Treatment research in prison: Problems and solutions in a randomized trial

Abstract

Opioid-dependent individuals are frequently incarcerated and relapse rates following their release are high. Evidence of the effectiveness of prison-based treatments and community aftercare is limited. We report the preliminary findings of a randomized controlled trial that encountered specific challenges inherent in prison research. Naltrexone implants were compared with methadone maintenance treatment among pre-release inmates. Naltrexone implants have not previously been evaluated in prison settings. Approximately 41% of eligible inmates volunteered to participate; 27 of the total 46 participants commenced treatment according to the protocol. Although most inmates intended to remain abstinent from heroin after their release, the relapse rates among individuals who had not commenced treatment were high. Naltrexone implants were regarded as a valuable treatment option. During the study, several problems were encountered regarding enrollment, treatment initiation and aftercare. These findings have important implications for future prison research. Random allocation to the fixed interventions was inefficient in limiting the risk of selection bias. Alternative approaches should be considered, such as sequential multiple assignment randomized trials.