Mental health clustering and diagnosis in psychiatric in-patients

Abstract

Aims and method This paper investigates the relationship between cluster (Mental Health Clustering Tool, MHCT) and diagnosis in an in-patient population. We analysed the diagnostic make-up of each cluster and the clinical utility of the diagnostic advice in the Department of Health's Mental Health Clustering Booklet. In-patients discharged from working-age adult and older people's services of a National Health Service trust over 1 year were included. Cluster on admission was compared with primary diagnosis on discharge. Results Organic, schizophreniform, anxiety disorder and personality disorders aligned to one superclass cluster. Alcohol and substance misuse, and mood disorders distributed evenly across psychosis and non-psychosis superclass clusters. Two-thirds of diagnoses fell within the MHCT 'likely' group and a tenth into the 'unlikely' group. Clinical implications Cluster and diagnosis are best viewed as complimentary systems to describe an individual's needs. Improvements are suggested to the MHCT diagnostic advice in in-patient settings. Substance misuse and affective disorders have a more complex distribution between superclass clusters than all other broad diagnostic groups.