Models for Medication-Assisted Treatment for Opioid Use Disorder, Retention, and Continuity of Care. APPENDIX 1: Literature Review

07/09/2020

APPENDIX 1. LITERATURE REVIEW

IBM Watson Health

July 2020

DISCLAIMER: The opinions and views expressed in this report are those of the authors. They do not reflect the views of the Department of Health and Human Services, the contractor or any other funding organization. This report was completed and submitted on January 11, 2019.

 

TABLE OF CONTENTS

ACRONYMS

EXECUTIVE SUMMARY

INTRODUCTION

OBJECTIVE

METHODS

FINDINGS

  • Treatment Settings and Models of OUD MAT
  • Defining Retention and Continuity of Care
  • Patient Characteristics Affecting Retention in SUD Treatment
  • Evidence-Based Methods to Address Treatment Retention in SUD and OUD Treatment
  • Promising Models of Psychosocial Support for Retention in OUD MAT
  • Reimbursement Policy, Effects on Services and Retention in SUD Treatment
  • Effects of Treatment Retention on Outcomes

GAPS IN THE LITERATURE

REFERENCES

APPENDIX A: Retention Rates from MAT-Focused Studies in this Review

TABLE A1-1: Preliminary Search of Peer-Reviewed Literature

 

ACRONYMS

The following acronyms are mentioned in this appendix.

ADHD Attention Deficit Hyperactivity Disorder
AHRQ HHS Agency for Healthcare Research and Quality
ASAM American Society of Addiction Medicine
AUD Alcohol Use Disorder
 
BHIVES Buprenorphine-HIV Evaluation and Support study
 
CBT Cognitive Behavioral Therapy
CEO Chief Executive Officer
CJS Criminal Justice System
CM Contingency Management
CMS HHS Centers for Medicare & Medicaid Services
COR-12 Comprehensive Opioid Response with the Twelve Steps
CPT Cognitive Processing Therapy
CRT Cognitive Rehabilitation Treatment
CSI Contracting with Staff Incentives
 
ED Emergency Department
 
FDA HHS Food and Drug Administration
FQHC Federally Qualified Health Center
 
HHS U.S. Department of Health and Human Services
HIV Human Immunodeficiency Virus
 
ICER Institute for Clinical and Economic Review
IOP Intensive Outpatient
 
MAT Medication-Assisted Treatment
 
NIATx Network for the Improvement of Addiction Treatment
NQF National Quality Forum
 
OBOT Office-Based Opioid Treatment
OBOT-B Office-Based Opioid Treatment with Buprenorphine
OUD Opioid Use Disorder
 
PCP Primary Care Provider
POATS Prescription Opioid Addiction Treatment Study
PTSD Post-Traumatic Stress Disorder
 
QI Quality Improvement
 
RCT Randomized Controlled Trial
 
SAMHSA HHS Substance Abuse and Mental Health Services Administration
SUD Substance Use Disorder
 
UDS Urine Drug Screen
 
VA U.S. Department of Veterans Affairs
 
XR-NTX Extended-Release Naltrexone

 

EXECUTIVE SUMMARY

Objective

This review summarizes the recent literature on retention in or continuity of treatment for opioid use disorder (OUD) that involves medication-assisted treatment (MAT) and, to a lesser extent, treatment for substance use disorders (SUDs) more generally. It establishes context for five case studies of programs, sites, or treatment approaches to OUD treatment that show promise in retaining individuals with OUD in treatment.

Methods

We performed a literature review that addressed retention in SUD treatment with a primary focus on OUD treatment. The review included peer-reviewed and grey literature. The peer-reviewed literature included English-language publications from 2014-2018, supplemented, where appropriate, with seminal literature prior to 2014. We culled the literature to determine components of SUD treatment that support retention and recovery, including psychosocial supports, reimbursement structures or payment models, and other factors. We also reviewed the literature to lay the groundwork for defining treatment retention and continuity. The grey literature review included searches of websites of government agencies, health systems, non-profit stakeholders, and research organizations. We searched for information on ongoing treatment models, programs, program evaluations, and reimbursement initiatives with an eye toward their influence on retention in treatment.

Findings

Our focus on retention or continuity of MAT treatment requires that those terms be defined. Retention involves remaining in treatment for some period of time. The concept of continuity of care may equate to retention, most frequently continuous possession of treatment medication, or may be used to identify successful transitions of care from intensive treatment, such as detoxification or residential, to less intensive outpatient (IOP) treatment.

Retention has been operationalized in many ways, with time frames ranging between 3 months and 7 years. Studies and measures may or may not allow for gaps in treatment when measuring retention. They rarely allow retention to include treatment re-entry after disengagement or relapse and most do not address the nature of treatment participation. Previous studies, using differing periods for measurement, reveal widely disparate retention rates.

Variables that affect retention in SUD treatment include patient characteristics, treatment-related variables, and payer policies. Patient characteristics such as age, sex, race/ethnicity, education, housing status, previous or current substance use, co-occurring psychiatric conditions such as PTSD, pregnancy status, geographic location, insurance, and health system use characteristics are identified as influential in recent literature.

Practice or system-focused interventions also may influence retention. A few examples addressed in this review include in-hospital induction to buprenorphine with facilitated entry into outpatient treatment, home-based delivery of extended-release naltrexone (XR-NTX), use of an onsite pharmacy, physician-pharmacist collaboration, low-threshold buprenorphine treatment, the nurse care manager model of buprenorphine treatment, integration of buprenorphine treatment into HIV clinics, rapid treatment with MAT in young adults newly diagnosed with OUD, and use of telehealth in the prescribing of buprenorphine. Evidence also shows that the dose of medication treatment influences retention. Recent studies also identify psychosocial supports that have been shown to positively influence retention in MAT for OUD, including contingency management (CM), cognitive behavioral therapy (CBT), trauma-specific Prolonged Exposure or cognitive processing therapy (CPT), and integration of a 12-step approach into buprenorphine treatment that also included Motivational Interviewing and CBT.

The literature on effects of reimbursement policy changes on retention in or continuity of SUD treatment is limited. A large-scale intervention in Delaware involving incentive and penalty-associated measures of "program completion" and "active participation," coupled with subsequent quality improvement (QI) initiatives, resulted in increased treatment length of stay. An initiative in Washington State focused on continuity of care after residential treatment or detoxification and included electronic reminders regarding patients in need of follow-up and financial awards based on continuity of care. The Washington State initiative resulted in improvements only for clients at residential agencies already performing at either a moderate or high level. Finally, a reimbursement-focused approach to improving continuity of care known as "contracting with staff incentives" or CSI helped improve transitions between a short-term intensive residential facility and outpatient care in the Baltimore area by providing incentives to the outpatient facility for intake and three follow-up visits.

There is historic evidence that retention in MAT treatment for OUD results in better outcomes, including medical morbidity, social functioning, rates of HIV transmission, and criminal activity. Recent research provides us with additional evidence of reduced mortality, substance use, and inpatient utilization.

Gaps in the Literature

There are certain subjects related to treatment retention in OUD MAT that are not adequately studied in the current literature. For example, there is not much research in understanding retention and outcomes over periods longer than 2-5 injections of XR-NTX. There also is need for evaluating certain models of MAT that are recently being used, such as the Medicaid Opioid Health Homes; treatment approaches that promote early use of MAT after initial diagnosis; and treatment approaches that promote retention for postpartum women. We also need a better understanding of the effects of providing recovery supports such as social support, housing support, employment support, childcare, or transportation on MAT treatment retention and outcomes, as well as improved understanding of the effects of long-term retention on functioning and quality of life. More research is also needed on the occurrence of multiple episodes of treatment, retention within and across episodes, and outcomes from recurring treatment.

 

INTRODUCTION

In recent years, the rate of opioid overdose and death among the American population has escalated to record levels. From 2016 to 2017, drug overdose deaths increased by an age-adjusted 9.6 percent. In 2017, there were 70,237 drug overdose deaths, and 47,600 or 67.8 percent of these deaths involved opioids (Centers for Disease Control and Prevention, 2018). In October 2017, the opioid crisis was declared a national public health emergency by the federal administration (HHS, 2017).

The most effective solution available for treating people with OUD is MAT (e.g., Fullerton et al., 2014; Mattick et al., 2009; Thomas et al., 2014), which is the integration of medication and psychotherapy. There are three federally approved medications to treat OUD: methadone, buprenorphine, and XR-NTX. Despite the availability of effective treatment, use of treatment services has remained extremely low. The National Survey on Drug Use and Health estimated that in 2017 approximately 20.7 million people aged 12 years or older needed treatment for a SUD and only 0.9 percent of those who needed treatment received it at a specialty facility (SAMHSA, 2018). Further, once treatment is initiated, rates of retention also are estimated to be quite low. In our own analysis of private insurance and Medicaid claims data, we found that only 13 percent of individuals with private insurance and 21 percent of Medicaid beneficiaries who initiated treatment for OUD continued it for 180 days or more (IBM Watson Health, 2017). Research has demonstrated that treatment of OUD with MAT improves social functioning and decreases drug use, infectious disease transmission, inpatient utilization, criminal activities, and the risk of relapse, overdose, and death (Ma et al., 2018; Manhapra et al., 2018; Shcherbakova et al., 2018; Stone et al., 2018), and better understanding approaches to improved retention is important in addressing the opioid emergency.

 

OBJECTIVE

The objective of this literature review was to summarize the recent literature on retention in or continuity of treatment for SUD, with a primary focus on MAT for OUD, examining patient characteristics, practice or system-based approaches, psychosocial supports, and reimbursement approaches that influence retention, as well as examining the effects of retention on patient outcomes. We also reviewed the literature to support defining treatment retention and continuity. The results will lay the groundwork and establish context for five case studies of programs, sites, or treatment approaches to OUD treatment that show promise in retaining individuals with OUD in treatment and, to the extent possible, in promoting positive treatment. In addition to providing context for the case studies, it will assist in developing criteria to select sites and inform our choices for recruitment.

 

METHODS

We performed a literature review that addressed retention in SUD treatment with a primary focus on treatment for OUD. The literature review included both peer-reviewed and grey literature.

Peer-reviewed literature. The peer-reviewed literature included English-language publications from the years 2014-2018, supplemented, where appropriate, with seminal literature prior to 2014. Our searches of the peer-reviewed literature used the PubMed and Google Scholar databases.

We culled the recent literature to determine what is known about the following: the components of SUD treatment that support retention and sustained recovery, including psychosocial supports, reimbursement structures or payment models that support retention and sustained recovery, and other factors that may support retention and sustained recovery and which may interact with or influence the development or application of treatment models. Initial search terms included the Level 1 search (Table A1-1) with additional search terms included as secondary searches to help discriminate between subtopics:

TABLE A1-1. Preliminary Search of Peer-Reviewed Literature
Level 1 Level 2 Options
Language: English Option 1: evidence OR evidence-based
AND Publication Dates: 2014-2018 Option 2: psychosocial OR counseling OR support* OR therapy
AND ((substance OR opioid) AND disorder) AND treatment Option 3: outcome*
AND (retain OR retention OR continuity) Option 4: reimburse* OR financing OR payment OR insurance*
  Option 5: setting* OR model* OR structure

We: (1) reviewed identified abstracts to determine if they were relevant to the research questions; (2) for abstracts identified as relevant, retrieved the full-text articles to determine if they provided material related to retention in SUD treatment; and (3) abstracted those articles for further use in the literature review. Key words were used to track pertinence to research questions, allowing sorting and filtering of literature as part of our synthesis. Initial key words included: evidence-based, psychosocial, outcome, retention, abstinence, overdose, reimbursement, setting, and model. These key terms were supplemented after abstracting began to include related terms (e.g., continuity, relapse), MAT medication types, and other terms relevant to our study. Based upon findings in the articles identified, we included additional literature referenced in the initial publications.

Grey literature review. The grey literature included searches of websites of federal and state government agencies (e.g., state health department and Medicaid programs, SAMHSA, CMS, AHRQ, VA), private payers and health systems, non-profit stakeholders, and research organizations. We searched for information on ongoing treatment models, programs, program evaluations, and reimbursement initiatives with an eye toward their influence on retention in treatment. We also reviewed reports and information supplied by key informants being interviewed as part of the larger study.

Approach to synthesis. The resources identified in the peer-reviewed and grey literature were reviewed with the objectives of: (1) describing the meaning of "retention" and "continuity" as used in the literature; (2) synthesizing the information collected in order to address each of the research questions for this study; (3) identifying gaps in the literature; and (4) beginning the process of developing criteria for selecting case study sites and starting to identify a subset of sites for further consideration. Our findings also will inform the development of the case study protocol in advance of site visits.

 

FINDINGS

In this section, we summarize our findings, including: (1) an overview of treatment settings and models of OUD treatment using MAT; (2) a discussion of how retention and continuity are defined and operationalized; (3) summaries of recent literature on patient characteristics, practice or system-focused approaches to treatment, psychosocial supports, and reimbursement approaches that affect retention; and (4) discussion of effects of retention on patient outcomes.

Practice-Based Models in Primary Care

  • OBOT
  • Buprenorphine HIV Evaluation and Support Collaborative Model
  • One-stop shop model (HIV/Hepatitis C virus management and buprenorphine)
  • Integrated prenatal care and buprenorphine

Systems-Based Models in Primary Care

  • Hub-and-spoke model
  • Medicaid health home model
  • Project ECHO
  • Collaborative opioid prescribing
  • Massachusetts nurse care manager model
  • ED initiation of OBOT
  • Inpatient initiation of OBOT
  • Southern Oregon model

Treatment Settings and Models of OUD MAT

As background for our discussion of approaches to treatment for OUD and other SUDs, we examined the literature to obtain an overview of current models of treatment. In 2016, AHRQ published a report identifying 12 representative models of OUD MAT treatment in primary care settings. The models were characterized as either practice-based (i.e., capable of implementation in an individual clinic) or systems-based (i.e., involving components across the health care system, including some that have origins in inpatient or emergency department settings). Key components of the models included: (1) pharmacological therapy (primarily buprenorphine); (2) provider and community educational interventions; (3) coordination or integration of SUD treatment with other care; and (4) psychosocial services, although not all models incorporate every component (AHRQ, 2016). The text box to the right lists the 12 models identified by AHRQ, many of which are referenced throughout this report if there are recently published studies related to retention. Models that do not have published evidence regarding retention might warrant further investigation. In addition to the models identified by the AHRQ report, this literature review suggests other models that might be added, including low-threshold models, physician-pharmacist collaborative models, models involving telehealth, home-based models, criminal justice models, and models integrating 12-step approaches into treatment with medication.

Defining Retention and Continuity of Care

Defining retention and continuity of care in SUD treatment is not straightforward. The first concept, retention, presupposes entry into treatment and requires continuation in treatment for some period of time. Retention also has been called "persistence" (Shcherbakova et al., 2018). As discussed below, retention may be operationalized many ways. The second concept, continuity of care, may equate to treatment retention, but most frequently refers to continuous or near-continuous possession of treatment medication (e.g., Saloner et al., 2017). A recently developed performance measure which has been endorsed by NQF considers 180 days with no more than a 7-day gap in medication possession to be continuity of pharmacotherapy for OUD (NQF, 2018). Continuity of care also may be used to identify successful transitions of care from intensive treatment, such as detoxification or residential, to less IOP treatment (e.g., Acevedo et al., 2018).

Studies looking at retention often define it based on the duration of data available if the study is claims-based, or a reasonable time within which data can be collected. Thus, retention has been defined as 3 months (e.g., Weintraub et al., 2018), as 7 years (Bhatraju et al., 2017), and as different time periods within that range (e.g., Timko et al., 2016). The NQF-endorsed measure of medication continuity uses a 180-day period, primarily based on FDA registration trials which have studied effectiveness over 3-month to 6-month periods (NQF, 2018). Concern has been expressed about unintended consequences of such measures, including that payers may begin to treat whatever period is measured as "sufficient" and decline to pay for additional treatment (ASAM, 2014).

Studies and measures on retention in care may allow for gaps in treatment. Some measure only continuous involvement in treatment (e.g., Manhapra et al., 2017; Manhapra et al., 2018; Riggins et al., 2017); others allow gaps, such as 30 days (e.g., Eibl et al., 2017; Franklyn et al., 2017; Shcherbakova et al., 2018; Wilder et al., 2015) or 90 days (Saloner et al., 2018). On the other hand, Bhatraju et al. (2017) excluded only those with a mean non-participation of greater than 18 weeks, judging that to be evidence of multiple treatment episodes. The 180-day measure of MAT pharmacotherapy permits a gap of only 7 days, in part based on the rationale that risk of mortality increases immediately after treatment cessation (NQF, 2018). The idea that treatment should be continuous has been criticized as not considerate of the chronic nature of SUD. Vogel et al. (2016) suggest that there are three approaches to retention: (1) a single continuous episode; (2) a provider's perspective that looks backwards from the current point and asks how long a patient has been in treatment; and (3) a public health perspective that asks how many days in the past year the person has been in treatment. Vogel et al. characterize the third approach as best addressing chronicity and cycles of relapse and remission.

Individuals Often Enter and Leave Treatment Repeatedly

Studies rarely examine the extent or course of repeated cycles of OUD or SUD treatment, including retention over the course of treatment participation. Studies that do, find repeated episodes to be not uncommon.

Studies rarely allow retention to include treatment re-entry after disengagement or relapse. A long-term follow-up study after the National Drug Abuse Treatment Clinical Trials Network Prescription Opioid Addiction Treatment Study (POATS), however, revealed that, even after the buprenorphine treatment in the original POATS ended, among those available for follow-up at 42 months (375/653), 29.4 percent were in opioid agonist therapy and did not meet current symptom criteria for OUD while 31.7 percent were abstinent and not on opioid agonist therapy (Weiss et al., 2015). Other studies have recognized the fact that treatment reoccurs. Shcherbakova et al. (2018) did, independently of retention, examine number of treatment episodes within 365 days of the first prescription and found that 20 percent of patients had two or more treatment episodes during the 1-year period. Similarly, Weinstein et al. (2017) looked both at a more conventionally defined retention and at the number of treatment episodes experienced during a lengthy study period.

Most studies also do not address the nature of treatment participation, although the outcome of retention in studies often is paired with an outcome of abstinence from all illicit substances and sometimes alcohol (e.g., Kumar et al., 2016; McLellan et al., 2008 (where the abstinence outcome adhered to state requirements for pay-for-performance)), suggesting that, for treatment to be considered successful, one cannot simultaneously be using other substances. This pairing is interesting as individuals are sometimes administratively discharged upon discovery of any substance use (A.R. Williams, personal communication, November 13, 2018), effectively precluding retention.

Studies, using differing periods for measurement, reveal widely disparate retention rates and at least four recent systematic reviews have addressed retention in MAT. A systematic review of 55 studies of retention in MAT treatment covers findings from studies published between 2010 and 2014 (Timko et al., 2016). One review included 35 studies of primary care-centered interventions involving buprenorphine or methadone with coordinated care that were published prior to 2016 (Lagisetty et al., 2017). Another examined 34 studies related to use of XR-NTX that were published between 2006 and 2017 (Jarvis et al., 2018). Finally, Wilder et al. (2015) reported on 15 studies published between 1973 and 2012 of buprenorphine and methadone treatment discontinuation among pregnant and postnatal women. Recent studies examined for this review, and in which MAT was a clearly identified part of treatment, reveal retention rates as shown in Appendix A.

Patient Characteristics Affecting Retention in SUD Treatment

Variables that affect retention in SUD treatment may include patient characteristics, treatment-related variables, and payer policies. In this section, we examine patient characteristics such as age, sex, other socio-demographic characteristics, previous or current substance use, co-occurring psychiatric conditions, and other patient-related factors identified as influential in recent literature. Effects of some of these characteristics vary depending on stratification, such as by sex, and are addressed under the relevant stratification characteristic.

Patient Characteristics Associated with Increased Retention

Being older, having more education, having stable housing status, not using cocaine, and living in the same county as the pharmacy where buprenorphine prescriptions are filled have all been associated with increased length of retention in treatment for OUD.

Age. Retaining younger adults in treatment has historically been more difficult than retaining older adults. In analysis using Medicaid claims data, Samples et al. (2018) examined factors associated with discontinuation of buprenorphine treatment. Adjusted logistic regression indicated that discontinuation of buprenorphine treatment before 180 days was significantly associated with being younger than 55 years of age, with discontinuation most pronounced among those ages 18-24 years. Multivariate analyses using an all-payer claims pharmacy database revealed that, compared to those ages 18-34 years, those who were older were more likely to remain in buprenorphine treatment for 180 days (Saloner et al., 2017). Schuman-Olivier et al. (2014) examined retention in buprenorphine treatment among emerging adults ages 18-25 years compared to those who were older. Emerging adults remained in treatment at significantly lower rates at 3 months (57 percent vs. 78 percent) and 12 months (17 percent vs. 45 percent).

Sex. In their analysis of Medicaid claims data, Samples et al. (2018) found that discontinuation of buprenorphine treatment before 180 days was significantly associated with being male. In contrast, multivariate analyses using an all-payer claims pharmacy database revealed that females were less likely than males to remain in buprenorphine treatment for 180 days (Saloner et al., 2017). Choi et al. (2015) note that, historically, women are more likely to remain in treatment if they have higher education and income and lower psychiatric severity. In contrast, in public outpatient treatment settings, men may be better retained. Choi explored retention in private for-profit residential treatment among individuals with co-occurring mental health disorders and SUDs, where the status of MAT use was unclear, by sex and found that, at 30 days, 40 percent of women remained in treatment, in contrast to 30 percent of men. Factors associated with increased retention among men included being older, having ADHD in contrast to mood disorder, and having less severe employment issues. Among women, increased retention was associated with not being cocaine dependent, greater severity on the Addiction Severity Index alcohol subscale, depression in contrast to another mood disorder, and being in the action or maintenance stages of readiness to change. In another study of an intervention that did not incorporate MAT, Braitman et al. (2016) examined influences on initiation and retention in 12 weekly outpatient sessions of Behavioral Couples Therapy. The couples all had at least one child, and one or both the spouses were already in SUD treatment elsewhere, which may or may not have included MAT. Couples were significantly more likely to attend a greater number of treatment sessions if the male was older when he noticed substance use problems, did not report being a victim of intimate partner violence, or had more obsessive-compulsive or phobic anxiety symptoms. There were no characteristics of the women that were significantly associated with couples attending treatment. Older research does indicate that gender-specific treatment has historically been linked with improved retention among females (see, e.g., references cited in Choi et al., 2015).

Other socio-demographic characteristics. Cui et al. (2016) found that both education and housing status were associated with 12-week and 24-week retention of veterans in a dual diagnosis clinic focused on psychosocial treatment for those with combined PTSD, depression, and SUD. Higher levels of education predisposed participants to remain in treatment at both time points. In contrast housing difficulties at baseline significantly interfered with treatment retention at 12 weeks. Samples et al. (2018) research using Medicaid claims data indicated that discontinuation of buprenorphine treatment before 180 days was significantly associated with being African-American or Latino in contrast to White.

Other substance use. Recent studies have investigated how substance use affects retention in OUD treatment. In analysis using Medicaid claims data, Samples et al. (2018) found that discontinuation of buprenorphine treatment before 180 days was significantly associated with having alcohol use disorder (AUD) or other non-OUDs identified on claims or having experienced an opioid overdose in the 6 months prior to buprenorphine induction. Several studies indicate that cocaine use at baseline is a predictor of poor retention in OUD treatment. In a nurse manager model of IOP with transition to outpatient buprenorphine treatment, cocaine-positive toxicology at intake predicted poor 3-month treatment retention (Schuman-Olivier et al., 2014). One-year retention in methadone/ buprenorphine clinics in Ontario was associated with not having a cocaine-positive urine screen at baseline (46 percent if negative (median days retained=302), 39 percent if positive (median days retained=212 days)) (Franklyn et al., 2017). A study of HIV-infected individuals with AUD or OUD who initiated XR-NTX while incarcerated found that receipt of the second injection within 30 days post-release was significantly less likely if the person had a cocaine-positive toxicology result after release (Springer et al., 2015). A study in Italy that did not address MAT concluded that patients with cocaine use disorder had lower retention rates in residential treatment than did those with heroin use disorders (Maremmani et al., 2017). In contrast, a study by Socias et al. (2018) of individuals in treatment in methadone/buprenorphine clinics in Vancouver found that those who used cannabis at least daily during treatment had greater odds of retention in OUD treatment at 6, 12, and 18 months than did those who did not.

Co-occurring behavioral health disorders. Several recent studies have examined the effect of co-occurring behavioral health disorders on treatment retention, as well as interventions focused on those disorders. Several indicate that patient trauma affects retention in treatment. Kumar et al. (2016) examined the effect of early childhood trauma on retention in outpatient buprenorphine treatment. Adjusted regressions revealed that those with moderate to severe physical or emotional neglect are significantly more likely to leave OUD treatment within 90 days. A study conducted at a Veterans Health Administration dual diagnosis clinic looked at characteristics associated with 24-week treatment retention among veterans with depression, trauma, and SUD (Cui et al., 2016). Type of SUD varied, and the study did not specifically address use of MAT. However, veterans who had experienced sexual trauma attended more individual treatment sessions than did those with combat-related or other traumas.

Therapeutic alliance and motivation to engage in treatment. Therapeutic alliance and patient motivation to participate in treatment have historically been associated with improved SUD treatment retention (Meier et al., 2005; Joe et al., 1997; Joe et al., 1998).

Health system factors and insurance. In analysis of Medicaid claims data, Samples et al. (2018) examined factors associated with discontinuation of buprenorphine treatment. Adjusted logistic regression indicated that, among other things, discontinuation before 180 days was significantly associated with having capitated Medicaid coverage and with inpatient service use in the 6 months before induction. Saloner et al. (2017) conducted multivariate analyses using an all-payer claims pharmacy database and found that those whose primary prescriber was other than a primary care provider (PCP) or psychiatrist were significantly less likely to remain in buprenorphine treatment for 180 days, compared to those seen by a PCP. Six-month retention in buprenorphine treatment also was significantly associated with cash payments for the prescription fill, in contrast to insurance payment (Saloner et al., 2017).

Criminal justice system (CJS) involvement. In a study of treatment outcomes among individuals receiving buprenorphine treatment integrated into HIV clinics, Riggins et al. (2017) examined whether incarceration in the 30 days before clinic entry influenced retention in treatment. There was no significant difference in either 6-month or 12-month retention in treatment between those who had and had not been recently incarcerated.

Pregnancy status. Wilder et al. (2015) undertook both a systematic literature review of factors influencing retention in buprenorphine or methadone treatment of pregnant and postpartum women, as well as an analysis of retention of a group of pregnant and postpartum women in methadone treatment. The literature review revealed several studies of varying ages that inconsistently separated the pregnant and postpartum periods, and did not all include MAT, making conclusions difficult to interpret. A separate analysis using data from the methadone treatment program revealed a prenatal discontinuation rate of 11 percent and an adjusted 6-month postpartum discontinuation rate of 56 percent. A longer prenatal connection to the clinic was significantly associated with slightly lower risk of postpartum discontinuation.

Geography. In a study of 6-month retention in buprenorphine treatment using an all-payer claims pharmacy database, those who crossed county lines to fill the prescription were significantly less likely to remain in treatment for 180 days. The state in which the pharmacy was located also had a significant effect on 6-month retention (Saloner et al., 2017).

Evidence-Based Methods to Address Treatment Retention in SUD and OUD Treatment

Approaches to SUD treatment have been developed that are intended to promote treatment retention and to provide other benefits. We discuss elsewhere psychosocial and reimbursement-related interventions. In this section, we address approaches that are practice or system-focused or that otherwise rely on the treatment system structure. We address separately the use of telehealth because it affects access and, consequently, retention. We also look at recent literature on medication dosing. The evidence supporting these influences on retention is variable. Studies are of various sizes, of different durations, and involve different settings and treatment approaches. Only some include comparison groups or earlier retention rates. We address, first, eight studies with some comparison; second, six studies or models without any comparison; third, three studies involving telehealth; and, finally, a selection of recent studies that examine dosing of buprenorphine or methadone.

Studies with Comparative Data

A Washington statewide intervention designed to promote continuity of care after detoxification or residential treatment was established as a randomized four-arm trial including: (1) weekly electronic reminders on recently discharged patients not receiving follow-up treatment; (2) a reimbursement option (discussed elsewhere); (3) both; or (4) no intervention. The weekly reminders identified clients discharged in the previous 2 weeks, their continuity of care deadline, and the number of days until the deadline, as well as graphs of the agency's quarterly performance. Tips for improving continuity of care and a link to resources also were provided. The continuity of care measure looked only at admission within 14 days to subsequent treatment following discharge from one of the two settings. The use of electronic reminders had a significant effect only for agencies that were already performing at a moderate or high level at baseline. Focus group interviews revealed some reasons that continuity of care did not improve more consistently. These included the need for wrap-around services, most commonly transportation; the complexity of patients, including many with co-occurring disorders and homelessness; technical problems related to opening encrypted emails or slow entry of service data; lack of treatment system capacity; and competing state-level transitions that impeded attention to improving continuity of care (Acevedo et al., 2018).

To assess effects of inpatient induction into buprenorphine treatment, patients with OUD who were hospitalized in a New England hospital were randomized to either: (1) in-hospital detoxification with buprenorphine and post-discharge referral information; or (2) in-hospital buprenorphine induction, a maintenance dose, and facilitated entry into an associated primary care opioid addiction treatment (OAT) program. Retention-related outcomes included entry into the OAT within 6 months and continuation in treatment at 6 months. Slightly less than 12 percent of the comparison group entered treatment in the OAT within the 6 months follow-up period, in contrast to 72.2 percent of the experimental group. At 6 months, 16.7 percent of the experimental group were still in treatment with 64.4 mean days of buprenorphine treatment, compared to 3.0 percent of the comparison group with 26.2 mean days of buprenorphine treatment (Liebschutz et al., 2014).

Rationales for Failure of Continuity of Care Interventions

Reasons cited by residential and withdrawal management providers that reminders of need for follow-up or incentives for ensuring patient continuity of care did not improve continuity into outpatient treatment included:

  • need for wrap-around services;
  • patient complexity;
  • technical problem;
  • slack of treatment system capacity;
  • competing demands.

(Acevedo et al., 2018)

A study of retention after being seen in the emergency department involved three alternate interventions: (1) referral information for further treatment; (2) brief intervention in the emergency department with linkage to a referral; and (3) brief intervention with buprenorphine induction in the emergency department, followed by take-home daily doses sufficient to last until they could be seen in the hospital's primary care center within 72 hours, where they received treatment for 10 weeks and referral to follow-on care. In the arm of the study where buprenorphine was started, those who requested it were also provided 2 weeks of detoxification. The group inducted on buprenorphine in the emergency department had significantly higher rates of self-reported participation in formal treatment at 2 months, compared to the other two groups (74 percent vs. 53 percent (referral) or 47 percent (brief intervention)) but not at 6 months (D'Onofrio et al., 2017).

Home-Based Delivery of XR-NTX

A Baltimore pilot program involving home-based delivery of XR-NTX showed promise for retention of young adults in treatment over a 16-week, 5-dose course of treatment (Vo et al., 2018).

In general, evidence has shown that successful induction of XR-NTX can be difficult (Lee et al., 2018). A couple of recent studies involved retention in treatment with XR-NTX. To address issues of retention of young adults at a Baltimore community-based treatment program with specialty programming for adolescents and young adults with OUD, a small pilot program was established for home-based delivery of XR-NTX to young adults (Vo et al., 2018). Residential detoxification and induction of either naltrexone or buprenorphine was followed by outpatient maintenance using XR-NTX or buprenorphine, with the first dose of XR-NTX given in the residential setting. The young adult outpatient program began with IOP care, including treatment of co-occurring conditions, comprised of groups and individual sessions. Fourteen patients, selected to include variable clinical status, were offered enrollment in the home-based XR-NTX program. Home visits occurred every 3-4 weeks and participants continued to receive usual counseling at the clinic, or at home if the person was not attending the clinic. Thus, the intervention included home-based delivery of XR-NTX and medication management services, assertive outreach, and case management, with decreased emphasis on psychosocial treatment or abstinence from non-opioid substances. The pilot group was compared to an historic group of 21 patients who received usual care XR-NTX over the prior year. Of the 14 youths enrolled in the home-based intervention, nine initiated outpatient treatment and received at least one dose at home. Over a 16-week period, the intervention group received a greater number of doses (66 percent vs. 40 percent of the maximum five doses), were more likely to receive all five doses (50 percent vs. 9 percent), attended a similar number of counseling sessions (fewer per week (mean 1.3 vs. 2.3) spread over a longer period of time (mean 7.9 weeks vs. 6.1 weeks)), and had a retention rate of any past-month attendance of 64 percent compared to 19 percent for the group receiving treatment as usual (Vo et al., 2018).

Induction and retention in treatment for individuals leaving the CJS is important, particularly given the high rates of opioid-related overdoses after release. Implementation of pre-release induction of XR-NTX has proven useful to encourage subsequent treatment participation. The Hampden County Correctional Center in western Massachusetts instituted a program of XR-NTX induction 7 days before release, with a link to a collaborating community program for continued treatment and counseling. Buprenorphine was also available at the community sites. Forty-seven prisoners were inducted pre-release and 20 were referred for induction post-release. Retention was as follows: week 4, 55 percent vs. 25 percent; week 8, 36 percent vs. 25 percent; week 24, 21 percent vs. 15 percent. At weeks 8 and 24, the difference was not statistically significant (Lincoln et al., 2018).

In Canada, methadone treatment initiation must occur in a specialized clinic and, although the drug must be prescribed by an addiction specialist, it can be dispensed by a nurse or pharmacist. Once stabilized, patients can be treated with observed dosing in a primary care setting or in a pharmacy and take-home doses eventually are allowed. Studies of methadone treatment initiatives in Canada reveal retention improvement from which we might draw lessons for MAT more generally in the United States. One study from Ontario was aimed at retention in multiple methadone clinics, testing dispensing in onsite pharmacies against dispensing from offsite pharmacies (Gauthier et al., 2018). Patients using an onsite pharmacy demonstrated a 1-year retention of 57.3 percent compared to 11.9 percent retention in the offsite pharmacy group. Multivariate regressions found that those who filled their prescriptions at onsite pharmacies were 77 percent less likely to withdraw from treatment before 1 year, compared to the offsite pharmacy group.

A Network for the Improvement of Addiction Treatment (NIATx) open-access model of rapid enrollment in methadone treatment was implemented at a community-based organization in New Haven, Connecticut (Madden et al., 2018). A study using 9 years of data found increased access without large changes in retention. The program undertook numerous modifications to promote ease of access to treatment, which included rapid access to group and individual treatment options. Among the structural changes implemented were minimizing barriers to rapid intake such as not requiring a tuberculosis skin test to be read before admission; providing walk-in same-day screening, intake, and initiation; providing treatment regardless of ability to pay and assistance in obtaining Medicaid; and making the primary mode of treatment drop-in groups with individual counseling available as needed or on request. The most immediate goal of the model was to improve access, which happened, with increased census, reduced wait times, and an increase in overall revenue. There were no major changes in rates of negative toxicology screens or mortality. The measure of 90-day retention at baseline was 89.3 percent. In the 9 years after baseline, mean retention ranged from 81.9 percent to 91.8 percent (tests of significance were not performed).

Timely treatment of young adults (ages 19-21 years) newly diagnosed with OUD was associated with improved retention in an analysis using Medicaid claims data. Timely treatment was defined as receipt of behavioral health services within 3 months of diagnosis with OUD. Among 4,837 youths diagnosed with OUD, 75.5 percent received any treatment within 3 months of diagnosis; 52 percent received only behavioral health services, while 23.5 percent received medication. Median retention in care among youths who received timely buprenorphine was 123 days; naltrexone, 150 days; and methadone, 324 days, compared with 67 days among those who received only behavioral health services. Adjusted multivariate analysis showed that timely receipt of each drug independently was associated with lower attrition from treatment compared with receipt of only behavioral health services (Hadland et al., 2018).

Studies without Comparative Data

Low-threshold or no-threshold OUD treatment is increasingly being discussed as an option to reduce impediments to retention in treatment. Bhatraju et al. (2017) reported on an initiative to treat OUD in an office-based primary care setting in a New York City public hospital. Although there was no comparison group, the study did have the advantage of providing 7 years of data. Program characteristics included unobserved buprenorphine-naloxone induction following a new patient visit, weekly and then less frequent follow-up, and a general recommendation but no requirement for additional psychosocial treatment. During the initial office visit, the patient was offered a prescription and provided with bilingual pictogram-based instructions to self-administer the initial induction and maintenance doses after leaving the clinic. Psychosocial support was primarily delivered during provider-patient medical management visits. The study examined retention from program inception in 2006-2013. Treatment retention was a median of 38 weeks (range 0-320 weeks) among induction patients (n=302); 110 weeks (0-353 weeks) among transfers (n=175), and 57 weeks for all patients (n=477). Treatment departure for those inducted to buprenorphine in the program was as follows: week 5, 25 percent; week 38, 50 percent; week 144, 75 percent. Among all patients, adjusted hazard ratios for earlier drop-out included induction and participation in an earlier time in the study period.

Another initiative that focused on reducing barriers to treatment was the Baltimore Buprenorphine Initiative/Advancing Recovery project. Advancing Recovery was a Robert Wood Johnson/NIATx initiative implemented in three Baltimore treatment agencies. Treatment included buprenorphine treatment combined with IOP counseling, followed by extended buprenorphine treatment and transfer to community physicians. Seeking to reduce wait time, the program greatly trimmed paperwork and other barriers to admission, reducing wait time considerably. The response was positive yet did not result in increased continuity in care. The rationale provided in a case study published by NIATx (2010) was that patients who wanted RAM were likely in acute withdrawal and not interested in IOP counseling. The program undertook focus groups and surveys to determine barriers to continuation and the top reasons for not remaining in treatment were transportation, work schedules, cost of services, and housing issues. Many other barriers were also identified. The focus groups also revealed that clients wanted individual counseling and help with housing and job skills as part of group counseling. The program responded to what they learned but the results of those changes are not available.

Massachusetts OBOT Model

A 12-year study of OBOT with buprenorphine in a large public hospital found that 53.7% of patients had at least 1 treatment period of a year or more (Weinstein et al., 2017). A 6-year study of the model as applied in community health centers showed increased 12-month retention rates as high as 65% (LaBelle et al., 2016

Weinstein et al. (2017) reported on retention within the Massachusetts model of Office-Based Opioid Treatment (OBOT) with buprenorphine, a model that uses a nurse care manager to promote collaborative care. The setting was a large urban safety-net hospital and data were available covering a 12-year period, beginning with model implementation. Both primary care and buprenorphine treatment were included in the hospital's primary care clinic. A nurse care manager saw patients weekly for the first month and the buprenorphine prescriber saw them approximately every 3 months, depending on stability. Weekly SUD counseling was required but was most often accessed outside the hospital. During the last 2 years of the study period, limited enhanced access to psychiatry was available at the primary care clinic. The study examined retention in OBOT for at least 1 year, allowing for up to a 60-days gap in treatment. Patients who left the program, but who re-enrolled, contributed repeated observations. Forty-five percent of all treatment periods were 1 year or longer and 53.7 percent of patients had at least one treatment period of a year or more.

A related study examined retention and other outcomes in the Massachusetts model of OBOT with buprenorphine (OBOT-B), but within primary care in the system of community health centers that participated throughout the Commonwealth. The health centers used nurse care managers to provide buprenorphine waivered physicians with clinical support to manage patients with OUD. Data from a 6-year period showed that patients remaining in treatment for longer than 12 months during 2010, 2011, and 2012 were 32 percent, 56 percent, and 65 percent, respectively. Data on retention at baseline in 2007 were not reported (LaBelle et al., 2016).

The Buprenorphine-HIV Evaluation and Support (BHIVES) study integrated buprenorphine treatment into HIV clinics in different parts of the United States. Sites included academic medical centers, community clinics, and a public hospital; only one had provided MAT before the study began. All sites provided comprehensive medical and social services, including substance use counseling and case management, and most provided follow-up outreach services, although approaches varied by site. Extensive technical assistance was provided as part of implementation and evaluation (Weiss et al., 2011). Early analysis of retention over a 1-year period showed that a small percentage of patients (8 percent) transferred to methadone and 3 percent went into inpatient or detoxification settings. Over the course of 1 year, 74 percent, 67 percent, 59 percent and 49 percent were retained at 3, 6, 9, and 12 months, respectively (Fiellin et al., 2011). Riggins et al. (2017) reported on one aspect of the larger BHIVES study, to determine if incarceration within 30 days before initiation affected retention. There was no significant difference in retention between those recently incarcerated and those not.

A small initiative was implemented in a suburban health department in the Baltimore area and involved physician-pharmacist collaboration. The patients were inducted on buprenorphine-naloxone by outside providers and referred to the collaborative program for continued treatment. The providers involved in the collaborative program included a PCP, a medical assistant, and a psychiatric pharmacist. Protocols and responsibilities were established in advance. Patients were initially monitored weekly and then monthly depending on treatment plan adherence and toxicology results. Participants were required to use one pharmacy and the preferred pharmacy was the one involved in the collaboration. Referrals were made for 19 patients, of which 12 participated. Mean duration in the pilot was 20 weeks (ranging from 2-52 weeks). Retention was defined as being enrolled and remaining in treatment for 6 and 12 months. Fifty percent of patients (6/12) successfully progressed from weekly to monthly monitoring. One hundred percent were retained for 6 months and 73 percent for 12 months (DiPaula & Menachery, 2015).

Telehealth and Retention

Access to care and retention are related. If access is difficult, initiation, engagement, and retention will be more difficult. For that reason, interventions designed to promote access also affect retention. Service delivery via telehealth is one such approach. Weintraub et al. (2018) studied the use of telehealth for prescribing buprenorphine in a drug treatment center for adults in rural Maryland. The program included IOP treatment and transitional housing. The intended treatment duration was 4 months and all patients presented already detoxified. Medications were provided by an affiliated local pharmacy and were placed by patients in locking bags kept in a locked room at the housing unit. The telehealth provider was at a university site and both the distant and originating/local sites had site coordinators. Retention out of 177 participants was as follows: 1 month, 91 percent, 2 months, 72.8 percent, and 3 months, 57.4 percent.

In a Canadian study (Eibl et al., 2017), clinics across Ontario provided both methadone and buprenorphine. Because of provider shortages, telehealth came to be used extensively in both rural and urban areas. The telemedicine practice guidelines required one in-person visit to occur within the first 6 weeks of treatment. The study categorized patients as predominantly in-person (<25 percent by telemedicine), predominantly telemedicine (>75 percent telemedicine), or mixed. Treatment discontinuation was defined as 30 continuous days without either methadone or buprenorphine. Patients using predominantly telemedicine were maintained for a median of 366 days and 50 percent were retained for 1 year; patients receiving care predominantly in-person or mixed care were maintained for a median of 207 days with 39 percent retained for 1 year and 317 days with 47 percent retained for a year, respectively. Patients who did not see their physician in-person within 6 weeks of beginning treatment were as likely or more likely to be retained than patients who did have an in-person visit within their first 6 weeks.

A second study from Ontario incidentally concluded that telehealth may be a factor influencing retention. Franklyn et al. (2017) sought to determine the effect of cocaine use on OUD treatment retention over a 1-year period. Non-retention was defined as non-receipt of either methadone or buprenorphine/naloxone over 30 consecutive days. Medical records for 3,835 patients were examined from 58 clinics providing opioid agonist treatment throughout Ontario. In general, baseline cocaine users or those who used cocaine at higher rates had a lower retention rate and patients in northern Ontario had higher rates of cocaine use. Despite this, patients from northern Ontario had better retention than those from the southern part of the province. The study authors hypothesized that the more common use of telehealth in the remote northern areas likely increased retention compared to that of more urban southern patients.

Buprenorphine Dosing

An initial buprenorphine dose <4 mg/day was associated with discontinuation of treatment before 180 days (Samples et al., 2018).

Dosing and Retention

There is considerable evidence that the dose of methadone or buprenorphine prescribed affects treatment outcomes, including treatment retention. A 2014 summary of the evidence on dose by the Institute for Clinical and Economic Review (ICER) indicates the threshold doses beyond which clinical outcomes do not improve are approximately 100 mg/day for methadone and 16-32 mg/day for buprenorphine/naloxone. Doses that are too low, however, can adversely affect retention. ICER's summary of the evidence references three case studies concluding that methadone doses of over 60 mg/day, precisely 96 mg/day, or up to but not exceeding 100 mg/day of methadone enhance retention. The evidence on buprenorphine dosing referenced by ICER includes randomized controlled trials (RCTs) indicating that less than daily dosing can be as effective as daily dosing. In more recent analysis using Medicaid claims data, Samples et al. (2018) examined factors associated with discontinuation of buprenorphine treatment. Results indicated that discontinuation of buprenorphine treatment before 180 days was significantly associated with having an initial dose of buprenorphine less than or equal to 4 mg/day.

Promising Models of Psychosocial Support for Retention in OUD MAT

Clinical guidelines recommend concurrent medication and psychosocial treatment or supports for those with OUD (e.g., ASAM, 2015; British Columbia Centre on Substance Use, 2017). The psychosocial treatment is provided to help patients control urges to use drugs, remain abstinent and also to assist patients in coping with the emotional strife that often accompanies addiction (Dutra, et al., 2018). There, however, are those who argue the psychosocial supports are not a necessity for everyone (Martin et al., 2018), and others suggest that a stepped-care model might be preferable, whereby the level of treatment is matched to the patient (Carroll & Weiss, 2017). One reason for this stance is that, over the past decade and a half, studies are not consistent in finding that concurrent treatment results in improved retention in treatment or other outcomes (see, e.g., studies referenced in Meshberg-Cohen et al., 2018; Carroll & Weiss, 2017). A 2016 systematic review on the use of psychosocial interventions with medication for treatment of OUD examined three literature reviews and 27 more recent publications; CM and CBT were most widely studied, and the medication most often studied was methadone (Dugosh et al., 2016). Dugosh et al. agreed that there was inconsistency of results but concluded that there were benefits and that the evidence was strongest in the studies with methadone treatment. Studies examining methadone maintenance found significant effects of psychosocial treatment (i.e., CM and general supportive therapy) on treatment attendance and drop out, while a smaller number of studies showed significant effects on attendance and retention in buprenorphine treatment (i.e., Intensive Role Induction). Positive effects on retention and attendance also were found with oral naltrexone (i.e., behavioral therapy and CM) and XR-NTX (i.e., CM). It has been noted that many of the studies not finding benefit from concurrent psychosocial treatment are conducted in primary care settings; exclude patients with varieties of clinical severity such as alcohol or other drug disorders, trauma, mental illness, or poor physical health; and may not address fidelity to treatment protocols. It has been suggested that studies focusing on patients with less clinical acuity may rule out those who might most benefit from psychosocial supports (Klein, 2017).

We discuss below recent findings regarding the effect of psychosocial supports on retention in OUD-MAT. Because claims analyses cannot identify types of psychotherapy received, we focused on studies of specific treatments to determine whether there are promising models of psychosocial support that assist in maintaining individuals in MAT for OUD. As background, however, a recent large-scale claims analysis did indicate that receipt of psychotherapy in conjunction with buprenorphine among the privately insured is associated with increased retention in MAT (Manhapra et al., 2018).

Recent studies have examined retention in MAT treatment for OUD with psychosocial supports that include CBT and buprenorphine, trauma-specific Prolonged Exposure treatment or CPT with buprenorphine, integration of a 12-step approach into buprenorphine treatment, and a study from Iran reporting on retention in treatment that included a program of cognitive rehabilitation treatment (CRT) coupled with methadone maintenance in a court-mandated setting. These are discussed below. We also address one recent study that examined retention in relation to psychosocial treatments without explicit incorporation of medication treatment for OUD. This was a VA study that examined treatment participation involving CBT and CPT for veterans with different SUDs. Apart from the Iranian study, each of these show promise for retention. The Iranian study is included because the intervention (CRT) is proving to address cognitive impediments that can be associated with opioid abuse and, in a setting that is not court-mandated, might have beneficial effects for retention in MAT for OUD.

An outpatient addiction treatment center affiliated with the University of Arkansas provides buprenorphine treatment coupled with weekly relapse prevention groups, medication education groups, and individual CBT. Study participants, the majority of whom used prescription opioids, could advance to Phase 2 biweekly sessions if they had at least 4 consecutive weeks of negative toxicology screens (for all illicit substances) and attendance at all scheduled group and individual treatment appointments. Evaluation looked at phase advancement within 90 days and retention, defined as ongoing participation in treatment with no absence longer than 8 days during the first 90 days (n=113). The study was focused particularly on the outcomes for participants who had experienced early childhood trauma as assessed at intake. Eighty-two percent of the sample remained in treatment during the 90-days and 76 percent advanced to Phase 2 during that time. Treatment retention was more common among those without a history of moderate to severe trauma. Multivariate regression revealed that those who had experienced physical or emotional neglect were significantly more likely to leave treatment early (Kumar et al., 2016).

Although not focused on delivery of MAT or OUD specifically, another study heavily populated with participants who had PTSD, as well as depression and different SUD diagnoses, also showed promise using CBT or CPT as alternate arms in a clinical trial over a 6-month period. A VA outpatient dual diagnosis clinic in San Diego implemented a trial in which 146 veterans with depression, SUD, and trauma received 12 weeks of group Integrated CBT for depression and substance use and then were randomized to 12 weeks of either individual CBT (n=62) or CPT (n=61) modified to address substance use and trauma. All had a SUD (alcohol, other drug, or polysubstance) and 82 percent had PTSD (combat-related, sexual, or other). The trial was not focused on OUD specifically and receipt of MAT for either alcohol or OUD was not an integral part of the study. The analyses, however, did include receipt of a "substance use medication prescription" as a variable of analysis, with 21 percent of participants having a prescription for a substance use treatment medication. Receipt of such a prescription was not significantly correlated with treatment attendance, nor was it significantly associated with attendance in multivariate regression. Retention over the 24-week period was not addressed but number of group and individual sessions attended was. Participants attended an average of 14.5 group sessions out of 24 and 6.6 individual sessions out of 12, or more than 50 percent of available treatment (Cui et al., 2016).

The Role of Trauma-Specific Treatment in MAT for OUD

A study found that adding trauma-specific treatments such as Prolonged Exposure or CPT to buprenorphine treatment resulted in improved 6-month retention in treatment, with veterans with PTSD and trauma treatment having 30.23 times the odds of retention compared to veterans with PTSD who received no trauma-specific treatment (Meshberg-Cohen et al., 2018).

A study conducted at a VA outpatient clinic examined the effectiveness of adding trauma-specific treatment (Prolonged Exposure or CPT) to buprenorphine treatment for veterans with PTSD (n=21), compared to veterans with PTSD who did not receive trauma treatment (n=46), and to veterans without PTSD (n=73). There is evidence that PTSD, as well as other psychiatric conditions, may impede response to SUD treatment and the objective was to determine if trauma-specific therapy could alleviate that. Veterans with PTSD who received trauma treatment had 6-month retention of 90.5 percent, while 46.6 percent of those without either PTSD or trauma treatment and 23.9 percent of those with PTSD but no trauma treatment were retained. When treatment was regressed on group alone, veterans with PTSD and trauma treatment had 30.23 times the odds of retention compared to the reference group of veterans with PTSD who received no such treatment. The odds of retention for those without PTSD and no trauma treatment were nearly three times the odds for those in the reference group (Meshberg-Cohen et al., 2018).

Twelve-step approaches such as Narcotics Anonymous have traditionally not been receptive to participation by individuals receiving medication treatment for OUD. The city of Baltimore encouraged outpatient drug treatment programs that were abstinence-based to incorporate buprenorphine treatment. This study (n=300) drew data from two such programs (an federally qualified health center [FQHC] and a community mental health center) which adopted buprenorphine treatment; encouraged participants to attend a 12-step program, with one offering the program onsite; and included counseling services. Some counselors required 12-step attendance (for 76 percent of participants). Multivariate analyses showed that group counseling attendance was negatively associated with retention at 6 months, while attendance at Narcotics Anonymous was positively associated with retention, although mandated attendance was not. Of the 300 initial participants, 93 percent were available for follow-up at 6 months and there was a 63 percent overall retention at that time (Monico et al., 2015). Another study conducted by the Butler Center for Research with the Hazelden Betty Ford Foundation (Hazelden) found that Hazelden's Comprehensive Opioid Response with the Twelve Steps™ (COR-12), which combines buprenorphine or naltrexone with individual and group counseling using motivational interviewing and CBT and 12-step approaches in residential treatment, reduced drop-out and resulted in longer stays in residential treatment among those in the COR-12 program (Klein, 2017).

Chronic use of opioids can lead to neurocognitive impairment which can impede treatment for OUD. To address this, a RCT (n=120) was conducted among recruits with OUD who were receiving treatment at a court-mandated methadone maintenance residential program in Tehran. The court mandate required participants to remain at the treatment program for at least 2 months. After completion of the 2-month residential program, participants were treated on an outpatient basis. Participants were randomized to treatment as usual (daily methadone and counseling) or to CRT plus treatment as usual. CRT is designed to improve cognitive functioning and is being applied to brain injuries, strokes, and other conditions, including treatment of AUDs. CRT was provided in a group setting for 1 hour, twice a week for 8 weeks. To mirror the contact provided by CRT, the control group attended group painting for equal periods. At 3 months, only biweekly visits were required, including to obtain take-home methadone doses. There was significant improvement in cognitive functioning in the experimental group at 3 months which continued among those assessed at 6 months. Retention in treatment at 3 months, however, was not significantly different between the two study arms (39 percent (CRT) and 47 percent (control), p=0.51). There was a low rate of retention between 3 months and 6 months, with no significant differences between the two groups (Rezapour et al., 2017).

Reimbursement Policy, Effects on Services and Retention in SUD Treatment

Reimbursement policy levers such as incentives or payment withholds for providers are typically implemented using a metric established or adopted for that purpose. The metrics and associated reimbursement policies are designed to encourage changes in provider behavior, with the ultimate intention of influencing outcomes such as retention in treatment. They may, effectively, be regarded as the provider counterpart to the use of CM for patients.

The published literature on effects of reimbursement policy changes on retention in SUD treatment is limited. We address here two large-scale interventions in Delaware and Washington State, and a smaller intervention in Maryland. The Washington State and Maryland studies consider a short time frame and address continuity of care between settings. All address SUD treatment services more generally, beyond the provision of MAT. There are other interventions underway that utilize reimbursement approaches such as bundled payments (e.g., the Vermont hub-and-spoke model (ASTHO, n.d.) and the Medicaid Opioid Health Homes established in Maryland, Rhode Island, and Vermont (CMS, 2015)) where research into outcomes is not complete.

To improve quality of care in SUD treatment, Delaware implemented contract requirements for outpatient SUD treatment facilities, including incentive payments and penalties (reduced base payments) on a monthly basis with other billing. Measures used were: (1) capacity utilization; (2) active participation in treatment; and (3) "program completion." The second measure set a standard number of treatment sessions per patient per week or month, depending on phase of treatment, with additional precautions to avoid measure manipulation. The third measure required active participation for at least 60 days, achievement of treatment plan goals, and at least four consecutive urine drug screens (UDSs) negative for alcohol or illicit drug use. Nearly all providers met the criteria for the third measure, requiring the state to cap those payments. The second measure resulted in an increase in active participation across all four phases of care (1-30, 31-90, 91-180, and 180+ days) but particularly in the last two phases, although each phase was calculated separately based on number at the beginning of that phase rather than on continuity from the prior phase. This meant that actual retention across phases was not being measured. As part of its initiative, Delaware provided technical assistance to providers as part of the initiative, including opportunities for providers to share strategies. The study did not provide extensive detail on program strategies but did note that they included one or more of the following: streamlined admission to improve early engagement, increased hours of operation, new satellite office in underserved areas, physical changes to facilities, sharing bonuses with clinical staff, and clinician training (primarily Motivational Interviewing and CBT) (McLellan et al., 2008).

Effects of a Reimbursement Initiative Coupled with QI Assistance on Outpatient SUD Treatment

A Delaware reimbursement and QI intervention focused on improving treatment session attendance and retention in SUD treatment resulted in increased length of stay in treatment (Stewart et al., 2013).

Delaware subsequently added a QI component to this initiative, with facilities participating in QI through NIATx and Advancing Recovery. Analysis of data from the longer-term Delaware initiative used data from Maryland to create a comparison group and looked at length of stay in treatment, subtracting admission date from discharge date and adding 1 day for those admitted and discharged on the same day. Outliers were omitted, and the researchers were unable to combine records to capture readmissions in Maryland, which did occur for about 2 percent of Delaware admissions. This analysis revealed that length of stay increased in Delaware after the introduction of the contracting component and increased further with the QI intervention. Interviews with program CEOs indicated that the contracting intervention was instrumental and led to increased attention both to strategies to increase treatment length of stay and to the QI interventions designed to assist in that effort (Stewart et al., 2013).

Electronic Reminders and Pay for Performance

These interventions were only effective in improving continuity of treatment between residential or withdrawal management programs and outpatient specialty services in residential programs that were already performing at a moderate or high level in ensuring continuity of care for patients they discharged (Acevedo et al., 2018).

An initiative in Washington State randomized residential and detoxification agencies receiving public funding into one of four trial arms: (1) weekly electronic reminders on recently discharged patients not receiving follow-up (discussed elsewhere); (2) financial awards based on patient continuity of care relative to either a benchmark or improvement; (3) both of the continuity interventions; or (4) no intervention. Continuity was defined as admission to a SUD treatment service within 14 days of discharge from residential or detoxification services. Baseline data indicated that the 50th percentile or achievement threshold for continuity at detox facilities was 29 percent and 40 percent for residential facilities. The benchmark 90th percentile was 37 percent and 56 percent, respectively. To earn up to 10 points, the agency had to exceed the 50th percentile and, to earn the maximum, attain the 90th percentile. Agencies also could earn up to 10 points for improvement on their individual baseline. Agencies were trained on the process and informed twice a quarter of their approximate performance. Incentive payments were based on number of discharges and points earned. The total amount the state could spend was $1.5 million over nine quarters for a possible 33 agencies. The interventions did not produce statistically significant results overall. Adjusted difference-in-difference results, however, revealed that clients at residential agencies already performing at either a moderate or high level at baseline had improved continuity of care, while those at lower-performing agencies did not. We discuss elsewhere some of the factors reported by providers in focus groups that may have interfered with additional improvement (Acevedo et al., 2018).

A separate but similar intervention in Washington State, involving performance-based contracting and reminders for specialty outpatient services, targeted engagement in treatment within 14 days of treatment initiation using the definition of engagement in the National Committee for Quality Assurance Initiation and Engagement of Alcohol and Other Drug Abuse or Dependence Treatment measure. No significant effect was found for any of the interventions, regardless of baseline performance, including when IOP services were examined separately (Garnick et al., 2017). Subsequent analysis of the residential, detoxification, and outpatient data, comparing clients with an SUD only to those with co-occurring psychiatric disorders, found that the interventions did not affect either engagement or continuity of care, other than a positive effect on continuity of care from residential treatment for those with co-occurring disorders (Lee et al., 2018).

Another reimbursement-focused approach to improving continuity of care is known as "contracting with staff incentives" or CSI. This was used to improve transitions between a short-term (21-28 day) medically monitored intensive residential facility and outpatient care in the Baltimore area. A counselor from the selected outpatient facility would meet the patient at the residential facility and, using a Motivational Interviewing approach, provide information about the outpatient program and the benefits of transition, schedule an intake appointment, and have the patient sign a continuing care "contract." The outpatient program received up to $100 for each such patient referred, with $25 paid for patient enrollment and for up to three sessions in the first 30 days (n=49). A separate study arm provided a client incentive for intake and attendance (n=97) and both were compared to a usual care group (n=114) that was simply referred to outpatient care. The CSI group had significantly higher rates of admission to outpatient care. However, for both the usual care group and other groups, outpatient intake was highest when it was at the clinic onsite with the residential facility (Aquavita et al., 2013).

Effects of Treatment Retention on Outcomes

There is historic evidence that retention in MAT treatment for OUD results in better outcomes, including medical morbidity, social functioning, rates of HIV transmission, and criminal activity (see, e.g., studies referenced in Manhapra et al., 2018). Recent research provides us with additional evidence of positive effects on mortality, reduced substance use, and service utilization. It is important to note that these studies are so different in terms of time frames studied and the nature of the medications and other treatments provided, that it is impossible to say definitively that a specific retention period is required for a given outcome.

A systematic review and meta-analysis by Ma et al. (2018) found that crude mortality rates, all-cause mortality, and overdose deaths were significantly lower for those retained in treatment for longer than a year, with 30 studies examined ranging 2-20 years in duration. During medication treatment, all-cause rates were lowest for those treated with naltrexone, followed by buprenorphine and methadone. All-cause mortality rates shortly after discharge from medication treatment are high but were lowest for those who had been taking buprenorphine. Ma et al. noted that many studies showed transitioning in and out of treatment, resulting in repeated exposure to times with high mortality risk. A recent study of retention in methadone maintenance treatment in an area of Rhode Island with high fentanyl exposure also found that those who remained in treatment during a 6-month follow-up period experienced no mortality, in contrast to those who were not retained (Stone et al., 2018).

Effects of Treatment Retention on Patient Outcomes

Recent research reveals that longer duration of MAT is associated with positive effects on mortality, substance use, and service utilization (e.g., Bhatraju et al., 2017; Lo-Ciganic et al., 2016; Ma et al., 2018; Monico et al., 2015; Shcherbakova et al., 2018; Stone et al., 2018).

UDSs are commonly used to assess substance use during treatment. Recent studies confirm that treatment retention is associated with abstinence although most studies do not compare those retained to follow-up with those not retained. One exception was a study by Stone et al. (2018), who found that 71 percent of all patients and 89 percent of those retained for 6 months in methadone treatment achieved abstinence from opioids within 6 months. Other studies look at changes over time within the patient population in treatment. A study at a primary care-based buprenorphine treatment program within a hospital found that positive drug screens for opioids were reduced over time, from 60 percent for those in treatment less than 12 weeks to 27 percent for those in treatment more than 52 weeks (Bhatraju et al., 2017). A study of buprenorphine treatment via telemedicine found that 86.1 percent of patients engaged in treatment at 3 months were opiate negative, an increase from 78.6 percent at 1 week (Weintraub, 2018). A multivariate logistic regression predicting abstinence from heroin or cocaine after 6 months of buprenorphine treatment integrated with 12-step Narcotics Anonymous group participation showed that those retained in treatment for 6 months had nearly seven times the odds of abstinence (p<0.001) (Monico et al., 2015). A systematic review of the literature on treatment with XR-NTX by Jarvis et al. (2018) also showed a limited association between extended retention in treatment and abstinence.

Retention in treatment has also been identified as a factor reducing inpatient admissions and emergency department use (Lo-Ciganic et al., 2016; Shcherbakova et al., 2018). One study of persistence in buprenorphine treatment revealed that all-cause inpatient admissions were 70 percent less likely among those retained in buprenorphine treatment for 1 year (Shcherbakova et al., 2018). Another analyzed Medicaid data in Pennsylvania to determine effects of different durations of buprenorphine treatment on all-cause inpatient admissions and emergency department use. Six trajectories for discontinuation were identified: 24.9 percent discontinued in less than 3 months; 18.7 percent discontinued between 3 months and 5 months; 12.4 percent discontinued between 5 and 8 months; 13.3 percent discontinued after 8 months, 9.5 percent refilled intermittently and 21.2 percent refilled persistently for 12 months. Persistent refills trajectories were associated with an 18 percent lower risk of hospitalizations and 14 percent lower risk of emergency department visits in the following year, compared with those discontinuing between 3 and 5 months (Lo-Ciganic et al., 2016).

 

GAPS IN THE LITERATURE

This review focused primarily on literature from the years 2014-2018 related to OUD and treatment involving MAT for OUD. There are subjects related to treatment retention in OUD MAT that are not adequately studied in the current literature. This includes retention and outcomes over periods longer than 2-5 injections of XR-NTX. There also is need for evaluation of certain models in recent use, such as the Medicaid Opioid Health Homes; treatment approaches that promote early use of MAT after initial diagnosis; and treatment approaches that promote retention for postpartum women. We also need a better understanding of the effects on retention and outcomes of providing recovery supports such as social support, housing support, employment support, childcare, or transportation, which AHRQ has identified as critical for treatment retention in MAT in rural areas (AHRQ, 2017). We believe these supports are important for improving retention and outcomes in both rural and urban areas. We also need more research on the effects of long-term retention on functioning and quality of life. Given the apparent frequency with which individuals with OUD and other SUDs cycle into and out of care, as well as the extreme risk of overdose at the immediate end of an episode of treatment for OUD, much more also needs to be known about the occurrence of multiple episodes of treatment, retention within and across episodes, and outcomes from recurring treatment.

 

REFERENCES

Acevedo A, Lee MT, Garnick DW, Horgan CM, Ritter GA, Panas L, Campbell K, Bean-Mortinson J. Agency-level financial incentives and electronic reminders to improve continuity of care after discharge from residential treatment and detoxification. Drug & Alcohol Dependence. 2018; 183: 192-200.

Acquavita SP, Stershic S, Sharma R, Stitzer M. Client incentives versus contracting and staff incentives: How care continuity interventions in substance abuse treatment can improve residential to outpatient transition. Journal of Substance Abuse Treatment. 2013; 45(1): 55-62.

Agency for Healthcare Research and Quality (AHRQ). Implementing Medication-Assisted Treatment for Opioid Use Disorder in Rural Primary Care: Environmental Scan. Vol. 1, Rep.No. 17(18)-0050-EF. 2017. https://integrationacademy.ahrq.gov/sites/default/files/mat_for_oud_environmental_scan_volume_1_1.pdf. Accessed December 3, 2018.

Agency for Healthcare Research and Quality (AHRQ). Medication-Assisted Treatment Models of Care for Opioid Use Disorder in Primary Care Settings. 2016. https://effectivehealthcare.ahrq.gov/sites/default/files/pdf/opioid-use-disorder_technical-brief.pdf. Accessed December 3, 2018.

American Society of Addiction Medicine (ASAM). The ASAM Performance Measures for the Addiction Specialist Physician. 2014. https://www.asam.org/docs/default-source/advocacy/performance-measures-for-the-addiction-specialist-physician.pdf?sfvrsn=5f986dc2_0. Accessed December 3, 2018.

American Society of Addiction Medicine (ASAM). The ASAM National Practice Guideline for the Use of Medications in the Treatment of Addiction Involving Opioid Use. 2015. https://www.asam.org/docs/default-source/practice-support/guidelines-and-consensus-docs/asam-national-practice-guideline-supplement.pdf?sfvrsn=24. Accessed December 3, 2018.

Association of State and Territorial Health Officials. Medication Assisted Treatment Program for Opioid Addiction. 2015. http://www.astho.org/Health-Systems-Transformation/Medicaid-and-Public-Health-Partnerships/Case-Studies/Vermont-MAT-Program-for-Opioid-Addiction/. Accessed December 3, 2018.

Bhatraju EP, Grossman E, Tofighi B, McNeely J, DiRocco D, Flannery M, Garment A, Goldfeld K, Gourevitch MN, Lee JD. Public sector low threshold office-based buprenorphine treatment: Outcomes at year 7. Addiction Science & Clinical Practice. 2017; 12(1): 7.

Braitman AL, Kelley ML. Initiation and retention in couples outpatient treatment for parents with drug and alcohol use disorders. Experimental & Clinical Psychopharmacology. 2016; 24(3): 174-184.

British Columbia Centre on Substance Use, British Columbia Ministry of Health. A Guideline for the Cinical Management of Opioid Use Disorder. 2017. http://www.bccsu.ca/wp-content/uploads/2017/06/BC-OUD-Guidelines_June2017.pdf. Accessed December 3, 2018.

Carroll KM, Weiss RD. The role of behavioral interventions in buprenorphine maintenance treatment: A review. American Journal of Psychiatry. 2017; 174(8): 738-747.

Choi S, Adams SM, Morse SA, MacMaster S. Gender differences in treatment retention among individuals with co-occurring substance abuse and mental health disorders. Substance Use & Misuse. 2015; 50(5): 653-663.

Center for Disease Control and Prevention. Opioid Overdose Deaths. 2018. https://www.cdc.gov/drugoverdose/data/statedeaths.html. Accessed January 4, 2019.

Centers for Medicare & Medicaid Services (CMS). Designing Medicaid Health Homes for Individuals With Opioid Dependency: Considerations for States. 2015. https://www.chcs.org/media/HH-IRC-Health-Homes-for-Opioid-Dependency.pdf. Accessed December 3, 2018.

Cui R, Haller M, Skidmore JR, Goldsteinholm K, Norman S, Tate SR. Treatment attendance among veterans with depression, substance use disorder, and trauma. Journal of Dual Diagnosis. 2016; 12(1): 15-26.

DiPaula BA, Menachery E. Physician-pharmacist collaborative care model for buprenorphine-maintained opioid-dependent patients. Journal of the American Pharmacists Association. 2015; 55(2): 187-192.

D'Onofrio G, Chawarski MC, O'Connor PG, Pantalon MV, Busch SH, Owens PH, Hawk K, Bernstein SL, Fiellin DA. Emergency department-initiated buprenorphine for opioid dependence with continuation in primary care: outcomes during and after intervention. Journal of General Internal Medicine. 2017; 32(6): 660-666.

Dugosh K, Abraham A, Seymour B, McLoyd K, Chalk M, Festinger D. A systematic review on the use of psychosocial interventions in conjunction with medications for the treatment of opioid addiction. Journal of Addiction Medicine. 2016; 10(2): 93-103.

Dutra L, Stathopoulou G, Basden SL, Leyro TM, Powers MB, Otto MW. A meta-analytic review of psychosocial interventions for substance use disorders. American Journal of Psychiatry. 2008; 165 2: 179-187.

Eibl JK, Gauthier G, Pellegrini D, Daiter J, Varenbut M, Hogenbirk JC, Marsh DC. The effectiveness of telemedicine-delivered opioid agonist therapy in a supervised clinical setting. Drug & Alcohol Dependence. 2017; 176: 133-138.

Fiellin DA, Weiss L, Botsko M, Egan JE, Altice FL, Bazerman LB, Chaudhry A, Cunningham CO, Gourevitch MN, Lum PJ, Sullivan LE, Schottenfeld RS, O'Connor PG, BHIVES Collaborative. Drug treatment outcomes among HIV-infected opioid-dependent patients receiving buprenorphine/naloxone. Journal of Acquired Immune Deficiency Syndromes. 2011; 56(suppl 1): S33-S38.

Franklyn AM, Eibl JK, Gauthier GJ, Marsh DC. The impact of cannabis use on patients enrolled in opioid agonist therapy in Ontario, Canada.PLoS One. 2017; 12(11): e0187633.

Fullerton CA, Kim M, Thomas CP, Lyman DR, Montejano LB, Dougherty RH, Daniels AS, Ghose SS, Delphin-Rittmon ME. Medication-assisted treatment with methadone: assessing the evidence. Psychiatric Services. 2014; 65(2): 146-157.

Gauthier G, Eibl JK, Marsh DC. Improved treatment-retention for patients receiving methadone dosing within the clinic providing physician and other health services (onsite) versus dosing at community (offsite) pharmacies.Drug & Alcohol Dependence. 2018; 191: 1-5.

Hadland SE, Bagley SM, Rodean J, Silverstein M, Levy S, Larochelle MR, Samet JH, Zima BT. Receipt of timely addiction treatment and association of early medication treatment with retention in care among youths with opioid use disorder. JAMA Pediatrics. 2018; 172(11): 1029-1037.

IBM Watson Health, unpublished analysis of IBM MarketScan Medicaid and private insurance claims data, 2017.

Institute for Clinical and Economic Review (ICER). Management of Patients With Opioid Dependence: A Review of Clinical, Delivery System, and Policy Options. Draft report. 2014. https://icer-review.org/wp-content/uploads/2016/01/CEPAC-Opioid-Dependence-Draft-Report_June-9-2014-FINAL_With-Corrections.pdf. Accessed December 3, 2018.

Jarvis BP, Holtyn AF, Subramaniam S, Tompkins DA, Oga EA, Bigelow GE, Silverman K. Extended-release injectable naltrexone for opioid use disorder: A systematic review. Addiction. 2018; 113(7): 1188-1209.

Joe GW, Simpson DD, Broome KM. Effects of readiness for drug abuse treatment on client retention and assessment of process. Addiction. 1998; 93(8): 1177-1190.

Klein A. What Does It Really Mean To Be Providing Medication-Assisted Treatment for Opioid Addiction? Hazelden Betty Ford Foundation; 2017. https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=2ahUKEwi7zsftwtHeAhVJh-AKHYBQD4UQFjAAegQIBxAC&url=https://www.hazeldenbettyford.org/~/media/files/bcrwhitepapers/matopioidaddicitonwp1710.pdf?la=en&usg=AOvVaw2FiZKskeVmeYSkYfGsMXxj. Accessed December 3, 2018.

Kumar N, Stowe ZN, Han X, Mancino MJ. Impact of early childhood trauma on retention and phase advancement in an outpatient buprenorphine treatment program. American Journal on Addictions. 2016; 25(7): 542-548.

LaBelle CT, Han SC, Bergeron A, Samet JH. Office-based opioid treatment with buprenorphine (OBOT-B): Statewide implementation of the Massachusetts Collaborative Care Model in community health centers. Journal of Substance Abuse Treatment. 2016; 60: 6-13.

Lagisetty P, Klasa K, Bush C, Heisler M, Chopra V, Bohnert A. Primary care models for treating opioid use disorders: What actually works? A systematic review. PLoS One. 2017; 12(10): e0186315.

Lee JD, Nunes EV Jr, Novo P, Bachrach K, Bailey GL, Bhatt S, Frakas S, Fishman M, Gauthier P, Hodgkins CC, King J, Lindblad R, Liu D, Matthews AG, May J, Peavy KM, Ross S, Salazar D, Schkolnik P, Shmueli-Blumberg D, Stablein D, Subramaniam G, Rotrosen J. Comparative effectiveness of extended-release naltrexone versus buprenorphine-naloxone for opioid relapse prevention (X:BOT): A multicentre, open-label, randomised controlled trial. Lancet. 2018; 27; 391: 309-318.

Lee MT, Acevedo A, Garnick DW, Horgan CM, Panas L, Ritter GA, Campbell KM. Impact of agency receipt of incentives and reminders on engagement and continuity of care for clients with co-occurring disorders. Psychiatric Services. 2018; 69(7): 804-811.

Liebschutz JM, Crooks D, Herman D, Anderson B, Tsui J, Meshesha LZ, Dossabhov S, Stein M. Buprenorphine treatment for hospitalized, opioid-dependent patients: A randomized clinical trial. JAMA Internal Medicine. 2014; 174(8): 1369-1376.

Lincoln T, Johnson BD, McCarthy P, Alexander E. Extended-release naltrexone for opioid use disorder started during or following incarceration. Journal of Substance Abuse Treatment. 2018; 85: 97-100.

Lo-Ciganic WH, Gellad WF, Gordon AJ, Cochran G, Zemaitis MA, Cathers T, Donohue JM. Association between trajectories of buprenorphine treatment and emergency department and in-patient utilization. Addiction. 2016; 111(5): 892-902.

Ma J, Bao YP, Wang RJ, Su MF, Liu MX, Li JQ, Degenhardt L, Farrell M, Blow FC, Ilgen M, Shi J, Lu L. Effects of medication-assisted treatment on mortality among opioids users: A systematic review and meta-analysis. Molecular Psychiatry. 2018; doi.org/10.1038/s41380-018-0094-5.

Madden LM, Farnum SO, Eggert KF, Quanbeck AR, Freeman RM, Ball SA, Schottenfeld RS, Shi JM, Savage ME, Barry DT. An investigation of an open-access model for scaling up methadone maintenance treatment. Addiction. 2018; 113(8): 1450-1458.

Manhapra A, Agbese E, Leslie DL, Rosenheck RA. Three-year retention in buprenorphine treatment for opioid use disorder among privately insured adults. Psychiatric Services. 2018; 69(7): 768-776.

Manhapra A, Petrakis I, Rosenheck R. Three-year retention in buprenorphine treatment for opioid use disorder nationally in the Veterans Health Administration. American Journal on Addictions. 2017; 26(6): 572-580.

Maremmani AGi, Pani PP, Trogu E, Vigna-Taglianti F, Mathis F, Diecidue R, Kirchmayer U, Amato L, Ghibaudi J, Camposeragna A, Saponaro A, Davoli M, Faggiano F, Maremmani I. The impact of psychopathological subtypes on retention rate of patients with substance use disorder entering residential therapeutic community treatment. Annals of General Psychiatry. 2016; 15: 29.

Mattick R, Breen C, Kimber J, Davoli M. Methadone maintenance therapy versus no opioid replacement therapy for opioid dependence. Cochrane Database of Systematic Reviews. 2009; (3): CD002209.

McLellan AT, Kemp J, Brooks A, Carise D. Improving public addiction treatment through performance contracting: the Delaware experiment. Health Policy. 2008; 87(3): 296-308.

Meier PS, Barrowclough C, Donmall MC. The role of the therapeutic alliance in the treatment of substance misuse: A critical review of the literature. Addiction. 2005; 100(3): 304-316.

Meshberg-Cohen S, Black AC, DeViva JC, Petrakis IL, Rosen MI. Trauma treatment for veterans in buprenorphine maintenance treatment for opioid use disorder. Addictive Behaviors. 2018; 89: 29-34.

Monico LB, Gryczynski J, Mitchell SG, Schwartz RP, O'Grady KE, Jaffe JH. Buprenorphine treatment and 12-step meeting attendance: conflicts, compatibilities, and patient outcomes. Journal of Substance Abuse Treatment. 2015; 57: 89-95.

National Institute on Drug Abuse. Overdose Death Rates. 2018. https://www.drugabuse.gov/related-topics/trends-statistics/overdose-death-rates. Accessed December 3, 2018.

National Quality Forum (NQF). Measures, Reports and Tools. Continuity of Pharmacotherapy for Opioid Use Disorder (Measure 3175). https://www.qualityforum.org/Measures_Reports_Tools.aspx. Accessed December 3, 2018.

NIATx. Getting Started with Medication-Assisted Treatment. 2010. https://niatx.net/PDF/NIATx-MAT-Toolkit.pdf. Accessed December 3, 2018.

Rezapour T, Hatami J, Farhoudian A, Sofuoglu M, Noroozi A, Daneshmand R, Samiei A, Ekhtiari H. Cognitive rehabilitation for individuals with opioid use disorder: A randomized controlled trial. Neuropsychological Rehabilitation. 2017; 29(8): 1-17.

Riggins DP, Cunningham CO, Ning Y, Fox AD. Recent incarceration and buprenorphine maintenance treatment outcomes among human immunodeficiency virus-positive patients. Substance Abuse. 2017; 38(3): 297-302.

Saloner B, Daubresse M, Caleb Alexander G. Patterns of buprenorphine-naloxone treatment for opioid use disorder in a multistate population. Medical Care. 2017; 55(7): 669-676.

Samples H, Williams AR, Olfson M, Crystal S. Risk factors for discontinuation of buprenorphine treatment for opioid use disorders in a multi-state sample of Medicaid enrollees. Journal of Substance Abuse Treatment. 2018; 95: 9-17.

Schuman-Olivier Z, Weiss RD, Hoeppner BB, Borodovsky J, Albanese MJ. Emerging adult age status predicts poor buprenorphine treatment retention. Journal of Substance Abuse Treatment. 2014; 47(3): 202-212.

Seth P, Scholl L, Rudd RA, Bacon S. Overdose deaths involving opioids, cocaine, and psychostimulants--United States, 2015-2016. Morbidity & Mortality Weekly Report. 2018; 67(12): 349-358.

Shcherbakova N, Tereso G, Spain J, Roose RJ. Treatment persistence among insured patients newly starting buprenorphine/naloxone for opioid use disorder. Annals of Pharmacotherapy. 2018; 52(5): 405-414.

Simpson DD, Joe GW, Rowan-Szal GA. Drug abuse treatment retention and process effects on follow-up outcomes. Drug & Alcohol Dependence. 1997; 47(3): 227-235.

Socias ME, Ahamad K, Le Foll B, Lim R, Bruneau J, Fischer B, Wild TC, Wood E, Jutras-Aswad D. The OPTIMA study, buprenorphine/naloxone and methadone models of care for the treatment of prescription opioid use disorder: Study design and rationale. Contemporary Clinical Trials. 2018; 69: 21-27.

Springer SA, Brown SE, Di Paola A, Altice FL. Correlates of retention on extended-release naltrexone among persons living with HIV infection transitioning to the community from the criminal justice system. Drug & Alcohol Dependence. 2015; 157: 158-65.

Stewart MT, Horgan CM, Garnick DW, Ritter G, McLellan AT. Performance contracting and quality improvement in outpatient treatment: Effects on waiting time and length of stay. Journal of Substance Abuse Treatment. 2013; 44(1): 27-33.

Stone AC, Carroll JJ, Rich JD, Green TC. Methadone maintenance treatment among patients exposed to illicit fentanyl in Rhode Island: Safety, dose, retention, and relapse at 6 months. Drug & Alcohol Dependence. 2018; 192: 94-97.

Substance Abuse and Mental Health Services Administration (SAMHSA), Center for Behavioral Health Statistics and Quality. Key Substance Use and Mental Health Indicators in the United States: Results From the 2017 National Survey on Drug Use and Health. HHS Publication No. SMA 18-5068, NSDUH Series H-53. 2018. https://www.samhsa.gov/data/sites/default/files/cbhsq-reports/NSDUHFFR2017/NSDUHFFR2017.pdf. Accessed December 4, 2018.

Thomas CP, Fullerton CA, Kim M, Montejano L, Lyman DR, Dougherty RH, Daniels AS, Ghose SS, Delphin-Rittmon ME. Medication-assisted treatment with buprenorphine: assessing the evidence. Psychiatric Services. 2014; 65: 158-170.

Timko C, Schultz NR, Cucciare MA, Vittorio L, Garrison-Diehn C. Retention in medication-assisted treatment for opiate dependence: A systematic review. Journal of Addictive Diseases. 2016; 35(1): 22-35.

U.S. Department of Health and Human Services (HHS). HHS Acting Secretary Declares Public Health Emergency to Address National Opioid Crisis. 2017. https://www.hhs.gov/about/news/2017/10/26/hhs-acting-secretary-declares-public-health-emergency-address-national-opioid-crisis.html. Accessed December 4, 2018.

Vo HT, Burgower R, Rozenberg I, Fishman M. Home-based delivery of XR-NTX in youth with opioid addiction. Journal of Substance Abuse Treatment. 2018; 85: 84-89.

Vogel M, Dürsteler KM, Walter M, Herdener M, Nordt C. Rethinking retention in treatment of opioid dependence--the eye of the beholder. International Journal of Drug Policy. 2017; 39: 109-113.

Weinstein ZM, Kim HW, Cheng DM, Quinn E, Hui D, Labelle CT, Drainoni ML, Bachman SS, Samet JH. Long-term retention in office based opioid treatment with buprenorphine. Journal of Substance Abuse Treatment. 2017; 74: 65-70.

Weintraub E, Greenblatt AD, Chang J, Himelhoch S, Welsh C. Expanding access to buprenorphine treatment in rural areas with the use of telemedicine. American Journal on Addictions. 2018; 27(8): 612-617.

Weiss L, Egan JE, Botsko M, Netherland J, Fiellin DA, Finkelstein R. The BHIVES collaborative: Organization and evaluation of a multisite demonstration of integrated buprenorphine/naloxone and HIV treatment.Journal of Acquired Immune Deficiency Syndromes. 2011; 56(suppl 1): S7-S13.

Weiss RD, Potter JS, Griffin ML, Provost SE, Fitzmaurice GM, McDermott KA, Srisarajivakul EN, Dodd DR, Dreifuss JA, McHugh RK, Carroll KM. Long-term outcomes from the National Drug Abuse Treatment Clinical Trials Network Prescription Opioid Addiction Treatment Study. Drug & Alcohol Dependence. 2015; 150: 112-119.

Wilder C, Lewis D, Winhusen T. Medication assisted treatment discontinuation in pregnant and postpartum women with opioid use disorder. Drug & Alcohol Dependence. 2015; 149: 225-231.

 

APPENDIX A. RETENTION RATES FROM MAT-FOCUSED STUDIES IN THIS REVIEW

Reference Retention Results
Bhatraju, 2017 Treatment with buprenorphine: median retention 38 weeks (range 0-320 weeks) among induction patients (n=302); 110 weeks (0-353 weeks) among transfers (n=175), and 57 weeks for all patients (n=477). Drop-out: week 5, 25%; week 38, 50%; week 144, 75%.
DiPaula, 2015 Treatment with buprenorphine: mean duration 20 weeks (2-52 weeks), 100% retained for 6 months and 73% for 12 months
D'Onofrio, 2017 Buprenorphine induction in the ED accompanied by brief intervention, take-home daily doses pending admission to the hospital's primary care center within 72 hours, where they received treatment for 10 weeks and referral to follow-on care: 74% at 2 months
Eibl, 2017 Treatment with buprenorphine or methadone: predominantly telemedicine retained for a median of 366 days and 50% retained for 1 year; predominantly in-person median of 207 days with 39% retained for 1 year; mixed care retained for a median of 317 days with 47% retained for a year
Fiellin, 2011 Treatment with buprenorphine in HIV clinics: 74%, 67%, 59% and 49% retained at 3, 6, 9, and 12 months, respectively
Franklyn, 2017 Treatment with buprenorphine or methadone: retained for 365 days, baseline cocaine users 39% and non-users 46%
Gauthier, 2018 Treatment with methadone: onsite pharmacy 1-year retention of 57.3% compared to 11.9% retention in offsite pharmacy group
Hadland, 2018 Treatment of young adults with buprenorphine, methadone, or naltrexone within 3 months of OUD diagnosis: median retention with buprenorphine, 123 days; naltrexone, 150 days; and methadone, 324 days, compared with 67 days for only behavioral health services
Kumar, 2016 Treatment with buprenorphine: retention 82% at 90 days
LaBelle, 2016 Community health center OBOT-B, patients remaining in treatment for longer than 12 months during 2010, 2011, and 2012 were 32%, 56%, and 65%, respectively
Lincoln, 2018 Pre-correctional release induction of XR-NTX vs. induction post-release. Retention rates: week 4, 55% vs. 25%; week 8, 36% vs. 25%; week 24, 21% vs.15%
Lo-Ciganic, 2016 Treatment with buprenorphine: 24.9% discontinued in less than 3 months; 18.7% discontinued between 3 and 5 months; 12.4% discontinued between 5 and 8 months; 13.3% discontinued after 8 months
Madden, 2018 Treatment with methadone: 90-day retention at baseline 89.3%; in 9 years after, mean retention ranged from 81.9 to 91.8%
Manhapra, 2017 (VA) Treatment with buprenorphine: mean duration=1.68 years, with 61.60% >1 year, 31.83% >than 3 years
Manhapra, 2017 (comm) Treatment with buprenorphine: 85% for 31-365 days, 45% for 1-3 years, 13.7% >3 years
Meshberg-Cohen, 2019 Treatment with buprenorphine: veterans with PTSD who received trauma treatment had 6-month retention of 90.5%; 46.6% of those without trauma treatment; 23.9% of those with PTSD but no trauma treatment
Monico, 2015 Treatment with buprenorphine: 6-month retention 63%
Riggins, 2017 Treatment with buprenorphine: 6-month retention 66%; 12-month retention if recently incarcerated at baseline 39% vs. 50% if not
Saloner, 2017 Treatment with buprenorphine: mean duration 266 days, median duration 118 days, 41% for 6 months or longer
Samples, 2018 Treatment with buprenorphine: 89.6% for 1 week, 71.6% for 1 month, 35.4% for 180 days. In analysis using Medicaid claims data, Samples et al. (2018) examined factors associated with discontinuation of buprenorphine treatment.
Schuman-Olivier, 2014 Treatment with buprenorphine: Emerging adults: month 3, 57%, month 6, 38%, month 9, 21%, month 12, 17%; other adults: month 3, 78%, month 6, 62%, month 9, 53%, month 12, 45%
Shcherbakova, 2018 Treatment with buprenorphine: mean duration of first treatment episode greater than 30 days = 206.4 days, 82.4% for 30 days, 53.6% for 6 months, 40.4 for 1 year
Socias, 2018 Treatment with buprenorphine and (primarily) methadone: 6 months, 52.6%; 12 months, 38.5%; 18 months, 31.5%
Stone, 2018 Treatment with methadone: 68% for 6 months
Vo, 2018 Treatment with XR-NTX with young adults: Over 16-weeks, in-home visits 50% received all 5 doses, clinic visits 9% did
Weinstein, 2017 Treatment with buprenorphine: 45% of treatment periods were a year or longer; 53.7% of patients had at least 1 such treatment period
Weintraub, 2018 Treatment with buprenorphine: 57.4% for 3 months
Wilder, 2015 Treatment with methadone among pregnant and postpartum women: prenatal retention rate of 89%, postpartum at 6 months retention rate of 38.1%