Most people with severe mental illness (SMI) may now live in the community, but few have jobs and many are socially isolated. Unemployment rates for people with serious mental health problems range from 60% to nearly 100%, and are particularly high if people have additional disadvantages in the labour market — being a member of an ethnic minority, having poor educational and employment history or possessing a criminal record. Unemployment is a cause of poverty, physical and mental ill health and is a cost to the community. Paid employment is central to human health and offers financial, psychological and social benefits to people with mental health problems: an income not derived from benefits, social contacts, a social role other than that of psychiatric patient, psychological recovery and possibly symptom reduction. These psychosocial and health gains may follow from any work — paid employment, low paid or unpaid work, training or education. Many mental health service users want jobs and alternatives to welfare dependency and traditional day centres. The Government wants to improve health and to reduce welfare spending and social exclusion. For deinstitutionalisation to achieve social integration and employment as well as relocation there is a need for a range of actions; public psycho-education, political reform and development and research of modern alternatives to sheltered work and industrial therapy.

This high unemployment is as much a product of social factors — discrimination and stigma, organisational policies and practices, the regulations of the benefits system and the economy — as of the personal consequences of mental health problems. Most employers and employees are not yet ready to work alongside people with mental health problems and modern organisations may be technologically demanding and stressful, themselves a cause of ill health. Mental health professionals may discourage or ignore employment needs. The social security system, while central to the well-being of unemployed people with disabilities, traps them with ‘penal levels of means testing’, making it difficult to return to anything but well-paid jobs, and precluding part-time earnings above a token allowance (Reference Bray, Hart and AnsonBray et al, 1997). The local and national labour markets, economic cycles and cultural factors all have powerful effects (Reference WarnerWarner, 1994).

The Government is using a combination of anti-discrimination legislation and reform to the welfare, tax, employment and health systems — in a context of (and probably dependant on) economic growth with relatively low unemployment and high job availability. The Disability Discrimination Act 1995 requires employers to make ‘reasonable adjustments’ to jobs and work places and has recently been reinforced by the Disability Rights Commission. This is combined with wider anti-stigma campaigns — as in the Department of Health's Impact (1999). There is an evolving process of change to the tax-benefit system and to employment services, with a number of pilot — and as yet, unevaluated — schemes including the New Deal for Disabled People, the Disabled Persons Tax Credit and the ONE service, which provides a “single gateway to benefits… a personal adviser… (and) a work focused interview” (Department of Social Security, 2000). The National Service Framework (NSF) for Mental Health (Department of Health, 1999) states that:

The NSF notes the importance of work to the wider community of people with mental illness, including those using primary services, and requires action within the enhanced Care Programme Approach, with paid employment an outcome indicator as “local circumstances allow”. This places vocational and occupational issues clearly within the remit of community mental health teams working with people with SMI.

Action by a community team will need expertise and knowledge — an understanding of the wider issues restricting the employment of the people they work with, as well as local knowledge of available vocational services and an approach to assessing changing vocational needs. Successful employment will depend on the local labour market, the national economy and the welfare system, as well as on the personal resources and choices of the person (Reference AnthonyAnthony, 1994). Assessment is difficult because specific symptoms and diagnoses do not predict employability or performance, although people's productivity can be disrupted by persistent features and the fluctuating and recurring course of much mental disorder and the side-effects of treatment. Assessment is best carried out in work settings, rather than with interviews or tests. Some people with mental health problems can get or keep jobs with good treatment and stability of their disorder and liaison with their employers. The unemployed may need contact with mainstream employment services or disability employment advisers at Job Centres. Others, including those with other disadvantages or with high benefit incomes, may not be able to compete in the labour market or deal with the demands of open paid employment. However, they can work in, gain skills from or be supported by specialist community vocational rehabilitation projects; work projects or employment schemes. These aim to enable people to work as much and for as much as they can within benefits regulations.

In the UK there is a diverse, confusing and vibrant vocational rehabilitation sector. Work projects are generally small community schemes or discrete programmes in traditional day centres, provided by health or social services or the voluntary sector, with short-term funding from multiple local and central Government sources and the European Commission (Reference Pozner, Ng and HammondPozner et al, 1996). There are model work projects nationally in a context of patchy local development, no statutory responsibility and no vocational rehabilitation professional training, with staff coming from a variety of backgrounds. Most areas will have at least one work project, but there will be limited provision both of places and of the various models that have been developed internationally. Service user involvement in the running and decision-making of work projects — and increasingly, their evaluation and research (Smith & O'Flynn, 2000) — is key to training in the complexities of working life and enabling recovery through empowerment. When people work in work projects they may recover, gain confidence and skills, increase their productivity, progress to more work and eventually may enter the labour market.

Work projects use one or more of several distinct approaches, which either create jobs for people or help them access mainstream opportunities. Jobs can be created in social firms; not-for-profit high street businesses (such as restaurants or shops) that operate in the open market, employing a workforce of staff both with and without disabilities and paying them equal market wages. People can progress to the open labour market after working in a social firm, with a track record of paid employment. Social firms need subsidies of 10-40% of their running costs to compensate for lower productivity — the higher the subsidy the more workers with disabilities can be employed (Reference Grove, Freudenberg and HardingGrove et al, 1997). There are more than 2000 work projects in Italy where they started as social cooperatives (part of the Italian Democratic Psychiatry of the late 1970s) and they have been developed across Europe, with about 50 now in the UK. They appear as consumer-run businesses in the USA where they have an emphasis on businesses run solely by service users.

Jobs in social firms are different to those traditionally created in sheltered workshops, which transferred from the industrial therapy units of the asylums. Sheltered work creates part-time jobs, with low ‘therapeutic’ pay not affecting benefits (‘benefits-plus’) in segregated workshops, but these are associated with poor quality products and conditions, reinstitutionalisation and infrequent transition to the labour market. The social enterprise covers a range of models in which sheltered work has been influenced by the social firm's emphasis on marketable products and services and meaningful jobs. In the German literature social forms and social enterprises are distinguished as being market orientated or rehabilitation orientated projects. Social enterprises create benefits-plus part-time work in more commercial and ‘real’ circumstances, but do not generate revenue to pay wages and may not have a mixed workforce. They rely on about 90% subsidy of running costs.

People can be employed in mainstream organisations on usual pay and conditions and off benefits, with support for them, their colleagues and employers by job coaches or employment consultants. These supported employment programmes may include reasonable adjustments (many of which are low cost and easy to implement) and a subsidy to the employer. They have been widely used and evaluated in the US and although initially for people with learning disabilities in the UK, increasingly employ mental health service users (Reference Perkins, Buckfield and ChoyPerkins et al, 1997). Some people will first need employment training (preparation of curriculum vitae or interview training, a job club or dedicated employment agency), specific skills (e.g. new technologies) or qualifications (learnt in segregated vocational training projects or with specialist support in mainstream further education colleges — supported education) to be able to enter employment.

The ‘clubhouse’, originating in New York, offers a combination of these approaches in centres where people with major mental health problems can drop in for social activities, receive training, work, be involved in the running and planning of the clubhouse and try time-limited (typically 6 months) unskilled jobs with local employers to gain employment experience in transitional employment programmes.

Quantitative evaluation of the different models of employment and work project is methodologically difficult because the models are often small — with different user populations and programme details, operating in different local circumstances. Most work project research is quasi-experimental or descriptive, but the international evidence suggests that each model helps different people at different times in their recovery and reintegration (Reference Pozner, Ng and HammondPozner et al, 1996) and that people need access to different models. The “pitfalls of considering any one model to be the magic bullet for all people with SMI” has been termed the single model fallacy in the USA (International Association of Psychosocial Rehabilitation Services, 1999). Work projects are also a cost-effective alternative to traditional day care and are preferred by service users (Reference SchneiderSchneider, 1998).

This is an exciting time for vocational rehabilitation. The political reforms may work to reduce the social barriers to employment, but the Government will need to plan long-term and buck the historical trend of neglecting the employment of all people with disabilities during economic recession. The Disability Rights Commission will need to establish the extent of ‘reasonable adjustments’ for people with mental health problems in the UK. Welfare to Work reforms must avoid the extremes of a system that only pays benefits in return for labour or creates lifelong welfare dependency regardless of “skills, experience or the desire to work” (Reference SchneiderSchneider, 1998). The Government has put a responsibility for action on employment needs in its Care Programme Approach. As each model seems to help different people at different times, and because people may need a series of approaches as they recover and progress, people's employment needs may be met best by developing a range of these models in each area as an integrated system facilitating long-term access to work and individual progression towards employment for even those with severe disabilities (Reference O'FLYNN and IngamellsO'Flynn & Ingamells, 1997). This will need local solutions to local conditions in the context of the wider issues and the involvement of service users and employers. If this strategy is accepted and if work projects can encourage recovery and reintegration (and avoid reinstitutionalisation) when they are used in the UK as mainstreamed services rather than model projects, there will be a considerable need for investment in local vocational services — and professional staff development. It is not clear where this funding will come from — although it may be from existing day care in the short-term and from savings in social security and other spending in the long-term. There is a lot of work to be done if the human and social costs of unemployment are to be reduced and if the mental health service user with a job and a career is no longer the exception.