Historically, studies of outcome in schizophrenia have tended to rely on unitary outcome measures, such as hospitalization, and reported a generally pessimistic outcome. However, single summary measures of global outcome are unlikely to capture the true diversity of patients' functioning over time and the assumption that such measures can serve as a proxy for both symptomatic and social outcome are no longer tenable in the era of community care. Further, studies of outcome over periods of up to 10 years reveal that while about half of those people receiving a diagnosis of schizophrenia remain chronically ill, the other half recover fully or improve substantially after the initial acute illness. Ultimately, about half are able to live independently, while 10% remain chronically institutionalized, and, over lifetime follow up, 10% will die by suicide. Evidence from recent, longer-term, outcome studies in Europe and the US, as well as meta-analyses of historical, global outcome data, suggests that the outcome is more heterogeneous and favourable than reported in studies earlier in the last century. Nevertheless, although schizophrenia does not inevitably progress to a deteriorated end stage, there are substantial and enduring adverse consequences for most patients. The concept of duration of untreated illness (the period between the emergence of psychosis and the initiation of effective treatment) has recently attracted interest because of its possible relationship with treatment outcome and implications for preventive efforts.