Statistics on Treating Opioid-Dependent Prisoners Offer Limited Guidance for Policymakers

Data from a study published in the Journal of Offender Rehabilitation (2023 May 24) establishes a history of injection drug use (IDU) as a factor predicting uptake of medications for opioid use disorder (MOUD) among former prisoners. The researchers’ analysis indicates potential populations where interventions could improve treatment and outcomes for prisoners with a history of opioid use and mental health issues. Yet, a lack of follow-up data limits the study’s implications.

“Our findings indicate that post-release MOUD treatment uptake has a strong association with also receiving other behavioral health pharmacology, specifically, being prescribed psychiatric medication,” said study author Grant Victor, PhD, MSW, an assistant professor at the Rutgers School of Social Work, in New Brunswick, N.J.

After recruitment among incarcerated adults (age, 18 years and older) with a history of opioid use and presenting mental health symptomatology, participants who consented to MOUD treatment were offered extended-release naltrexone (Vivitrol, Alkermes) while incarcerated and referred to one of three post-incarceration treatment programs in Michigan that offered MOUD. Among the 160 people with co-occurring OUD and a mental health disorder entering post-incarceration treatment between May 1, 2017, and April 30, 2020, 46.25% reported IDU indicating a higher risk for overdose and other health concerns (e.g., infectious disease and serious bacterial infection).

The most significant correlate of IDU (12 months before incarceration) was whether a participant had contracted hepatitis C virus (HCV). Participants identifying as white and having a history of nonfatal overdose were also more likely to report IDU. The subgroup of patients who were HCV-positive and had a history of overdose had a 92.10% likelihood of reporting IDU, compared with the least likely subgroup to report IDU (26.51%): those who were HCV-negative, with no history of overdose and identified as a person of color. “Our findings suggest that there’s a high overlap between injection drug use and HCV and prior nonfatal overdose. It is important that carceral settings screen and triage medical services for those who meet this criterion to mitigate the spread of infectious disease in their facilities,” Victor said.

In the study’s second model, researchers used IDU as a predictive factor to assess whether users were linked to MOUD services, which have shown to be associated with injection cessation and a lowered risk for overdose. Of the 45 study participants who responded to a one-month post-incarceration follow-up survey, 88.89% (n=40) were taking MOUD. As the researchers expected, those taking MOUD were more likely to have a history of IDU, but it was the second most significant predictor of MOUD engagement (ME) among the study population. The receipt of psychiatric medication was the most significant predictor of ME. The third most significant factor predicting ME was participation in substance use treatment.

“We interpret this finding as that the re-entry program in which our sample was enrolled likely improved these treatment linkages post-release and this is a model that should be expanded via partnerships between state prisons and local behavioral health providers in lieu of decriminalizing drug use and drug paraphernalia,” the authors wrote.

The research was limited by the fact that only 45 of 160 individuals (28.13%) responded to the one-month post-incarceration survey. The researchers believed their use of chi-square automatic interaction detection models was well equipped to deal with missing data and largely unaffected by reductions in sample size. It was found that those who did not respond to the second survey were not random and shared several characteristics:

• an increased number of peer recovery support services post-release;
• an earlier starting age for opioid use;
• greater use of opioids in the prior year; and
• more likely to be non-white.

Despite these findings, Victor noted, “we simply don’t have data on those who didn’t respond [but] it stands to reason that those who haven’t overdosed don’t have HCV; don’t inject drugs; have a lower ‘risk profile’ relative to others in this sample; and they were simply more interested in attending to other needs after their release.” Yet, he cautioned that, “I don’t think it’s fair to assume that those who didn’t respond gave up or returned to use, only because we don’t have any data to say one way or another.”

Victor further noted that because the researchers were unable to control the type of MOUD dispensed at reentry, the study’s findings should not be generalized to specific antagonists or agonist pharmacotherapies. Furthermore, “it could not account for causality or the temporal relationship between health services in the post-release period.”

Lynn Webster, MD, a senior fellow at the Center for U.S. Policy, in Washington, D.C., who was not associated with the research, expected the reported relationship between IDU and HCV, but was surprised that patients with a history of IDU were more likely to receive MOUD. “I would have thought they would be less likely to be interested in MOUD.” Webster, a member of the Pain Medicine News editorial advisory board, offered a possible explanation of the results” “Two-thirds of the [IDU] subjects were in the missing data group. The missing data IV drug use group had a much higher overdose history, which may be why the correlation of those with non-missing data looked better.”

Since the original study took place, some carceral facilities in Michigan have expanded their MOUD offering to include buprenorphine and methadone as well as extended-release naltrexone. There is a plan to collect data on the outcomes of each enrollee based on the type of MOUD they are administered as well as the feasibility of implementing harm reduction services within carceral systems.

—Myles Starr

Victor and Webster reported no relevant financial disclosures.