BackgroundThe societal costs of problematic class A drug use in England and Wales exceed £15B; drug-related crime accounts for almost 90% of costs. Diversion plus treatment and/or aftercare programmes may reduce drug-related crime and costs.ObjectivesTo assess the effectiveness and cost-effectiveness of diversion and aftercare for class A drug-using offenders, compared with no diversion.PopulationAdult class A drug-using offenders diverted to treatment or an aftercare programme for their drug use.InterventionsProgrammes to identify and divert problematic drug users to treatment (voluntary, court mandated or monitored services) at any point within the criminal justice system (CJS). Aftercare follows diversion and treatment, excluding care following prison or non-diversionary drug treatment.Data sourcesThirty-three electronic databases and government online resources were searched for studies published between January 1985 and January 2012, including MEDLINE, PsycINFO and ISI Web of Science. Bibliographies of identified studies were screened. The UK Drug Data Warehouse, the UK Drug Treatment Outcomes Research Study and published statistics and reports provided data for the economic evaluation.MethodsIncluded studies evaluated diversion in adult class A drug-using offenders, in contact with the CJS. The main outcomes were drug use and offending behaviour, and these were pooled using meta-analysis. The economic review included full economic evaluations for adult opiate and/or crack, or powder, cocaine users. An economic decision analytic model, estimated incremental costs per unit of outcome gained by diversion and aftercare, over a 12-month time horizon. The perspectives included the CJS, NHS, social care providers and offenders. Probabilistic sensitivity analysis and one-way sensitivity analysis explored variance in parameter estimates, longer time horizons and structural uncertainty.ResultsSixteen studies met the effectiveness review inclusion criteria, characterised by poor methodological quality, with modest sample sizes, high attrition rates, retrospective data collection, limited follow-up, no random allocation and publication bias. Most study samples comprised US methamphetamine users. Limited meta-analysis was possible, indicating a potential small impact of diversion interventions on reducing drug use [odds ratio (OR) 1.68, 95% confidence interval (CI) 1.12 to 2.53 for reduced primary drug use, and OR 2.60, 95% CI 1.70 to 3.98 for reduced use of other drugs]. The cost-effectiveness review did not identify any relevant studies. The economic evaluation indicated high uncertainty because of variance in data estimates and limitations in the model design. The primary analysis was unclear whether or not diversion was cost-effective. The sensitivity analyses indicated some scenarios where diversion may be cost-effective.LimitationsNearly all participants (99.6%) in the effectiveness review were American (Californian) methamphetamine users, limiting transfer of conclusions to the UK. Data and methodological limitations mean it is unclear whether or not diversion is effective or cost-effective.ConclusionsHigh-quality evidence for the effectiveness and cost-effectiveness of diversion schemes is sparse and does not relate to the UK. Importantly this research identified a range of methodological limitations in existing evidence. These highlight the need for research to conceptualise, define and develop models of diversion programmes and identify a core outcome set. A programme of feasibility, pilot and definitive trials, combined with process evaluation and qualitative research is recommended to assess the effectiveness and cost-effectiveness of diversionary interventions in class A drug-using offenders.Funding detailsThe National Institute for Health Research Health Technology Assessment programme.