Many people with severe schizophrenia have increased cerebral ventricular size and diffuse reduction in cortical volume; recent attention has focused on subtle malformations of the cytoarchitecture in the hippocampus and parahippocampal cortex. Sufferers also show an excess of dermatoglyphic and minor physical abnormalities, and a significant proportion had psychomotor deficits, cognitive or behavioural problems as children. Such findings suggest that the form of schizophrenia most akin to Kraepelin's original description of dementia praecox results from neurodevelopmental impairment. This may have its origin in genetic defects in the control of early brain growth, or in early environmental hazards such as prenatal exposure to maternal influenza or perinatal complications. How foetal or neonatal lesions produce hallucinations and delusions two or three decades later remains a mystery, but maturational changes in the brain may be important.

A group of 114 long-stay patients in Friern and Claybury Hospitals was assessed while in hospital, and then again one year and five years after discharge to community placements. Neurotic symptoms, verbal and non-verbal behaviour and, most notably, negative symptoms all improved between the two community follow-up interviews. Patients were living under much less restrictive conditions in the community, and their appreciation of this freedom continued to grow over the years. Their social networks were enriched by an increase in the number of friends in the first year after discharge and in the number of confidants in the subsequent four years. There were no adverse outcomes, but these conclusions cannot yet be said to apply to all long-stay residents of mental hospitals.

The total direct cost of treating schizophrenia in the UK is £397 million, or 1.6% of the total health care budget. Hospital-based and community-based residential care accounts for nearly three-quarters of these costs, while drugs account for only 5%. A conservative estimate of the indirect annual costs of lost production is in the region of £1.7 billion. The heterogeneity of the disease and its outcome means that average treatment costs per person with schizophrenia should be treated with caution; 97% of direct costs are incurred by less than half the patients. Therefore, treatments which reduce the dependence and disability of those most severely affected by schizophrenia are likely to have a large effect on the total cost of the disease to society and may therefore be cost-effective, even though they appear expensive initially.

The development of antipsychotic drugs has followed two complementary approaches, either towards highly specific actions (e.g. on the dopamine receptor) or targeting a broad range of receptors. The properties of ‘atypical’ agents challenge the original dopamine hypothesis and suggest roles for a variety of dopamine receptors and for other pathways, such as serotonin. Older drugs, despite their proven efficacy in relieving many schizophrenic symptoms, have several drawbacks, being ineffective in some patients, relatively ineffective against negative symptoms, and causing adverse neurological effects which may, in turn, be associated with poor compliance. Among newer agents, currently available ones, such as clozapine and risperidone, offer the possibility of more effective control of negative symptoms and an improved side-effect profile, while others are in earlier stages of development. However, much still remains to be understood about their mechanisms of action.