Weight change among patients engaged in medication treatment for opioid use disorder: a scoping review

Abstract

Background:

Medication treatment for opioid use disorder (MOUD) is an instrumental tool in combatting opioid use and overdose. Excess weight gain associated with MOUD initiation is a potential barrier that is not well understood.

Objectives:

Conduct a scoping review of available studies investigating the effect of MOUD on weight.

Methods:

Included studies consisted of adults taking any type of MOUD (e.g. methadone, buprenorphine/naloxone, naltrexone) with data on weight or body mass index for at least two time points. Evidence was synthesized using qualitative and descriptive approaches, and predictors of weight gain including demographics, comorbid substance use, and medication dose were examined.

Results:

Twenty-one unique studies were identified. Most studies were uncontrolled cohort studies or retrospective chart reviews testing the association between methadone and weight gain (n = 16). Studies examining 6 months of methadone treatment reported weight gain ranging from 4.2 to 23.4 pounds. Women appear to gain more weight from methadone than men, while patients using cocaine may gain less. Racial and ethnic disparities were largely unexamined. Only three case reports and two nonrandomized studies examined the effects of either buprenorphine/naloxone or naltrexone, and potential associations with weight gain were not clear.

Conclusion:

The use of methadone as an MOUD appears to be associated with mild to moderate weight gain. In contrast, there is little data supporting or refuting weight gain with buprenorphine/naloxone or naltrexone. Providers should discuss the potential risk for weight gain with patients as well as prevention and intervention methods for excess weight gain.

Introduction

Opioid use disorder (OUD) greatly increases risk of morbidity and mortality, most notably overdose. In 2021, over 75,000 people died from opioid overdose (1). An instrumental tool in combatting opioid use and overdose are medications such as methadone and buprenorphine plus naloxone, collectively referred to as medication treatments for opioid use disorder (MOUD). MOUD have been shown to reduce opioid and other drug use, criminal activity, HIV risk behaviors and transmission, opioid overdose, and all-cause mortality (2). Despite these positive effects, MOUD can carry risk of certain side effects, including weight gain.

Preventing obesity is an important treatment target among a multitude of patient groups as obesity broadly confers increased risk of both morbidity and mortality independent of OUD (3,4). Certain medical comorbidities associated with MOUD could be exacerbated by obesity, including chronic pain (5), sleep apnea (6), and diabetes (7). Furthermore, obesity and the stigma associated with obesity are also associated with poorer quality of life (8,9). Weight gain and consequent obesity may also represent a barrier to MOUD treatment retention, but this has yet to be empirically investigated. Data on psychotropic medication generally suggests the potential for reduced adherence if the medication causes significant weight gain (10,11). Overall, the data indicate that understanding the association between different types of MOUD and weight gain or obesity may be critical in improving long-term outcomes associated with MOUD treatment.

The goals of the current study were: 1) to identify all available research examining the association between weight and MOUD treatment; 2) characterize the available research with respect to study design and sample composition; 3) synthesize weight outcomes across studies using qualitative and descriptive approaches, with particular attention paid to studies that could identify the percent of patients that transition from either underweight or healthy weight to overweight or obese; and 4) summarize available data on predictors of weight gain including gender, race and ethnicity, and medication dose using qualitative and descriptive approaches.

Methods

The reporting of this scoping review is guided by the Joanna Briggs Institute framework (12) and the PRISMA Extension for Scoping Reviews (13).

Eligibility criteria

The following inclusion criteria were used to identify sources of evidence: 1) studies of adults with OUD; 2) studies with participants taking at least one medication treatment for opioid use disorder including methadone or dolophine, buprenorphine and naloxone combinations including oral, sublingual, or injectable forms, and naltrexone in oral or injectable form; 3) studies including data on weight or body mass index (BMI) at two or more time-points; and 4) studies written in English or with an available English translation. Studies were excluded if they exclusively focused on a pregnant or perinatal sample or if they were taking methadone, buprenorphine plus naloxone, or naltrexone for a condition other than OUD.

Data sources and data searches

The search strategy was developed by an experienced medical librarian (MCF) and peer-reviewed by a second medical librarian not otherwise associated with the project. The search strategy included a combination of controlled vocabulary terms (e.g. medical subject headings; MeSH) and related synonyms on the concepts of “MOUD” and “weight gain.” A total of 6 databases were searched including: MEDLINE, Embase, APA PsycInfo, APA PsycExtra, Web of Science, and Cochrane Library. In addition to published academic literature, a targeted Google search strategy was used. No date or language restrictions were applied. Regarding language, articles written in a language other than English were not initially excluded to allow for further investigation if possible translations could be identified. See Supplement 1 for search strategies.

Study selection

Study tracking and data extraction were managed through Covidence software. The first stage was title and abstract review using provisional inclusion and exclusion. For example, title and abstracts may not have indicated if weight was measured at two timepoints if the primary focus of the article was something other than weight. All titles and abstracts were reviewed by two team members (GMG, RL) trained by the first author (MMC), who also resolved any conflicts that arose. The next step involved full-text review which was completed by the first author (MMC).

Data extraction and summary

A single author (MMC) extracted and compiled data using Covidence and Microsoft Excel, including author, publishing year, journal, type of publication, country in which data collection occurred, study design, sample size, participant characteristics (gender, race, age), inclusion and exclusion criteria, type of MOUD, and any reported weight outcomes, including raw weight gain, BMI change, or BMI category change. For BMI, categories outlined by the CDC (14) were used where a BMI < 18.5 is underweight, 18.5–24.99 is healthy weight, 25.00–29.99 is overweight, and ≥30.00 is obesity.

Results

The final search retrieved a total of 3,694 references which were pooled in EndNote 20 and deduplicated [www.endnote.com]. This set was uploaded to Covidence [www.covidence.org] for screening, which identified additional duplicates, leaving 2185 for screening. In addition, 1 article was found outside the database search through analysis of references. In total, 21 sources with 21 unique samples were identified [see PRISMA flow diagram in Figure 1]. Table 1 includes all available studies with a primary outcome of pounds gained or lost, and Table 2 includes studies in which change in BMI is the primary outcome. Tables are organized by type of MOUD treatment and then ordered by length of observation period (shortest to longest observation period).

Figure 1.

PRISMA flow diagram of identified, screened, and included studies.

Table 1.

Studies examining the effects of MOUD treatment with pounds gained/lost as primary outcome.

Study: Source	Country	Observation Period	Participants Gender: % Race:% Age: M(SD) Avg. Baseline BMI cat.	MOUD	Outcomesa,b	Gender Outcomes	Study Description
Umbricht 2015: Abstract	United States	4.60 m	MOUD N = 62 Gender: 52% M/48% F Race: 34% B Age: 42 Baseline BMI: NR	Meth.	↑ 13.7 lbs*	NR	Sub-study of an RCT testing topiramate as an adjunctive treatment for comorbid OUD and cocaine use disorder. Placebo patients gained significantly more than topiramate patients.
Parvaresh 2015: Journal article	Iran	6 m	MOUD N = 200 Gender: 91% M/10% F Race: NR Age: NR Baseline BMI: NR	Meth.	↑ 4.2 lbs*	Weight gain by gender not statistically compared M: ↑ 4.2lbs 6 m F: ↑ 4.1 lbs 6 m	Cohort study examining sleep, sexual functioning, and weight gains.
Montazerifar 2014:Journal article	Iran	6 m	MOUD N = 25 Gender: 80% M/20% F Race: NR Age: 37.4 Baseline BMI: Normal	Meth.	↑ 11.9 lbs* ↑ 4.8 BMI*	NR	Case control study examining weight and hormonal changes, including leptin. Leptin was low at the start of treatment for OUD patients and increased to match controls.
Montazerifar 2014:Journal article	Iran	2 m/6 m	MOUD N = 55 Gender: 80% M/20% F3 Race: NR Age: 31.6 Baseline BMI: Normal	Meth.	↑ 8.6 lbs 2 m* ↑ 23.4 lbs: 6 m* ↑ 1.6 BMI: 2 m* ↑ 4.1 BMI: 6 m*	NR	Cohort study examining weight gain.
Staub 2009: Other	United States	9 m/12 m	MOUD N = 154 Gender: NR Race: NR Age: NR Baseline BMI: NR	Meth.	↑ 11.6 lbs: 9 m* ↑ 22.1 lbs: 12 m*	Weight gain not sig. different between men and women	This was an observational cohort study followed by a naturalistic experiment introducing nutritional education. Gender did not predict weight gain prior to the intervention, post-intervention women gained about twice as much as men.
Stein 2008: Abstract	United States	12 m	MOUD N = 107 Gender: 66% M/34% F Race: 15% B/76% Lat. Age: 40 Baseline BMI: NR	Meth.	↑ 10 lbs*	NR	Retrospective chart review examining weight gain.
Sadek 2016: Journal article	Canada	12 m	MOUD N = 29 Gender: 45% M/55% F Race: NR Age: 29.3 7.0 Baseline BMI: Normal	Meth.	↑ 14.3 lbs* ↑ 2.2 BMI*	NR	Cohort study testing changes in anthropomorphic measures. Observed weight gain was due increased body fat.
Malik 2021: Journal article	Israel	NA	MOUD N = 3944 Gender: 78% M/22% F Race: NR Age: 43.7 Baseline BMI: Normal	Meth.	Avg.change in weight or BMI NR	NR	Retrospective chart review examining weight, vitamin B12, and vitamin D change. Vitamin B12 was unrelated. Vitamin D did not change over time, but people with higher Vitamin D more weight gain was observed (r = .26).
Fenn 2015: Journal article	United States	Avg. 21.6 m	MOUD N = 96 Gender: 64% M/36% F Race: 98% W Age: 38 Baseline BMI: Overweight	Meth.	↑ 17.8 lbs* ↑ 2.9 BMI*	Women gained sig. more weight M: ↑ 1.7 BMI, 12 lbs avg. 21.6 m F: ↑ 5.2 BMI, 28 lbs avg. 21.6 m	Retrospective chart review with variable treatment lengths examining weight gain.
DeVido 2015: Journal article	United States	3 m	MOUD N = 1 Gender: 100% F Race: NR Age: 31 Baseline BMI: NR	Bupe + nalox	↑ 50 lbsNR	NR	Case series with 1 patient report of weight gain.
Baykara 2019: Journal article	Turkey	4 m	MOUD N: 107 Gender: NR Race: NR Age: 24.55 Baseline BMI: NR	Bupe + nalox	↑ 10.2 lbs*	NR	Cohort study examining weight gain.
Varghese 2006: Journal article	India	6 m	MOUD N = 1 Gender: 100% F Race: NR Age: 35 Baseline BMI: Normal	Nal.	↑ 33.1 lbsNR	NR	Case report of 1 patient with weight gain.
Mysels 2011: Journal article	United States	3 m/6 m	MOUD N = 36 Nal Gender: 80% M/20% F Nal Race: 50% W/20% B/5 25% Lat/5% Other Nal Age: 37.45 Meth Gender: 69% M/31% F Meth Race: 56% W/13% B/31% Lat Meth Age: 41.18 Baseline BMI: NR	Meth & nal	Meth: ↑ 1.9% 3 m Nalt: ↑ 4.6% 3 m Meth: ↑ 3.7% 6 m Nalt: ↑ 6.7% 6 m	NR	Outcomes were % increase but discussion indicated approximately 10 pound weight gain in both groups over 6 months.

Notes. avg: average; B: Black; bupe+nalox: buprenorphine and naloxone; Lat: Latine/Hispanic; M: male; meth: methadone; MOUD; patients engaged in medication treatment for opioid use disorder; nal: naltrexone; NR: not reported; Oth: other racial group; RTC: randomized controlled trial; W: white; F: female.

^aoutcomes reported in kg were converted to pounds to increase interpretability across studies. Any studies reporting days were converted to weeks.

^bfor this column w = weeks, m = months, y = years, ↑ = increase, ↓ = decrease, * indicates that the study reported the difference as statistically significant, whereas NR means it was not reported if it was or was not statistically significant.

^cfor these results are pulled from 2 papers. Montazefir 2014 is the primary paper because it includes two time points, but some demographic and weight data was only reported in Montazefir 2012.

Table 2.

Studies examining the effects of MOUD treatment with BMI change as primary outcome.

Study: Source	Country	Observation Period	Participants Gender: n(%) Race: n(%) Age: M(SD) Avg. Baseline BMI category	MOUD	Outcomesa,b	Gender Outcomes	Study Description
Montazerifar 2014b: Journal article	Iran	6 m	MOUD N = 25 Gender: 80% M/20% F Race: NR Age: 37.4 Baseline BMI: Normal	Meth.	↑ 11.9 lbs* ↑ 4.8 BMI*	NR	Case control study examining weight and hormonal changes, including leptin. Leptin was low at the start of treatment for OUD patients and increased to match controls.
Montazerifar 2014a: Journal article	Iran	2 m/6 m	MOUD N = 55 Gender: 80% M/20% F3 Race: NR Age: 31.6 Baseline BMI: Normal	Meth.	↑ 8.6 lbs: 2 m* ↑ 23.4 lbs: 6 m* ↑ 1.6 BMI: 2 m* ↑ 4.1 BMI: 6 m*	NR	Cohort study examining weight gain.
Preston 2011: Abstract	United States	6.90 m	MOUD N = 50 Gender: 74% M/26% F Race: 38% W/54% B Age: NR Baseline BMI: Normal	Meth.	↑ 2.5 BMI*	NR	Cohort study examined weight and lipid profiles. Increase in weight was correlated with increase in cholesterol.
Reimer 2011: Journal article	Germany	12 m	MOUD N = 5003 Gender: 80% M/20% F Race: NR Age: 36.6 Baseline BMI: Normal	Meth.	↑ 1.2 BMI*	NR	RCT comparing injectable diacetylmorphine treatment to methadone. BMI also slightly increased in diacetylmorphine group (↑ 1.0).
Peles 2011: Journal article	Israel	6 m/12 m	MOUD N = 23 Gender: 83% M/17% F Race: NR Age: 43.9 Baseline BMI: Normal	Meth.	↑ 1.7 BMI: 6 m* ↑ 3.1 BMI: 12 m*	NR	Cohort study examined sleep and weight gain.
Sadek 2016: Journal article	Canada	12 m	MOUD N = 29 Gender: 45% M/55% F Race: NR Age: 29.3 Baseline BMI: Normal	Meth.	↑ 14.3 lbs* ↑ 2.2 BMI*	NR	Cohort study examined weight and anthropomorphic measures. Increases in weight was chiefly due to increased body fat.
Housova 2005: Journal article	Czech Republic	12 m	MOUD N = 12 Gender: 66% M/25% F Race: NR Age: 26.2 Baseline BMI: Normal	Meth.	↑ 0.3 BMI	NR	Case control study examined weight and hormone changes. MOUD patients initially had reduced levels of leptin and adiponectin and higher levels of resistan compared to controls. After 1 year of methadone treatment leptin levels were normalized.
Peles 2016: Journal article	Israel	Avg. 8.88 m Avg. 18.72 m Avg. 15.94 m	MOUD N = 114 Gender: 76% M/24% F Race: NR Age: 43.6 Baseline BMI: Normal	Meth.	↑ 1.8 BMI: avg. 9.97 m* ↑ 3.1 BMI: avg. 18.93 m* ↑ 3.2 BMI: avg. 16.12m*	NR	Cohort and retrospective study examining weight, predictors of weight gain, nutrition knowledge, and eating preferences. Patients with lower nutrition knowledge and higher sweet preferences had a higher BMI.
Fenn 2015: Journal article	United States	Avg. 21.6 m	MOUD N = 96 Gender: 64% M/36% F Race: 98% W Age: 38 Baseline BMI: Overweight	Meth.	↑ 17.8 lbs* ↑ 2.9 BMI*	Women gained sig. more weight M: ↑ 1.7 BMI, 12 lbs avg. 21.6 m F: ↑ 5.2 BMI, 28 lbs avg. 21.6 m	Retrospective chart review that followed patients variable lengths of time and examined weight gain.
Sweeney 2019: Journal article	United States	36 m	MOUD N = 74 Gender: 55% M/45% F Race: 32% W/65% B/1% Lat./1% Other Age: 40.9 Baseline BMI: Overweight	Meth.	↑ 5.4 BMI*	Women gained sig. more weight M: ↑ 4.5 BMI 36 m F: ↑ 6.6 BMI 36 m	Retrospective chart review examined changes in cardiovascular risk factors. Identified large increases in the proportion of patients with a BMI≥28.8, hypertension, high cholesterol, and to a smaller extent diabetes.
Kolarzyk 2005: Journal article	Poland	48 m	MOUD N = 30 Gender: 77% M/23% F Race: NR Age: 30.4 M/27.4 F Baseline BMI: Normal	Meth.	Men: ↑ 2.6 BMI* Women: ↓ .5 BMI	Men gained a sig. amount of weight. Women’s weight loss was not sig. See main outcomes	Cohort study examined weight gain over time. Women described more likely to report stimulant use, but no quantitative data was presented.
Li 2016: Journal article	UK	4 m	MOUD N = 25 Gender: 60% M/40% F Race: 80% W/20% Other Age: 41.5 M/33.4 F Baseline BMI: Overweight M/Obese F	Meth & Bupe	Men: ↓ 1.1 BMINR Women: ↑ 1.8 BMINR	Weight gain by gender not stat. compared see main outcomes	Cohort study examined weight and diet. Patients taking methadone and buprenorphine were combined in analyses.

Notes. avg: average; B: Black; bupe: buprenorphine; Lat: Latine/Hispanic; M: male; meth: methadone; MOUD; patients engaged in medication treatment for opioid use disorder; NR: not reported; Oth: other; RTC: randomized controlled trial; W: white W: Women; F: female.

^aoutcomes reported in kg were converted to pounds to increase interpretability across studies. Any studies reporting days were converted to weeks.

^bfor this column w = weeks, m = months, y = years, ↑ = increase, ↓ = decrease, * indicates that the study reported the difference as statistically significant, whereas NR means it was not reported if it was or was not statistically significant.

^cData for these results are pulled from 2 papers. Montazefir 2014 is the primary paper because it includes two time points, but some demographic and weight data was only reported in Montazefir 2012.

Study characteristics

Studies were published in manuscripts (n = 17; 15–31) or conference abstracts (n = 3; 32–34), and one study was summarized in a special report associated with a professional organization (35). With respect to study design, most studies were either cohort studies (n = 8; 15,16,18,21,25,29,30,33) or retrospective chart reviews (n = 5; 19,23,28,31,32). Less frequently, randomized controlled trial (RCT; n = 2; 24,34), case control (n = 2; 17,27), or case report (n = 2; 20,22); designs were used. Regarding RCTs, one study tested topiramate to treat comorbid cocaine use disorder among patients in methadone treatment (34). The data showed that participants taking methadone and placebo gained 13.7 pounds in about 5 months, while participants taking methadone and topiramate gained 4.18 pounds in the same period. The only other RCT investigated differences in treatment outcomes between injectable diacetylmorphine treatment and methadone (24). This was the largest available study, with 500 participants taking methadone. The change in weight was modest, with a 1.2 increase in BMI in 12 months. The length of studies varied from a minimum of 3 months (20) to a maximum of 4 years (29).

In addition to heterogenous study designs, studies were completed in a range of geographic regions. Most studies were conducted in the United States (n = 8; 20,23,28,31–35) while a sizable minority were conducted in Israel (n = 3; 19,25,26) or Iran (n = 3; 15–17). Studies conducted in the United States demonstrated some of the largest increases in weight and BMI, while data from other countries such as Germany (24), Czech Republic (27), and Poland (29) reported comparatively smaller gains.

Participant characteristics/study criteria

For some studies, participant characteristics were difficult to ascertain, and inclusion and exclusion criteria were often not explicitly stated. When inclusion and exclusion criteria were described, common criteria included minimum treatment durations (range 6 months to 3 years (19,23,26,28,31,35); medical ruleouts such as pregnancy, hypertension, or diabetes (15,21,25,31,32); or available data within the study window (19,23,31). Very few studies reported on the rates of other psychotropic medication, though a small study by Li, et al. (30) indicated high rates of antidepressant use (40%) and relatively high use of antipsychotic medications (12%).

The issue of substance use disorder comorbidity was also heterogeneously characterized and sometimes absent from the existing studies. When substance use beyond opioids was described, it was most often cohort studies that included data on substance use upon admission. The data indicates that cocaine use was common (24–26,28), with estimates generally 40% or higher.

Excluding case studies, the samples were predominantly men, with the proportion of men ranging from a minimum of 52% (34) to a maximum of 91% (15). The racial and ethnic composition of the study samples was reported much more inconsistently; fourteen studies did not provide any information on race and/or ethnicity. Of the remaining studies that reported race and ethnicity, there was a mix of studies with predominantly White samples (30,31), predominantly Black and Hispanic samples (28,32,33), and some studies included about equal numbers of White participants and participants of Color (i.e, Black, Hispanic, “Other” (23).

An additional participant consideration was baseline BMI category. Fourteen of the included studies reported baseline BMI. Eleven studies reported an average baseline BMI in the normal range (16–19,22,24–27,29,33), three reported average BMI in the overweight or obese range (28,30,31), and no studies reported average baseline BMI in the underweight range.

Study outcomes

BMI category change

A total of six studies reported data related to the proportion of participants in specific BMI categories before and after treatment (not included in tables). All studies reporting BMI category changes examined the effects of methadone. BMI category changes were sometimes described using pre and post measurement of overweight and/or obesity only, whereas some studies described the changes for all four BMI categories. Preston, et al. (33) reported 42% of the sample had overweight or obesity at baseline and 68% had either overweight or obesity after about 7 months of methadone treatment. Stein, et al. (32) reported that 30% of patients initially had obesity and 42% had obesity after 12 months of treatment. Sweeney, et al. (28) reported an even greater increase with more than 40% of the sample developing overweight or obesity for a total of 87.8% of the sample after 36 months of treatment. For studies that considered all BMI categories, there were significant differences between studies. Montazerifar, et al. (36) reported a relatively high proportion of patients with underweight BMI (27%), and the data indicated that after 12 months of treatment about half of these individuals transitioned to the normal weight category. In the same sample, the rates of overweight and obesity from baseline to 6 months were relatively stable. In contrast, two studies conducted in the United States with longer observation periods reported relatively minimal levels of underweight both pre and post treatment (31,35). Moreover, both studies reported increases in the rates of overweight and an increase in the rates of obesity. In particular, Staub, et al. (35) followed participants for 9 months and the percent of patients with normal weight decreased 10% and the percent of participants with overweight and obesity increased 9% and 5%, respectively. Fenn, et al. (31) followed participants for a total of about 22 months, and among women, the rate of overweight increased 20% to a total of 31% and the rate of obesity increased 14% for a total of 51%. The increases were smaller, though still notable among men; the rate of overweight increased 13% to a total of 36%, and the rate of obesity increased 8% for a total of 34%.

Study outcomes: pounds gained/lost

For studies examining the association between methadone and pounds gained/lost, two studies included observation periods less than 6 months (weight increase ranged from 8.6–13.7 pounds (16,34), three studies had observation periods of 6 months (weight increase ranged from 4.2–23.4 pounds (15–17), three studies had observation periods between 9 and 12 months (weight increase ranged from 10.0 to 22.1 pounds (18,32,35), and a single study following patients for over 12 months (17.8 pound increase in 21.6 months (31). All studies reported statistically significant weight gain. Figure 2 summarizes the available data on pounds gained during methadone treatment of varying lengths. Among studies with 6 month observation periods, the reported weight gain was highly variable: 4.2 pounds (15), 11.9 pounds (17), and 23.4 pounds (16). The amount of weight gain reported in studies with 12-month observation periods was somewhat similar to the amounts observed in the 6-month studies, including 10 pounds (32), 14.3 pound (18), and 22.1 pound (35) gains. Only one study followed patients for more than 1 year, and the study follow-up period varied between patients. The average was almost 2 years, and patients gained almost 20 pounds in the follow-up period (31). Finally, a single study reported outcomes based on percent weight gain and both methadone and naltrexone were studied. However, the discussion indicated that patients taking methadone gained an average of about 10 pounds, which adds additional data to the 6-month treatment window (23).

Figure 2.

Summary of studies investigating the effect of methadone on weight reported in pounds gained.

Notes. Figure 2 displays all available studies reporting the effects of methadone on weights in pounds. Studies are ordered according observation period and further sorted in ascending order for amount of weight gain. ^a This studies had two observation periods, and it is repeated to demonstrate the findings for 0–3 months and 4–6 months. ^b This study had two observation periods, both of which were greater than 6 months. The reported value is the highest reported weight gain observed in the study, which occurred after 12 months of methadone treatment.

Studies examining pounds gained or lost associated with MOUD treatments other than methadone were infrequent in the literature. Two studies were case reports with one patient reporting a 50-pound increase in 3 months from buprenorphine plus naloxone treatment (20). The other case report described a 33-pound weight gain from naltrexone (22). A single cohort study examined the effects of buprenorphine plus naloxone and reported a 10-pound gain in 4 months (21).

Study outcomes: BMI change

For studies examining the association between methadone and BMI change, four studies included observation periods of 6 months (BMI change range 1.7–4.8 (16,17,25,33), four studies included observation periods of 12 months (BMI change range 0.2–3.1 (18,24,25,27), and four studies followed patients for more than 12 months (BMI change range −0.5–5.4 (26,28,29,31). Nine out of 11 studies indicated significant increases in BMI (16–18,24–26,28,31,33), and one study showed a significant increase for men and a nonsignificant increase for women (29). However, the results were heterogenous both within and across observation periods. The BMI change associated with about 6 months of methadone treatment included increases of 1.7 (25), 2.5 (33), 4.1 (16), and 4.8 (17). In contrast, some of the reported changes for 12 months of methadone treatment included very small increases (i.e., 0.2 (27), but increases of up to 3.1 were also observed (25). One study followed patients for variable lengths of time with multiple follow-up points and generally found an upward increase in BMI starting with an increase of 1.8 in about 9 months (26). In the same study, BMI further increased about 3 points but was essentially stable between 16- and 19-month observation timepoints (26). Finally, two studies reported much longer follow up periods with somewhat divergent findings. Sweeney, et al. (28) followed patients over 3 years and reported a BMI increase of 5.4. Kolarzyk, et al. (29) reported findings were stratified by sex, and women experienced a nonsignificant change in BMI over 4 years, while the BMI of men in the study increased 2.6 points in the same period.

There was only a single study that examined BMI change for a combined sample of individuals taking methadone and buprenorphine plus naloxone; Li, et al. (30) reported that men’s BMI decreased 1.1 points and women’s BMI increased 1.8 points.

Predictors of weight gain

Few studies examined any predictors of weight gain, but a minority of studies examined sex/gender differences (n = 6; 15,28–31,35). Among studies of methadone treatment, two studies reported that women gained more more weight (28,31), one studys showed no differences between men and women (15), and one reported that men gained a significant amount of weight and women’s weight gain was nonsignificant (29). In the study by Staub, et al. (35) there were no overall differences between men and women when the data was aggregated across sites. However, for three of the sites that monitored weight after instituting nutrional counseling, weight gain was double or more for women as compared to men. The remaining study reported BMI change in a combined sample of participants taking methadone and buprenorphine plus naloxone, and men lost 1.1 BMI points and women gained 1.8 BMI points (30). One study investigated whether White and Black patients’ weight gain was significantly different and found that Black patients gained significantly more (an increase in BMI of 6.36 in 36 months) than White patients (an increase in BMI of 3.37 in 36 months (28).

Several studies investigated the effects of ongoing cocaine use on weight outcomes, with most showing that cocaine use predicted less weight gain (32–34). For example, Stein, et al. (32) compared individuals with consistent cocaine use to individuals with either no or inconsistent cocaine use and found that individuals using cocaine regularly gained an average of 1.5 pounds whereas individuals not consistently using cocaine gained 12.4 pounds over 12 months.

Three studies investigated if weight gain was associated with methadone dose (26,28,31). Sweeney, et al. (28) found that after stabilization, methadone dose was significantly correlated with BMI increase (r = 0.27). In contrast, Peles, et al. (26) found no differences in the average weight gain associated with methadone doses of <100 mg/day, 100–150 mg/day, and >150 mg/day. Fenn et al (31) also did not find an association between methadone dose and amount of weight gain.

Discussion

Twenty-one unique studies investigated the effects of MOUD treatment on weight. Most studies examined the effects of methadone treatment and indicated the potential for significant weight gain. The amount of weight gain was highly variable across studies, but among studies reporting weight gain for 6 months of treatment, weight gain ranged from 4.2 to 23.4 pounds. Factors likely to influence outcomes included study region and comorbid cocaine use during treatment. Other factors will require further empirical investigation, but it is possible that women experience more significant increases in weight than men. Racial and ethnic inequities were largely unexamined. The results indicate that weight should be regularly monitored throughout MOUD treatment. Additionally, providers may want to initiate conversations with patients about available prevention and intervention methods for excess weight gain.

When the issue of weight gain in MOUD treatment arises, many hypothesize that weight gain could represent a restoration of weight that was lost excessively during drug use and a positive shift toward improved nutrition. The available data at least partially contradict this notion. The majority of patients were not underweight at the start of MOUD treatment. A subset of studies further refutes this theory by examining the proportion of patients who develop overweight or obesity during methadone treatment (no studies of other MOUD treatments examined BMI category change). Most studies reported a significant increase in the proportion of patients with overweight or obese BMIs. When exclusively considering obesity, which carries the strongest association with poor health outcomes, the percent increase with 1 year or less of treatment ranged from a low of 3.7% to a high of 27%. Even more concerning, of the available studies with longer follow-up periods, the risk of developing overweight and obesity appears substantial. Sweeney followed patients 36 months and found that the proportion of patients with overweight or obesity increased 47.8% to a high of 87.8%. It is unknown if this extreme increase is due to something unique about this sample, the more extensive follow-up period, or a combination. What also remains unknown is if and how eating behaviors could represent a response to malnutrition. It is possible that weight gain could be both in excess of what is necessary for health reasons and the increased eating could be a biologically driven attempt to address vitamin and mineral deficiencies. In addition to monitoring weight in MOUD treatment, studying changes in blood levels of important vitamins and minerals could further clarify the types of changes (positive and negative) associated with MOUD treatment.

A notable limitation of the current literature is the heterogeneity in sample populations and observation periods, which make drawing firm conclusions difficult. Moreover, the majority of studies were conducted using small, uncontrolled cohort studies. Regarding sample size, about half (48%) of the included studies had sample sizes under 50 people, and 71% of studies included under 100 people. Small sample sizes increase the possibility outlying cases having undue influence as well as between study heterogeneity. Small studies can also be underpowered (37). Accordingly, the few available studies with larger samples can be helpful in contextualizing likely outcomes for patients on methadone treatment, and some of the largest studies show the most modest results (15,24). For example, the study by Reimer and colleagues (24) randomized 500 patients to methadone treatment and 550 patients to injectable diacetylmorphine treatment. Both the size and rigor of the study is unique, and this study demonstrated minimal change in BMI. But these larger studies may be confounded with regional effects. Studies from the United States demonstrated some of the largest effects, and studies from Poland, Germany, and the Czech Republic demonstrated comparatively smaller effects. The studies from Iran were highly heterogenous (15–17). The differences generally mirrored trends in the prevalence of obesity in various countries, particularly among men (38). If there are geographic trends, continued research across geographic settings will allow for an improved understanding of how existing trends in health and weight are or are not mirrored in the data examining MOUD treatment.

What is largely absent from the literature is the study of other MOUD treatments. Of particular interest would be direct comparisons of methadone, buprenorphine/naloxone, and naltrexone. Any differences in risk of substantive weight gain could have profound treatment implications. A cross-sectional study using retrospective chart review found that patients engaged in methadone treatment were significantly more likely to have diabetes as compared to patients in buprenorphine/naloxone treatment, but there were no differences in the average BMI between groups (7). Future research should investigate if the three types of MOUD treatment have similar or dissimilar effects on weight over time. The lack of randomized designs also limits any firm conclusions comparing treatments. For example, psychiatric comorbidity is both associated with obesity and may increase the likelihood of receiving methadone treatment as compared to other treatments. Only two studies included randomized designs; the study by Reimer and colleagues and the study by Umbricht and colleagues (34). Regarding the Umbricht trial, it was part of a larger study randomizing to topiramate (versus placebo) to treat comorbid cocaine use disorder. Topiramate, like naltrexone, is included as one component in FDA-approved antiobesity drugs (39). Unlike the majority of retrospective and observational studies included, this was one of the only studies that speaks to a possible treatment option to address weight gain. Future research should: 1) include randomized designs to compare MOUD treatments; 2) replicate and extend findings related to weight change; and 3) explore possible behavioral and pharmacologic treatment options for patients in MOUD experiencing excessive weight gain.

A significant gap in the literature also exists in understanding what patients are at risk for gaining weight. While the current review provides important aggregated data on the average amount of weight gained, it remains unknown what percentage of patients gain a clinically significant amount of weight. It would be helpful to know the range of weight gain observed specifically among patients who gain in treatment. Moreover, information is needed related to what puts an individual at increased risk of such weight gain. Participant characteristics potentially relevant to the magnitude of the effect were sex and/or gender. Potential sex-effects could include the biological underpinnings that broadly confer increased risk of obesity among females (40). With respect to gender, research indicates that women in MOUD treatment have an increased risk of psychiatric comorbidity (41,42), which could in turn be associated with greater weight gain in MOUD. However, a recent meta-analysis investigated sex/gender differences in weight gain from MOUD (studies n = 4), and the authors did not find a significant effect, though the authors caution that the available data is quite limited (43). Other demographic differences, such as race and ethnicity were largely unexamined. Fourteen studies failed to report on the race and ethnic makeup of the sample. Rates of obesity (44) and receipt of methadone versus buprenorphine (45,46) vary by racial and ethnic groups due to complex interacting systems that contribute to health disparities. Accordingly, it is critical that greater information related to race and ethnicity is captured in future studies.

The data on methadone treatment indicated that on average patients gain weight in the first year of treatment, with the range spanning from 4 to 23 pounds. Figure 2 depicts the significant variability in amount of weight gain across studies and observation periods, though also clearly indicates that a significant risk is present. Similar increasing trends were observed in studies examining BMI. The Agency for Healthcare Research and Quality defines a meaningful change in BMI as 0.2, which corresponds to an increase of 0.5 kg for an individual with a BMI of 27 (47). Aggregating across all studies reporting BMI change during the first year of treatment, the range was 0.3–4.8, which indicates significant heterogeneity in the magnitude of the effect but also the potential for highly significant increases in BMI.

There were only five studies that reported weight changes associated with an MOUD treatment other than methadone. As such, the evidence regarding weight gain associated with these treatments remains unknown. One unexpected finding were the studies reporting weight gain associated with naltrexone, including Mysels, et al. (23) and a single case report by Varghese and Sagar (22). Naltrexone is currently a compound included in anti-obesity medications such as Contrave (48) and thought to promote weight loss. If weight gain among patients with OUD taking naltrexone is replicated, it will be critical to examine other mechanisms of action beyond the pharmacokinetics of the MOUD drug.

Even among studies of methadone, which are more numerous, the mechanism of action for weight gain is not understood. Research dating back to the 1970s documents impaired insulin response (49,50), and a recent small cohort study compared buprenorphine and methadone treatment on a number of metabolic indices. Both treatments were associated with insulin resistance and methadone specifically was associated with elevated fasting glucose (51). However, the role of endogenous and exogenous opioids in food intake and body weight is highly complex. It is possible that methadone and/or buprenorphine is associated with excess intake of sugary foods, which could lead to weight gain (52,53). Experimental studies that blind participants to nutritional content of taste-matched foods or beverages (e.g. 54), could be helpful in elucidating a potential link. Alternatively, social determinants of health, which are unequally distributed in the two patient populations, may have contributed to the observed differences. National data collected by SAMHSA indicate that patients taking methadone are significantly less likely to be employed and have lower levels of education as compared to patients taking buprenorphine (55). Similar trends have been observed in epidemiologic studies of obesity (56).

Several studies also highlighted the increased risk of medical comorbidities in tandem with the weight gain. Sweeney and colleagues (28) found that rates of high blood pressure, high cholesterol, and type 2 diabetes also increased substantially during the course of methadone treatment. Data also indicates that increasing BMI among patients in methadone treatment is linked to increased sleep breathing disorders, such as obstructive and central sleep apnea (6), which influenced quality of life as well as morbidity and mortality.

In addition to increasing understanding of the potential mechanisms explaining MOUD treatment weight gain, it is essential to consider implications for clinical practice. At minimum, these data support monitoring weight within MOUD treatment. An additional consideration is the potential for prevention and/or treatment of obesity simultaneously with MOUD treatment. In the general population, long term weight loss is challenging to obtain and even more challenging to maintain (57). Moreover, studies of other clinical populations undergoing weight loss treatment in the context of medication associated weight gain have shown very modest results (58,59). Accordingly, a focus on prevention of significant weight gain may be a better target. Obesity prevention is necessarily multifaceted and should include a focus on increasing nutrition knowledge, changing eating habits, and increasing physical activity. Prevention of weight gain in MOUD settings will require interdisciplinary teams with a shared goal of considering the whole health of individuals engaged in MOUD treatment. Best practices for obesity prevention include community engagement prior to program development and implementation (60). An important first step could be working with MOUD patients to understand what they think about weight, nutrition, and exercise in the context of MOUD treatment. Results from this could more effectively guide the development and implementation of prevention programs.

Study findings should be considered within the context of several limitations. First, the current methods cannot eliminate the possibility of publication bias, and the possibility that studies with nonsignificant weight gain are unpublished. Second, studies were excluded if an English translation was not available, which may limit understanding of the phenomenon in non-English speaking countries. Third, there was significant heterogeneity in the study observation periods and how outcomes were defined, which makes definitive conclusions difficult. Additionally, the lack of controlled studies increases confounds and decreases the specificity of findings. An important confound in multiple studies is the requirement that participants be in treatment for a certain length of time. It is possible, likely even, that patients who stay in treatment have important differences from patients that do not. As a result, there are significant limitations on generalizability.

These limitations notwithstanding, this is one of the first comprehensive scoping reviews of MOUD treatment and weight gain. A single prior review was identified (61) that focused on randomized and nonrandomized clinical trials. The current review included all available research irrespective of study design, which substantially increased the number of studies reviewed. The current scoping review highlights several important avenues for future work, including identifying specific patient subgroups most at risk for excessive weight gain, understanding the mechanisms of weight gain in MOUD treatment, and perhaps most importantly identifying if weight gain is negatively associated with MOUD treatment outcomes.