The paper by Huang and colleagues  adds to the already accumulating literature that provides strong evidence that newly released prisoners are at disproportionately high risk of death in the first week after release, particularly from drug overdose . It also provides further evidence, building on the work of Dolan et al. , that methadone maintenance treatment (MMT) provided to former prison inmates can reduce mortality significantly following release to the community. Moreover, in addition to providing support for already well-documented findings, it is especially important to ask what questions this paper raises as well as what directions for future research it provides.
Another important observation, and seemingly unique to this study, is that the harm reduction program available for former prisoners in Taiwan includes not only methadone maintenance, but needle exchange, free human immunodeficiency virus (HIV) testing and treatment for HIV-positive individuals, counseling, and education about the health risks of opiate injection. Because evaluations of the effectiveness of drug treatment programs seldom report on the effectiveness of programs' various components [4,5], an important analysis to be conducted is particularly which components of this program, in addition to methadone maintenance, were related to reduced mortality from what specific causes, and in what combinations of services? In addressing this question, in addition to the types of services provided, what were their frequency, intensity and duration in relation to reduced mortality?
It was particularly helpful that the authors investigated how soon after release from prison the participants entered methadone treatment. It was particularly disturbing that most of the participants did not enter MMT until 30–150 days post-release; and that only 6% attended in the first 30 days after release, which is the period in which newly released prisoners with histories of opioid dependence are disproportionately susceptible to overdose death [2,6]. Such individuals rarely receive treatment for their addictions during incarceration and upon release to the community [4,7] and typically relapse to opioid addiction within the first month of release from incarceration [8–10]. Therefore, in view of these problems, and the observation that both jail- and prison-initiated MMT facilitates community-based treatment entry shortly after release and retention in that treatment as well as reduced heroin use [3,11–15] and reduced mortality from overdose death , it would seem beneficial to examine the effectiveness of the current harm reduction program if it were initiated shortly before release from prison.
The authors reported that they ‘have done serial interviews of many of the subjects in the cohort since their amnesty’. They report that there were no differences in the levels of drug use at baseline among those who continued MMT and dropouts, but that those who continued had more social support. These interview data need to be described in much more detail—what was the sample, how many participants were followed-up, at which time(s) after amnesty were they interviewed, how many assessments per participant were completed and what was the response rate, how drug use and social support were measured, and what additional baseline and outcome measures were examined? Because opiate-dependent individuals vary considerably on a number of important dimensions, such as the age of onset, type, frequency and severity of criminal activity; education, employment, psychological functioning and motivation for treatment, in addition to drug use and social support, these results, and additional long-term follow-up studies of this nature, may be able to help address an important but infrequently studied question: what types of treatment work best for what types of clients [4,8,9,16]?