Half of Young People Start Opioid Treatment. One in 32 Is Still on Medication Six Months Later.

A patient of Dr. Scott Hadland’s died of an overdose in Massachusetts last month. Hadland, who runs adolescent and young adult medicine at Mass General Brigham for Children, told STAT that “I have had patients die as recently as this past month here in Massachusetts, from overdose.” Hadland said it while explaining a new study of his that, on the surface, contains good news.

We fixed the front door of opioid treatment for young people and forgot to build the hallway behind it.

The study, published in JAMA Network Open, followed 229,847 Medicaid enrollees between 13 and 25 who were diagnosed with opioid use disorder from 2016 through 2023. About half, 50.7 percent, started treatment within 14 days of diagnosis, according to the study’s numbers as reported by Medical Daily. Roughly a third, 32.9 percent, kept showing up: at least two more visits within 34 days. Of those who kept showing up, 46.3 percent were prescribed a medication, which works out to about 15 percent of the whole group. And 3.1 percent, about one in 32, were still taking a medication for opioid addiction 180 days after they started.

Read that last number slowly. Of every 32 teenagers and young adults the system found, one is still on the medication that cuts overdose risk six months later. Hadland’s summary of where the field stands: “There have been some real positive changes, and then there have been some really catastrophic changes.”

Getting in the building was never the same as staying in treatment

The pattern in the study is a staircase with a missing step at every landing. A diagnosis is a billing code, not a person walking through a door. Half of those young people made it to a first appointment within two weeks, and that deserves a real acknowledgment: it is not nothing for a teenager with an opioid problem to show up at all. Then the stairs narrow. A third engaged. Fewer than one in six ended up with a medication in hand. Fewer than one in 30 still had it at the six-month mark.

You may know this staircase from the inside. If you have ever walked into a clinic ready, been handed a form, a referral to somewhere else, and an appointment three weeks out, you know what three weeks is when the thing you are trying to leave is available tonight. The study cannot tell us why any individual stopped. It is built on billing records, so it cannot see the reason someone missed the second visit. It covers only Medicaid enrollees, so it says nothing certain about privately insured families. The data end in 2023, so they predate later changes in prescribing rules and telehealth. Those are real limits, and the team’s own accounting of them matters.

What the numbers can say is where the losses cluster. The authors found Black young people and other racially minoritized groups had worse drop-off at every stage than white peers, per STAT’s reading of the paper. And the methadone line is nearly empty: fewer than 10 minors received methadone across the entire eight-year window, a count the researchers suppressed to protect privacy. Methadone is not the right medicine for everyone. But a medicine that almost no one under 18 can get is not a menu item, it is a rumor.

Sarah Bagley, an internist and pediatrician at Boston Medical Center, put the gap in one line to STAT: “It doesn’t matter how good the treatment we have actually is if we haven’t created a clinical environment that is appealing.” She added that the “engagement piece is everything.” Her point is not about pharmacology. It is about whether a 19-year-old feels, in the first ten minutes, like the building was built for them.

It is about whether a 19-year-old feels, in the first ten minutes, like the building was built for them.

Six months is not an arbitrary line

Why does the six-month mark matter enough to build a study around? Because the same research group has already shown what stopping early costs. In a Pediatrics paper published in November 2025, Hadland and colleagues tracked 11,649 Massachusetts youth ages 13 to 26 who started buprenorphine between 2014 and 2020. Only about one in four, 24 percent, took it reliably for a full year. Compared with those young people, those who stopped between three and nine months had an 82 percent higher risk of overdose; those who stopped in under three months had a 76 percent higher risk; those who took it on and off for the year had a 46 percent higher risk.

That is an association in health records, not a controlled trial. People who stay on a medication may differ from people who do not in ways the records cannot see, such as housing, family support, and how much money is in their pocket. The authors say as much. Still, the direction matches what clinicians see. Hadland’s summary of the finding is blunt: “Amid the fentanyl era, short-term treatment is simply not enough.”

Put the two studies side by side and a picture forms. In the Massachusetts data, the elevated risk sits with everyone who left before a year, and in the national Medicaid data most young people are gone long before six months. If you are in that stretch right now and thinking about stopping because you feel fine, feeling fine is what working looks like. It is a reason to talk to your prescriber before you change anything, not a reason to quit alone.

The people who say “just stop” are often the ones in the room

Here is the part that never shows up in a billing code. A 2018 case series from Hadland, Tae Woo Park and Bagley described a young man in his early 20s with severe opioid use disorder, depression and PTSD who had gotten into therapy, found work and started a relationship while on buprenorphine. The resistance he met was not from the street. Residential staff questioned whether he needed the medication. The director of his sober house questioned his dose and raised it with his parents. His parents questioned it too. His clinic’s answer was to see him regularly and keep telling him the medication was helping his recovery. The authors report he stayed abstinent.

He is a de-identified case in a journal, not a poster child, and one clinic’s case series proves nothing about prevalence. But anyone who has worked in this field recognizes him. Dr. Sivabalaji Kaliamurthy, a child and adolescent psychiatrist who treats addiction, told STAT that “Once you turn 18, a whole bigger field of people are open to helping you.” Before 18, the options narrow: many specialized methadone clinics still do not take minors, and few adolescent residential programs offer buprenorphine, according to the research STAT cited. State law can also be stricter than federal law, as a Vital Strategies explainer on the rule notes. The federal 42 CFR Part 8 final rule of February 2024 dropped the old requirement that a minor document two failed attempts at drug-free or detox treatment before starting methadone maintenance, but it did not override stricter state rules. Whether a 16-year-old in your county can get methadone still depends on where the clinic sits.

The medication itself is no longer the bottleneck. For 22 years, a prescriber who wanted to treat opioid addiction with buprenorphine needed a special federal waiver and could only treat a capped number of patients, a system created by the Drug Addiction Treatment Act of 2000. People who needed the medicine waited in the gap between how many patients the law allowed and how many needed help. Congress ended that in the Consolidated Appropriations Act of 2023, signed December 29, 2022: no waiver, no patient cap, a standard DEA registration is enough where state law allows. We took away the cap and assumed that would fix it. The new study says what happened next. We opened the supply and the stairs behind it were still missing.

We opened the supply and the stairs behind it were still missing.

What a clinic can do before the next team meeting

A diagnosis count does not save anyone. A phone call on day 12 might. If you run or work in a clinic that sees people ages 13 to 25, here is a concrete week.

Pull the list of every patient in that age range who started buprenorphine in the last 180 days. Mark who has not been seen in 14 days. Call those people, and ask one question rather than five: “What is making you want to stop, or what’s getting in the way?” Put a day-14, day-30 and day-90 check-in on the schedule for every new start, and give a named person ownership of those calls. Tell every patient in plain words, at the first visit, that a missed week or a slip is not a discharge and that they can restart without a lecture. Ask your sober-house and residential referral partners whether they accept residents on buprenorphine, and stop sending young people to the ones that do not. The 2018 case series includes a 20-year-old woman who was told at a 12-step meeting that she was “not truly sober” on her medication, stopped it, bought pills from a friend, and arrived in crisis with fentanyl and benzodiazepines in her system. She restarted and has not used other opioids since. Her story ends with a clinic that restarted her medication.

Look at who is dropping at each stage, too. The study found worse drop-off for Black and other minoritized young people. Check your own clinic’s numbers by race at each landing, and bring it to the team.

Why the national numbers do not settle this

The country is in a third straight year of falling overdose deaths. The CDC’s provisional count is 69,973 overdose deaths in 2025, down from 81,313 in 2024, almost 14 percent. Those are provisional figures, incomplete and subject to change, and they hide the states that went the other way: the same CDC release lists New Mexico, Arizona and Colorado with increases of 10 percent or more. A falling national number is a reason to keep going, not a reason to stop counting the people the system is still losing at 90 days.

If you are the person this is about: you are not the failure in these numbers. Thirty-one of every 32 young people in this data were off the medication by six months, and what we built for you is a prescription pad with no phone number attached. Buprenorphine can now be prescribed by any clinician with a standard DEA registration, and naloxone has been available over the counter since 2023. Both are still yours tonight. Restarting after a slip is allowed.

Hadland’s patient did not die for lack of a medicine. There was one, and a clinic, and a doctor who writes about this for a living. The question the study leaves is the one the lede starts with: what happens in the six months after a young person says yes. Right now, in the biggest data set we have, the answer is: almost nothing, and then almost nobody.

Further reading: the treatment and recovery pillar, our opioids hub, and how peer-recovery funding is supposed to supply the human at the other end of that day-14 call.