Letters

Psychiatric home treatment

Vigorous, well designed trials are needed

Home treatment works

Treatment at home is nationwide and successful

Vigorous, well designed trials are needed

EDITORSmyth and Hoult ask why the implementation of home treatment has been "delayed" in the United Kingdom and call for more "sophisticated evaluations" while continuing the same tired (and tiresome) polemic masquerading as science that holds up such progress.¹ They present a categorical view of home treatment versus "the rest," where only one of "the 12 features of an effective home treatment team" (several visits daily) is not available in some form in routine British mental health practice. How do Smyth and Hoult know that each of these 12 features is necessary or makes a difference? They do not. The series of studies they quote are all "black box" trials of complex and often poorly defined experimental services against even more poorly defined control services (often simply called standard care). To prove effectiveness, carefully controlled trials that vary only one component are needed. Smyth and Hoult did not quote our study (of which they were aware), which is one of the first community care studies to do this.²

Crisis intervention makes intuitive sense to physicians and surgeons used to myocardial infarcts and obstructed hernias. It does not sustain close scrutiny in mental healthbreakdowns take days and weeks, not hours. Mental health services that are well linked to primary care and that offer reasonable access soon find that "crises" become a small part of their work. (The exception is in inner cities with many homeless mentally ill patients.) As Pelosi and Graham remark, crisis intervention services soon evaporate.³ These services are either unsuccessful and collapse or they are successful and staff build up good relationships with other service providers and do themselves out of a job. To describe Madison, with its fixed caseload of patients (many receiving treatment for more than a decade), as a crisis service is almost mischievous. Pelosi and Jackson are right to point out that the references cited by Smyth and Hoult are out of date and irrelevant. These studies' control groups receive mainly poorly coordinated, outpatient care from isolated, office based practitioners. Home treatment teams and crisis intervention need to show their sustainable superiority over well coordinated modern care.

Community psychiatry is the victim of too many strong opinions. We need to take a more humble, practical approach to establishing knowledge, and we need to learn from the rest of medicine. Individual components of complex interventions should be identified and subjected to rigorous, well designed trials before we call them "effective." We also need to acknowledge changes outwith our discipline that may make earlier research findings redundant. Conventional British mental health treatment already contains and delivers most of the features of "home treatment" proposed by Smyth and Hoult.

Tom Burns, head, section of community psychiatry. 
Department of General Psychiatry, St George's Hospital Medical School, University of London, London SW17 0RE tburns{at}sghms.ac.uk

1.	Smyth MG, Hoult J. The home treatment enigma. BMJ 2000; 320: 305-308[Free Full Text]. (29 January.)
2.	Burns T, Creed F, Fahy T, Thompson S, Tyrer P, White I, for the UK 700 Group. Intensive versus standard case management for severe psychotic illness: a randomised trial. Lancet 1999; 353: 2185-2189[CrossRef][Medline].
3.	Pelosi AJ, Jackson GA. Home treatmentenigmas and fantasies. BMJ 2000; 320: 308-309. (29 January.)

Home treatment works

EDITORThe polarised debate about home treatment presented by Smyth and Hoult and Pelosi and Jackson is frankly depressing: as usual the truth lies somewhere between the two extremes. ^{1 2} Smyth and Hoult argue that the research base for home treatment is well established and leads to a two thirds reduction in hospital admissions, but their definitions of service models are imprecise. If home treatment is defined as an acute, short term intervention to avoid hospital admission (as in the authors' north Birmingham model) then three of the eight original references cited are not of home treatment at all: two took place outside the United Kingdom, and two were conducted more than 20 years ago.

The only recent, well conducted study of home treatment in the United Kingdom is that of Minghella et al,³ but this is not a randomised controlled trial and needs replication. However, as clinicians working in a well developed home treatment service in inner Manchester,⁴ we agree with many of the advantages of the model cited by Smyth and Hoult and dispute most of Pelosi and Jackson's criticisms.

In our experience, if hospital admission is to be avoided for people who are acutely ill, it must be possible for staff to visit at least three times a day and to be available to patients and carers at all times. Intensity and continuity of this kind is hardly ever available from community mental health teams or primary care.

We agree that it is critical for community workers to remain in contact with their patients throughout periods of illness, and we successfully work alongside our community team while patients are in home treatment. The work is intensive and may be more demanding of medical time but staff seem to prefer working in a service of this kind than in an inpatient setting, and we have not found recruitment or retention to be a problem.

Pelosi and Jackson criticise Smyth and Hoult for their use of anecdote, but the satisfaction involved for patients, carers, and staff in successfully treating a patient with a first onset psychosis without recourse to hospital can be enough to convert even the most diehard cynic. Pelosi and Jackson should try home treatment.

Judy Harrison, consultant psychiatrist. 
John Marshall, service manager. 
Home Options Service, Department of Psychiatry, Central Manchester Healthcare NHS Trust, Manchester M13 9WL

1.	Smyth MG, Hoult J. The home treatment enigma. BMJ 2000; 320: 305-308. (29 January.)
2.	Pelosi AJ, Jackson GA. Home treatmentenigmas and fantasies. BMJ 2000; 320: 308-309. (29 January.)
3.	Minghella E, Ford R, Freeman T, Hoult J, McGlynn P, O'Halloran P. Open all hours: 24 hour response for people with mental health emergencies. London: Sainsbury Centre for Mental Health, 1998.
4.	Harrison J, Poynton A, Marshall J, Gater R, Creed F. Open all hours: extending the role of the psychiatric day hospital. Psychiatr Bull 1999; 23: 400-405[Abstract/Free Full Text].

Treatment at home is nationwide and successful

EDITORIn their response to Smyth and Hoult,¹ Pelosi and Jackson have dismissed an anodyne but comprehensive presentation of empirical research regarding home treatment, and with it the case for the inclusion of crisis care provision in the NHS.² Pelosi and Jackson discount published evaluation studies, countering with anecdotal evidence concerning the practice of crisis care.² We wish to offer new evidence, gleaned from a nationwide survey of crisis service provision and concerning the scale and character of this provision, that adds to the debate and contextualises the argument.

Firstly, Pelosi and Jackson describe an out of hours crisis team that experienced a low level of take up of the service. We too have carried out a case study that shows a low level of take up (K Hogan et al, unpublished report for Walsall Health Authority, 1997). This was in part occasioned by the fact that few general practitioners (the only means of referral) knew that the service existed. Pre-existing patterns of patient management, particularly referrals, take time and commitment from senior professionals to adapt to the provision of new services.

Secondly, the staff of Pelosi and Jackson's cited project became involved with clients' emotional and social problems and were therefore distracted from the needs of people with severe mental illness. Our research suggests the contrary, in that many crisis systems specifically do not deal with such problems but rather concentrate on providing support for those with severe mental illness (over 55% of services reported targeting this client group). These data are based not on a case study but on a survey of all services extant at 1 May 1999.

Thirdly, Pelosi and Jackson commended the work of general practitioners as the people who have known clients for years, and they noted general practitioners' enthusiasm for care of patients with mental illness. Our work shows that of 150 crisis services sampled, general practitioners were the major source of referrals in most cases. Clearly, large numbers of general practitioners value and make use of crisis services as an integral part of their care of the mentally ill.

Finally, Pelosi and Jackson referred to inexperienced clinicians setting up services. However, hundreds of services are in operation and each has to have a responsible medical officer. From our record, crisis teams have been operating for an average of 28 months, giving 308 years of service operation and hence experience.

Moreover, we would point out that the expertise contained in crisis services in the United Kingdom does not reside wholly, or even largely, in psychiatry. Rather, the majority of crisis teams (55%) are staffed by nurses only, and although a minority (45%) are multidisciplinary, psychiatrists rarely figure as a significant element of service provision.

Kevin Hogan, leader. 
k.hogan{at}wlv.ac.uk

Sarah Orme, research fellow. 
Psychology Division, University of Wolverhampton, Wolverhampton WV1 1SB

1.	Smyth MG, Hoult J. The home treatment enigma. BMJ 2000; 320: 305-308. (29 January.)
2.	Pelosi AJ, Jackson GA. Home treatmentenigmas and fantasies. BMJ 2000; 320: 308-309. (29 January.)