A new systematic review of psychological interventions in prisons published in the Lancet this week finds that there is very little evidence to support their effectiveness in reducing reoffending. The research, Effectiveness of psychological interventions in prison to reduce recidivism: a systematic review and meta-analysis of randomised controlled trials was undertaken by Gabrielle Beaudry, Rongqin Yu, Amanda E Perry, and Seena Fazel.

The authors searched for randomised controlled trials (RCTs) that evaluated the effect of psychological interventions, delivered to adolescents and adults during incarceration, on recidivism outcomes after release. They excluded studies of solely pharmacological interventions and of participants in secure psychiatric hospitals or special residential units, or attending therapies mainly delivered outside of the prison setting.

The studies

As is usual in systematic reviews, the authors identified a large number of studies but were left with a relatively small number which were conducted with sufficient rigour to stand the tests of a meta-analysis. In this case, 6345 articles were retrieved but just 29 randomised controlled trials met the key inclusion criteria for the study. Crucially, the authors then excluded 15 of these studies (many of which suggested positive outcomes) for the (simple and entirely appropriate) reason that they were very small scale – each had fewer than 50 people in each “experimental arm” – i.e. the treatment group and the control group. This left 14 studies which followed up a total of 6446 participants in different studies around the world.

Findings

The researchers’ primary conclusion was that the impact on reoffending rates was “at most, modest”.

The authors undertook additional analysis of cognitive behaviour therapy based prison programmes and found no strong evidence of reduced reoffending. They compare these findings with a 2007 systematic review of both prison-based and community-based interventions that reported reduced reoffending rates of between 20-30%. The authors suggest two reasons for this discrepancy.

One potential explanation for no clear effectiveness of such CBT interventions found in the current systematic review is
that these interventions are not linked with psychosocial support upon release.

The other proposed explanation is that the psychological therapies under review, which were developed for mental health problems, do not address the accommodation, employment, and financial difficulties after release that
contribute to the risk of further offending.

More positive news is report on a small subgroup analysis of two rigorous RCTs conducted on therapeutic communities in prisons. Both TCs  linked people released from prison to voluntary post-prison services. In one study links to community services were associated with a lower return to custody rate than for participants without such links.

The authors note that their findings align with those of another systematic review of psychoeducational programmes for reducing prison violence which suggests that therapeutic communities, as programmes tailored to specific needs (eg, drug use) were associated with reduced institutional violence. These findings were also supported by another systematic review of any people who offended and had co-occurring drug and mental health problems, in this review three out of four therapeutic communities were associated with reductions in recidivism.

Conclusions

In addition to identifying an overall lack of robust evidence in this area, the researchers highlight two key issues:

1. The lack of reliable aftercare and continuity of healthcare on release may be undermining psychological interventions in prison.
2. Interventions may be based on general community interventions and are not tailored to address risk factors specific to offending.

We have known about the importance of aftercare for many years and it is worrying that one of the possible adverse side effects of the reunification of the probation service is the probable loss of the enhanced through the gate service.

Thanks to Andy Aitchison for kind permission to use the header image in this post. You can see Andy’s work here.