Study	Reason for exclusion
Fenton 1998	Allocation: randomised. Participants: people with persistent mental illness. Interventions: residential alternative care to acute admission to a general hospital, not 24 hour care.
Malm 2003	Allocation: randomised. Participants: people with 'schizophrenic disorders'. Interventions: integrated community care versus standard traditional community care, not 24 hour residential care.
Nuttbrock 1998	Allocation: randomised. Participants: people with problems of addiction + mental illnesses (˜50% schizophrenia). Interventions: treatments focusing on management of substance misuse, not 24 hour residential rehabilitation.
Sellwood 1999	Allocation: randomised. Participants: people with chronic schizophrenia. Interventions: home‐based versus outpatient‐based rehabilitation, did not fulfil the remit of 24 hour residential care.
Shern 2000	Allocation: randomised. Participants: homeless people with severe mental illness. Interventions: outreach approach to rehabilitation, not 24 hour residential rehabilitation.
Tyrer 1998	Allocation: randomised. Participants: psychiatric patients. Interventions: community multidisciplinary teams versus hospital‐based care, not 24 hour care.