Skip to content

Website cookies

This website uses cookies to help us understand the way visitors use our website. We can't identify you with them and we don't share the data with anyone else. If you click Reject we will set a single cookie to remember your preference. Find out more in our privacy policy.

News 19 July 2019

Forward’s alcohol treatment programme evaluated by the Justice Data Lab

1-year re-offending rate of 37%, significantly lower than predicted reoffending rates for alcohol dependent offenders who do not get access to treatment

The Forward Trust is committed to evidencing the effectiveness of our services through research and evaluation and works closely with the Justice Data Lab (JDL), a unique service from the Ministry of Justice that helps organisations to assess the impact of their work on reducing reoffending. To date, we are one of only four providers from the substance misuse sector to engage with the JDL and to be open to public scrutiny on the results they produce.

Our first submission related to our work with male offenders with a history of alcohol dependence. It was found that participants of our Alcohol Dependence Treatment Programme (ADTP) – who engaged at nine different prisons between 2007 and 2015 – reported a 1-year reoffending rate of 37%. Whilst the results from this evaluation were statistically inconclusive compared to the comparison group used by the JDL, we are nevertheless encouraged that the reoffending rate associated with this programme was relatively low.

Read the full JDL study.

The significance of dependence

When interpreting the findings from this evaluation, it’s important to note that the JDL is currently limited in its ability to match like-with-like on the important factor of alcohol dependence (a limitation that we have discussed with them and that they acknowledge). The data at the JDL’s disposal stems from the standard risk assessment carried out for anyone sentenced to 12 months or more in prison, while Forward uses a validated screening tool to identify dependence as a core criterion for admission to the ADTP.

As such, the JDL comparison group generated for this study was matched with our programme participants on variables such as self-reported ‘problems with binge drinking’, ‘problems with past misuse of alcohol’ and, ‘problems with current use of alcohol’ – not on level of dependence. (The World Health Organisation (WHO) and Public Health England (PHE) distinguish dependence from hazardous or harmful drinking, defining it as: “a cluster of behavioural, cognitive and physiological phenomena that may develop after repeated use of alcohol use”.)

Informed by our experience of supporting thousands of people in prison over the last three decades, it is our working hypothesis that those suffering with alcohol dependence are at greater risk of relapse and recidivism than those who report problematic use or problems with binge drinking. If this hypothesis is true, the predicted reoffending rate of an alcohol dependent comparison group is likely to be significantly higher than the 36% reported by the JDL.

What is known about alcohol dependence and offending behaviour?

There is currently an absence of clear, comparable evidence specifically on the re-offending rates of people released from prison who have been identified as alcohol dependent. Research from the MoJ, however, points to re-offending rates of between 58% (for participants of accredited drug/alcohol programmes in prison) and 76% (for ex-prisoners who reported using Class A drugs post-release) for potentially substance dependent offenders, though these studies have other limitations in terms of comparability with the ADTP study (e.g. sentence length). We also know from a recent review of the literature that those who are alcohol dependent are more likely to commit violent crime with significantly higher rates of offending associated with increased symptom severity.

In the absence of definitive evidence, The Forward Trust calls for more research, debate and discussion on the extent to which dependence is a determinant of reoffending outcomes, relative to other criminogenic factors, and measures of drug or alcohol use. In the meantime – and in the interests of contributing to the wider evidence base around substance misuse treatment and reoffending – we will continue to work with the JDL and to share its findings.

The full Justice Data Lab report can be found here

Forward’s response to criticism on the ‘Drug and Alcohol Findings’ website

In a ‘cell’  published on the Findings website in March 2021, several statements were made by the anonymous author that criticised Forward for its presentation of research findings from the Ministry of Justice Data Lab (JDL), into the reducing re-offending impact of our Alcohol Dependence Treatment Programme (ADTP) in English prisons.

The overall theme of the cell is the paucity of global research evidence demonstrating positive impacts of alcohol interventions with offenders. This overall analysis is one we agree with – over 50 years, there has been surprisingly little high quality research in this area, and none of it can clearly describe a causal link between interventions and outcomes.

The Findings author rightly drew our attention to a misleading headline on one of our website research pages, which claimed that the research in question had ‘proved’ the effectiveness of our programme. This was a case of overenthusiastic drafting, and we agreed to change it as soon as we were alerted to it.

But the subsequent text on our website, which we’ve further clarified, was a genuine attempt to discuss and understand the issues raised by the JDL research. Specifically, the JDL reported a 37% re-offending rate amongst programme participants, compared with 36% amongst a matched comparison group. Their conclusion was, therefore, that the study did not show a statistically significant impact on post-release re-offending.

We made the point at the time that the comparison group was not adequately matched in one important respect – it was assembled on the basis of questions on frequency of drinking and self-reported effects, rather than on internationally recognised measures of alcohol dependence (e.g. DSM-IV criteria and AUDIT scores). We argued that, as all of our programme participants were by definition alcohol dependent, their predicted re-offending rate should be higher than the 36% reported in the JDL comparison group.

We have been in correspondence with JDL officials on this point for years, and it is not easy to resolve – there are no clear research findings on the likely re-offending rate of alcohol dependent prisoners who do not receive treatment while in prison, so a direct comparison is difficult. But our position on this question is transparent and genuine, and has been openly discussed in writing and in public discussions.

In a market where the majority of commissioned interventions have never been subjected to any form of outcome evaluation, it is ironic that Forward Trust’s genuine attempt to evaluate its programme outcomes, and subject the results to open discussion, are characterised in such negative terms by the Findings author. The text in the Findings ‘cell’ gives the reader the impression that our website text gave a misleading account of these methodological debates, that was only corrected when we were challenged by the Findings team. While this was true regarding the website page headline, it is certainly not true of the discussion that followed.