coercive measures are not only applied due to safety concerns, but also due to the specific local tradition of an institution. We examined the association of the use of coercive measures and the admission to a locked ward with person-related characteristics compared to the admission to a specific clinic. In this 15-year, naturalistic observational study, we examined 230,684 admissions to 14 German psychiatric inpatient clinics from Jan 1, 1998, to Dec 31, 2012. To analyze the degree to which admission to a locked ward and coercive measures (received vs. not received) were connected with person- and clinic-specific factors, two-step logistic regression analyses were applied. 27% of the variance of the admission to a locked ward were explained by person-related characteristics (Nagelkerke r² = 0.269). By adding the clinic the person was admitted to, the explained variance increased by 15% (Nagelkerke r² = 0.418). 36% of the variance of the use of coercive measures were explained by person-related characteristics (Nagelkerke r² = 0.364). By adding the clinic the person was admitted to, the explained variance increased by 4% (Nagelkerke r² = 0.400). The local tradition of a psychiatric clinic seems to play a more prominent role for the decision to admit a person to a locked ward than for the decision to use coercive measures. Clinicians should be made aware of the connection of local traditions with clinical pathways in acute psychiatry to avoid unnecessary admissions to locked wards.