A Systematic Review of the Efficacy of Alcohol Interventions for Incarcerated People

Dorothy Newbury-Birch; Jennifer Ferguson; Sarah Landale; Emma L. Giles; Grant J. McGeechan; Charlotte Gill; Kelly J. Stockdale; Aisha Holloway


Alcohol Alcohol. 2018;53(4):412-425. 

In This Article


Search Strategy and Selection Criteria

We carried out a systematic review of the international literature, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines on reporting of systematic reviews (Moher et al., 2009). The systematic review protocol was registered on the PROSPERO Register at the University of York (CRD42016039895). We included Randomised Controlled Trials (RCTs) or matched group trials. As well as examining the efficacy/effectiveness of alcohol interventions, the current review adds to other reviews in the criminal justice setting (Graham et al., 2012; Newbury-Birch et al., 2016b) by including evidence around the timing of screening and interventions within the offender journey and information about the type and nature of the interventions themselves.

We included studies with control groups comprising: treatment as usual; information-only; assessment only; no assessment or another intervention. Studies eligible for this review were peer-reviewed trials of any alcohol interventions carried out in the jail/prison setting (including remand). We included interventions that were categorised as brief interventions as well as extended longer alcohol psychosocial interventions (extended brief interventions). We included individuals aged 16 years or over and any outcome measure. We excluded studies that included a drug and alcohol intervention where alcohol information could not be isolated.

We searched: MEDLINE; PsychINFO; Web of Science; Cochrane Library; EBSCO; CINAHL and the Campbell Collaboration Library. We included all dates in the search. The search was conducted in August 2017. Citations were scanned and we contacted experts in the field to minimise selection bias. The search terms used were: 'alcohol OR alcoholism OR alcohol abuse OR alcohol misuse OR binge drinking' AND 'crim* OR prison* OR offend* OR correctional OR penitentiary OR incarc* OR remand' AND 'randomised controlled trials OR randomise OR control OR trial OR random* OR quazi* OR quasi* OR matched' with Boolean/phrase searches. Searches were tailored to the search functionality of each database. We also searched grey literature including google scholar and did a 360°-citation check of included papers.

All authors were involved in the data sifting. Two reviewers on the team conducted eligibility assessments of titles and abstracts independently. Disagreements between reviewers were resolved by consensus or by discussing with a third person. Data were extracted into a Microsoft Excel spreadsheet independently by J.F., E.L.G., G.M., S.L. and A.H. These researchers jointly reviewed the extracted data and all studies were double extracted by D.N.-B. Data were extracted on characteristics of trial participants, type and nature of interventions (including content, duration, frequency, provider, setting), type of outcome measure as well as information relating to the interventions itself. We used the Template for Intervention Description and Replication (TIDieR) checklist to ascertain how interventions are reported in the included studies (Hoffman et al., 2014). One researcher completed the checklist (D.N.-B.) and it was checked by another (J.F.).

Quality Assessment

The relevant screening tools from Critical Appraisal Skills Programme were used by one researcher (C.G.) and checked by another (D.N.-B.) (MKPC Trust, 2002). High risk of bias was recorded if 'no' or 'unsure' was recorded for 6 or more of the 11 questions on the tool. Medium risk of bias was assigned if 'no' or 'unsure' was recorded for 4–5 questions and low risk for 1–3 questions (Table 3).

Data Synthesis

We grouped intervention content into two categories. The first was short interventions that were categorised as brief interventions and included up to three sessions. The second was extended brief interventions delivered over more than three sessions (Table 1).