Evidence-based vocational rehabilitation

Successful forms of vocational rehabilitation for people with severe mental illness, supported employment schemes, have been developed since the 1980's and have been internationally evaluated in randomised controlled trials across a range of countries including the USA, Canada, Australia, Europe and Hong Kong. ^{Reference Drake, Bond and Becker3} One particular approach, Individual Placement and Support (IPS), has proved particularly effective, delivering rates of open employment of above 60% (compared with rates of around 25% in the control services). ^{Reference Bond, Drake and Becker4} Individual Placement and Support is a form of supported employment and differs from other types of vocational rehabilitation which have traditionally taken a ‘train and place’ approach. Traditional vocational services typically focus time and resources on training and supporting people to develop new skills in segregated and sheltered environments. In contrast, the primary goal of the IPS approach is to directly find a job and then provide continued support - a ‘place and train’ approach (Appendix). Adhering to the principles of IPS (fidelity to the model) is a key factor in ensuring success of the programmes and the skills of vocational workers and the quality of their relationship with patients is crucial. Individual Placement and Support schemes have low drop-out rates and generate positive outcomes across several employment outcomes (into work quicker, work more hours per week, longer job tenure), give good personal outcomes, fewer hospital admissions, quicker recovery and are cost-saving. ^{Reference Drake, Bond and Becker3-Reference Knapp, Patel, Curran, Latimer, Catty and Becker5} These schemes are well regarded by patients and are effective in routine practice ^{Reference Cook, Blyler, Leff, McFarlane, Goldberg and Gold6,Reference Rinaldi, Montibeller and Perkins7} and first-episode psychosis.

Despite this high level of evidence, unusual in vocational rehabilitation services, IPS schemes have proved difficult to implement internationally. ^{Reference Becker, Lynde and Swanson8} Why is this the case?

Attitudinal barriers

The most consistently noted individual predictors of success in achieving open employment are motivation, self-efficacy and previous employment history. Attitudinal and structural factors provide more significant barriers. Many patients report being given pessimistic messages about the possibility of work, and although many clinicians may believe that many people with psychoses may be capable of work, they may not see this applying to those on their case-loads. ^{Reference Marwaha, Balachandra and Johnson9} Employers may deny that people with mental health problems can work and prefer not to employ people with a history of mental illness. ¹⁰ The attitudes of clinicians may make it difficult to integrate vocational workers in clinical teams and hampers the development of referral systems. ^{Reference Shepherd, Lockett, Bacon and Grove11} Broader prejudices limit liaison with employers and employment agencies. ^{Reference Shepherd, Lockett, Bacon and Grove11} Clinicians and vocational workers may be protective of their patients and avoid rapid job placement for fear of placing too much stress on them.

Contextual factors

Structural aspects of the labour market are influential and historically affect the employment and rehabilitation prospects of people with severe mental illness. ^{Reference Warner12} High local unemployment rates reduce the effectiveness of the IPS programmes as shown in the multicentre trials. ^{Reference Cook, Blyler, Leff, McFarlane, Goldberg and Gold6,Reference Burns, Catty, Becker, Drake, Fioritti and Knapp13} However, even in these conditions the IPS schemes perform better than control conditions. Although unemployment rates may reduce the likelihood of attaining open employment, obtaining work is still possible and we need not abandon important vocational schemes in times of austerity as the effects of this barrier should even out across the economic cycle. ^{Reference Rinaldi, Montibeller and Perkins7}

Most of the early trials of supported employment were conducted in the USA and several commentators have highlighted the problem of transferring these results to countries with different labour market conditions. The regulation of employment and social security policies provide considerable barriers to employment outside the USA and comparison of IPS trials carried out in the USA show higher overall rates of open employment (62%) than those undertaken in other countries (47%). ^{Reference Bond, Drake and Becker4} However, these international differences may be overemphasised, as although these contextual differences may be important, the USA and the UK labour markets are similar across several key characteristics and can be loosely grouped together to illustrate a single ‘Anglo-Saxon’ model which contrasts with the continental European approach. ^{Reference Ngai and Pissarides14} We may need to re-examine factors that may explain the differences in results between the USA and other countries.

The IPS literature has tended to look at disincentives to employment for individuals and ignored those factors that reduce the likelihood of employers taking on workers with mental health problems, such as the levels of tax on labour or regulations on job security. Several studies have noted the effects of USA social security systems on employment for people with schizophrenia. ^{Reference Cook, Blyler, Leff, McFarlane, Goldberg and Gold6} Most people who obtain work through IPS programmes in the USA are employed in part-time jobs, and of those people who were on social security benefits, only 4% earned enough to allow them to leave these. ^{Reference Bond, Drake and Becker4,Reference Cook, Blyler, Leff, McFarlane, Goldberg and Gold6} This ‘benefits trap’ is seen in other countries, ^{Reference Burns, Catty, Becker, Drake, Fioritti and Knapp13} but in the USA movement off social security also means loss of health insurance payments, something that does not happen in the UK. Of course, these contextual factors exist internationally and overall they do impose a significant influence on the success of the IPS schemes and their implementation, but they will vary specifically from country to country depending on local welfare and employment policies and rules. In general, these benefit systems need to be sufficiently flexible to encourage people with fluctuating conditions to enter employment while protecting them in periods of crisis and avoiding unnecessary coercion and problems such as in-work poverty. With thought and consideration given to local systems it has been shown that some of these structural issues are not insurmountable. ^{Reference Rinaldi and Perkins15}

Fidelity to the model

Bond et al ^{Reference Bond, Drake and Becker4} have recently pointed out the effects of low fidelity of the IPS model in determining the success of IPS programmes and their dissemination. Provision of technical support to new schemes for implementation and fidelity determines their success ^{Reference Becker, Lynde and Swanson8,Reference Becker, Drake, Bond, Nawaz, Haslett and Martinez16} and use of the fidelity scale can help the adoption and effective implementation of IPS. ^{Reference Becker, Lynde and Swanson8,17} Other commentators have noted that barriers to implementation and success relate to the essential elements of the IPS model. ^{Reference Latimer18} For example, integration of vocational and clinical services, lack of support and follow-up of patients by vocational staff and lack of integration of workers into the clinical team. Scrutiny of the principles of IPS themselves (Appendix) give clear pointers to the factors governing successful implementation of IPS schemes in organisations.

Organisational factors

The effect of organisational context on clinicians' attitudes, work performance and supervision is crucial to implementing IPS schemes with sufficient fidelity. In England, ^{Reference Boyce, Secker, Floyd, Grove, Johnson and Schneider19} organisations with low-fidelity IPS schemes had eligibility criteria for access to the schemes, limited the extent of follow-up and outreach and developed high case loads and protective attitudes to patients. Those with high fidelity had organisational policies that supported the IPS approach and provided training for their vocational workers. The quality and success of supported employment depends on an acceptance and understanding of the principles of the IPS model and a reflection of these in the delivery of the services. Supported employment is an innovation with a strong evidence base, but it is subject to the same barriers as many other innovative schemes for its diffusion and its widespread adoption will require top-to-bottom changes to attitudes, structure and practice throughout our health and employment services. Support for these schemes requires a shift in our approach to working with people with severe mental illness. ²⁰

Changes to organisational culture are central to the development of new and innovative services. The Implementing Recovery through Organisational Change (ImROC) project in the UK, the Johnson & Johnson - Dartmouth Community Mental Health Programme in the USA ^{Reference Becker, Lynde and Swanson8,20} and the Janssen-Cilag Catalyst IPS programme are examples of programmes set up to support organisations to implement new approaches. New services require changes in attitudes and working practices; organisational structures that support change; provision of relevant training; and new mechanisms of finance. In the UK, commissioners of services need to be made aware of the already existing commissioning guidance for vocational rehabilitation. ²¹ The undertaking of a multicentre UK trial and inclusion of the evidence for IPS in National Institute for Health and Care Excellence guidance is now overdue and essential.

Creation of new IPS services requires account to be taken of local conditions and will need broader policy changes to encourage more flexible employment regulation and welfare benefit schemes. Broadly, facilitating the development of improved and evidence-based vocational schemes for people with mental illness will require integration of mental health, welfare benefits, social care and employment support at every level, from local services to governmental departments. ^{Reference Perkins, Rinaldi, Gregg and Cooke22} The recent Schizophrenia Commission report has revealed yet again the parlous state of services for people with schizophrenia and recommended a greater focus on rehabilitation and recovery-based services. ²³ Increasingly, international mental health policy strives to implement a recovery-focused approach within mental health services, yet within this employment is often marginalised or ignored. Given the importance that patients place on employment as a key element of their recovery, IPS needs to move from being seen as a ‘scheme’ to a core integrated facet of future services. If we are to achieve the outcomes and quality desired by patients, then we really do need to consider employment as an outcome measure of effective mental health services.

Appendix