Skip to Main Content

Secretary of Health and Human Services Tom Price sparked a firestorm last week with his comments about medication-assisted treatment, saying that “if we’re just substituting one opioid for another, we’re not moving the dial much” in the nation’s opioid epidemic. Notably, the former surgeon general, Dr. Vivek Murthy, took him to task on Twitter for, as he put it, moving away from evidence-based treatment protocols.

A spokesman for HHS later said that Price supports expanding access to various treatment and recovery services, including medication-assisted treatment, and was only arguing that the treatment that’s right for one person isn’t necessarily right for another.


But how much do we know about the effectiveness of medication-assisted treatment, or MAT, and what do we not know? Here’s STAT’s guide to the science.

What is MAT? How does it work?

Understanding what heroin does in the brain and in the body is crucial to understanding why medication-assisted treatment — sometimes called medication-assisted therapy— works, said Dr. Nora Volkow, the director of the National Institute on Drug Abuse. (NIDA is one agency of the National Institutes of Health, which is part of HHS.)

Opiates work by crossing the blood-brain barrier and attaching to receptors on brain cells, which triggers a cascade of neurotransmitters and brain activity and produce the high that people feel. That brain activity can contribute to physiological dependence — and, only if combined with genetic and psychological factors that modify the way a drug is perceived, addiction.


For those struggling with opioid addiction, breaking the habit can be demanding. Dr. Scott Lukas is researching and developing drugs to treat the condition. Alex Hogan

MAT can work in one of two ways. Doctors can give people opiates that activate the same receptors but are absorbed into the blood over a longer period of time — staving off withdrawal symptoms and breaking a psychological link between taking a drug and immediately feeling high. Doctors can also give someone an opioid antagonist — a non-opioid drug that sits on those same receptors and blocks them — so that if someone relapses, he or she won’t feel anything. (If someone goes off the drug and relapses, however, that can have deadly consequences. A person’s tolerance for the drug decreases after long periods of sobriety — especially after antagonist treatment.)

That means most of these treatments just substitute one opiate for another, as Price correctly noted. Doctors think this is a good thing?

Yes, they do. “The drug that we’re replacing is a dangerous one that will kill you, and we’re replacing it with a drug that allows you to go back to work and have money in your pocket and allow you to live normally again,” said Dr. Stuart Gitlow, past president of the American Society of Addiction Medicine.

Price’s argument that “what’s right for one person isn’t necessarily right for another person” is a valid point, experts say; data have shown that MAT is effective for the general population, but that doesn’t mean it will be equally effective for every individual, given the particulars of someone’s life and someone’s addiction.

How strong is the body of evidence behind MAT?

“I don’t think that there’s any areas where the data is shaky. It clearly shows better outcomes with medication-assisted therapy than without it,” Volkow said. “Studies have shown that outcomes are much better when you are on medication-assisted therapy. For one, it decreases risk of relapse — significantly. Second, MAT has also been shown to be effective in preventing infectious diseases like HIV. Third, medication-assisted therapy has been shown to be effective in preventing overdoses,” she said.

Though it’s hard to point to one seminal paper in the field, there is a substantial body of literature. For example, one paper found deaths from opioid overdoses fell as treatment with buprenorphine, one of three drugs often used for MAT, became more popular in Baltimore — a “statistically significant and strong” correlation. Multiple meta-analyses compared MAT to plans without any replacement therapies. (Methadone wasn’t correlated with reducing deaths in the paper or in a 2009 meta-review, but it did seem to decrease heroin use.)

While there’s plenty of evidence supporting the efficacy of MAT, there are still gaps in the scientific literature that researchers want to see addressed.

What are those gaps?

One big gap in the research is how to pick the right drug and the right dose for an individual patient. According to 2015 national practice guidelines from the American Society of Addiction Medicine, “there is some evidence supporting the relative efficacy of one medication over another, but in many cases, there are no good-quality studies comparing the relative benefits of one medication over another.”

Each drug does carry different risks and benefits. People can overdose and die on methadone, whereas they can’t with buprenorphine or naltrexone. Buprenorphine and naltrexone can be taken orally; naltrexone also comes in an injectable form. And people who need surgery can be taken off buprenorphine more quickly than they could be if they took methadone.

Ultimately, the choice will be based on a doctor’s conversation with his or her patient, Gitlow said.

Also, there’s been no clinical research done about MAT for people who have become addicted to fentanyl. Volkow thinks she knows which drugs should be most effective, but there will be an NIDA meeting on July 6 to discuss how to do the kinds of studies needed in this area.

What about talk therapy or psychotherapy?

Some studies, including one funded by NIDA, have shown that MAT resulted in some benefits even without psychotherapy. According to reviews published in the American Psychological Association’s journal, Psychiatric Services, studies found that adding psychotherapy or cognitive behavioral therapy had no effect on the outcomes of buprenorphine- and methadone-based treatment. Still, support programs are often a standard part of addiction treatment in addition to MAT.

“For every major, chronic, life-threatening disease, there’s always a combination of therapeutic modalities that are used. For instance, in hypertension, we tell people to exercise and lose weight and quit smoking and don’t eat salt in their diet and by the way, take this antihypertensive [drug],” Gitlow said. “For opioid use disorder, it’s really the same thing.”

  • That’s interesting that you could supplement the traditional methods of treating addiction with medication to make it easier. I have heard that breaking an addiction is really hard to do, so I could see how having as much help as possible with that would be good. I should recommend that my buddy tires that out since I think that he might have an issue with addiction.

  • As others have rightly defended so, OUD can be as iatrogenically-acquired as Type II DM and Hypertension. Harm-reduction therapy in OUD is the same as applied in both of the etiologies referenced above. The same biases and stigmas that can be assigned to one disease entity can be assigned to another, as so will accompany our failures. No one chooses homelessness or addiction. The patient inappropriately prescribed narcotic pain meds who naturally becomes dependent and tolerant who then becomes addicted has less of a choice to tail-spin, as does the non-compliant Type II diabetic or hypertensive who then suffers a heart attack; the consequence and sequelae of disease comes for both. Ignorance is optional.

  • First, long term suboxone use is not good. Second, it does not cure anyone. It is merely a harm reduction model that prolongs the life of someone who will end up dying without comprehensive treatment. I also wish people would stop comparing it to a chronic illness. People with sickle cell or juvenile diabetes did not choose their disease or acquire it through the same manner as drug addicts. Just more junk science

  • For Lea,
    Understand that MAT has taken us from 10-15% recovery to 50% after a year. Very good, but we only bat .500. Good news is that we often do get another chance with the patient after relapse. Bad news is that we have to be disappointed for the patient too frequently. We have a ways to go and improve with research, but things are looking up. Dr. Rust

Comments are closed.