Office-Based Methadone Treatment for Opioid Use Disorder and Pharmacy Dispensing: A Scoping Review

We searched Ovid MEDLINE and the Cochrane Database of Systematic Reviews from database inception to July 2020. For a description of the search strategies, see Table S1 in the online supplement. We also reviewed reference lists of relevant articles to identify additional studies. The analysis was conducted as part of a larger scoping review of policy research on methadone treatment for opioid use disorder that covered additional research areas (Figure 1) (11). In the present analysis, we focused on the literature on methadone in office-based settings and U.S. studies of pharmacy dispensing of methadone.

Studies were selected using a hierarchical approach. Randomized trials and controlled observational studies were prioritized; observational and descriptive studies were included when the priority evidence was unavailable. Studies addressing methadone policy were eligible if they focused on office-based settings and included outcomes (e.g., treatment initiation, treatment receipt, retention in care, or drug use) and were conducted in the United States or in other highly developed countries. One investigator abstracted key information (clinical setting, country, sample size, intervention/comparison, and main findings) into a table and a second investigator verified data. Because this was a scoping review that broadly surveyed the relevant literature, we used a descriptive approach to summarize the literature and did not formally synthesize or grade the quality of the evidence. The literature flow diagram is presented in Figure 1.

The first clinical trial of office-based methadone treatment (medical methadone maintenance) was a 1993 initiative in Chicago that recruited 130 stable patients from 14 OTPs (15). Patients were selected for study inclusion if they had a year or more of methadone treatment and if for the past 6 months had no positive urine screens, were employed, had plans to remain on methadone for the next year, and were adherent to treatment plans. Participants were randomly assigned in a 2:1 ratio to medical methadone (N=87) or continuing care at the OTP (N=43). Medical methadone patients had monthly physician visits and two medication visits per month at their original OTP and received 13 days of take-home medication. The OTP continued to provide counseling services. The comparison group remained in their OTP with two medication visits per week (N=43) (i.e., two or three take-home medications per visit). After 6 months, participants in the comparison group were enrolled in the medical methadone program. The two groups had similar completion rates at six (89% compared with 85%) and 12 months (both groups, 73%), and rates of positive urine screens were less than 1% in both groups (16). All patients were satisfied with the services, and none wanted to return to treatment as usual (two medication visits per week) (15).

In the second trial, conducted in New Haven, Conn., 46 stable methadone patients (i.e., patients were required to have been in an OTP for more than 1 year, have no positive urine screens for opioids or cocaine in the past year, have plans to remain in the OTP for 6 months, have no co-occurring medical, psychiatric, or substance use disorder, and have transportation to the clinic or office and legal income and stable housing) were recruited from an OTP and randomly assigned to either six primary care internists who provided office-based methadone agonist therapy (N=22) (with methadone dispensed in the physician’s office) or to usual care in the OTP (N=24) for 6 months (12). The groups had similar rates of opioid-positive urine screens (9% compared with 11%) and clinical instability (i.e., defined a priori as two consecutive urine drug screens detecting cocaine or opioid metabolites or detecting no methadone) (18% compared with 21%) during the 6-month study period; the office-based patients were more likely to rate the care as excellent (73% compared with 13%) (12). A secondary qualitative analysis (included in Table 1 as an observational study) reviewed the office records and interviewed physicians to identify barriers to office-based methadone treatment. Physicians reported that the urine drug monitoring and extra paperwork required for methadone dispensing were barriers to continued office-based care and recommended that studies be conducted using pharmacies for dispensing (30). Additional barriers were patient access to psychiatric services, communication with the OTP, and patient nonadherence to medication (30).

In a third trial, based in Baltimore, 92 stable methadone patients (i.e., in the past year, the patients had an uninterrupted episode of methadone maintenance, no positive urine screens for any drug, verified employment, no failed methadone medication checks [the patient was recalled to the OTP, where the remaining doses were counted to verify that the medication was being used correctly], and plans to remain in care for the duration of the study) were randomly assigned to one of three groups: 28-day methadone treatment in a physician’s office (office-based, N=32), 28-day methadone maintenance in an OTP (clinic-based, N=33), or regular methadone maintenance in an OTP with one or two visits per week for medication (routine care, N=27) (13). For all study participants, an adaptive treatment model was used that intensified services (additional counseling required) if urine screens were positive for illicit drugs or the weekly medication checks were incorrect (14). After 6 months of the 12-month study period, the three study groups had similar outcomes: 1% of urine screens detected illicit drug use, drug diversion was not observed, methadone misuse was limited, patients were satisfied or very satisfied (patients receiving 28-day supplies of methadone were more satisfied), and patients accepted the treatment intensification (14). At the 12-month follow-up, 77 participants were still in care. Few patients tested positive for illicit drug use or missed medication recalls (the groups did not differ significantly) (13). The three groups also did not differ in the use of treatment intensification (36% of patients), self-reported drug use was minimal, and all patients reported being satisfied or very satisfied with the services. However, patients in both of the 28-day medication groups (office-based and clinic-based) were more satisfied than those in routine care and were more likely to report improved employment and social activities (13).

The fourth trial, a pilot study conducted in New Mexico, tested office-based primary care and methadone medical management with dispensing from community pharmacies, and a social worker provided counseling and case coordination. Women (N=26) who were enrolled in an OTP and were stable were randomly assigned to either continued care in the OTP (N=12) or office-based care (N=14) (17). A treatment manual specified roles and responsibilities for the five pharmacists, five medical practitioners (four physicians and one nurse practitioner), and one master’s-level social worker with experience treating drug use disorders. The pharmacies were registered with the U.S. Drug Enforcement Administration (DEA) and dispensed methadone as medication units for the OTPs. The social worker provided clinical coordination and at least monthly psychosocial services (in her office or at the participant’s home) and participated in weekly case conferences. The medical practitioners were listed as OTP staff and met with patients at least monthly for medical and medication management and urine drug screens (16, 17). Patients eligible for participation in the study had at least 6 months of methadone treatment, 6 months of stable methadone dosage, at least two take-home doses per week, and plans to remain on methadone for the 12-month study period. The analysis was restricted to participants who completed the 12-month follow-up period (N=22) (office-based care, N=13; OTP care, N=9). Two office-based care participants were returned to the OTP early in the study period for illicit drug use and mental health problems. Patients in office-based care were less likely to use illicit opioids (23% compared with 78%; χ²=6.4, df=1, p<0.05) during the 1-year study period (16, 17). A qualitative analysis of a satisfaction survey suggested that patients receiving community services were pleased with the services and were treated well (16). The small sample generated imprecise estimates.

There were two non-U.S. randomized trials, one conducted in France and the other in Australia. The French trial (N=221) compared methadone initiation in primary care compared with specialty care; it found that methadone initiation in primary care was feasible and acceptable to physicians and patients, with similar rates of abstinence and retention compared with specialty care induction (18). Office-based methadone treatment is not typically available in France, and this was the first evaluation of office-based care. The Australian randomized trial (N=139) recruited stable methadone patients and active heroin users and randomly assigned participants to receive buprenorphine in a methadone clinic (N=66) compared with buprenorphine in an office-based setting (with pharmacy dispensing) (N=73) (19). Self-reported heroin use and retention in care were similar in both settings (19).

Observational studies conducted in the United States echoed results from the clinical trials; stable methadone patients responded to office-based methadone treatment with high levels of retention in care and low rates of drug use (21–27). With one exception (21), the studies described only patients who received office-based methadone treatment and did not include an OTP comparison. Across studies, patients receiving office-based methadone treatment generally had monthly physician visits with observed dosing and urine tests. In most of the studies, patients received take-home medication directly from the physician (pharmacies delivered medication to the office setting the day prior to scheduled appointments).

In 1983, Dr. Marie Nyswander used an Investigational New Drug approval for methadone by the U.S. Food and Drug Administration (FDA) to open the first medical methadone maintenance program for stable methadone patients in New York City; the initial report on the program described 1-year results for 28 patients (23). The study was restricted to patients who met 10 eligibility criteria: 5 years of conventional methadone treatment, 3 years of stable employment, no criminal involvement for the past 3 years, no illicit drug use and no alcohol abuse for the past 3 years, reliable clinic attendance, a need for continued maintenance, no need for psychiatric medication, no social ties to drug users, clinical recommendations to participate, and a willingness to participate in research interviews. Take-home medication was dispensed during the medical visits. Subsequent analyses expanded the size of the program and the duration of follow-up from 40 patients with 12 to 55 months of follow-up (25) to 100 patients with up to 9 years of follow-up (24) and 15 years with 158 study participants (26). Treatment retention rates declined slightly over time, rates of discharge for drug use were low, and over 15 years, 16% of the medical methadone patients were returned to methadone clinics.

A second New York City study, also conducted under an FDA Investigational New Drug protocol, compared patients receiving office-based methadone treatment with patients treated in an OTP (21). Participating patients were employed, had no evidence of illicit drug use in the past 3 years, and were psychiatrically stable. When the medical maintenance patients (N=127) were compared with the OTP patients (N=3,342), patients in the office-based setting were older (mean age, 52 years compared with 44 years) and more likely to be male (72% compared with 59%) and White (59% compared with 17%). Within the medical methadone group, the rate of illicit opioids and cocaine detected in monthly urine drug tests was less than 1%, and unannounced recalls found no evidence of diversion or inappropriate use. Patients received take-home medication in a central pharmacy near the physician’s office (21).

The other studies reported outcomes of patients receiving office-based methadone treatment without comparisons to OTP patients. A Baltimore-based analysis described 21 stable methadone patients enrolled in a medical maintenance program with take-home methadone dispensed in the physician’s office (27). Participating patients were employed and had 5 years of abstinence from illicit drug use, no evidence of alcohol use disorder, and no concurrent psychiatric disorders. Urine drug tests were conducted monthly. After 12 years, 71% of the patients remained in care, with 0.5% urine tests positive for drug use and no methadone overdoses or diversion (27). Participants reported significant improvement in quality of life and ability to travel.

A small study (N=10) conducted in rural Pennsylvania described the use of a community physician and a community pharmacy to monitor and dispense methadone to stable patients (i.e., with a median of 3 years in OTP care without evidence of current drug use) living 40 miles from the OTP (20). The first pharmacy dispensing occurred in July 2003 (despite objections from the state alcohol and drug authority). All patients were enrolled in a local internal medicine clinic; the clinic’s director was certified in addiction medicine and wrote monthly medication orders that could be filled only at the participating pharmacy. The pharmacy observed one dose and provided weekly or biweekly take-home medication. The primary care practice provided counseling and urine testing. Participating patients paid pharmacy charges out of pocket. Five of 216 (2.3%) urine tests were positive for drugs other than methadone (three for prescribed medications and two for illicit drugs). Study participants felt comfortable in the pharmacy and liked being treated like a customer; they were pleased with the physician and his skills and grateful for the program. Pharmacy staff reported that the bottle return requirement was unnecessary and that the lockbox was stigmatizing (20). Despite the results, the state alcohol and drug authority remained uncomfortable with the program and in July 2005 ordered the patients to return to the original OTP 40 miles away. Most patients switched to buprenorphine and remained in care with the internal medicine clinic (20).

Harborview Medical Center in Seattle partnered with Evergreen Treatment Services (an OTP) and requested SAMHSA and DEA exceptions and state exemptions from opioid treatment regulations (22). The OTP continued to provide counseling, and the hospital provided pharmacy and medical clinic services. The OTP patients were screened for medical maintenance eligibility, and of the 49 individuals who met study eligibility criteria (i.e., at least four take-home medications each week, with reliable attendance, monthly urine screens negative for drug use, no evidence of alcohol use disorder, no legal issues or unpaid program fees, and no psychiatric disorders), 30 agreed to participate (22). Participants picked up methadone medication once or twice per month from the hospital pharmacy, and pharmacists assessed patient stability and primary care needs, collected a monthly urine sample, observed patients take their regular dose, and dispensed take-home doses. Monitoring included random callbacks, which required patients to return to the pharmacy within 24 hours to confirm pill counts. Physician visits were adjusted according to clinical need. When participants entered medical maintenance, their mean age was 45 years, and most were male (70%), White (83%), and employed (83%), with a mean of 12 years on methadone, 7 years of take-home medication privileges, and a mean daily dose of 63 mg of methadone. Eighty percent had Medicaid or other health insurance. In the first 12 months of program participation, less than 1% of urine tests were positive for illicit drugs, callbacks found no discrepancies, and two patients were transferred back to the OTP. Scores on the Addiction Severity Index medical subscale improved, other subscale scores were low and stable, and 87% of patients were very satisfied with the medical maintenance program (22).

The search found two non-U.S. observational analyses: one conducted in England (28) and one in Ireland (29). England’s National Treatment Outcome Research Study compared patients with methadone prescriptions from a general practitioner (N=79) with patients enrolled in specialized drug clinics (N=161) (28). At 12 and 24 months after treatment entry, patients in both groups reported reduced illicit drug use, injecting, syringe sharing, mental health symptoms, medical problems, and criminal activity (28). Finally, a study of a random sample of patients in Ireland’s Central Methadone Treatment List who began care in 1999, 2001, and 2003 (N=1,269) assessed retention in care and patient characteristics associated with retention in care (29). Sixty-one percent of patients remained in care for more than 1 year, and patients in primary care were twice as likely as patients in specialty methadone clinics to be retained for the full year (29).

Four U.S. studies of office-based methadone treatment (three observational studies [20–22] and one clinical trial [17]) used pharmacies as medication units to dispense methadone take-home medication. When compared with studies with office dispensing, outcomes appeared to be similar. A small trial used community pharmacies (with special approvals from the DEA) (17). An observational study described how the hospital pharmacy conducted observed doses and dispensed take-home methadone before business hours (22). A second study used a central pharmacy near the physician offices (21). In the third analysis, a community pharmacy dispensed methadone for rural patients who initially enrolled in an OTP 40 miles from where they lived (20).

A 1970 initiative in San Antonio, Tex., utilized pharmacy dispensing to expand the OTP’s caseload by 30% (31). Each pharmacy purchased, stored, and dispensed its own supply of methadone (1970–1972) until the 1973 FDA regulations required the OTP to purchase, store, and deliver methadone as needed (31). The study described a group of 96 patients who received 12 months of methadone dispensed in pharmacies. The study did not include a comparison group (31). An ecological, before-after study conducted in the United Kingdom found a decrease in methadone deaths following implementation of office-based methadone treatment with pharmacy dispensing, despite increased methadone prescribing (32).

Two analyses using census tract data observed that patients’ travel burden for daily dispensing could be reduced if pharmacies dispensed methadone. The drive time to an OTP was greater than the drive time to a pharmacy (20 minutes compared with 4 minutes); the difference increased in rural communities (4, 5).

The U.S. randomized trials were unblinded, with small samples in most of the studies that may have been underpowered to detect differences in outcomes. Study samples were also too small to detect infrequent adverse events, and follow-up periods varied across the studies. The observational studies were primarily descriptive, and most did not control for possible confounding characteristics or include comparison groups. Few studies, moreover, have been conducted in the past 20 years; results, therefore, are dated and may not generalize to the current environment. Nonetheless, the studies provide evidence that stable methadone patients benefit from office-based care.

However, patients were carefully selected, and although the operationalization of “stable patient” varied, at least 6 months of OTP care was required, along with no evidence of drug use. No U.S. studies tested office-based care with patients new to care. In the French study, however, methadone patients successfully initiated methadone and were stabilized in office-based settings. Finally, while office-based methadone treatment is relatively common in other countries, the international studies relied on retrospective analyses using historical data, and the clinical trials tested facets of methadone services (i.e., switching to buprenorphine) that may be of limited applicability to U.S. settings.

The association between access to opioid agonist therapy and improved patient outcomes is well described in the literature. Our findings suggest that modifications to current methadone policies to allow for health services delivery changes (specifically, office-based prescribing of methadone for stabilized patients) could enhance access to care. Pharmacy dispensing of methadone would further facilitate access to effective treatment for opioid use disorder, especially in rural communities. Pharmacies are an integral part of methadone treatment in Canada and Europe. U.S. pharmacies already dispense methadone when it is prescribed for pain, and there are no obvious barriers to using pharmacies in the United States, aside from current DEA regulations that prohibit physicians from prescribing methadone for opioid use disorder and pharmacists from filling methadone prescriptions for opioid use disorder. Recent analyses highlight the distance many patients in rural areas travel to access methadone services and advocate for the use of rural pharmacies as medication units (3, 5).

An expansion of methadone services may also begin to address persistent racial disparities in care, which are observed in the current design of the opioid agonist treatment system (35, 36). Black patients, for example, are less likely to have access to and receive office-based care with buprenorphine (37). Implementation of office-based methadone treatment should include a critical assessment of its impact on access, with intentional efforts to increase access to minoritized populations.

Some physicians, moreover, appear to be interested in caring for their patients using methadone for opioid use disorder. Interviews with 71 primary care and HIV providers in 11 New York City practices determined that most (85%) had provided medical care for methadone patients, 70% were comfortable managing medical care for drug users in the primary care setting, and 66% were willing to prescribe methadone for OTP patients in their caseload if training and support were available. However, half the respondents were concerned that methadone patients often have multiple needs that could be difficult to address in their practice setting (38). It should be noted, however, that relatively few physicians have the DATA 2000 waiver to prescribe buprenorphine for opioid use disorder and that among those who have the waiver, many do not use it. There is widespread hesitancy to address opioid use disorder in clinical practice.

It is noteworthy that aside from recent modeling studies of drive time to pharmacies compared with OTPs and two international analyses, all of the studies were published in the 1980s, the 1990s, and the first decade of the 21st century. New trials of office-based methadone treatment using contemporary patients with opioid use disorder would help verify the continued applicability of office-based methadone treatment.

In addition, all of the U.S. studies were limited to stable OTP patients. As a result, the lack of data on patients new to care is a major gap in the research literature. There are no data on the risks and benefits of office-based methadone treatment for patients new to care. While clinical evidence suggests that patients continue to use drugs while new to care, the evidence also suggests that for many patients, opioid agonist therapy helps them to stabilize; they respond well to care, and their drug use dissipates over time. It is likely that some federally qualified health centers have the resources and skills to work effectively with methadone patients who are new to care. Given the need for more access to methadone services in nonmetropolitan communities, it is critical to test office-based methadone treatment for patients seeking treatment for opioid use disorder. One option would be to extend the buprenorphine hub-and-spoke model to include methadone patients (patients are stabilized on medication in an OTP and transferred to office-based practices when stable [39–41]).

Despite the potential advantages of methadone dispensing in pharmacies, the research on it is limited. Our search found only five studies that used pharmacy dispensing, and one study was initiated prior to the 1973 methadone regulations (31); in two studies, the pharmacies operated as medication units (17, 22). Qualitative implementation research would be useful to understand perceived barriers to the use of pharmacies to dispense methadone and development of strategies to address potential barriers. Clinical trials and other studies evaluating clinical or administrative data are needed to document the safety of pharmacy dispensing and potential effects on reducing burdens on patients, particularly for those located in areas requiring lengthy travel to an OTP. Additional research could inform changes to regulations to encourage greater use of pharmacies for methadone dispensing.

Another strategy to increase access to methadone is use of mobile medication units: methadone vans. Since 2006, the DEA has not approved new mobile medication units. Aside from a handful of studies (42, 43), little systematic research has been conducted to test facets of mobile medication units. In February 2020, the DEA released a draft of changes in regulations that would permit a resumption of approvals. Renewed approvals will provide opportunities for implementation research on the development and use of mobile medication units.