Key messages
1. Methadone may keep more people in treatment than buprenorphine.
2. People reported less opioid use with methadone than buprenorphine, although when testing urine for opioid use there was no difference between groups.
3. Buprenorphine maintenance probably keeps more people in treatment and may be better at helping people reduce opioid use than non-opioid treatments.
What is dependence on opioid pain medicines?
Use of pharmaceutical opioids (medicines that are used to treat pain) has increased dramatically in some parts of the world since the mid-1990s. With the increased use, there has been increasing numbers of people seeking treatment for dependence (addiction) on pharmaceutical opioids. Currently, most treatment guidelines are based on research that was conducted in people who were dependent on heroin (a highly addictive opioid). People who use pharmaceutical opioids may differ from people who use heroin in important ways, such as having a higher prevalence of chronic pain and mental health symptoms.
What did we want to find out?
This review sought to compare different opioid agonist maintenance treatments (i.e. treatments such as methadone or buprenorphine that are given for at least 30 days to help the person to reduce their unsanctioned medicine use) for the treatment of pharmaceutical opioid dependence. We also compared results from maintenance treatment to short-term treatments such as detoxification (removal of the drug from the body) or psychological treatments (e.g. talking therapy, counselling).
What did we do?
We examined the scientific literature up to January 2022. We identified eight randomised controlled trials (studies where people were allocated at random to one of two or more treatment or control conditions) involving 709 adults and adolescents who were dependent on pharmaceutical opioids. Seventy percent of the people in the studies were male, and had an average age of 32.0 years. The average duration of the studies comparing different opioid maintenance treatments (four studies that compared methadone to buprenorphine) was 21 weeks, and the average duration of studies comparing a maintenance treatment (four studies with buprenorphine maintenance) to detoxification, an opioid antagonist, or psychological treatment was 14 weeks. Seven of the eight studies were conducted in the USA, with one study from Iran.
The main outcomes we examined were opioid use and leaving treatment early.
The National Institutes of Health (USA) funded seven studies, with one study not reporting the funding source. Five studies reported that a pharmaceutical company provided the medicine.
What did we find?
We found that when comparing methadone with buprenorphine maintenance treatments, methadone may keep more people in treatment than buprenorphine. People on methadone may report less opioid use than people on buprenorphine, although when testing urine for opioid use there was no difference between methadone and buprenorphine. When comparing buprenorphine maintenance to other non-opioid treatments such as detoxification, opioid antagonists like naltrexone, or psychological treatments, buprenorphine probably keeps more people in treatment, and may be better at helping people reduce opioid use.
What were the limitations of the evidence?
Overall, the evidence was of low to moderate quality. All studies put people into treatment groups randomly, but the participants and researchers knew which medication the participants were taking, which could bias the results and lower the quality of the evidence. In some studies, many people did not finish the study, leading to a meaningful amount of missing data which may bias the results. In some studies, there were more missing results in one arm of the study than the other. Most of the studies were similar in design and results were collected in a way that allowed them to compare the main outcomes of opioid use and number of people completing the study.
How up to date is this evidence?
The evidence is current to January 2022.
There is very low- to moderate-certainty evidence supporting the use of maintenance agonist pharmacotherapy for pharmaceutical opioid dependence. Methadone or buprenorphine did not differ on some outcomes, although on the outcomes of retention and self-reported substance use some results favoured methadone. Maintenance treatment with buprenorphine appears more effective than non-opioid treatments.
Due to the overall very low- to moderate-certainty evidence and small sample sizes, there is the possibility that the further research may change these findings.
There are ongoing concerns regarding pharmaceutical opioid-related harms, including overdose and dependence, with an associated increase in treatment demand. People dependent on pharmaceutical opioids appear to differ in important ways from people who use heroin, yet most opioid agonist treatment research has been conducted in people who use heroin.
To assess the effects of maintenance opioid agonist pharmacotherapy for the treatment of pharmaceutical opioid dependence.
We updated our searches of the following databases to January 2022: the Cochrane Drugs and Alcohol Group Specialised Register, CENTRAL, MEDLINE, four other databases, and two trial registers. We checked the reference lists of included studies for further references to relevant randomised controlled trials (RCTs).
We included RCTs with adults and adolescents examining maintenance opioid agonist treatments that made the following two comparisons.
1. Full opioid agonists (methadone, morphine, oxycodone, levo-alpha-acetylmethadol (LAAM), or codeine) versus different full opioid agonists or partial opioid agonists (buprenorphine) for maintenance treatment.
2. Full or partial opioid agonist maintenance versus non-opioid agonist treatments (detoxification, opioid antagonist, or psychological treatment without opioid agonist treatment).
We used standard Cochrane methods.
We identified eight RCTs that met inclusion criteria (709 participants). We found four studies that compared methadone and buprenorphine maintenance treatment, and four studies that compared buprenorphine maintenance to either buprenorphine taper (in addition to psychological treatment) or a non-opioid maintenance treatment comparison.
We found low-certainty evidence from three studies of a difference between methadone and buprenorphine in favour of methadone on self-reported opioid use at end of treatment (risk ratio (RR) 0.49, 95% confidence interval (CI) 0.28 to 0.86; 165 participants), and low-certainty evidence from four studies finding a difference in favour of methadone for retention in treatment (RR 1.21, 95% CI 1.02 to 1.43; 379 participants). We found low-certainty evidence from three studies showing no difference between methadone and buprenorphine on substance use measured with urine drug screens at end of treatment (RR 0.81, 95% CI 0.57 to 1.17; 206 participants), and moderate-certainty evidence from one study of no difference in days of self-reported opioid use (mean difference 1.41 days, 95% CI 3.37 lower to 0.55 days higher; 129 participants). There was low-certainty evidence from three studies of no difference between methadone and buprenorphine on adverse events (RR 1.13, 95% CI 0.66 to 1.93; 206 participants).
We found low-certainty evidence from four studies favouring maintenance buprenorphine treatment over non-opioid treatments in terms of fewer opioid positive urine drug tests at end of treatment (RR 0.66, 95% CI 0.52 to 0.84; 270 participants), and very low-certainty evidence from four studies finding no difference on self-reported opioid use in the past 30 days at end of treatment (RR 0.63, 95% CI 0.39 to 1.01; 276 participants). There was low-certainty evidence from three studies of no difference in the number of days of unsanctioned opioid use (standardised mean difference (SMD) −0.19, 95% CI −0.47 to 0.09; 205 participants). There was moderate-certainty evidence from four studies favouring buprenorphine maintenance over non-opioid treatments on retention in treatment (RR 3.02, 95% CI 1.73 to 5.27; 333 participants). There was moderate-certainty evidence from three studies of no difference in adverse effects between buprenorphine maintenance and non-opioid treatments (RR 0.50, 95% CI 0.07 to 3.48; 252 participants).
The main weaknesses in the quality of the data was the use of open-label study designs, and difference in follow-up rates between treatment arms.