Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Aug 1.
Published in final edited form as: Drug Alcohol Depend. 2018 Jun 19;189:166–171. doi: 10.1016/j.drugalcdep.2018.05.010

Tapering off and returning to buprenorphine maintenance in a primary care Office Based Addiction Treatment (OBAT) program

Zoe M Weinstein a,*, Gabriela Gryczynski b, Debbie M Cheng c, Emily Quinn d, David Hui e, Hyunjoong W Kim e, Colleen Labelle a, Jeffrey H Samet a,f
PMCID: PMC6139651  NIHMSID: NIHMS1503890  PMID: 29958128

Abstract

Background:

Guidelines recommend long-term treatment for opioid use disorder including the use of buprenorphine; however, many patients desire to eventually taper off. This study examines the prevalence and patient characteristics of patients that voluntarily taper off buprenorphine.

Methods:

This is a 12-year retrospective cohort study of adults on buprenorphine in a large urban safety-net primary care practice. The primary outcome was completion of a voluntary buprenorphine taper, which was further characterized as a medically supervised or unsupervised taper. The secondary outcome was re-engagement in care after taper. Descriptive statistics and estimated proportions of both taper completion and re-engagement in treatment were calculated using Kaplan-Meier estimates.

Results:

The study sample included 1308 patients with a median follow-up time of 316 days; 48 patients were observed to taper off buprenorphine during the study period, with an estimated proportion of 15% (95%CI: 10%−21%) based on Kaplan Meier analyses. Less than half of the tapers, 45.8% (22/48), were medically supervised. Thirteen of the 48 patients subsequently, reengaged in buprenorphine treatment (estimated proportion 61%, 95%CI: 27%−96%), based on Kaplan-Meier analyses with median follow-up time of 490 days.

Discussion:

Despite the fact that many patients desire to discontinue buprenorphine, a minority had a documented taper. Among those who tapered, more than half did so unsupervised by the clinic and a majority of those who tapered off returned to buprenorphine treatment within two years. As many patients are unable to successfully taper off buprenorphine, the medical community must work to address any barriers to long-term maintenance.

Keywords: buprenorphine, opioid use disorder, patient dropout

1. Introduction

Buprenorphine is an effective treatment for opioid use disorder and guidelines recommend long term or even life-long treatment (Kraus et al., 2011); however, a majority of patients are interested in eventually tapering off buprenorphine (Kleber, 2007; Winstock et al., 2011). Some patients avoid starting medications altogether due to concern about their ability to eventually taper off (Appel et al., 2004; Peterson et al., 2010). Currently, physicians lack sufficient evidence to answer common patient questions about the frequency and feasibility of discontinuing buprenorphine for stable patients without causing relapse (Amato et al., 2011). To date, published literature has not described characteristics associated with elective taper off of buprenorphine among long-term maintenance patients.

The majority of prior research focused on the tolerability of short term buprenorphine detoxification (Dunn et al., 2015; Sigmon et al., 2013; Weiss and Rao, 2017) involved small numbers of patients in an experimental setting (Dakwar and Kleber, 2015) or extrapolated from methadone literature (Calsyn et al., 2006; Mannelli et al., 2009). Some characteristics that have been associated with successful methadone taper include longer length of time in treatment and a slow taper schedule (Calsyn et al., 2006; Magura and Rosenblum, 2001; Milby, 1988; Nosyk et al., 2012); however, completed methadone tapers are rare, with one large 10-year cohort study finding only 2.5% of patients successfully tapered off (Nosyk et al., 2012).

Tapering off buprenorphine among long-term maintenance patients has only been described in relatively smaller trials over shorter periods, and has not highlighted any specific characteristics associated with taper completion.(Alford et al., 2011; Dakwar and Kleber, 2015; Fiellin et al., 2008) The goal of this study is to describe a long term experience among a large cohort of patients on buprenorphine within a highly structured collaborative-care model, the proportion of patients who taper off buprenorphine, the patient characteristics of those who taper, and the proportion who return to buprenorphine treatment after the taper.

2. Materials and methods

This study is part of a larger retrospective cohort study (DROP, Disenrollment and Reengagement in an OBOT Program) (Weinstein et al., 2017), which evaluated patients maintained on buprenorphine at Boston Medical Center’s Office Based Addiction Treatment (OBAT) Program from January 1, 2002 to February 28, 2014.

2.1. Study setting

This Office Based Addiction Treatment (OBAT) Program integrates primary care and opioid use disorder treatment using a collaborative care nurse care manager model and has been previously well described and dubbed the Massachusetts Model (Substance Abuse and Mental Health Services Administration, 2014; LaBelle et al., 2016; Alford et al., 2011; Weinstein et al., 2017). This clinic supports patients in indefinite maintenance on buprenorphine and in no way encourages patients to taper as long as patients are meeting their treatment goals. There are no formal taper protocols and if a patient chooses to electively taper off buprenorphine he/she does so at an individualized pace.

2.2. Data sources and collection

Data was abstracted from the Electronic Medical Record (EMR) including OBATtemplate nursing notes, outpatient physician notes, active medication lists and laboratory data. Manual chart review of de-identified clinic notes, by trained reviewers (D.H. and H.K.) and a physician (Z.M.W.), was used to detail patients’ substance use history, prior opioid use disorder (OUD) treatment, and reasons for leaving clinic.

2.2.1. Primary outcome.

The primary outcome was the number of patients who completed a voluntary taper and time to taper off of buprenorphine maintenance. Treatment began with buprenorphine induction, as documented by receipt of the first buprenorphine prescription. Treatment ended when the patient had neither an active buprenorphine prescription nor any clinic contact for at least 60 days. Participants were censored at loss to follow-up or end of study, whichever occurred first. The study included prescription data collected through February 2014, with the final date of the medication extrapolated from the end date of the prescription, to account for refills. At the end of each treatment period, the reason for disengagement was coded by the research team, based on a review of the three notes prior to the end of buprenorphine treatment and the three notes after the end of buprenorphine treatment (e.g., primary care visits or telephone notes), if available.

Taper was one of eleven possible reasons for disengagement coded based on consensus by the research team using content analysis (Alford et al., 2011; Fingerhood et al., 2014; Gryczynski et al., 2014; Weinstein et al., 2017). For patients to be documented as tapering off, this had to be explicitly documented in the medical record. The patients categorized as tapering off did not include patients who left care for reasons unknown and did not include patients who had a documented relapse prior to discontinuing buprenorphine.

The taper event was further characterized as either 1) medically supervised (i.e., clinic staff wrote out taper protocol and prescriptions) or 2) unsupervised (i.e., patient self-tapered and did not notify the clinic until near or after taper completion). All of the unsupervised taper patients explicitly reported self-tapering to a medical professional who documented the taper in the medical record. However, this was a heterogeneous group of patients, including patients who continued regular medical care and reported in these visits they had tapered off buprenorphine on their own, as well as patients who stopped following up for months or years in OBAT, but upon return to care reported that they had tapered off.

As the clinical setting of this study supported indefinite maintenance and encouraged patients to complete at least six months of maintenance prior to taper (Alford et al., 2011), a second analysis of time to taper was performed, but was limited to patients who had completed at least six months of buprenorphine treatment, to focus on this subgroup of patients, which might be most likely to successfully complete a taper.

2.2.2. Secondary outcome.

The secondary outcome was time from voluntary taper off of buprenorphine maintenance to re-engagement in care with the same OBAT treatment program after taper. Re-engagement in care at OBAT was defined as receiving a new induction prescription after a disengagement event. Similar to the primary analysis as described above, participants were censored at loss to follow-up or end of study, whichever occurred first.

2.2.3. Exploration of patient factors.

The relationship between individual patient characteristics and time to taper was assessed, as well as the relationship of patient characteristics and time to re-engagement among those who tapered. Patient characteristics included gender, race/ethnicity, education, employment status, history of heroin use, prior treatment with methadone or buprenorphine, median time in treatment before taper and buprenorphine dose during treatment. Patients’ lowest prescribed daily buprenorphine dose that may have occurred as part of a supervised taper was grouped into three categories: >8mg, 8–5mg or ≤4 mg, to capture patients who had attempted to complete a taper. The average daily dose for patients in the OBAT program is 16mg per day (Weinstein et al., 2017), thus having a dose less than 8mg represents a possible taper, and a dose less than 4mg represent a likely taper.

2.3. Statistical analyses

Descriptive statistics were used to characterize all study participants, overall as well as stratified by voluntary taper status, based on age, gender, race/ethnicity, education, employment status, history of heroin use, prior treatment with methadone or buprenorphine, median time in treatment before taper and buprenorphine dose. The proportion of all clinic patients who tapered by the end of the study period (full follow-up of 2361 days) was estimated using the Kaplan-Meier estimator. These analyses take into account the different lengths of follow-up among patients and incorporate the partial information from ‘censored’ subjects who did not have complete follow-up (Rich et al., 2010). Separate unadjusted Cox regression models were used to assess the association between individual patient characteristics and time to taper. The exact method was used to handle tied event times (Kalbfleisch and Prentice, 2002). Adjusted analyses were performed using a priori selected covariates: age, history prior buprenorphine treatment and lowest average buprenorphine dose (divided into >8, 8–5mg or ≤4 mg). Secondary analyses were conducted for time to taper on the subset of patients (n=860) who were engaged in care for at least six months, to capture patients who were most likely to be successful in their taper.

For the secondary outcome of time from voluntary taper to re-engagement in care, the proportion of patients who re-engaged after tapering was estimated with the Kaplan Meier estimator. Descriptive statistics were used to characterize those patients who tapered and then reengaged. Multivariable analyses were not performed given the small sample size. All analyses were completed using SAS 9.4 software (Cary, NC). The Boston University Medical Campus Institutional Review Board approved this study.

3. Results

3.1. Frequency of taper

The study included 1308 patients with a median follow-up of 316 days. Of these 1308 patients, 48 were observed to taper off buprenorphine during the study period, with an estimated proportion of 15% (95%CI: 10%−21%) after the full follow-up period of 2361 days based on Kaplan Meier analyses. These 48 patients were in treatment a median of 490 days (IQR: 242–1402) before completing the taper. Less than half of the tapers, 45.8% (22/48), were medically supervised (Table 1).

Table 1:

Characteristics of the Patients (n=48) Who Tapered Off Buprenorphine Maintenance and Hazard Ratio of Time to Taper

Characteristicsa Overall N=1308 Tapered Off N=48 Did Not
Taper Off
N=1260		 Unadjusted
Hazard Ratiob
(95% CI)		 Adjusted
Hazard
Ratioc (95%
CI)
Age at Enrollment, Mean (SD) 38 (11) 38 (13) 38 (11) 1.00
(0.97,1.02)		 1.00 (0.98,
1.03)
Age of First Opioid Use, Mean (SD)
N=1232		 22 (8) 22 (8) 22 (8) 0.99 (0.96,
1.03)
Female Gender, N(%) 501 (38.8%) 18 (37.5%) 483 (38.8%) 0.75 (0.42,
1.34)
White Race/Ethnicity
N=1270		 860 (67.7%) 35 (74.5%) 825 (67.5%) 0.97 (0.50,
1.87)
<High School Education
N=1045		 374 (35.8%) 12 (30.8%) 362 (36.0%) 0.76 (0.38,
1.49)
Employed/Student N=1240 293 (23.6%) 17 (37.0%) 276 (23.1%) 1.43 (0.78,
2.60)
History of Heroin Use 1138 (87.0%) 37 (77.1%) 1101 (87.4%) 0.78 (0.40,
1.54)
Prior Buprenorphine Treatment 442 (33.8%) 20 (41.7%) 422 (33.5%) 1.24 (0.70,
2.20)		 1.20 (0.66,
2.19)
Prior Methadone Treatment 424 (32.4%) 12 (25.0%) 412 (32.7%) 0.77 (0.40,
1.48)
Lowest
Buprenorphine
Dose		 >8mg 899 (68.7%) 20 (41.7%) 879 (69.8%) Reference Reference
5–8mg 270 (20.6%) 13 (27.1%) 257 (20.4%) 1.62 (0.81,
3.26)		 1.65 (0.82,
3.35)
≤4mg 139 (10.6%) 15 (31.3%) 124 (9.8%) 2.75 (1.40,
5.38) 		2.73 (1.39,
5.37)
Open in a new tab
a

All variables represent data from an N=1308 unless otherwise noted; history of heroin-reported use of heroin, with or without opioid pills, as compared to use of opioid pills only; prior buprenorphine treatment – patient self-report of prior buprenorphine treatment with any buprenorphine provider on initial intake

b

Unadjusted Hazard Ratios represent: female vs. male gender; White vs. non-White race/ethnicity; <high school education vs. ≥ high school; employed/student vs. unemployed/retired

c

Adjusted Hazard Ratio model includes the variables: age at enrollment, prior buprenorphine treatment and lowest buprenorphine dose

3.2. Characteristics of patients who tapered

Among the 48 patients who tapered, the mean age of first opioid use was 22 years, while the mean age of enrollment in OBAT clinic was 38 years. The majority were of white race/ethnicity (74.5%) and approximately a third were female (37.5%), had not completed high school (30.8%) and were employed or a student (37.0%). The majority (77.1%) had used heroin in the past as opposed to only opioid pills. Over a third (41.7%) of the patients had been on buprenorphine before and a quarter (25.0%) had been previously treated in a methadone clinic. In the adjusted Cox model, there was no significant association between age at OBAT enrollment (Adjusted Hazard Ratio [AHR] 1.00, 95% CI: 0.98 – 1.03) or prior buprenorphine treatment (AHR: 1.20, 95% CI: 0.66 – 2.19) and time to taper (Table 1).

As for the lowest buprenorphine dose achieved, over a third of patients who tapered (41.7%) had a lowest mean daily dose >8mg prior to taper completion, less than a third achieved a minimum dose between 8mg and 5mg (27.1%) and the final third (31.3%) of patients had a minimum daily dose ≤4mg prior to completing the taper. Patients had a dose of ≤ 8mg for a median of 40 days (IQR: 0–222) and a dose of ≤ 4mg for a median of 0 days (IQR: 0–67) prior to taper. In the adjusted Cox model, patients achieving a minimum mean daily dose ≤4mg was associated with a shorter time to taper (AHR: 2.73, 95% CI: 1.39 – 5.37) as compared to patients whose lowest mean dose was >8mg (Table 1).

3.2.1. Supervised taper patient characteristics.

A supervised taper occurred in 22 of the 48 patients who tapered off buprenorphine. Supervised taper patients had a median length of time in treatment of 774 days (IQR 271, 1579) as compared to 433 days (IQR 183, 799) among the unsupervised taper group. A higher proportion of the supervised tapers were female (45.5% vs 30.8%) and were either employed or a student (52.4% vs. 24.0%), while a lower proportion of supervised taper patients were of white race/ethnicity (66.7% vs 80.8%). Supervised taper patients were prescribed a buprenorphine dose less than 8mg for a median of 161 days, compared to a median of 0 days for the unsupervised group. Among the patients with a supervised taper, 45.5% had a mean daily dose ≤4mg prior to taper completion, compared to 19.2% among those with an unsupervised taper (Table 2).

Table 2:

Characteristics of Patients with a Supervised as Compared to Unsupervised Taper Off of Buprenorphine Treatment

Characteristicsa Overall N= 48 Supervised
N=22		 Unsupervised N=26
Time to taper, Median (IQR) 490 (242, 1402) 774 (271, 1579) 433 (183, 799)
Age at Enrollment, Mean (SD) 38 (13) 39 (13) 37 (12)
Age of First Opioid Use
N=46		 22 (8) 24 (10) 21 (7)
Female Gender N (%) 18 (37.5%) 10 (45.5%) 8 (30.8%)
White Race/Ethnicity
N=47		 35 (74.5%) 14 (66.7%) 21 (80.8%)
<High School Education
N=39		 12 (30.8%) 6 (35.3%) 6 (27.3%)
Employed/Student N=46 17 (37.0%) 11 (52.4%) 6 (24.0%)
History of Heroin Use 37 (77.1%) 17 (77.3%) 20 (76.9%)
Prior Buprenorphine Treatment 20 (41.7%) 7 (31.8%) 13 (50.0%)
Prior Methadone Treatment 12 (25.0%) 6 (27.3%) 6 (23.1%)
Number of Days with
Buprenorphine Dose <8, Median (IQR)		 40 (0, 222) 161 (0, 398) 0 (0,104)
Lowest
Buprenorphine
Dose, N (%)		 >8mg 20 (41.7%) 6 (27.3%) 14 (53.8%)
5–8mg 13 (27.1%) 6 (27.3%) 7 (26.9%)
≤4mg 15 (31.3%) 10 (45.5%) 5 (19.2%)
Open in a new tab
a

All variables represent data from an N=48 unless otherwise noted; history of heroin-reported use of heroin, with or without opioid pills, as compared to use of opioid pills only; prior buprenorphine treatment – patient self-report of prior buprenorphine treatment with any buprenorphine provider on initial intake

3.2.2. Subgroup of patients in care for at least six months.

There were 860 patients who were engaged in treatment with the OBAT program for at least 6 months, among whom only 39 patients were observed to taper off (estimated proportion 14%, 95% CI: 9%−20%, based on Kaplan-Meier analyses). In the Cox regression analyses of this subgroup, there were no statistically significant associations between patient demographics and tapering, except between a mean daily dose ≤4mg and shorter time to taper (AHR 2.79, 95% CI: 1.32 – 5.90) (data not shown).

3.3. Re-engagement in OBAT

Thirteen of the 48 patients who tapered off of buprenorphine subsequently re-engaged in buprenorphine treatment (estimated proportion by end of follow-up: 61%, 95%CI: 27%−96%, based on Kaplan-Meier analyses, median 490 days of follow-up time among the 48 patients). The median time to re-engagement was 1062 days (IQR: 565 – 1783). Of the 13 patients who reengaged, the majority, 76.9% (10/13), had completed an unsupervised taper in their prior engagement and only 3 (23.1%) had completed a supervised taper in their prior engagement (Table 3).

Table 3:

Characteristics of Patients (n=13) who Re-engaged with Buprenorphine Treatment after Tapering Off

Characteristicsa Re-engaged N=13 Did not Reengage N=35
Medically Supervised Taper 3 (23.1%) 19 (54.3%)
Age at Enrollment, Mean (SD) 36 (12) 38 (13)
Age of First Opioid Use
N=46		 23 (10) 22 (8)
Female Gender N (%) 4 (30.8%) 14 (40.0%)
White Race/Ethnicity
N=47		 10 (76.9%) 25 (73.5%)
<High School Education
N=39		 4 (33.3%) 8 (29.6%)
Employed/Student N=46 5 (38.5%) 12 (36.4%)
History of Heroin Use 10 (76.9%) 27 (77.1%)
Prior Buprenorphine Treatment 7 (53.8%) 13 (37.1%)
Prior Methadone Treatment 2 (15.4%) 10 (28.6%)
Lowest
Buprenorphine
Dose		 >8mg 8 (61.5%) 12 (34.3%)
5–8mg 3 (23.1%) 10 (28.6%)
≤4mg 2 (15.4%) 13 (37.1%)
Open in a new tab
a

All variables represent data from an N=48 unless otherwise noted; history of heroin-reported use of heroin, with or without opioid pills, as compared to use of opioid pills only; prior buprenorphine treatment – patient self-report of prior buprenorphine treatment with any buprenorphine provider on initial intake

In terms of prescribed buprenorphine dosing, the majority of patients (61.5%) who reengaged had their lowest dose >8mg as compared to only a third (34.3%) of those who tapered but did not re-engage. A minority (15.4%) of those who tapered and then re-engaged had achieved an average daily dose ≤4mg prior to taper completion, as compared to over a third (37.1%) of those who tapered but did not re-engage (Table 3).

4. Discussion

Despite the fact that many patients desire to taper off buprenorphine (Teruya et al., 2014; Woodcock et al., 2015), a small minority (an estimated 15%) had a documented taper among this large sample of patients over a 12 year period. This rate was similar (14%) even among the subgroup of patients who were in care for at least six months and may have been more likely to successfully complete a taper. In addition, 61% of those who tapered re-entered treatment during the study follow-up, which is likely an underestimate of the total number who returned to treatment, as this is a single site study. This low rate of completed tapers and high rate of return to treatment likely reflect the challenge of tapering completely off opioid agonist treatments due to symptoms of withdrawal and cravings (Calsyn et al., 2006; Nosyk et al., 2012). However, these numbers likely represent an underreporting of the frequency of self-taper. For example, within the larger patient cohort, approximately a third (30.6%) of patients had an unknown reason for leaving treatment, which may have included unreported self-tapers (Weinstein et al., 2017).

Among the few patients who tapered off buprenorphine, more than half (54.8%) tapered off on their own, without direct clinic supervision. It is possible, as this clinic is structured for indefinite maintenance, that patients did not feel supported by the medical team to taper, and so did so alone.(Winstock et al., 2011) This is suggested by the fact that only a little more than half of patients (58.4%) had a formal prescription for a daily dose ≤ 8mg and only a third (31.3%) had a prescription for a daily dose of ≤ 4mg despite reporting that their treatment ended with tapering completely off, implying many patients likely reduced their dose further on their own. Most tapers were not done slowly, with patients being on a prescribed daily dose less than 8mg for a median of only 40 days prior to taper completion. This is not in line with recommendations that slower tapers are more likely to result in clinical stability after opioid agonist treatment (Bentzley et al., 2015; Calsyn et al., 2006; Chang et al., 2017; Nosyk et al., 2012) and again may speak to the tension between patient and provider preferences for taper, especially as maintenance is much more successful than taper in achieving positive clinical outcomes (Weiss and Rao, 2017). Notably, however, those with a supervised taper had much longer time in treatment overall (median 774 vs .443 days) and time with a prescribed buprenorphine dose less than 8mg (median 161 vs. 0 days). Thus, it is important to engage patients in a collaborative taper plan, if possible, to help patients have a more gradual taper. There is room for qualitative work to further explore patient and provider attitudes and experiences about tapering, so as to enhance communication and a more collaborative approach. In addition, in this sample there was a wide range of taper lengths, reflecting the lack of formal guidelines for taper schedules for long-term maintenance patients to assist providers in making recommendations about tapering buprenorphine. The range of taper lengths may also reflect variability in patient preference (Amato et al., 2011; Teruya et al., 2014), as well as variability in underlying tolerance and severity of the patient’s opioid use disorder (Bentzley et al., 2015).

The overall sample of patients, as well as those who tapered or re-engaged, are similar to patients in general who engage in opioid agonist treatment, as this group has historically been younger, male and of white race/ethnicity (Substance Abuse and Mental Health Services Administration, 2015). Given the low frequency of tapers, this study was likely underpowered to detect significant demographic characteristics unique to those who tapered.

In the adjusted Cox model, having a daily dose of ≤4 mg was associated with a shorter time to taper as compared to those who never were maintained on these lower doses. This result was found both in the full sample, as well as the subgroup of patients with at least six months of treatment. This result suggests that a formal provider-patient collaboration with prescribed dose decreases, may allow patients to complete a taper more successfully if the patient desires. Providers should be actively eliciting patients’ plans around taper and offering to proactively partner to help patients be more successful in their goal. As in all other clinical care, a strong therapeutic alliance and shared decision-making are supportive of successful addiction treatment.

An estimated 61% of those who taper off returned to buprenorphine within the study period. This proportion is likely an under-estimate as this only captures the patients who returned to the same OBAT program and does not account for those who engaged in care elsewhere. This high rate of re-engagement suggests that the majority of patients may truly require indefinite maintenance and that taper may not be a feasible goal. While providers may support a patient’s stated goal to taper off buprenorphine as laudable, just as provider would support diet and lifestyle changes for a patient with diabetes in an effort to come off insulin therapy, this is with the acknowledgement that this goal will be possible for some but not for others. Providers must continue to emphasize to patients the myriad of benefits associated with treatment retention, such as improved quality of life (Schackman et al., 2012), in addition to decreased morbidity and mortality (Bart, 2012; Bentzley et al., 2015) and promote the view that long-term recovery, like long-term good glycemic control, is the ultimate goal, whether with or without medication. In addition, further research is needed to assess barriers and facilitators to treatment retention, from the patient, provider and health care system perspective in order to enhance retention.

With the realization that many patients may not be able to successfully taper off of buprenorphine, the medical community must work to address any and all barriers to long-term maintenance. This likely involves addressing systemic barriers such as ensuring that there are flexible and variable clinic structures that can meet the needs of patients in early recovery, as well as those stable on long-term medication (Gryczynski et al., 2014). Patients who are engaged in treatment may have re-gained employment or re-connected with family and be less able to attend multiple, frequent or daytime appointments and thus ensuring integrated patient-centered care is essential to facilitate retention. Studies exploring different models of care, including staged models that help long-term patients graduate to a more flexible program may be necessary. In addition, it is essential that the public health community addresses the stigma of medications for OUD, as this stigma motivates many patients to discontinue treatment (Appel and Oldak, 2007).

Those who returned to buprenorphine did so after years (median 1062 days). This length of time out of care is concerning if it represents a relapse, given the risk of overdose and death (Bentzley et al., 2015). Buprenorphine clinics may benefit from enhanced case-management resources to continue to follow patients even after they discontinue treatment and facilitate rapid re-engagement if needed. Rapid access for all patients is essential to minimize time on waiting lists, which has been shown to be fatal (Peles et al., 2013). In addition, the use of shared decision-making and a strong therapeutic alliance throughout treatment may allow patients feel more safe and comfortable re-engaging in care after tapering.

4.1. Strengths and limitations

Although this study included a long follow-up time, documented tapers were uncommon, limiting the extent and power of the current analyses. As a single site retrospective study, we were unable to assess for treatment provided at other clinics. As many patients left for unknown reasons from this clinic, this study also likely represents an undercount of unsupervised tapers. In addition, the documentation of an unsupervised taper relied on both patient self-report and provider documentation of this in the medical record. Patients may report a taper, but this may include a wide range of dose changes, from an abrupt stopping of the medication, to the slow dose reduction over months, and so thus the term taper in the context of these patients represents their individual understanding of the term taper, and not a single definition. Nonetheless, the large number of patients, long follow-up time, consistent collection of patient data and wide range of variables offer a unique perspective on long-term buprenorphine treatment outcomes.

4.2. Conclusions

Despite the fact that many patients desire to taper off buprenorphine, a small minority had a documented taper. Among those who tapered, more than half did so unsupervised by the clinic and a majority of those who taper off returned to buprenorphine treatment within a few years. The addiction field needs additional research about patients who desire to taper off buprenorphine and tools to help them do so successfully. In addition, more longitudinal supports are needed to facilitate rapid re-engagement for patients when they have a need to return to pharmacological treatment. As completely discontinuing buprenorphine is unlikely to be a feasible goal for many patients, it is essential to address all barriers to long-term treatment, especially the stigma of medications for opioid use disorder.

Highlights.

  • Completing a taper off long-term buprenorphine treatment is uncommon.

  • Tapering off buprenorphine is often done by the patients on their own.

  • A majority of those who taper off returned to buprenorphine within two years.

Acknowledgements

The project described was supported in part by grant R25DA033211 from the National Institute on Drug Abuse, grant R25DA0123582 from the National Institute on Drug Abuse, grant T32AI052074 from the National Institute of Allergy and Infectious Diseases and grant 1UL1TR001430 from the National Center For Advancing Translational Sciences.

Footnotes

Role of the Funding Source

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute on Drug Abuse, National Institute of Allergy and Infectious Diseases or the National Center For Advancing Translational Sciences. No funding sources had any role in study design; in the collection, analysis and interpretation of data; in the writing of the report; or in the decision to submit the manuscript for publication.

Author Disclosures

Conflict of Interest

No conflict declared.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Alford DP, LaBelle CT, Kretsch N, Bergeron A, Winter M, Botticelli M, Samet JH, 2011. Collaborative care of opioid-addicted patients in primary care using buprenorphine: Five-year experience. Arch. Intern. Med 171 10.1001/archinternmed.2010.541. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Amato L, Minozzi S, Davoli M, Vecchi S, 2011. Psychosocial and pharmacological treatments versus pharmacological treatments for opioid detoxification, in: Cochrane Database of Systematic Reviews. John Wiley and Sons, Ltd. [DOI] [PubMed] [Google Scholar]
  3. Appel PW, Ellison AA, Jansky HK, Oldak R, 2004. Barriers to enrollment in drug abuse treatment and suggestions for reducing them: Opinions of drug injecting street outreach clients and other system stakeholders. Am. J. Drug Alcohol Abuse 30, 129–153. [DOI] [PubMed] [Google Scholar]
  4. Appel PW, Oldak R, 2007. A preliminary comparison of major kinds of obstacles to enrolling in substance abuse treatment (AOD) reported by injecting street outreach clients and other stakeholders. Am. J. Drug Alcohol Abuse 33, 699–705. 10.1080/00952990701522641. [DOI] [PubMed] [Google Scholar]
  5. Bart G, 2012. Maintenance medication for opiate addiction: The foundation of recovery. J. Addict. Dis 31, 207–225. 10.1080/10550887.2012.694598. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bentzley BS, Barth KS, Back SE, Book SW, 2015. Discontinuation of buprenorphine maintenance therapy: Perspectives and outcomes. J. Subst. Abuse Treat 52, 48–57. 10.1016/j.jsat.2014.12.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Calsyn DA, Malcy JA, Saxon AJ, 2006. Slow tapering from methadone maintenance in a program encouraging indefinite maintenance. J. Subst. Abuse Treat 30, 159–163. 10.1016/j.jsat.2005.11.007. [DOI] [PubMed] [Google Scholar]
  8. Chang DC, Klimas J, Wood E, Fairbairn N, 2017. A case of opioid overdose and subsequent death after medically supervised withdrawal: The problematic role of rapid tapers for opioid use disorder. J. Addict. Med 12, 80–83. 10.1097/ADM.0000000000000359. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Dakwar E, Kleber HD, 2015. Naltrexone-facilitated buprenorphine discontinuation: A feasibility trial. J. Subst. Abuse Treat 53, 60–63. 10.1016/j.jsat.2015.01.004. [DOI] [PubMed] [Google Scholar]
  10. Dunn KE, Saulsgiver KA, Miller ME, Nuzzo PA, Sigmon SC, 2015. Characterizing opioid withdrawal during double-blind buprenorphine detoxification. Drug Alcohol Depend. 151, 47–55. 10.1016/j.drugalcdep.2015.02.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Fiellin DA, Moore BA, Sullivan LE, Becker WC, Pantalon MV, Chawarski MC, Barry DT, O’Connor PG, Schottenfeld RS, 2008. Long-term treatment with buprenorphine/naloxone in primary care: Results at 2–5 years. Am. J. Addict 17, 116–120. 10.1080/10550490701860971. [DOI] [PubMed] [Google Scholar]
  12. Fingerhood MI, King VL, Brooner RK, Rastegar DA, 2014. A comparison of characteristics and outcomes of opioid-dependent patients initiating office-based buprenorphine or methadone maintenance treatment. Subst. Abuse 35, 122–126. 10.1080/08897077.2013.819828. [DOI] [PubMed] [Google Scholar]
  13. Gryczynski J, Mitchell SG, Jaffe JH, O’Grady KE, Olsen YK, Schwartz RP, 2014. Leaving buprenorphine treatment: Patients’ reasons for cessation of care. J. Subst. Abuse Treat 46, 356–361. 10.1016/j.jsat.2013.10.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Kalbfleisch JD, Prentice RL, 2002. The Statistical Analysis of Failure Time Data, 2nd edition. ed. Wiley-Interscience, Hoboken, N.J. [Google Scholar]
  15. Kleber HD, 2007. Pharmacologic treatments for opioid dependence: Detoxification and maintenance options. Dialogues Clin. Neurosci 9, 455–470. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Kraus ML, Alford DP, Kotz MM, Levounis P, Mandell TW, Meyer M, Salsitz EA, Wetterau N, Wyatt SA, 2011. Statement of the American Society of Addiction Medicine consensus panel on the use of buprenorphine in office-based treatment of opioid addiction. J. Addict. Med 5, 254–263. 10.1097/ADM.0b013e3182312983. [DOI] [PubMed] [Google Scholar]
  17. LaBelle CT, Han SC, Bergeron A, Samet JH, 2016. Office-Based Opioid Treatment with Buprenorphine (OBOT-B): Statewide implementation of the Massachusetts collaborative care model in community health centers. J. Subst. Abuse Treat 60, 6–13. 10.1016/j.jsat.2015.06.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Magura S, Rosenblum A, 2001. Leaving methadone treatment: lessons learned, lessons forgotten, lessons ignored. Mt. Sinai J. Med. N. Y 68, 62–74. [PubMed] [Google Scholar]
  19. Mannelli P, Patkar AA, Peindl K, Gorelick DA, Wu L-T, Gottheil E, 2009. Very low dose naltrexone addition in opioid detoxification: A randomized, controlled trial. Addict. Biol 14, 204–213. 10.1111/j.1369-1600.2008.00119.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Milby JB, 1988. Methadone maintenance to abstinence. How many make it? J. Nerv. Ment. Dis 176, 409–422. [DOI] [PubMed] [Google Scholar]
  21. Nosyk B, Sun H, Evans E, Marsh DC, Anglin MD, Hser Y-I, Anis AH, 2012. Defining dosing pattern characteristics of successful tapers following methadone maintenance treatment: Results from a population-based retrospective cohort study. Addiction 107, 1621–1629. 10.1111/j.1360-0443.2012.03870.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Peles E, Schreiber S, Adelson M, 2013. Opiate-dependent patients on a waiting list for methadone maintenance treatment are at high risk for mortality until treatment entry. J. Addict. Med 7, 177–182. 10.1097/ADM.0b013e318287cfc9. [DOI] [PubMed] [Google Scholar]
  23. Peterson JA, Schwartz RP, Mitchell SG, Reisinger HS, Kelly SM, O’Grady KE, Brown BS, Agar MH, 2010. Why don’t out-of-treatment individuals enter methadone treatment programmes? Int. J. Drug Policy 21, 36–42. 10.1016/j.drugpo.2008.07.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Rich JT, Neely JG, Paniello RC, Voelker CC, Nussenbaum B, Wang EW, 2010. A practical guide to understanding Kaplan-Meier curves. Otolaryngol. Head Neck Surg 143, 331–336. 10.1016/j.otohns.2010.05.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Schackman BR, Leff JA, Polsky D, Moore BA, Fiellin DA, 2012. Cost-effectiveness of long-term outpatient buprenorphine-naloxone treatment for opioid dependence in primary care. J. Gen. Intern. Med 27, 669–676. 10.1007/s11606-011-1962-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Sigmon SC, Dunn KE, Saulsgiver K, Patrick ME, Badger GJ, Heil SH, Brooklyn JR, Higgins ST, 2013. A randomized, double-blind evaluation of buprenorphine taper duration in primary prescription opioid abusers. JAMA Psychiatry 70, 1347–1354. 10.1001/jamapsychiatry.2013.2216. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Substance Abuse and Mental Health Services Administration (SAMHSA), 2015. Treatment Episode Data Set (TEDS) 2003–2013: State admissions to substance abuse treatment services. BHSIS Series S-80, HHS Publication No. (SMA) 16–4964.
  28. Substance Abuse and Mental Health Services Administration (SAMHSA), 2014. Medicaid coverage and financing of medications to treat alcohol and opioid use disorders. HHS publication no. SMA-14–4854.
  29. Teruya C, Schwartz RP, Mitchell SG, Hasson AL, Thomas C, Buoncristiani SH, Hser Y-I, Wiest K, Cohen AJ, Glick N, Jacobs P, McLaughlin P, Ling W, 2014. Patient perspectives on buprenorphine/naloxone: A qualitative study of retention during the Starting Treatment with Agonist Replacement Therapies (START) study. J. Psychoactive Drugs 46, 412–426. 10.1080/02791072.2014.921743. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Weinstein ZM, Kim HW, Cheng DM, Quinn E, Hui D, Labelle CT, Drainoni M-L, Bachman SS, Samet JH, 2017. Long-term retention in Office Based Opioid Treatment with buprenorphine. J. Subst. Abuse Treat 74, 65–70. 10.1016/j.jsat.2016.12.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Weiss RD, Rao V, 2017. The prescription opioid addiction treatment study: What have we learned? Drug Alcohol Depend. 173, S48–S54. 10.1016/j.drugalcdep.2016.12.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Winstock AR, Lintzeris N, Lea T, 2011. “Should I stay or should I go?” Coming off methadone and buprenorphine treatment. Int. J. Drug Policy 22, 77–81. 10.1016/j.drugpo.2010.08.001. [DOI] [PubMed] [Google Scholar]
  33. Woodcock EA, Lundahl LH, Greenwald MK, 2015. Predictors of buprenorphine initial outpatient maintenance and dose taper response among non-treatment-seeking heroin dependent volunteers. Drug Alcohol Depend. 146, 89–96. 10.1016/j.drugalcdep.2014.11.016. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES