Implementing community-based diversion programs (DPs) as an alternative to incarcerating people who use drugs has yielded potentially beneficial, yet conflicting evidence regarding cost-effectiveness. This study systematically reviewed the economic evaluations (EEs) of DPs by synthesizing findings on cost-effectiveness and recidivism, while strengthening the appraisal of causal validity, risk of bias, and reporting quality to inform future research and policy development. A systematic literature review of EEs in the DPs was conducted using seven databases. We extracted characteristics including country, EE type, perspective, cost-effectiveness, recidivism, and limitations. We assessed causal validity and evaluated risk of bias using ROBINS-I, reporting quality using the Quality of Health Economic Studies (QHES) and Consolidated Health Economic Evaluation Reporting Standards (CHEERS), and strength of evidence on recidivism using the Maryland Scientific Methods Scale (SMS). Of the 2,246 records, 16 studies covering 12 DPs were included. Most originated in the US (n = 9), with five from Australia and two from the UK. Over half adopted a societal perspective, followed by a taxpayer perspective, and most were funded by nonprofit organizations. The study types included cost-utility (n = 3), cost-effectiveness (n = 4), cost-minimization (n = 4), and cost-benefit analyses (n = 5), predominantly incorporating criminal justice and healthcare costs. ROBINS-I indicated a moderate to serious risk of bias in most studies, driven primarily by confounding and selection issues, and three-quarters cited data-related limitations. The QHES and CHEERS scores showed a wide variability in reporting quality. Evidence from the three SMS-assessed studies among prison releasers suggested that DPs may lower recidivism and be cost-effective; however, confidence in these findings was limited by a moderate-to-serious risk of bias, mainly due to confounding, selection, and author-reported data limitations (75%), including heterogeneity and inconsistency in cost components and measurement. DPs demonstrate heterogeneous cost-effectiveness due to inconsistent analytical approaches, high methodological bias, and pervasive author-reported data limitations. These issues, particularly confounding and selection bias, undermine confidence in causal and economic conclusions. Standardized, bias-minimized cost-utility analyses with rigorous designs are urgently required to inform policy decisions. PROSPERO number: CRD420251186011
Nam et al. (Fri,) studied this question.