The routine assessment of severity amongst people with mental illness

Specialist mental health services are required to prioritise their work. To help this process numerous definitions of severe mental illness have been suggested. Such definitions vary, and are not necessarily valid or reliable. This investigation examined whether there was agreement over who constituted the most severely ill patients, amongst the case load of a community mental health team (CMHT). Suggested guidelines for the prioritisation of patients were adapted after consultation and pilot reliability studies, and were then used by CMHT staff to rate their case loads (n=299). Test re-test, and inter-rater reliability studies were then conducted. A random sub-sample (n=120) was selected for further analysis to measure concurrent validity with respect to assessment of need, functioning and quality of life; and criterion validity. There was consistency in individual key worker decisions over time, and key worker ratings were valid in terms of disability, need and quality of life. Patients with a psychotic diagnosis were more likely to be rated as a high priority than those with a non-psychotic diagnosis. Agreement amongst different staff (inter-rater reliability) was poor, especially when ratings from CMHT staff were compared to external ratings. Our findings highlight the difficulties inherent in trying to agree on who constitute the severely mentally ill, and warn against the indiscriminate use of guidelines to determine access to services.