Patients in Scotland who received employment-related support as part of their addiction treatment package were three times more likely later to find work. The findings suggest that such support does improve the employment prospects of at least a minority of patients, lending weight to the current UK policy emphasis on the provision of these services.
FINDINGS The figures derived from the Drug Outcome Research in Scotland study (DORIS). In 2001 this sampled 1033 patients starting treatment in different modalities and observed what happened as they went through the normal treatment process. Though using many other drugs, most saw their main problem as heroin.
33 months later the 695 who could be reinterviewed were asked about their legal paid employment; casual and cash-in-hand work was disregarded.1 Just 1 in 10 were working but a fifth had worked between their previous interview 17 months earlier and the 33-month interview (the employment assessment window). An omnibus analysis of 25 factors which might have influenced this outcome found that having been helped by the initial treatment agency to find work or gain employment-related skills or education was the one most closely related to employment. After taking in to account the other factors, patients who recalled They had been asked about this at an earlier interview eight months after treatment had started, establishing that the help had preceded (so could have contributed to) employment. such help2 were over three times more likely later to have worked than those who did not.
The only other variables significantly related to employment were age, severity of dependence in the three months preceding the employment assessment window (highly dependent patients half as likely to have worked), and whether patients had committed crimes in the three months before the final interview (those who had were half as likely to have worked).
Equally important was what was not related to employment. These factors included employment-related help from agencies other then the initial treatment service, the treatment modality (prison-based, residential rehabilitation, or methadone prescribing), and whether the patient had altogether avoided heroin in the three months leading up to the employment assessment window.
Faced with these surprising negative results, the researchers tested whether these influences might have been obscured by the other factors simultaneously taken in to account. An analysis which did not compensate for other factors found that preceding In the three months preceding the employment assessment window. abstinence3 from illegal drugs other than cannabis was significantly related to employment: 27% who had been abstinent later worked, 18% who had not. Further such analyses established that patients who had started the study in residential rehabilitation were over twice as likely to have received employment-related help, yet were not significantly more likely to have found work – 29% had done so, but so had 20% in methadone services or other non-residential treatments. Outside prison, the biggest gap in receipt of help was between residential rehabilitation (38%) and methadone services (13%), but this 25% gap converted to just a 9% gap in the attainment of employment.
IN CONTEXT These findings suggest that receipt of employment help is an important influence on later employment, that the treatment modality is less influential, and that whether treatment eliminates heroin use is less important than whether it reduces dependence and the crime that often comes with it. The fact that similar help from outside agencies did not enter the frame possibly indicates that on-site help from familiar and trusted faces is most likely to be acted on, or that external help was sought only after prompting from the treatment service. The link between employment and abstinence from illegal drugs other than cannabis makes sense, but was not tested in an analysis which took other influences in to account. As a result, it remains unclear whether abstinence was in itself influential. When abstinence from the sample's main problem drug (heroin) was tested in this way, it fell out of the frame.
Several features of the study hinder interpretation of its findings. The most important is that it observed normal treatment processes rather than deliberately allocating patients to receive or not receive employment-related help. This makes it impossible to be sure that the help actually caused the elevated employment rates it was associated with. Patients' pre-treatment employment assets and their desire and belief in their ability to work were not included in the analyses. As in other studies,4 5 6 these might have been the decisive influences over whether they found work and perhaps too over whether they sought help, creating a spurious relationship However, this seems unlikely to have been the full story. On starting treatment 4% of the non-prison sample were currently employed; 33 months later, about 10% who could be reinterviewed were. between the two.7 Significantly more residential rehabilitation patients received employment help yet this did not result in a significant advantage in employment, suggesting that employment prospects are heavily influenced by other factors.
A high hurdle was set: paid, 'on the books' employment. In one US study an intensive employment intervention for methadone patients significantly increased access to paid employment overall, but formal employment remained rare.5 Finally, an analysis based on whether patients had been offered help – rather than whether they recalled receiving it – might have been more relevant to service provision and produced a different impression of how available help was and how effective it had been.6
PRACTICE IMPLICATIONS Despite the doubts, the findings are compatible with the proposition that providing employment-related help during treatment means more patients later solidify their recovery through work, strengthening the case for services to take the initiative in helping patients on this journey. In this light it is worrying that just 17% of patients said they had received this support (more may have been offered it6), raising the issue of how many more might have found work with appropriate help. However, converting this help in to success in a tough job market would not have been easy. Even those who overcame their dependence would often have been held back by the stigma of drug use and criminal histories, lack of qualifications, poor health, underdeveloped work discipline, lack of confidence, and a benefit system which makes entering the job market financially risky. Attaining competitive employment may require a long-term, supportive and incremental introduction to work, and employers willing to take what to many will seem an unnecessary risk.8 Barriers like these probably account for the general failure of treatment itself, or vocational interventions during treatment, to increase entry to competitive employment.4 5 6 But this general failure masks positive studies, some showing that people previously considered unemployable can be helped to find work.9 Innovative schemes which actually provide work during treatment rather than just helping patients find it in the job market have also proved feasible.
Thanks for their comments on this entry in draft to Mick Bloor of the Centre for Drug Misuse Research at the University of Glasgow, coordinator of the DORIS project. Commentators bear no responsibility for the text including the interpretations and any remaining errors.
1 FEATURED STUDY McIntosh J. et al. Drug treatment and the achievement of paid employment. Addiction Research & Theory: 2008, 16(1), p. 37–45.
2 They had been asked about this at an earlier interview eight months after treatment had started, establishing that the help had preceded (so could have contributed to) employment.
3 In the three months preceding the employment assessment window.
4 Magura S. et al. The effectiveness of vocational services for substance users in treatment. Substance Use & Misuse: 2004, 39(13–14), p. 2165–2213.
5 Magura S. et al. An innovative job placement model for unemployed methadone patients: a randomized clinical trial. Substance Use & Misuse: 2007, 42, p. 811–828.
6 Dunlap L.J. et al. Do treatment services for drug users in outpatient drug-free treatment programs affect employment and crime? Substance Use & Misuse: 2007, 42(7), p. 1161–1185.
Last revised 09 April 2008
STUDY 2002 The grand design: lessons from DATOS