Homelessness and the use of acute psychiatric beds: findings from a one-day survey of adult acute and low-level secure psychiatric patients in North and South Thames regions

This payer describes the impact of homelessness on the use of adult acute a nd lo iv-level secure psychiatric beds in North and South Thames regions. W e specifically focus on those homeless patients who no longer required the acute psychiatric facilities in order to determine the complement of accomm odation and alternative services necessary, to enable them to be discharged , a point prevalence survey of all patients occupying adult acute and low-l evel secure psychiatric beds in 54 mental health providers and seven privat e psychiatric units in North and South Thames regions were included in the survey. Demographic, admission, and diagnosis data were collected. Homeless patients were identified as those patients not living in permanent accommo dation. Ward managers (nurses) were also asked to identify patients who, in their opinion, no longer required acute care and to identify more appropri ate provision for these patients. Of the 3978 adult and low-level secure pa tients enumerated 20.5% (817) were recorded as homeless. A total of 306 hom eless patients were identified by ward staff as inappropriately located, of wham 58.2% of were unsuitable to bo discharged into the community but stil l required some form of alternative health provision, for example staffed-h ousing in the community and in-patient rehabilitation services based in hos pitals. However, 51.8% of patients were suitable to be discharged back into the community and the majority required housing as a key component within their 'package of care'. Homeless psychiatric in-patients have contributed to the increasingly precarious state of mental health services provision an d represent a failure to provide a sufficient level of community alternativ es for those who can no longer access the long-stay hospital services, and for whom the pool of acute psychiatric beds is ever reducing.