Sir: Protheroe and Carroll (Psychiatric Bulletin, November 2001, 25, 416-417) are clearly enthusiasts for crisis services. They relate that they are struck by the lack of awareness of and hostility towards the development of crisis services. Could it be that it is not lack of awareness of such services, but an awareness of the lack of up-to-date evidence for or against such services? The evidence base from randomised controlled trials for crisis intervention services is weak. Most of it is around 20 years old and only one study is derived from the UK. The older research studied hybrids of crisis intervention and assertive community treatment, rather than pure crisis intervention. Even the most up-to-date research of a service that approximates to crisis intervention, a study of the Daily Living Programme (Reference Muijen, Marks and ConnolyMuijen et al, 1992), did not compare home treatment with the cornerstone of modern day community care — the community mental health team, using the framework of the Care Programme Approach. What is more, the terminology of crisis intervention, or home treatment as it is otherwise known, is inadequate and confusing and prevents adequate conclusions being formed.

Protheroe and Carroll complain that UK-based psychiatrists are hostile towards the development of crisis services. If indeed this is the case, such hostility is not reflected in the sentiments of health authority chairs and trust chief executive officers who responded to a recent questionnaire study: all health authorities and 97% of trust chief executive officers were in favour of the principle of providing home treatment (Reference Owen, Sashihadran and EdwardOwen et al, 2000). It would be interesting to know just how prevalent such hostility actually is among UK-based psychiatrists.

The authors note that the public continues to fear care in the community despite the evidence that de-institutionalisation has not increased the low risk of homicide by those with mental illness. This is a specious argument. Homicide is an uncommon event, violence on the other hand is not and its consequences can be very serious. Between 10% and 40% of patients commit assault before admission to hospital and 28% of discharged patients have been found to have committed at least one violent act within a year of discharge (Reference Monahan, Faigman, Kaye and SaksMonahan, 1997; Reference Steadman, Mulvey and MonahanSteadman et al, 1998). As with intensive case management (Reference Walsh, Gilvarry and SameleWalsh et al, 2001), crisis intervention has not so far been demonstrated to reduce the frequency of violent episodes committed by patients. This is neither argument for nor against crisis intervention, but simply a statement that we just don't know what the impact of crisis intervention is on violence.

Finally, with regard to issues relating to the detainment of patients under the Mental Health Act, the authors' views may be too radical for liberal-minded UK psychiatrists. Our current system of detention of patients may be considered too slow and unwieldy by the authors, but the alternative proposal of a single individual (a crisis assessment and treatment team worker) alone being able to swiftly effect the deprivation of an individual's liberty is surely much more open to abuse than the English and Welsh system: surely our more elaborate processes of application are meant to serve as a safeguard for patients.