NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
National Academies of Sciences, Engineering, and Medicine; Health and Medicine Division; Board on Health Sciences Policy; Committee on Medication-Assisted Treatment for Opioid Use Disorder; Mancher M, Leshner AI, editors. Medications for Opioid Use Disorder Save Lives. Washington (DC): National Academies Press (US); 2019 Mar 30.
National Academies of Sciences, Engineering, and Medicine; Health and Medicine Division; Board on Health Sciences Policy; Committee on Medication-Assisted Treatment for Opioid Use Disorder; Mancher M, Leshner AI, editors.
Washington (DC): National Academies Press (US); 2019 Mar 30.U.S. Food and Drug Administration–approved medications to treat opioid use disorder are effective and save lives. Long-term retention on medication for OUD is associated with improved outcomes. A lack of availability of behavioral interventions is not a sufficient justification to withhold medications to treat OUD.
Methadone, buprenorphine, and extended-release naltrexone are the three medications currently approved by the U.S. Food and Drug Administration (FDA) for treating opioid use disorder (OUD). Box 2-1 provides a full list of them. All three medications reduce opioid cravings and help to sever the ties between opioid use and established situational or emotional triggers. These medications work by targeting the mu-opioid receptor within the endogenous opioid system, although each has a distinct mechanism of action. Their safety and efficacy profiles differ due to their differing pharmacological, pharmacodynamic, and pharmacokinetic properties (Connery, 2015; Kleber, 2007). This chapter examines the evidence base for the effectiveness of these three medications as well as identifying gaps in knowledge and future research needs. The chapter also explores the use of behavioral interventions in conjunction with medications to treat OUD.
U.S. Food and Drug Administration–Approved Medications for the Treatment of Opioid Use Disorder.
Methadone is a synthetic, long-lasting opioid agonist (Kreek, 2000). Methadone fully activates the mu-opioid receptors in the brain through the same mechanism of action as prescription or illicit opioids. In persons with OUD, methadone occupies those mu-opioid receptors and has the effect of lessening the painful “lows” of opioid withdrawal, and, at therapeutic doses, it attenuates the euphoric “highs” of shorter-acting opioids such as heroin, codeine, and oxycodone. Because it is an agonist treatment and individuals do not have to go through opioid withdrawal before initiating it, methadone can be started at any time during OUD treatment. However, it does require days to weeks to achieve a therapeutic dose, which needs to be individualized to decrease cravings and prevent return to other opioid use (NIH, 1998).
By law in the United States, outpatient methadone treatment can only be administered to people enrolled in state- and federally certified opioid treatment programs (OTPs), historically called methadone clinics (CRS, 2018). Methadone can also be provided when patients are admitted to a hospital for treatment of other conditions or in emergencies. 2 Most patients are required to visit an OTP in person every day to receive their daily dose. Eventually, stable patients may receive take-home doses if they meet certain criteria, such as having had a stable period of good functioning without illicit drug use. In addition, patients on methadone are required to attend regular counseling sessions with clinic providers.
As an agonist, methadone sustains the opioid tolerance and physical dependence of the patient, so missing doses can cause opioid withdrawal. The major risk to patients on methadone—opioid overdose death—is elevated within the first 2 weeks of methadone treatment (Degenhardt et al., 2009), after which the risk of overdose death is significantly lower than for people with OUD who are not in treatment (Degenhardt et al., 2011; Sordo et al., 2017). The risk of overdose is higher among patients who are also taking other sedatives, but FDA has advised that “methadone should not be withheld from patients taking benzodiazepines or other drugs that depress the central nervous system” because overdose risk is even higher for people who are not on medication for OUD (FDA, 2017). This is also true of buprenorphine (see below). The other potential harms of methadone include hypogonadism (low testosterone), which is a common side effect of chronic use of any opioid (Bawor et al., 2015), and an increase in the electrocardiographic corrected QT interval, although the clinical significance of the latter is unclear (Bart et al., 2017). No special training is required for physicians working within an OTP to prescribe methadone.
Buprenorphine is a high-affinity partial opioid agonist as well as an antagonist of the kappa-opioid receptor and an agonist of the opioid like-1 receptor (Kleber, 2007). As a partial agonist, buprenorphine does not fully substitute for other opioids on the mu receptor (e.g., heroin, codeine, and oxycodone). Like methadone, buprenorphine can bring relief to a patient in opioid withdrawal. Through its partial agonist effect, it can also reduce the rewarding effect if the patient uses opioids while taking buprenorphine. Because it is a partial agonist, buprenorphine also has less of an effect on respiratory depression, so it has a lower risk of overdose than methadone and other opioids (Dahan et al., 2006), and a therapeutic dose may be achieved within a few days (Connery, 2015).
The most widely available forms of buprenorphine in the United States are tablets or films that are absorbed under the tongue (see Box 2-1). In these formulations, buprenorphine is combined with the opioid antagonist naloxone to discourage injection, because naloxone is not well absorbed sublingually but will rapidly reduce the rewarding effect if the product is injected. Buprenorphine is also available in implantable and extended-release subcutaneous formulations, which are more difficult to divert 3 and theoretically increase adherence to treatment.
In the United States, buprenorphine can also be provided at an OTP, but it is most commonly prescribed in an office-based setting (e.g., a primary care clinic) to patients who fill the prescription at regular pharmacies. Patients can then administer buprenorphine sublingually to themselves, as with most other medications for chronic disease. Patients are often seen by providers frequently at first, but as the treatment progresses patients who do not use other opioids are usually able to reduce the frequency of the required office visits (Fiellin et al., 2006). In order to treat OUD with buprenorphine, prescribers in the United States must undergo additional training and obtain a waiver from the Drug Enforcement Administration. Only a limited number of providers pursue these waivers. In fact, until recently only 2 to 3 percent of physicians in the United States were waivered to provide buprenorphine, most of whom are based in urban areas (Rosenblatt et al., 2015). Many physicians who are waivered do not prescribe to their maximum patient limit (Jones et al., 2015). In 2016 nurse practitioners and physician assistants became eligible to apply for training to obtain waivers. 4 Chapter 5 includes a more detailed discussion on this issue.
As with methadone, buprenorphine sustains opioid tolerance and physical dependence in patients, so discontinuation can lead to withdrawal—although buprenorphine's withdrawal syndrome may be less severe. The most prominent risk of buprenorphine to patients with OUD is precipitation of non-life-threatening opioid withdrawal at first dose. The risk of opioid overdose death declines immediately when patients with OUD initiate buprenorphine treatment (Sordo et al., 2017). Hypogonadotropic effects are less with buprenorphine than with methadone and buprenorphine is not associated with QTc prolongation or cardiac arrhythmias (Fareed et al., 2013).
It is important to note that since methadone and buprenorphine are opioids, they can be misused. As with other opioids, buprenorphine and methadone can result in physical dependence and a diagnosable OUD, which demands that these medications be safely stored and not be taken by anyone other than the individual for whom they are prescribed.
Naltrexone is not an opioid but rather is a full antagonist of the mu-opioid receptor and completely blocks the euphoric and analgesic effects of all opioids (Kleber, 2007). Naltrexone does not cause physical dependence, nor does it produce any of the rewarding effects of opioids. It is not uncommon for patients to try to use opioids while on extended-release naltrexone, but it is exceedingly rare that using an opioid can override the effect of naltrexone to the extent that the opioid yields rewarding effects. Ideally, patients on extended-release naltrexone learn quickly not to use the opioids that caused their addictive behaviors, and, after sustained use of the medication, their cravings decline (Krupitsky et al., 2011; Lee et al., 2018; Tanum et al., 2017).
Treatment initiation with extended-release naltrexone is complicated by its mechanism and long duration of action. Because naltrexone can trigger severe withdrawal symptoms, naltrexone treatment initiation typically requires medically supervised withdrawal followed by at least 4 to 7 days without any opioids, including opioids used in medication-based treatment like methadone and buprenorphine (Sullivan et al., 2017). This remains a key barrier to naltrexone use, although shorter outpatient initiation protocols have shown some promise (Sullivan et al., 2017). Risk of overdose for patients being treated with extended-release naltrexone may be reduced compared to treatment with a placebo, non-medication-based treatments, and treatment with oral naltrexone (Kelty and Hulse, 2017; Lee et al., 2016). Emerging evidence suggests that patients can experience an increased risk of overdose when they approach the end of the 28-day period of the extended-release formulation (Binswanger and Glanz, 2018).
Naltrexone is currently available both in a once-daily oral formulation and in a once-monthly, extended-release depot injection. The oral formulation was found to be no better than a placebo in retaining patients in treatment or eliminating their opioid use (Minozzi et al., 2011) and patients treated with oral naltrexone have an increased risk of overdose compared to methadone (Degenhardt et al., 2015). Thus, only the extended-release formulation has been approved for OUD by FDA. No special training is required for medical providers to prescribe naltrexone.
Naltrexone may be most appropriate for patients who need to avoid opioid agonists of any kind (including methadone and buprenorphine); patients who have not returned to use in 2 or more weeks but are at heightened risk of relapse; and patients who use opioids sporadically or at low levels. Naltrexone, unlike other OUD therapies, is not appropriate for the treatment of severe, acute pain—like that caused by a fractured bone or necessary surgery—because the medication completely blocks the effects of opioids. Depression is a relatively rare adverse effect of naltrexone and not a contraindication to its use (Dean et al., 2006).
The opioid antagonist naloxone is not a medication for OUD per se, but it has been approved by FDA to diagnose or treat the respiratory depressive symptoms of opioid use that can cause fatal opioid overdose. Naloxone is safe and effective, and it is the standard medication administered to reverse opioid overdose. The broader provision of naloxone has been shown to prevent opioid overdose morbidity and mortality (Bird et al., 2016; Coffin et al., 2016). In every state and the District of Columbia, naloxone can be obtained from a pharmacy without having to see a prescriber (Davis and Carr, 2017; Green et al., 2015; NASPA, 2019), and it is available from many community-based organizations and health departments for low or no cost (Wheeler et al., 2015). Notably, guidance from the U.S. Department of Health and Human Services 5 urges that all patients receiving medications for OUD be co-prescribed naloxone (HHS, 2018).
A wealth of evidence about medications to treat OUD has been amassed over the past half century from clinical studies, randomized controlled trials, systematic reviews, and meta-analyses. The verdict is clear: effective agonist medication used for an indefinite period of time is the safest option for treating OUD. According to a recent review of medications to treat OUD, “the evidence for efficacy both in reducing opioid use and retaining patients in care is strongest for agonist treatment” (Connery, 2015, p. 64).
People with OUD are less likely to die when they are in long-term treatment with methadone or buprenorphine than when they are untreated. Treatment using agonist medication is associated with an estimated mortality reduction of approximately 50 percent among people with OUD (Degenhardt et al., 2014; Larochelle et al., 2017; Ma et al., 2018; Pierce et al., 2016; Sordo et al., 2017). Both methadone and buprenorphine treatment retention have been linked to substantially decreased risks of both all-cause and overdose-related mortality among people with OUD (Sordo et al., 2017). Increased access to treatment using agonist medication is associated with reduced opioid overdose deaths (Schwartz et al., 2013). Studies of extended-release naltrexone have not had sufficient power or duration of follow-up to detect a mortality benefit (Jarvis et al., 2018).
Treatment with methadone or buprenorphine is also associated with lower rates of other opioid use (Kakko et al., 2003; Mattick et al., 2009, 2014; Thomas et al., 2014), improved social functioning (Bart, 2012), decreased injection drug use (Woody et al., 2014), reduced HIV transmission risk behaviors (Gowing et al., 2011), reduced risk of HIV diagnosis (MacArthur et al., 2012), reduced risk of hepatitis C virus (HCV) infection (Peles et al., 2011), and better quality of life compared to individuals with OUD not in treatment (Ponizovsky and Grinshpoon, 2007). Methadone is also associated with reduced levels of criminality for individuals with OUD (Bukten et al., 2012; Gearing, 1974; Schwartz et al., 2009, 2011; Sun et al., 2015). Limited evidence suggests that, compared with a placebo, extended-release naltrexone may be associated with reduced opioid use, but more rigorous studies are needed (Jarvis et al., 2018).
Compared with a placebo, both buprenorphine alone and buprenorphine in combination with naloxone administered in office-based treatment settings significantly reduce opioid use and opioid cravings (Fudala et al., 2003). In women who are pregnant, buprenorphine treatment has been linked to improved maternal and fetal outcomes; infants also tend to have less severe symptoms of neonatal abstinence syndrome when their mothers were treated with buprenorphine during pregnancy (Thomas et al., 2014).
Treatment retention with agonist medications is dose related, with meta-analyses indicating that methadone doses must exceed 60 mg and that smaller doses may be no better than placebo (Bao et al., 2009; Faggiano et al., 2003). Buprenorphine dosing at 12–16 mg increases treatment retention (Bart, 2012), and higher doses result in better outcomes (Hser et al., 2014). Retention in treatment with naltrexone is dependent on formulation rather than dose. A meta-analysis of trials found that oral, short-acting natrexone was not superior to a placebo in retaining people in treatment (Minozzi et al., 2011).
On the other hand, the optimal duration of medication for OUD has not been established. All studies of tapering and discontinuation demonstrate very high rates of relapse, although some patients may be able to successfully taper off without a return to use. Few definitive studies have been conducted because long-term treatment—particularly with methadone or buprenorphine—is complicated by stigma and misconceptions among patients and providers alike (see also Chapter 5). Nevertheless, multiple studies with longer-term follow-up indicate that extending treatment for years allows individuals to increase their opportunities to return to work, to regain their health, to avoid involvement with the criminal justice system, and to establish supportive networks of non-drug-using individuals (Eastwood et al., 2017; Goldstein and Herrera, 1995; Gossop et al., 2003; Hser et al., 2001; Simpson et al., 1982).
While a large proportion of people with OUD return to use at some point in their lives, the risk of death is mitigated by remaining in treatment. Given the consequences of returning to use without the protective effect of either a high opioid tolerance or treatment with an antagonist, most people would likely benefit from long-term maintenance treatment (Kleber, 2007). The period immediately after treatment discontinuation is a particularly high overdose risk period, as is the first 4 weeks of methadone treatment (with risks for the latter decreasing substantially after week 4), underlining the significance of efforts to enhance retention (Manhapra et al., 2017; Sordo et al., 2017). A recent systematic review found substantial variability in retention rates across treatment settings, with studies showing that 37–91 percent of individuals initiating treatment with medication for OUD were retained in treatment after the 12-month follow-up (Timko et al., 2016). Discontinuation rates are high across all medications, and most discontinuation occurs early after starting treatment.
Systematic reviews of comparative studies suggest that methadone is associated with better retention in treatment and greater patient satisfaction than other medications for OUD (Ali et al., 2017). A review of 11 randomized controlled trials found that, compared with a placebo or non-pharmacological therapy, people who received methadone were more than four times more likely to stay in treatment and had significantly lower rates of heroin use (Mattick et al., 2009). The evidence base for buprenorphine—in particular, the extended-release formulations—is not as extensive as for methadone, but it suggests that treatment with buprenorphine may have an overall mortality benefit that is slightly less than treatment with methadone (Sordo et al., 2017), possibly driven by the lower rate of retention in buprenorphine treatment. While buprenorphine maintenance treatment is at least as effective as methadone in suppressing the use of illicit opioids among people who remain in treatment, it appears to be slightly less effective than methadone maintenance treatment at retaining people in treatment (Hser et al., 2014; Mattick et al., 2014).
In contrast to methadone and buprenorphine, there have been fewer randomized controlled trials and thus less evidence about the effectiveness of extended-release naltrexone at retaining patients in treatment. Clinical studies demonstrate that oral naltrexone tends to have poorer long-term treatment adherence (Dunn et al., 2015) as well as higher mortality rates after treatment discontinuation than methadone (Degenhardt et al., 2015). A recent systematic review of 34 studies of extended-release naltrexone (Jarvis et al., 2018) reported that in controlled trials in which individuals had not already undergone opioid detoxification, only 63 percent of individuals randomized to extended-release naltrexone successfully received even a single dose of medication—the equivalent of 4 weeks of treatment. In real-world community treatment settings, only 10.5 percent of patients were adherent to extended-release naltrexone at 6 months (Jarvis et al., 2018). The only controlled trial from the United States comparing extended-release naltrexone to buprenorphine found that “in the intention-to-treat population of all patients who were randomly assigned, XR-NTX [extended-release naltrexone] had lower relapse-free survival than BUP-NX [buprenorphine-naloxone]” (Lee et al., 2018, p. 315). In the intention-to-treat analysis, the proportion of opioid-relapse events was 65 percent for extended-release naltrexone compared with 57 percent for buprenorphine treatment; the authors attribute this difference to a lower rate of patients successfully beginning the treatment in the extended-release naltrexone group, because relapse-free survival rates were similar across the groups for patients who received at least one dose (Lee et al., 2018). Among patients who have already been withdrawn completely from opioids, retention among patients randomized to buprenorphine or extended-release naltrexone is similar (Lee et al., 2018; Tanum et al., 2017), but in the real world, patients offered all three medications seldom select extended-release naltrexone (Green et al., 2018; Vermont Department of Health, 2018). A recent report on the use of naltrexone and buprenorphine in a large U.S. commercially insured population reported that 52 percent of individuals treated with extended-release naltrexone and 31 percent of individuals treated with sublingual buprenorphine discontinued treatment after only 1 month (Morgan et al., 2018). Strategies to improve retention are needed across all forms of medication-based treatment.
In spite of the extensive evidence supporting the use of medications to treat OUD, there remain major gaps in knowledge about which medication works best and for whom as well as how the medications compare over the long term. Additionally, as with all medical disorders, there is a need to expand the OUD treatment toolkit to help individuals who do not respond well to the current options.
A recent RAND systematic review of functional outcomes for individuals with OUD who were treated with medications found only 30 randomized trials and 10 high-quality observational studies that reported on at least one functional outcome in the five areas targeted: cognitive, physical, social/behavioral (including criminal), occupational, and neurological outcomes (Maglione et al., 2018). Maglione and colleagues noted that the lack of high-quality trials precluded a full meta-analytic approach to the available data or the ability to infer strong conclusions regarding the effects of medications for OUD on these important areas. Most of the evidence emerged from studies of methadone or buprenorphine treatment; few studies of functional outcomes after naltrexone were available. Moreover, the majority of the studies were cross-sectional with no follow-up data, and other reviews reporting patient-reported functional outcomes (e.g., health-related quality of life measures) are uncommon (Maglione et al., 2018).
Additional research and head-to-head trials are needed on the FDA-approved medications for OUD, particularly studies comparing the new formulations of the medications over the long term. Research is needed to assess more fully the medications' relative effects on brain functions (e.g., executive function, working memory, mood regulation, sleep architecture) and social outcomes, including those related to work, education, and family relationships. There is also a need for research focusing on optimal strategies for induction (for extended-release naltrexone) and retention (for all three medications) to improve the percentage of people retained in treatment (Kimber et al., 2015). For example, clonidine and lofexidine are alpha-2-adrenergic agonists administered to relieve opioid withdrawal symptoms after abrupt discontinuation of opioid use. Clonidine is suggested for use in conjunction with naltrexone or buprenorphine to reduce opioid withdrawal symptoms (Kleber, 2007; O'Connor and Kosten, 1998). Although it is not approved in the United States for treating opioid withdrawal, clonidine is used extensively off-label and the American Society of Addiction Medicine has recommended its inclusion in practice guidelines for managing withdrawal symptoms (Kampman and Jarvis, 2015). In 2018, lofexidine became the first non-opioid medication approved by FDA for reducing opioid withdrawal symptoms (Doughty et al., 2019; Fishman et al., 2018). Lofexidine could be used to support patients during naltrexone induction or to treat withdrawal symptoms in patients who are not yet ready to begin an opioid agonist medication-based treatment for OUD. Extended-release medications for OUD have the potential to help overcome some of the problems of poor treatment adherence to daily medications. Potential research directions could include further investigations of how the real-world effectiveness of subcutaneous or implantable buprenorphine compares with extended-release naltrexone and which have the potential to substantially extend the dosing window and eliminate the burden of daily oral dosing. More fundamentally, comparing the effectiveness of sublingual and extended-release buprenorphine formulations would test the assumption that daily dosing is inferior.
Patients' preferences about medications to treat OUD are fundamental in determining whether they start and stay on treatment for OUD, but those preferences have yet to be fully explored. Some informative data about patients' medication preferences are available from Rhode Island's correctional system and the state of Vermont. In both populations, methadone is the most common choice among people receiving medication for OUD (between 60 and 70 percent), with buprenorphine preferred by the remainder of patients. Only two people in Vermont and four people in Rhode Island's prison system have chosen treatment with extended-release naltrexone, according to recent data (Green et al., 2018; Vermont Department of Health, 2018). With extremely limited access to medications for OUD, however, patients may not be offered medication at all, much less be offered a choice between the FDA-approved medication options. Real-world evidence could help to elucidate the role of patient choice in the success of long-term treatment. Patients entering treatment often have strong preferences for one medication or another (Uebelacker et al., 2016), although many individuals entering treatment have limited knowledge regarding the available medications to treat OUD (Alves and Winstock, 2011). Increasing medication access, uptake, and retention will require taking patients' beliefs and preferences about medications into account (Uebelacker et al., 2016). Through shared decision making, a patient's preferences, goals, and motivations can be used to guide the choice of medication for OUD treatment.
Expanding the treatment toolkit for OUD has the potential to increase treatment rates and provide more effective, individualized care for people with OUD. Treatment options that warrant further exploration include slow-release oral morphine (SROM), supervised injectable opioid agonist therapies (siOAT), cannabinoids, and anti-opioid vaccines, to name a few. Many of these options would require not just approval by FDA, but also changes to the Harrison Narcotics Tax Act of 1914. 6
SROM is a full agonist opioid with a slow-release oral formulation that has been proposed as an alternative maintenance therapy for people who do not respond adequately to the other available types of medications to treat OUD. As yet, no definitive evidence indicates that SROM is equivalent or superior as a treatment option, but SROM appeared to be similar in effectiveness to methadone in one study that directly compared the two for maintenance treatment (Beck et al., 2014). A systematic review found that although the evidence is very limited, SROM had similar retention rates to methadone in one study, and in other studies it was associated with improving quality of life, relieving withdrawal symptoms and cravings, and reducing other drug use (Jegu et al., 2011). Another systematic review suggested that SROM may reduce opioid use and depressive symptoms, but adverse effects were more frequent than for other types of medications for OUD (Ferri et al., 2013).
siOAT has been demonstrated to be efficacious in treating people who have severe OUD that has not been well managed by other medications. The treatment is administered under strictly monitored, medically supervised conditions, typically via injection of diacetylmorphine—i.e., pharmaceutical-grade heroin—or of hydromorphone (Drug Policy Alliance, 2016), another opioid currently approved as an analgesic. Evidence demonstrates that among people who have previously been unsuccessful on methadone maintenance therapy, siOAT can significantly improve treatment retention while reducing the use of illicit opioids (Strang et al., 2015). Several countries have carried out studies with mixed results (Fischer et al., 2007). For example, a randomized controlled trial conducted in Canada found that injectable hydromorphone was as effective as injectable diacetylmorphine and had similar treatment outcomes for people with severe OUD (Oviedo-Joekes et al., 2016). Modeling suggests that over a patient's lifetime, siOAT with hydromorphone may provide greater benefit to patients at a lower lifetime cost than methadone maintenance therapy alone (Bansback et al., 2018). Despite this encouraging evidence and the opportunity it represents to engage more people with severe, treatment-resistant OUD in care, siOAT remains unavailable in the United States because it is hampered by political and regulatory barriers (Oviedo-Joekes et al., 2016). However, the growing interest in this modality may lead to the development of better oral, intranasal, or inhalable formulations that could circumvent the stigma associated with injectable opioid medications, even when administered under medical supervision (Klous et al., 2005).
Emerging evidence suggests that cannabinoids might be useful as a component of treatment for OUD. “Medical marijuana” has received significant attention because many OUD patients consume recreational marijuana either as a reward substitution in attempts to reduce overdose risk (because cannabis has low mortality risk) or to alleviate anxiety symptoms during opioid withdrawal (Wiese and Wilson-Poe, 2018). Consistently, results from the National Survey on Drug Use and Health (2008 to 2013) indicate that cannabis use in the general population is associated with reduced risk of past year opioid abuse in those with a history of illicit opioid use (Pisano et al., 2017). However, epidemiological evidence (National Epidemiologic Survey on Alcohol and Related Conditions) also suggests an increased risk of prescription opioid misuse and OUD with cannabis use (Olfson et al., 2018), and tetrahydrocannabinol (THC), the psychoactive component of cannabis, has been demonstrated in preclinical models to enhance opioid sensitivity (Ellgren et al., 2007), raising concerns about the potential of THC (and THC-rich medical marijuana strains) to be a viable treatment option for OUD. Dronabinol is an FDA-approved THC analog that has been studied as a treatment for opioid withdrawal, but it had modest efficacy and induced several side effects including tachycardia (Lofwall et al., 2015). More research is needed to compare the effectiveness of dronabinol or other cannabinoids (such as Sativex, a cannabis-based oral spray) to treatment with methadone, buprenorphine, or extended-release naltrexone.
A new line of research has recently focused on the potential of cannabidiol (CBD, a non-intoxicating cannabinoid) to help reduce the risk of opioid relapse by inhibiting drug-seeking behavior (Ren et al., 2009). CBD therapy is also known to relieve some of the psychological and physiological symptoms that are associated with OUD, such as anxiety, insomnia, and pain (Hurd et al., 2015). Unlike medications for OUD that target the endogenous opioid system directly, CBD represents a new way to indirectly affect systems that control opioid-seeking behavior and to support other medications by reducing the craving and anxiety that contribute to relapse (Hurd et al., 2015). CBD has the advantage of negligible abuse potential because it does not produce euphoria, and it has minimal side effects. To date, the data about CBD and its benefits are mostly preclinical. Animal studies suggest that CBD inhibits trigger-induced heroin-seeking behavior with prolonged effects that may last for weeks, even after relapse (Ren et al., 2009). Pilot studies with humans have demonstrated that CBD is safe to co-administer with the potent injectable opioid fentanyl (Manini et al., 2015), that CBD can induce a decrease in craving for heroin that persists for up to 1 week, and that CBD can reduce anxiety.
Vaccination against opioids to prevent OUD and its consequences is a relatively new avenue of research. Such vaccines work by causing a person's body to create its own antibody response to a specific opioid, thus blocking the psychoactive effects of that opioid in the brain if it is ingested (Bremer et al., 2017). Because mu-opioid receptors are required in order to develop compulsive opioid-taking behavior, it is hypothesized that people will not develop OUD if opioids do not reach the brain. Given that in people with OUD it is common to use more than one type of opioid, the vaccine would need to be effective for the different forms (e.g., fentanyl, heroin). A major drawback is that in order to be effective, vaccines may also inhibit the effects of opioids for critical pain relief or palliative care treatments (Olson and Janda, 2018). Another concern is that a vaccine may also interfere with the use of naloxone as an overdose reversal medication (Raleigh et al., 2017). In addition, ethical considerations may outweigh the prophylactic benefit of vaccination in high-risk populations. Such cases might include offering vaccination as an alternative to incarceration or parents seeking to vaccinate their children against future drug use before they can give consent (Shen et al., 2011). Finally, vaccines do not treat the underlying psychosocial or behavioral correlates of OUD and therefore could lead to the unintended consequence of developing another type of substance use disorder.
Additional research on the neurobiology of addiction and opioid signaling will be needed to advance the development of new medications. Current treatment options target only the opioid reward pathway, but new treatment options targeting other neural systems related to craving, negative affect, and cognitive control will expand our understanding of addiction and therapeutic interventions (Koob and Volkow, 2010). For instance, new mu-opioid receptor agonists that are biased toward specific downstream signaling pathways—and thus do not mediate the rewarding effects of opioids—could result in medications with lower misuse potential. Developing novel, non-opioid treatments that can help to relieve short- and long-term opioid withdrawal symptoms and cravings would require in-depth research into the interaction of the opioid systems with cellular and molecular mechanisms within discrete neural circuits that maintain long-term maladaptive processes and regulate opioid-seeking behavior. The development of such novel treatments has the potential to facilitate treatment induction, to improve retention in care, and to lengthen remission.
Behavioral interventions are often used in conjunction with medications in treating OUD, for two primary reasons. The first is to target a broad range of problems and issues not addressed by the medications themselves (e.g., comorbid psychiatric symptoms, concurrent use of other drugs, the need for social support, HIV risk behaviors, behavioral changes, motivation). The second is to address limitations associated with each form of medication (e.g., high attrition rates). See Box 2-2 for a description of different types of behavioral interventions that have been used with medication-based treatment for OUD. However, the evidence about the efficacy of different behavioral interventions used to complement each of the FDA-approved medications is limited to date, and the evidence that has been reported is mixed. Interpreting the outcomes is complicated by differences in the outcomes targeted in most studies—that is, retention in treatment and reducing opioid use versus addressing comorbid problems such as other drug use, psychosocial functioning, and HIV risk behaviors.
Types of Behavioral Interventions.
It is generally accepted that the best outcomes are typically achieved through a combination of pharmacological and behavioral therapies (NIDA, 2018), but there is evidence that some individuals may respond adequately to medications plus medical management alone (e.g., evaluation of medication safety and adherence, monitoring, or advice by the prescribing provider) (Gruber et al., 2008; McLellan et al., 1993; Schwartz et al., 2007, 2012; Weiss et al., 2011; Yancovitz et al., 1991). Given the resource limitations and the lack of empirical evidence about specific behavioral interventions to improve outcomes from medications for OUD, some have argued that clinicians should not be dissuaded from initiating medications for OUD simply because evidence-based behavioral therapies are not available (beyond medical management with monitoring) (Schwartz, 2016). At the same time, while medications to treat OUD prevent death and stabilize patients so that their comorbid psychiatric, medical, and social problems can be identified and addressed, these medications alone do not address the many complex problems that many individuals with OUD may have. Therefore, it is critical to take individual differences into account and select a treatment plan that is best suited to each patient's needs (Carroll and Onken, 2005). Provision of behavioral interventions can and often do occur in the medical management encounter with the prescriber.
There is robust evidence that contingency management interventions that reward positive behaviors are effective as behavioral adjuncts to methadone treatment. Furthermore, treatment retention improves when patients are permitted to take the medication home. Take-home medication privileges based on drug-free urine specimens have consistently been shown to reduce illicit drug use (Carroll and Onken, 2005), as have incentive programs using vouchers for goods and services to reward time without drug use (Silverman et al., 1996). Low-cost contingency management interventions (in which individuals earn chances to win prizes rather than earn vouchers) have also demonstrated efficacy and may be suitable and more acceptable for resource-constrained treatment settings (Petry and Martin, 2002).
A systematic review examining 14 recent studies indicated that, with the exception of contingency management, behavioral therapies themselves do not generally improve retention or reduce opioid use in individuals with OUD receiving methadone treatment (Dugosh et al., 2016). However, results from studies that target “secondary” outcomes such as psychosocial functioning and other drug use generally support the addition of behavioral interventions. Studies have also examined the effectiveness of the counseling that patients are required to receive in real-world OTPs. The results do not demonstrate differences in treatment retention or opioid use among patients who were randomized to receive little or no interaction with OTP drug counselors versus patients who received the federally mandated level of counseling (Gruber et al., 2008; Schwartz et al., 2006, 2012; Yancovitz et al., 1991). When considered in aggregate, these data suggest that the psychosocial supports required at OTPs should be recalibrated.
A systematic review of eight randomized controlled trials found mixed results with respect to the additional benefit of adding behavioral intervention to medical management in office-based buprenorphine treatment (Carroll and Weiss, 2017). Four of the trials found no additional benefit of behavioral therapy interventions that included varying the intensity of medical management (Fiellin et al., 2006; Ling et al., 2013; Weiss et al., 2011); cognitive behavioral therapy (Fiellin et al., 2014; Ling et al., 2013); contingency management with or without cognitive behavioral therapy (Ling et al., 2013); and medical management plus drug counseling (Weiss et al., 2011). The other four randomized controlled trials demonstrated some additional benefit of adding the behavioral interventions, particularly those that used contingency management as the intervention (Bickel et al., 2008; Christensen et al., 2014; Miotto et al., 2012; Schottenfeld et al., 2005). The authors suggested that research design may have played a role in these opposing outcomes. The four trials that reported no additional benefit of behavioral intervention all featured relatively intensive medical management in addition to the behavioral intervention under evaluation, while three of the four positive studies did not offer structured medical management.
To date, no trials have evaluated the efficacy of buprenorphine alone, without medical management, as the minimal standard of care. Thus, there are no data on the number or types of individuals who may respond to buprenorphine without medical management and monitoring (Carroll and Weiss, 2017). Another recent systematic review examined group-based therapy for OUD combined with buprenorphine, finding multiple methodological problems with most of the studies (e.g., small sample size, varying theoretical focus, weak control groups) that make it difficult to draw conclusions regarding efficacy (Sokol et al., 2018). However, some evidence suggests that patients with other comorbid addictions or psychiatric disorders have better outcomes when behavioral interventions are included in their treatment regimens (Arias and Kranzler, 2008; Kelly and Daley, 2013).
A recent systematic review found that relatively few robust studies meeting the criteria for inclusion had investigated behavioral interventions used with naltrexone (Dugosh et al., 2016). Most of the high-quality studies involved the use of contingency management to improve adherence to naltrexone and the submission of opioid-free urine specimens. Several trials have evaluated contingency management strategies with oral—not extended-release—naltrexone (Carroll et al., 2002; Dunn et al., 2013; Nunes et al., 2006; Preston et al., 1999), finding positive effects on treatment retention, attendance, and compliance in the short term, but poor treatment retention in the longer term. Two studies of injectable extended-release naltrexone in conjunction with contingency management found that the combination was effective in improving treatment retention and in increasing the number of naltrexone injections received (DeFulio et al., 2012; Everly et al., 2011).
Apart from contingency management, it is difficult to say which behavioral intervention will be most effective with a given medication or a given outcome in a given patient. Relatively few studies have investigated the comparative or differential effectiveness of different types of behavioral interventions in treating OUD at different points in the continuum of care, among different populations, or in different treatment settings (Dugosh et al., 2016). Given the mortality benefit of the medications, more research into behavioral interventions that result in improved treatment adherence is critical; behavioral techniques also have promising potential to assist patients in achieving good long-term functional outcomes. Investigating behavioral techniques to facilitate improvements in psychiatric, legal, interpersonal, and occupational functioning may support sustained remission (Carroll and Weiss, 2017). Other techniques may reduce HIV and HCV risk behaviors, regardless of other treatment outcomes (Edelman et al., 2014; Meade et al., 2010).
Another knowledge gap in OUD behavioral treatment innovations pertains to the provision of peer support to enhance treatment. Peer support is “the process of giving and receiving nonprofessional, nonclinical assistance from individuals with similar conditions or circumstances to achieve long-term recovery from psychiatric, alcohol, and/or other drug-related problems” (Tracy and Wallace, 2016, p. 143). Peer-based recovery support has a long history in addiction treatment and was advocated for by Dole and Nyswander, who developed the methadone maintenance treatment model for OUD (White, 2009). Peer support groups to supplement treatment for addiction have promising potential to increase treatment engagement and to reduce substance use and risk behaviors for infectious disease transmission, but more rigorous studies are needed (Tracy and Wallace, 2016). Peer providers with lived experience related to addiction may be able to contribute positively to other people's OUD treatments and to help address the vast workforce shortages in behavioral health. However, the inclusion of peer providers gives rise to important concerns about their training, certification, methodological consistency across programs, opportunities for career advancement, and fair compensation. Despite the high degree of public investment in these programs nationally, there are no data from well-controlled trials evaluating peer support. More research is needed to explore how peer providers may be able to support OUD treatment and to establish the effect size of such interventions (Chapman et al., 2018; Reif et al., 2014).
Conclusion 2: U.S. Food and Drug Administration–approved medications to treat opioid use disorder are effective and save lives.
Conclusion 3: Long-term retention on medication for opioid use disorder is associated with improved outcomes.
