Table 2 Summary of included study characteristics and findings

1

Gordon et al., 2018

RA

N = 199

CJ

Initiating buprenorphine treatment prior to versus after release from prison.

No significant differences.

Treatment condition did not predict likelihood of arrest.

Number of rearrests may have been biased. During the following 12 months after release, many remain detained.

2

Busch et al., 2017

RCT

N = 329

ED

Cost-effectiveness of ED-initiated buprenorphine.

Patient costs significantly lower in ED-initiated treatment group.

ED-initiated buprenorphine treatment is cost-effective.

Comparability of data. Length of follow-up was 30-days post-randomization.

3

Lee et al., 2017

RCT

N = 72

O

Predictors of retention in office-based treatment after hospitalization.

Prior treatment, older age, and non-minority status were associated with more time in office-based opioid treatment.

Linking hospitalized patients to office treatment may improve addiction treatment.

Small sample size; no measures of mental disorders other than PTSD.

4

Gordon et al., 2017

RCT

N = 211

CJ

Initiating buprenorphine treatment prior to versus after release from prison.

In-prison group had higher number of treatment days after release than those who without treatment in prison.

In-prison buprenorphine was correlated with more days of treatment after release.

Fewer women and mostly African American population; results may not be generalizable.

5

Riggins et al., 2017

Cohort

N = 305

CJ

Buprenorphine treatment retention among HIV-positive patients with a history of incarceration.

No significant differences in groups

Recently incarcerated were more likely to be homeless, unemployed, and previously diagnosed with mental illness.

As an observational study, clear causative relationships could not be established.

6

Finlay et al., 2016

RA

N = 48,689

CJ

Likelihood of US Veterans to receive treatment for opioid use disorder at Veteran Health Association hospitals.

Veterans exiting prison receive lowest rates of treatment among all justice-involved US Veterans.

Targeted efforts to reach prison-involved veterans necessary as they have lowest odds of receipt.

Study limited to veterans who received treatment at VHA facilities.

7

Sigmon et al., 2015

Pilot study

N = 10

O

Feasibility of interim buprenorphine treatment to bridge delays during patient navigation.

Opioid abstinence:70% of participants retained through 12-week treatment program.

Interim treatment might reduce illicit drug use and drug-related risk behaviors among waitlisted.

Unrandomized pilot trial with limited sample size.

8

D’Onofrio 2015

RCT

N = 329

ED

Determine success of three intervention options for ED patients with OUD.

After 30 days, group receiving buprenorphine reported greatest reduction of illicit opioid use per week.

ED-initiated buprenorphine vs. brief interventions and referral significantly increased engagement.

Study involved only physicians approved to prescribe buprenorphine,. May not be reflective most ED physicians.

9

Liebschutz et al., 2014

RCT

N = 139

O

Methods of treatment among hospitalized patients post-discharge.

Linkage (intervention) more likely to enter treatment in office setting than those in detox group (72% vs. 11.9%).

Initiation to treatment is effective for hospitalized patients not initially seeking addiction treatment.

Study conducted as single institution with an associated buprenorphine outpatient treatment program.

10

Gordon et al., 2014

RCT

N = 211

CJ

Success of buprenorphine treatment to addicted prison inmates nearing release versus after release

In-prison treatment group more likely to continue treatment post-release; women more likely to complete prison treatment than men (86% vs 53%)

Buprenorphine appears feasible and acceptable to inmates who are NOT opioid-tolerant

Study not generalizable to all geographic locations; 70% of participants were male.

11

Zaller et al., 2013

Pilot study

N = 44

CJ

Initiating treatment prior to release from incarceration and linking participants to community treatment.

Eleven of 32 participants remained in treatment for entire 6 months.

Initiating buprenorphine treatment during incarceration; continuing in community is feasible; may increase retention post-release.

Small sample size; self-report nature of data, particularly drug use and criminal history.

12

Schwarz et al., 2012

RA

N = 209

O

Effect of treatment retention on reducing ED utilization among treatment seeking patients.

Treatment retention was strongly correlated with a decline in ED visits (1 month = 1.6% decline per person).

Buprenorphine maintenance treatment significantly reduces ED utilization.

Lack of randomization does not allow for control of selection.

13

Lee et al., 2012

Cohort

N = 142

CJ

Comparing treatment retention and opioid misuse among those seeking treatment after release from jail.

Treatment retention over time was similar between groups.

Primary care appears to a feasible model of opioid treatment once released from incarceration.

Study participants were largely uninsured but received treatment through the study; whereas uninsured community referrals had no assistance.

14

Cropsey et al., 2011

RCT

N = 36

CJ

Efficacy of buprenorphine for relapse prevention among women in criminal justice system transitioning to community.

Treatment was effective in maintaining abstinence compared to placebo (92% placebo vs 33% buprenorphine were opioid positive per urinalysis).

Initiating buprenorphine in prison prior to release appears to reduce opioid use when participants reenter community.

Small sample size; limited generality as participants were women with criminal justice involvement.

15

Wang et al., 2010

RA

N = 166

CJ

Determine whether history of incarceration affects response to primary care office-based treatment.

Participants with history of incarceration have similar treatment outcomes with primary care office-based treatment than those w/o history of incarceration

Formerly incarcerated patients ar emore likely to have been treated with methadone, but do not have substantially different outcomes than those without prior incarceration.

Measurement of incarceration was self-reported and time incarcerated was grouped (patients with one month and multi-years were in same group).

16

Marzo et al., 2009

Cohort

N = 507

CJ

Describe the profile of imprisoned French opioid-dependent patients

77% of pts. received MAT at imprisonment, these patients were in poorer health & were more isolated than other population; 238/478 pts. were re-incarcerated within 3 years

MAT has increased in the criminal justice system in France, but maintenance therapy not associated with lower rate of reincarceration.

Conclusions on mortality are not well-supported as study was not designed for mortality analysis; pt. selection not random

17

Magura et al., 2009

RCT

N = 116

CJ

Test the efficacy of buprenorohine versus methadone while incarcerated and follow-up.

Patients in buprenorphone group reported to treatment significantly more than patients taking methadone.

There were no significant differences between groups for re-incarceration, relapse, re-arrests.

Findings may not be generalizable in other nations where methadone distribution protocols vary.

18

D’Onofrio et al., 2017

RA

N = 290

ED

Outcomes assessment of previous RCTs to determine long-term outcomes.

Patiengts in the buprenorphine group showed greater engagement in treatement at 2 months which was statistically significant.

Gains did not persist after 2 months when measure at the 6 and 12 month time points.

Buprenorphine treatment initiatied in the ED was associated with increased engagement during 2 month interval when treatment was continued at PCP.

19

Vocci et al., 2015

RA

N = 104

CJ

Assessed prior RCT to examine if induction into buprenorphine during incareceration was associated with seeking treatment post-release.

Participants were rapidly inducted onto buprenorphine with no serious side effects whle incarecerated.

Buprenorphine administered to non-opioid tolerant adults may be used to reduce rates of withdrawal and re-use post-incarceration.

None noted.

20

Cushman et al., 2016

RA

N = 113

O

To assess whether inpatient initiation to buprenorphine and linkage to counselling reduces illicit opioid use.

Patients who were linked to outpatients ervices versus patients in detox (inpatient) were more successful in the short term.

Differences did not persist between groups (linking versus detox) as far as injection opiate use at 1, 3, or 6 month timepoints.

May not be generalizable with a small population.