By Mark Gold, MD
47,600 people died from drug overdoses involving opioids in 2017. Between 2012 and 2018, the number of fentanyl-induced fatal overdoses rose dramatically, accounting for a majority of overdose deaths. While preliminary data from the Centers for Disease Control shows a marginal decline in fatal overdoses in 2018, from 70,237 to 68,557, it also reveals that fentanyl is still the primary cause of fatal overdoses.1 Naloxone is a non-opioid wonder drug that can reverse an opioid overdose. It is short-acting, and by temporarily reversing the effects of opioids, it gives a person with an opioid use disorder (OUD) a second chance—an opportunity to receive treatment. As a result of campaigns by, among others, the Surgeon General2 and the CDC to improve naloxone access, retail pharmacies increased naloxone dispensing from 2012 to 2018. Despite the increase in dispensation by pharmacies, only one naloxone prescription was dispensed for every 69 high-dose opioid prescriptions in 2018.3 In the old days, I remember patients saying that they felt stigmatized at the pharmacy when they heard, "Mr. Jones, your Elavil is ready." Stigma kept many depressed patients from filling prescriptions. But in this case, is the challenge both stigma and the lack of pharmacist or health care provider education?4 It is tough to pinpoint a cause explaining this data.
The CDC and Surgeon General encourage us to improve naloxone access at the local level, including through prescribing and pharmacy dispensing. Widespread distribution of naloxone is an essential component of the public health response to the opioid overdose epidemic. Unfortunately, the lowest rates of naloxone dispensing are in the areas with the highest opioid overdose rates. We are in the third phase of the opioid epidemic, with pain clinics’ overprescribing practices overtaken first by heroin and, more recently, by fentanyl. Individuals who overdose often overdose again, and many patients treated in addiction programs or health providers' offices through MAT often relapse.
For the treatment community, adverse outcomes and continued overdose deaths are, naturally, extremely upsetting. OUD treatment program employees frequently complain of burnout. High turnover in many programs is a major problem. Some are frustrated by recidivism rates, others because some patients are not offered MAT. Some patients cease their MAT course, and others drop out of treatment altogether. Yet experts have consistently agreed that while MAT, due to a lack of options, has not helped us combat cocaine or methamphetamine use disorders, it can be enormously helpful in managing OUD. A recent review, written by James Bell and John Strang, looks at the overall evidence on MAT and compares the relative benefits of different medications, helping to shed light on this critical public health issue. It is important to keep in mind, however, how contentiously and frequently some of our evidence is debated. We lack, for example, prospective, long-term, oncology-like, 5-year studies on the subject. MAT is debated—and so are standards measuring patient "outcomes." For a physician with an OUD, the relevant outcome standards may include a return to practice, 5-year sobriety, urine testing, and fitness for duty. In other cases, outcome standards may include coming to a treatment program, or following an MAT course, or simply not overdosing or dying.
What did this review find about MAT’s effectiveness?
Bell and Strang’s review shows that for OUD, evidence on certain outcomes favors MAT over non-MAT approaches. It points, for example, to a study that used Medicaid data to track outcomes for patients with OUD. This study measured how long patients under 22 stayed in treatment, also called “retention,” on the basis of their Medicaid claims. It also tracked patients using different MAT options: buprenorphine, naltrexone, and methadone. The “non-medical” treatment group in the study participated in behavioral health services instead of MAT. The median retention rates were 324 days in methadone, 150 in naltrexone, 123 in buprenorphine, and 67 in behavioral health services. The review points out that while this study was not randomized and had an ambiguous final measure of patient well-being, it is a “real-world” one. The review also found evidence that buprenorphine and methadone can enhance treatment for pregnant women with OUD. MAT in this context works best with other assistance, such as social support, specialist teams, and obstetrics.
Which MAT medication works best? Can We Know?
This review found that methadone works best for patient retention—again, simply how long patients remain in treatment. The question of which treatment medication works "best", however, is not easily answered. Bell and Strang call methadone the “gold standard” for MAT comparisons, a view with which many other experts would agree, including Herb Kleber, who has conducted four decades of research on the subject.5 Yet this review also points out that there is no evidence indicating which medication is best suited to particular patients, and asserts that, "The main guide is patient preference." I have worked in MAT programs since the 1970s and studied methadone, buprenorphine, and naltrexone. The reality is that each has worked for some patients, but not for others.
Methadone's higher retention rates are very important for treating a disease so tightly linked to overdoses, deaths, and co-morbidities. Results on its benefits in this regard have held across many studies and analyses of randomized trials, as Bell and Strang's review observes. One trial measured participants over 24 weeks and observed that 74% of patients stayed on methadone, compared to 46% of patients on sublingual buprenorphine-naloxone (BNx). Over half of the BNx patients were lost in just 24 weeks. Interestingly, this trial showed that in the first 9 weeks of their MAT course, patients on methadone used more non-prescribed opioids, measured by urine tests, than patients on BNx. But this didn’t continue after 9 weeks. The review suggests that this result in the trial makes sense, as buprenorphine blocks the effects of illicit opioids quickly, while methadone builds up tolerance over time. Furthermore, patients taking the highest BNx dosage in the trial still had a lower retention rate than patients on methadone. 30% of patients taking the highest BNx dosage still used illicit opioids. Researchers created a follow-up study to the trial after five years, involving 73% of the trial’s participants, and found that those who had used BNx now used considerably more opioids than those who had used methadone. Buprenorphine, and buprenorphine plus naloxone, also have abuse and diversion potential to consider.
The review also notes that while trials may not necessarily observe mortality differences, an Australian study examined 32,033 individuals on buprenorphine or methadone. In the Australian study, individuals starting methadone had a higher mortality rate than those starting buprenorphine. But this study did not find a difference during the course of treatment, and mortality was much lower for individuals who had used methadone a month after treatment than it was for those who had used buprenorphine.
While reviewing positive outcomes, Bell and Strang also describe large numbers of patients for whom all MAT modalities are unsuccessful. It is too easy to blame patients for "failing" in MAT treatment. For some patients, MAT may just work poorly. Buprenorphine may help the brain after exposure to fentanyl and not just prevent an overdose.6 Buprenorphine or other MAT courses may reverse opioid withdrawal and many somatic signs of distress in animals but, unfortunately, human brains and behaviors are not easily modeled on those of rats. Difficulties with MAT can be attributed in part to learning, the nature of the human brain, the nature of the disease, and adherence to treatment regimens. Some patients agree to treatment and find it very difficult to continue as long as recommended. Others distrust their providers and programs. Inconsistency in the administration of medications means patients do not always receive their treatment course according to best practices, or at the most effective dosage levels. Others have untreated co-morbid illnesses. Still others may have a disease that looks like OUD but is in fact, because of individual neurobiology or experiences, actually something quite different.
Higher doses of methadone can lead to depression and diminished sex drive. Patients may become focused on these problems and lose sight of the fact that higher doses lead to longer abstinence than medium or low doses. Methadone is given in methadone programs. Buprenorphine is often given in new clinics or private health care provider offices. To many patients, these differences matter a great deal. Unfortunately, relative to other peer countries, the United States' licensing and zoning laws limit provision of and access to methadone. MAT should be widely available. We might also reconsider laws and regulations restricting methadone to clinics.7 Dr. Jeffrey Samet and coauthors call for allowing methadone dispensing in pharmacies and prescriptions in primary care. This would be a promising and positive development.
Some patients need and want psychotherapy, group support, and peer counseling. Sadly, some methadone and newer MAT courses are often administered as if they were medication vending machines, with a glaring lack of psychosocial and behavioral treatment options. Many patients have sleeping problems, anxiety, depression, PTSD, and other psychiatric conditions. Psychiatry and psychological treatments may be available by referral, but they are often seen as independent of the "real" treatment, MAT. To me, this is not logical. Therapeutic Communities in the 60s had success with NA and abstinence re-learning models. The long-term life change approach of IOPs or RTCs could be paired with MAT. NA and AA have important roles to play, as well.
In the most recent Mayo Clinic Proceedings, I co-authored a review on MAT and noted that there simply is no silver bullet. Particular treatment options each have their place for different patients, and our review recommends a carefully targeted and individualized approach. Medical professionals should give close attention to patients’ concerns about treatment locations, and to their broader social support networks. Without the patient as a partner, medications used in OUD treatment will fail, as treatment is directly related to taking medications. Patients receiving MAT should also have a moderate-to-severe OUD, as outlined in the DSM 5 criteria, and be screened for co-morbid issues, such as infectious disease, trauma, heart disease, and psychiatric illness. It is important to remember that OUD patients often smoke, drink, and use other substances. It is common to see OUD alongside other substance use disorders. If a patient with OUD has an overlooked alcohol use disorder (AUD), he or she may then be discovered to suffer from alcoholic hepatitis or related AUD conditions. Not testing for infectious diseases, or not vaccinating them in the MAT or SUD clinic, may be a life-changer. Rarely do our patients have a medical "home", or receive routine evaluations, examinations, or vaccinations. Yet they need these to stay healthy, diagnose problems, and treat them accordingly. Ignoring cardiac and other diseases can also shorten patient lives, even when patients are following their MAT course and treatment program plan.
If patients are in need of a tapering or detoxification plan to transition to naltrexone, those working with them should design such plans safely and appropriately, and ensure access to naloxone for patients, their families, friends, and loved ones. Medical professionals should think about long-acting injectable naltrexone treatment for strongly motivated patients, health care providers, or those seeking to avoid an agonist treatment course. Motivational interviewing can also play an important role in revealing patient preferences and negotiating and encouraging optimal treatment plans.