Medical training tends to focus on acute episodes of illness and on ways of helping people to overcome them. Mental health training also emphasises this aspect of treatment. Most of the technologies and the greater part of the evidence base in psychiatry are orientated towards the short term. However, the acute illness model is only applicable to a small part of the work of mental health services, and clinicians can find it altogether more difficult to help people to stay well in the long term. Helping people to avoid relapse and to develop self-management strategies calls upon a wide range of skills in areas of clinical endeavour with a sparse evidence base.

The simplest model of mental health intervention involves a clinician doing something of a technical nature with a clear-cut start and finish. The problem is permanently resolved, and there is no need for the person to do anything more about his mental health. This model of illness–intervention–cure is based on the surgical paradigm that dominates acute health services. It exists as an assumed model in the funding arrangements for mental health services in many countries, but it is based on a false premise. Even if one has a strong attachment to a disease model, it is difficult to deny that the relationship between illness and contextual factors is profound. This is just as true of cancer or arthritis as it is of mental disorder.