opinion & debate Researchable questions to support evidence-based mental health policy concerning adult mental illness

Policy makers find much mental health research irrelevant to their concerns. What types of research would directly assist those who formulate policy? The two purposes of this paper are (i) to identify important gaps in completed research, particularly in relation to the National Service Framework (

The evidence base for such a programme is poorly developed, as shown in Table 1, and no relevant systematic reviews have been completed. In our view, an evidence base is needed that is not confined to measures to combat discrimination, but which will cover desirable employment policies, the effectiveness of health improvement programmes and the possibilities for prevention.

Disability and discrimination
Numerous problems under this heading might include studies providing answers to the following questions: What are the outcomes of providing employment services aimed at helping employers to take a more flexible joint decision to use the Disability Discrimination Act, and to assist general practitioners (GPs) in their use of sickness certificates?; What is the relative cost-effectiveness of forms of vocational placement schemes for people severely disabled by mental illness?; Are interventions that use the Disability Discrimination Act's 'reasonable adjustment' cost-effective for people with mental health problems?; What methods are necessary to assess the costs and outcomes of anti-discrimination activities in mental health; and How do such interventions affect knowledge, attitudes and behaviours?

Mental illness and employment
We suggested three possible studies under this general heading: Which interventions reduce sickness absence Thornicroft et al Evidence-based mental health policy due to mental health problems, among the general workforce?; What are the costs and benefits to the individuals and the organisation of policies for the employment of service users?; What are the clinical, social and economic outcomes of changes to welfare benefit regulations for people with severe mental illness; and How can disincentives to return to work be reduced? Prevention Which methods of implementing early intervention for psychosis in England are more cost-effective? Which interventions reduce the incidence of mental illness in high-risk groups?

Health improvement programmes
What methods need to be developed to make assessments of the mental health impact of health improvement programmes? Do health improvement programmes improve mental health of the population? Do improvements in the built environment lead to improved mental health? Do interventions to improve social capital affect mental health?

Standard 2: Primary care services
This standard requires that any service user who contacts their primary care team with a common mental health problem should have their mental health needs identified and assessed, and be offered effective treatments -including referral to the specialist services, should they require it (Department of Health, 1999a: p. 28).
Much of the basic research into this problem has already been addressed, but there remain some important gaps in our knowledge, including the best use of non-medically qualified staff.
Mental health skills in primary care Some fairly fundamental questions still need to be addressed here: What proportion of individuals with common mental disorders obtain effective treatment?; How can the mental illness detection skills of GP trainees be improved?; How can primary care workers who detect depression then assess its severity?; and How can GPs improve their skills in the psychological management of depression?

Use of non-medical staff
The key questions appear to be: Can primary care staff other than GPs be trained to identify mental distress?; and How cost-effective are training courses in treating common mental disorders for practice nurse health workers? In addition, there are problems that need to be solved concerning the use of non-medically qualified staff to assist with the work: Can 'gateway' workers carry out mental health work in primary care as cost-effectively as other staff?; and Can the new graduate mental health workers be trained to deliver mental health treatments cost-effectively?

New treatments
Can patients be helped in primary care to give up or reduce alcohol consumption? Can computerised treatments for mental disorders commonly encountered in primary care be used cost-effectively?

Indications for referral
Can referrals to specialists be made using evidence-based criteria?
Standard 3: Access to services around the clock, 365 days a year This standard requires that any individual with a common mental health problem should be able to make contact round the clock with local services necessary to meet their needs and receive adequate care. It is envisaged that NHS Direct might provide first level advice and referral to specialised help lines or local services (Department of Health,1999a: p. 28).

NHS Direct
We need a study to address the mental health component of the work of NHS Direct: How cost-effective is NHS Direct for callers with mental health problems?
Standard 4: Severe mental illness: care in the community for those on the Care Programme Approach (CPA) All those on the CPA should receive care which optimises engagement, anticipates or prevents a crisis, and reduces risk. Care should be available 24 hours a day, 365 days a year.They should have a copy of a written care plan which includes the action to be taken in a crisis by the service user, their carer, and their care co-ordinator, and advises their GP how they should respond if the service user needs additional help.This plan should be regularly reviewed by their care co-ordinator (Department of Health,1999a: p. 41).
The proper treatment and care of people living with severe mental illness, although more thoroughly investigated than many other areas (see Table 1), also leaves more unknown than known. The best organisation of resources to enable care to be offered 365 days a year round the clock has already been mentioned. There are also unsolved problems in assessing the form of community care: How is assertive outreach implemented in practice, and what is its impact on patient outcome?; Where crisis plans are concerned, what are the effects of such plans on rates of admission to hospital, and on staff, carer, and user satisfaction?; What are the effects of giving copies of written care plans to people subject to the provision of the CPA?; One of the most difficult groups of patients are those with dual diagnosis (psychosis and substance misuse) -which interventions are most cost-effective in helping such patients?
Standard 5: In-patient care for severe mental disorders This standard states that all those requiring care away from their home should have timely access to a hospital or alternative place that is in the least restrictive possible environment, and as near to home as possible.They should also be given a copy of a written after care plan agreed on discharge which sets out the care and rehabilitation to be provided, identifies the care co-ordinator, and specifies the action to be taken in a crisis (Department of Health,1999a: p.14).
The shortage of beds and intensive alternative forms of support in our inner cities means that there are still unsolved problems of resource allocation: What is the most equitable method for allocating resources to the mental health services? There is also a need to answer important questions about the desirable practices within hospital units: Which active treatment components are cost-effective in general adult and in intensive care inpatient settings?; Are restraint and seclusion clinicallyeffective in-patient settings?; How can levels of violence on in-patient units be reduced?; What are the costs and outcomes of advocacy services?
Extra-mural residential care What types of accommodation, and in what capacity, are necessary for patients who are inappropriately placed in general adult and in intensive care in-patient settings in England? Which types of high-support residential care are cost-effective for which type of patient?

Standard 6: Caring about carers
All individuals who provide regular and substantial care for a person on the CPA should have an assessment of their caring, physical and mental health needs, repeated on an annual basis, and have their own written care plan which is given to them and implemented in discussion with them (Department of Health, 1999a: p. 64).
The national strategy (Caring about Carers; Department of Health, 2000) has been accompanied by additional resources, and one systematic review has been conducted (Pharoah et al, 2002) that showed that family and relative support was useful and reduced relapse rates. Szmukler and Holloway (2001) concluded that evidence has been best evaluated in respect of brief interventions and the studies conducted to date show burden, distress and coping are largely unaffected. We need to know how care coordinators can best be trained: Does specific training for them in care planning for carers improve carers' mental health status, quality of life or improve their financial situation?; Which interventions are effective to improve the impact of caregiving upon carers?; How often do carers of people with mental illness request carer's assessments, and what are the consequences?; and Does active participation in the care planning process improve the carer's mental health, quality of life and service satisfaction?

Standard 7: Preventing suicide
This standard expects local health and social care communities to prevent suicides by taking action on appropriate sections of the previous standards (Department of Health1999a: p.76).
The most recent report of the confidential enquiry Safer Services (Department of Health, 1999b) highlighted the large number of suicides by in-patients (accounting for 4% of the national total). The confidential enquiry identified particular patient characteristics and made specific Thornicroft et al Evidence-based mental health policy opinion & debate recommendations regarding the environment in inpatient settings and clinical practice. A study of suicide and self-harm in in-patient units (Gournay, 2001) indicated that at least some of the suicides could be prevented by specific training initiatives for in-patient staff.
Studies are necessary to establish which measures are effective in enabling in-patient units to reduce suicide rates, and to discover which measures reduce suicide rates among patients discharged to the outside world?
Dissemination of evidence-based methods for minority ethnic groups It is established that Black service users receive less acceptable treatment than White service users in the following ways. They: . are less often identified in primary care as suffering from common mental disorders . receive psychological treatments less often . show significantly lower rates of satisfaction with mental health services . are more often brought to services by police . are more often treated coercively, using the Mental Health Act . are more likely to be treated in locked wards.
Investigation is, therefore, necessary not to further describe these difference, but to establish effective and acceptable methods to provide treatment and care to these populations. Which types of service provide acceptable forms of access for minority ethnic groups? How can psychological and social therapies be provided in culturally appropriate ways? What are the needs for provision of mental health services among the different groups of refugees and asylum seekers?

Conclusions
Fifty researchable questions have been proposed that, if addressed by research, could provide a more consolidated evidence-base upon which to develop future mental health policy in England. These questions are intended not to be prescriptive or exhaustive, but rather to stimulate debate on how policy may best be informed by research findings in the coming years.