Supported employment for people with severe mental illness

Aug 30 2016

The employment rate amongst people with severe mental illness such as schizophrenia or bipolar disorder is extremely poor (Marwaha et al., 2007). Re-entering the workforce after suffering mental illness can be daunting, whether due to stigma in the workplace, poor educational attainment, or the distress caused by symptoms themselves (Glozier, 1998).

Nonetheless, people with severe mental illness consistently show a desire to work, and the benefits of employment to such individuals are considerable. Benefits include improved self-esteem and general wellbeing, a sense of social support and independence, and reduced use of mental health services (Bush et al., 2015).

Sadly, mental health services are accused of focusing too much on symptom reduction, and too little on helping people regain employment (Secker, 2001). Traditional models of work rehabilitation emphasise lengthy training before work placement. This delay may exacerbate feelings of social exclusion (Waghorn et al., 2009). A more modern approach, supportive employment, rapidly places individuals into work while providing support and training alongside. The most successful iteration of supportive employment is ‘Individual Placement and Support’ (IPS). IPS places competitive employment as the primary goal by integrating vocational and clinical services towards this end. IPS is at its most successful when fidelity to its principles are high (Drake et al., 2012).

IPS is now seen as the ‘gold standard’, but only in the USA. There has been doubt as to whether IPS can generalise to other countries, given variation in economic growth, availability of employment, and health services (Boardman & Rinaldi, 2013). Published in the British Journal of Psychiatry, a recent study examined whether IPS is effective in different countries by carrying out a meta-analysis of available international studies (Modini et al., 2016).

This recent international review meta-analysed the published RCTs on supported employment for people with severe mental illness.

Methods

Two researchers independently identified relevant studies from medical databases, and one author extracted the following data for analysis: sample characteristics, country of origin, length of follow-up and rate of competitive employment achieved for the experimental and control groups.

Only randomised controlled trials (RCTs) were included, although true masking of group allocation for all involved was impossible given the nature of IPS. Only studies scoring moderate to high on an IPS fidelity scale were selected. Study quality rated as ‘poor’ resulted in exclusion. Severe mental disorders included schizophrenia, bipolar, and depression with psychotic features.

Three meta-analyses were carried out using the random effects model, which assumes that the true effect varies across studies. This was appropriate given the large variation in populations and socioeconomic factors.

The first was to determine the effectiveness of IPS compared with traditional rehabilitation. Meta-regressions analysed associations between IPS and region of origin, unemployment rate and GDP (gross domestic product) growth (total value of goods produced and services provided in a country annually, obtained from World bank database).

The second and third meta-analyses determined whether IPS effects would remain stable over a 2-year period. Main outcome was measured as binary (yes/no), so could be calculated as Risk Ratios (RRs). This essentially calculated the likelihood of successfully entering competitive employment using IPS relative to traditional rehabilitation models.

Results

In total, 19 studies were included:
- 4 studies in Asia and Australia
- 6 studies in Europe
- 9 studies in North America
Two authors found that two studies were of ‘fair’ quality, and 17 were ‘good’, with good inter-rater reliability (Cohen’s Kappa=0.74)
GDP growth ranged from 0.7% to 6.6% across countries, indicating a wide variability
A funnel plot used to visualise publication bias (i.e. when only positive results get published), found some slight bias, with smaller studies showing greater estimates of the effectiveness of IPS than is likely
Meta-analysis 1:
- Pooled risk ratio (PRR) for gaining competitive employment for IPS compared with traditional vocational rehabilitation = 2.40 (95% CI 1.99 to 2.90)
- Meta-regressions did not indicate that unemployment rate or geographic region accounted for this finding, although IPS was relatively more effective in higher GDP growth (b=0.13, s.e.=0.59 95% CI 0.00 to 0.25, P= 0.045)
- However, even when only including low GDP growth studies, IPS remained more effective, with PRR of 1.82 (95% CI 1.37 to 2.40)
Meta-analysis 2:
- PRR for gaining employment within first year after IPS started, compared with traditional vocational rehabilitation was 2.59 (95% CI 1.88 to 3.56)
Meta-analysis 3:
- PRR for gaining competitive employment within the second year was 2.41 (95% CI 1.96 to 2.97).

These results suggest that Individual Placement and Support is more than twice as likely to lead to competitive employment when compared with traditional vocational rehabilitation.

Discussion

This is the first meta-analysis examining the effectiveness of IPS in different countries and economies. Overall, Individual Placement and Support was effective in helping individuals with severe mental illness to gain competitive employment across multiple settings. In fact, those in the IPS group were more than twice as likely to get employment, regardless of country, unemployment rate, or GDP. Furthermore, these effects remained stable over a 2-year period. Although IPS was slightly less effective during times of economic downturn, it still remained significantly more effective than traditional rehabilitation.

Individuals do not need to be symptom-free before returning to work. The longer someone is out of work the less likely they are to return, partly due to deterioration in skills, but more likely owing to increased anxiety and perceived vulnerability (Henderson et al., 2011). IPS emphasises getting people back to work quickly with support from both clinical and employment services. Current barriers to the wider uptake of IPS include unhelpful attitudes by clinicians and employers, the way in which labour market and welfare are structured, and how mental health services are organised (Boardman, 2013). The authors argue these results should give ammunition to stakeholders wishing to pressure policy-makers into addressing these barriers.

Individual Placement and Support was significantly more effective than traditional rehabilitation, even in hard economic times.

Conclusions

The authors conclude that:

This study provides strong evidence that IPS is effective in a variety of international settings… policy-makers and clinicians need to begin addressing the barriers preventing wide-scale use of supported employment principles, to ensure access to high-fidelity IPS is made available to those with severe mental illness regardless of where they live or the prevailing economic conditions.

Strengths and limitations

This is the first study to examine the effectiveness of supported employment internationally
There was variation in what each study deemed a sufficient length of employment as a successful outcome, ranging from 1 day to 1 month
Study selection was limited to those investigating high fidelity IPS, so no data on less efficient implementation of IPS was examined
Some publication bias was present in smaller studies; sadly this is an issue across all research
No successful blinding/masking of anyone involved in each study was possible due to nature of IPS
It is unclear at present exactly which components of IPS are the most effective.

Further work will be required to ascertain which components of Individual Placement and Support are the most effective.

Links

Primary paper

Modini M, Tan L, Brinchmann B, Wang M-J, Killackey E, Glozier N, Mykletun A, Harvey SB. (2016) Supported employment for people with severe mental illness: systematic review and meta-analysis of the international evidence. The British Journal of Psychiatry, bjp. bp. 115.165092. [Abstract]

Other references

Boardman J, Rinaldi M. (2013) Difficulties in implementing supported employment for people with severe mental health problems (Editorial). The British Journal of Psychiatry, 203(4), 247-249.

Bush PW, Drake RE, Xie H, McHugo GJ, Haslett WR. (2015) The long-term impact of employment on mental health service use and costs for persons with severe mental illness. Psychiatric Services. [Abstract]

Drake RE, Bond GR, Becker DR. (2012) Individual placement and support: an evidence-based approach to supported employment: Oxford University Press. [OUP book page]

Glozier N. (1998) Workplace effects of the stigmatization of depression. Journal of Occupational and Environmental Medicine, 40(9), 793-800. [PubMed abstract]

Henderson M, Harvey S, Øverland S, Mykletun A, Hotopf M. (2011). Work and common psychiatric disorders (PDF). Journal of the Royal Society of Medicine, 104(5), 198-207.

Marwaha S, Johnson S, Bebbington P, Stafford M, Angermeyer MC, Brugha T, Azorin J-M, Kilian R, Hansen K, Toumi M. (2007) Rates and correlates of employment in people with schizophrenia in the UK, France and Germany. The British Journal of Psychiatry, 191(1), 30-37.

Secker BG, Patience Seebohm J. (2001). Challenging barriers to employment, training and education for mental health service users: The service user’s perspective. Journal of Mental Health, 10(4), 395-404. [Abstract]

Waghorn G, Lloyd C, Clune A. (2009) Reviewing the theory and practice of occupational therapy in mental health rehabilitation (PDF). The British Journal of Occupational Therapy, 72(7), 314-323.