Background There has been a significant reduction in the number of people with severe mental illness who spend extended periods in long-stay hospitals. District health authorities, local authorities, housing associations and voluntary organisations are jointly expected to provide support for people with severe mental disorder/s. This 'support' may well involve some kind of special housing. Objectives To determine the effects of supported housing schemes compared with outreach support schemes or 'standard care' for people with severe mental disorder/s living in the community. Search methods For the 2006 update we searched the Cochrane Schizophrenia Group Trials Register (April 2006) and the Cochrane Central Register of Controlled Trials (CENTRAL, 2006 Issue 2). Selection criteria We included all relevant randomised, or quasi-randomised, trials dealing with people with 'severe mental disorder/s' allocated to supported housing, compared with outreach support schemes or standard care. We focused on outcomes of service utilisation, mental state, satisfaction with care, social functioning, quality of life and economic data. Data collection and analysis We reliably selected studies, quality rated them and undertook data extraction. For dichotomous data, we would have estimated relative risks (RR), with the 95% confidence intervals (CI). Where possible, we would have calculated the number needed to treat statistic (NNT). We would have carried out analysis by intention-to-treat and would have summated normal continuous data using the weighted mean difference (WMD). We would have presented scale data for only those tools that had attained pre-specified levels of quality and undertaken tests for heterogeneity and publication bias. Main results Although 139 citations were acquired from the searches, no study met the inclusion criteria. Authors' conclusions Dedicated schemes whereby people with severe mental illness are located within one site or building with assistance from professional workers have potential for great benefit as they provide a 'safe haven' for people in need of stability and support. This, however, may be at the risk of increasing dependence on professionals and prolonging exclusion from the community. Whether or not the benefits outweigh the risks can only be a matter of opinion in the absence of reliable evidence. There is an urgent need to investigate the effects of supported housing on people with severe mental illness within a randomised trial.