There has been much debate about effective treatments, service configurations and costs within Britain's mental health care system, but it has largely taken place in academic and management circles. We were interested in the views of those providing care. We organised a meeting of community psychiatric nurses, general practitioners and consultant psychiatrists (funded with an educational grant from Zeneca Pharmaceuticals). Participants worked in various parts of Great Britain, including rural and inner city areas. The authors facilitated the discussion, the emphasis of which was on participants' clinical experience.

Although much of the discussion related to the full range of serious mental illnesses, the primary focus was schizophrenia, the associated costs of which are wideranging and high (Reference KnappKnapp, 1997). Not only are there costs to the NHS, but other public sector agencies are regularly involved (housing, social services, criminal justice), as are the voluntary and private sectors. There are additionally the more indirect costs of lost productivity due to unemployment, impaired work performance, sick leave, premature mortality and family burden, and together these could outweigh the direct costs (Reference Davies and DrummondDavies & Drummond, 1994). Economic costs apart, schizophrenia is associated with considerable intangible costs in terms of suffering and impaired quality of life.

The high and enduring costs of schizophrenia have attracted growing attention from policy-makers, care professionals and others. Evaluations have been commissioned to discover whether costs can be reduced or contained while maintaining or (better still) improving patient and other outcomes. Although there are still comparatively few completed economic evaluations of schizophrenia treatment, they nevertheless offer numerous pointers to cost-effectiveness improvements (Reference Knapp, Almond, Percudani, Maj and SartoriusKnapp et al, 1999). That is, there are proven ways to use existing resources to greater effect. One of the themes running through the discussion at our meeting was whether these cost-effectiveness gains could be realised in everyday practice, rather than in the ‘experimental’ environment of the research setting.

Two key issues emerged from this part of the discussion. First, care for schizophrenia can be worryingly ‘fragmented’ due to the involvement of multiple agencies and professionals, each with their own objectives, constraints, performance requirements and values. Second, there are major inequalities in services and in the needs of populations. Both factors were seen as contributing to variations in standards of care and clinical problems, hampering the pursuit of greater effectiveness and cost-effectiveness, but also reminding us that these two criteria are by no means the only objectives of a publicly-funded care system.

Multiple agencies and fragmentation

Health inequalities

Although the UK has recently celebrated more than 50 years of a tax-funded, national health service, complete with in-built resource allocation formulae to redistribute resources from better to less well-provided areas, a great many inequalities persist. Participants in the discussion were concerned by local inequalities. They emphasised how service availability and quality can vary quite markedly even within small areas:

Often services were described as woefully inadequate:

Some participants were having to ‘borrow beds’ from other districts:

Participants working in inner cities appeared to experience particular problems, as has been noted elsewhere (Johnson et al, 1997). In-patient admission rates are highest in inner city areas (Reference Faris and DunhamFaris & Dunham, 1939; Reference Giggs and CooperGiggs & Cooper, 1987), largely reflecting social drift, although being born in a city confers a slightly higher risk of developing schizophrenia (Reference Lewis, David and AndreassonLewis et al, 1992).

Inequalities in health provision partly reflect differences in morbidity within populations, as well as differences in the accessibility of services. In Western countries the prevalence of schizophrenia is 2-3 times greater among those of low socio-economic status compared with the general population. Areas of high residential mobility are associated with higher rates of schizophrenia (Reference Dear and WolchDear & Wolch, 1987). Most studies of ethnicity in Britain report higher prevalence rates for schizophrenia among African—Caribbean people compared with other groups, and admission rates have been reported to be 3-5 times higher (Reference HarrisonHarrison, 1988).

Ethnic minority groups face particular problems in accessing services. In some communities psychiatric illness carries great stigma, and people in need may avoid psychiatric services, perhaps under family pressure. Some find services unacceptable (Parkman et al, 1997). Mixed gender wards can be a particular source of difficulty. When African—Caribbean and Asian people do consult they are less likely than White people to have their mental health problems diagnosed by their general practitioner, and referral rates from primary to secondary care are also lower. This may partly explain why African—Caribbean people are over-represented in hospital admissions and are more likely to be detained under the Mental Health Act (Reference Davies, Thornicroft and LeeseDavies et al, 1996) and admitted to intensive care facilities. In one study the proportion of African—Caribbean people with a poor outcome of schizophrenia at one year was 2.5 times greater than for White people (Reference Bhugra, Leff and MallettBhugra et al, 1997).

The consequences of health and health care inequalities should be self-evident. When people with psychiatric illness cannot or will not access the volume or standard of care that their symptoms and needs suggest they require, their quality of life may be impaired and their prognosis worsened. Poorly resourced services may find it difficult to recruit staff, potentially setting up a vicious circle.

Ways forward

The views of health professionals have a place in assessing services, and in this paper we have sought to summarise some of their concerns. The views in this paper came from a small group of doctors and nurses; other stakeholders, including patients, relatives and voluntary agencies, were not represented. Nevertheless, the views of this group appear congruent with these expressed more widely and find support in the research literature (Reference Mauthner, Naji and MallisonMauthner et al, 1998).

Participants' concerns about our current mental health care system centred mainly around the themes of fragmentation and inequalities. Their criticisms were not of multi-agency work per se but of poor coordination without adequate time being allocated for working in this way. How can this be overcome? The formation of community mental health teams (CMHTs) can facilitate working between different professions and improve patient care, although it is not a panacea (Reference Mauthner, Naji and MallisonMauthner et al, 1998). Where separate agencies still operate, each needs to be clear about the services it provides, how referrals are accepted and how to communicate with other agencies. Better use of information technology can facilitate communication and improve time and resource management. Regularly updated directories of local services are essential. An increase in the number of psychiatrists and other health professionals would allow higher quality care and could be cost-effective. However, funding issues apart, there are currently insufficient trainees to fill existing consultant psychiatry posts let alone fill additional ones, a situation that also applies to psychiatric nurses (Reference StorerStorer, 1998).

The CPA should in principle enable better coordination but in many areas has proved difficult to implement. A common criticism is that the CPA was not targeted at those who would benefit most (i.e. those with severe and enduring mental illness) but was applied to all those coming into contact with secondary services. Furthermore, inadequate allowance has often been made either for the financial support needed to set systems up or professionals' time needed to run the system.

Joint training can improve understanding and cooperation between agencies. It has been suggested that there should be a clearer policy framework to provide greater incentives for inter-agency working. There has been talk of amalgamating health and social services budgets for mental health, and of course there is at least as much opposition to it. But even such amalgamation would not remove the gulf between agencies and individuals incurring the direct and indirect costs of schizophrenia. The widespread lack of awareness of, and responsibility for the overall economic consequences of schizophrenia and its treatment has arguably been most influential in the worryingly myopic emphasis on short-term, narrowly-conceived economic concerns.

Health professionals need to be more aware of the influence of socio-economic factors on morbidity levels, and also of the problems faced by ethnic minority groups in obtaining access to services. National guidelines should be available on how to research and assess commissioning needs. The allocation of central funding across the country needs to take better account than heretofore of the wide variations in morbidity and provision.

The 1997 White Paper, The New NHS (Secretary of State for Health, 1997), heralded wide-ranging changes in health provision with important implications for psychiatry. Setting service standards through National Service Frameworks and clinical guidelines through the National Institute for Clinical Excellence (NICE) may reduce some problems highlighted in this paper. These developments may also widen the availability of treatments for which there is robust evidence of both effectiveness and cost-effectiveness. In schizophrenia this includes family interventions and atypical antipsychotics both of which (broad) treatment modes have been restricted for apparently non-clinical reasons (Reference Knapp, Almond, Percudani, Maj and SartoriusKnapp et al, 1999). Negative aspects of the reforms include a failure to provide incentives for primary care to become more involved in psychiatric care and the potential for primary care groups to undermine clinical freedom.

Schizophrenia has many deleterious consequences, but there are treatment and support modalities available which can reduce symptoms, improve personal functioning, raise quality of life and lessen the burden on families. Research shows that many of these are also cost-effective. But there are barriers in the way of achieving these desirable interventions — notably those associated with fragmentation and inequalities — which national policy attention and local action urgently need to address.