Suicides and absconding by inpatients are both situations that present serious challenges for medical institutions. For this reason, many psychiatric wards use permanently locked doors to help promote patient safety. However, previous studies have shown that the overall therapeutic atmosphere of some psychiatric wards can be disrupted by the presence of permanently locked doors, leading to potentially greater health risks for patients.
To evaluate the role of hospital types (hospitals with and without locked wards) and ward types (day clinics, partially locked, and permanently locked) in rates of suicide, suicide attempts, and absconding by patients.
Data documented between 1998 and 2012 were gathered from 21 different German hospitals. These data were gathered through routine surveys completed by professionals from each institution. Sixteen of these hospitals featured at least one permanently locked ward, four had no locked wards, and one hospital had no locked wards at the beginning of the study period, but eventually introduced locked wards. A total of 271,128 cases within locked wards and 78,446 cases in wards with open door policies were analyzed.
Patients in wards with open door policies tended to be generally older, female, unmarried, and housed in assisted living residences. These patients were diagnosed mostly with various organic mental disorders, less frequently with substance use disorders, and were often admitted voluntarily. Absconding with return occurred more often in hospitals without locks, while absconding without return was less frequent compared to hospitals with locks. Ultimately, hospital type (i.e., locked wards versus open door wards) did not significantly predict suicide rates, suicide attempts, and absconding with or without return. However, being treated in a ward with an open door policy was correlated with a significantly lower suicide attempt probability and lower rates of absconding with or without return.
The authors note that the institutions included in this study may not be representative of university hospitals or private clinics, and that these hospitals likely admitted patients with lower risks for absconding and suicide. Information concerning patient treatment, therapeutic relational quality, and overall therapeutic atmospheres were not included in the study. No data were included concerning instances of self-harm or harm to others during cases of absconding without return.