Evidence that Labour is ignoring: Employment Advisers in IAPT

This is the fourth in my series of blog posts looking at employment support programmes for sick and disabled people. So far, the most successful has been Work Choice, which achieved 37% of participants getting work and staying in work for 13 weeks, and 23% staying in work for 26 weeks. However, Work Choice predominantly helped people who were fit for work, due to the criterion that people be likely to be able to work 16 hours/week within six months. Work Choice also provided much more intensive and personalised support than other support programmes, with weekly meetings of an hour or more for six months.

This blog looks at Employment Advisers in the ‘Improving Access to Psychological Therapies’ services.

The support was for people in work but struggling, off sick from work, or looking for work. It therefore excluded those who were long-term unfit for work, although the ‘off sick’ could include people who would not be able to return to work. All three groups are likely to be, on average, healthier than people who have been assessed as unfit for work.

Participation was voluntary, and the majority of people offered employment support did not take it up. Of those who did take up support, 72% said it was because “they felt it could help them get back into work”. 32% liked “the idea of receiving additional help and support”, and 24% took up support “because their IAPT therapist encouraged them to”. People were more likely to take up support if they were having more problems in work, were older, or had a higher depression score. Those who were economically inactive (homemaker/carer, 5%; full-time student, 3%; or long-term sick or disabled, 9%) were least likely to take up support.

There was no control group, in which people requested but did not receive support, so matched controls were used for statistical analysis instead. Whilst these could match people on characteristics such as sex, age, benefit receipt, and health, it could not match people based on other factors such as motivation to remain in or find work, confidence, or experiences in work or in seeking work. It is plausible that people who did not take up employment support have difference experiences of work which impact their likelihood of remaining in or taking up work.

Those who were in work at the start of IAPT were less likely to be in work at the end than their matched controls, at 88% compared to 93%. They were also more likely to report being on SSP, at 7% vs 3%. There was no impact on depression (PHQ-9 ‘caseness’: 89% on entry to IAPT; 35% and 36% at final therapy session), anxiety (GAD-7 ‘caseness’: 90%; 39% and 40%), or quality-adjusted life years.

In-work clients who saw an EA were more likely to say, at their final therapy session, that their mental illness impaired their ability to work. Whilst both the treatment group and matched controls had improved on this score, the control group improved more. However, the lack of difference in mental health and in daily functioning suggests that this was not because the EA group had worse health, but may instead be because seeing an EA helped these people to become more aware of the difficulties they faced in work. This runs contrary to DWP Theory of Change, which assumes that many sick and disabled people over-state, rather than under-state, their level of difficulty.

Of those on SSP, more of those who had seen an EA reported being unemployed (11% vs 5%). This was matched by fewer people reporting being in work, long-term sick or disabled, or ‘other’; but these results did not reach statistical significance. They were more likely to be in receipt of out-of-work benefits 12 months later (17% vs 11%), which suggests that the higher jobseeking levels (i.e., reporting being ‘unemployed’ rather than ‘long-term sick or disabled’ or ‘other’) did not translate into obtaining paid work. By 18 months this had dropped to 15% vs 11% and was not statistically significant.

For people on SSP, 67% of those who saw an EA experienced a reliable improvement in depression compared to 60% of matched controls, and 72% vs 65% experienced a reliable improvement in anxiety. However, improvements in mental health compared to matched controls were not large enough to reach statistical significance on IAPT reliable improvement, recovery, or reliable recovery measures. There was also no statistically significant difference on quality-adjusted life years. As with the in-work group, those who saw an EA were more likely to say that their ability to work was impaired by their mental illness compared to the matched control group. This is despite reporting better mental health (on some measures) than the matched controls.

People who were unemployed had improved mental health (depression: 93% at IAPT entry, 50% vs 56% at final therapy session; anxiety: 91%, 53% vs 57%) and were more likely to be in work after 12 months (26% vs 21%) if they saw an EA. There was no change in their overall daily functioning but they were less likely to say that they were severely impaired at work (26% vs 32%), and less likely to report being long-term sick or disabled (13% vs 23%), though they were also less likely to report no work impairment (6% vs 9%).

The results from this programme are ambivalent. The work-search group was more likely to be in work after 12 months if they had seen an EA, but the in-work group was less likely to be in work. The work-search and off-sick groups saw a greater improvement in mental health if they saw an EA, but for the off-sick group this was only the true for the ‘reliable change’ measure, and not for ‘caseness’ or ‘IAPT reliable improvement’, ‘IAPT recovery’, or ‘IAPT reliable recovery’. None of the groups had an improved Work and Social Adjustment score, and the off-sick and in-work groups had a worse ‘work impairment’ score. None of the changes, in any direction, were very large: they remained within single digit percentage points.

Interestingly, I think this is the only report I have read which explicitly states that work may not be a positive outcome. The authors say that, for some people, leaving work temporarily or permanently, or moving into long-term sickness/disability, may be a positive outcome.