Only 3% of Young Patients Stay on Addiction Medication
A JAMA Network Open study of 229,847 Medicaid-enrolled youths found just 3.1% stayed on opioid addiction medication for 180 days. What it means for New York.
Getting a teenager into addiction treatment is one problem. Keeping them there is another, and a new study puts a precise number on the second one.
Researchers at Mass General Brigham tracked 229,847 adolescents and young adults who were diagnosed with opioid use disorder while enrolled in Medicaid, following eight years of insurance claims. About half started treatment within two weeks. By six months, 3.1 percent were still taking a medication that clinicians consider the standard of care. The study, published October 2 in JAMA Network Open, is the most detailed accounting to date of where young patients fall out of the system, and it arrives as New York — home to one of the nation's largest Medicaid programs — heads into a coverage shake-up that could make retention harder still.
A Funnel That Narrows at Every Stage
The study looked at people ages 13 to 25 with a Medicaid claim for opioid use disorder between 2016 and 2023. Rather than asking whether treatment exists, the researchers measured how far each patient traveled through it, treating care as a sequence of steps instead of a single event.
How the Researchers Measured Care
They drew on Medicaid enrollment files and insurance claims, which record what a clinician actually billed — a visit, a prescription, a refill — rather than what a patient reported. The team then sorted each person into a stage: no treatment, treatment initiated, treatment engaged, medication started, medication continued.
The attrition compounds quickly.
| Stage of care | Share of youths with opioid use disorder |
|---|---|
| Started treatment within 14 days of diagnosis | 50.7% |
| Attended two or more follow-up visits | 32.9% |
| Began a medication (of those who engaged) | 46.3% |
| Still on medication at 180 days | 3.1% |
Read across that table and the shape of the problem becomes clear. Half of the young people diagnosed never returned for a second contact. Of the third who did, more than half never received buprenorphine, methadone or extended-release naltrexone — the three medications the study tracked. And only about one in 32 was still on a prescription when the six-month mark arrived.
Scott Hadland, the chief of adolescent and young adult medicine at Mass General Brigham for Children and the study's lead author, framed the gap bluntly. "Our findings suggest large gaps in care that are keeping this population vulnerable to premature death and poor health outcomes," he said in a statement accompanying the research.
Why Half a Year Is the Threshold
Six months is not an arbitrary cutoff. It has become the field's standard quality measure because treatment duration tracks outcomes. The National Institute on Drug Abuse puts it plainly: people with opioid use disorder who are treated with methadone or buprenorphine are less likely to die or overdose than those who receive no medication at all.
Youth-specific evidence points the same direction. A separate Mass General Brigham study published in Pediatrics in November, drawing on more than 11,000 adolescents and young adults in Massachusetts, found that those who stayed on buprenorphine for twelve months had almost half the risk of an overdose or hospitalization compared with those who stopped early. That was an association, not proof of cause, but it is the strongest signal available that continuity — not the first prescription — is what protects young patients.
Hadland told STAT that the stakes are not abstract. He described treating patients who died of overdose as recently as the past month. "There have been some real positive changes, and then there have been some really catastrophic changes," he said.
The Rules That Steer Teenagers Away From Methadone
Part of the drop-off is structural rather than clinical. For most of the period the study covered, federal rules required minors to document two unsuccessful attempts at short-term treatment before they could qualify for methadone maintenance. A 2024 federal rule removed that requirement, but the practical barriers did not disappear. Minors still generally need a parent or guardian's consent, and many specialty clinics simply will not enroll patients under 18.
The consequence shows up in the data. Fewer than ten minors in the entire study received methadone, according to STAT's reporting. Buprenorphine is easier to obtain but still uncommon: most adolescent residential programs do not offer it, which means a young person can complete a stay of treatment without ever being started on the medication that carries the strongest evidence base.
Turning 18 flips a switch. "Once you turn 18, a whole bigger field of people are open to helping you. And it's just a number if you think about it," Sivabalaji Kaliamurthy, a child and adolescent psychiatrist who treats addiction, told STAT. The system, in other words, is built around an age cutoff that has little to do with when someone is ready to stop using drugs.
Who Falls Out First
The averages hide an uneven picture, and the unevenness is the part policymakers are most likely to act on. Younger adolescents and racially minoritized youths moved through the stages of care at consistently lower rates than older or white patients, the researchers found. STAT reported that Black youths left care at higher rates than their white peers, a pattern that echoes disparities documented across adult addiction treatment.
Sarah Bagley, an internist and pediatrician at Boston Medical Center, argued that the field has underinvested in the part of care that keeps people coming back. "It doesn't matter how good the treatment we have actually is if we haven't created a clinical environment that is appealing, safe, compassionate, and nonjudgmental," she told STAT. "That engagement piece is everything, because without it, nothing."
That observation reframes retention as a design problem rather than a patient failing. A clinic that cannot see a 16-year-old without a guardian present, that offers no evening hours, or that treats a missed appointment as a discharge produces the numbers this study found.
What This Means in New York
New York does not appear as a separate line in the study, but the state sits squarely inside its findings. Its Medicaid program is among the largest in the country, and roughly 3.6 million of its enrollees live in New York City alone, according to the city Health Department. Medicaid is also the single biggest payer for addiction treatment in the state, which means any churn in enrollment translates directly into interrupted prescriptions.
A Coverage Deadline in January
That churn is scheduled. Federal work requirements for Medicaid begin January 1, 2027, for new applicants and March 1, 2027, for renewals. The city estimates that up to 900,000 enrollees could be at risk of losing coverage under the new documentation rules, which require proof of at least 80 hours a month of work, school, volunteering or job training, or a household income of at least $580 a month. Some people with substance use disorder will qualify for exemptions, but exemptions only help if the paperwork gets filed. Every lapse in coverage is a potential gap in a buprenorphine prescription, and this study shows how fragile that chain already is.
There is a counterweight worth naming. New York's overdose deaths have fallen for three consecutive years, a decline state officials have credited to naloxone distribution, mobile medication units and expanded treatment access. Those gains rest on the same treatment infrastructure that the retention data suggest is leaking patients at the front end. Getting someone started is the achievement. Keeping them started is the unfinished work.
What Families Should Ask
The study's most practical lesson is that a diagnosis or a first appointment is not protection. Parents and caregivers evaluating a program can ask a short set of questions that get at continuity directly.
- Does the program prescribe medication for opioid use disorder on site, or does it refer elsewhere?
- Does it treat patients under 18, and what does it require for consent?
- How does it stay in touch between visits — phone check-ins, telehealth, a case manager?
- What happens after a missed appointment: a discharge, or a call?
- If coverage lapses, will the program help with re-enrollment?
For help finding care, New Yorkers can call the OASAS HOPEline at 1-877-846-7369, a 24-hour line for addiction and problem gambling, or search FindAddictionTreatment.ny.gov. The federal SAMHSA National Helpline, 1-800-662-4357, offers free, confidential referrals in English and Spanish, and FindTreatment.gov lists programs that offer medication. Families can also keep naloxone on hand; it is available over the counter, and it reverses an overdose regardless of whether a treatment slot has opened. In a crisis involving substance use or mental health, 988 reaches the Suicide and Crisis Lifeline.
The Limits of the Evidence
The study is observational and built from billing records, which shapes what it can and cannot say. Claims data show where patients disappear from the system but not why. Care that was never billed to Medicaid — a school counselor, a free clinic, a telehealth visit paid out of pocket — would not appear at all. The sample covers only Medicaid enrollees, so the results may not match those of privately insured families. And the data end in 2023, before the newest prescribing rules and the expansion of telehealth had time to register.
Those caveats leave the central finding intact. Young people with opioid use disorder are being connected to care in growing numbers, and then losing it. Whether New York's treatment system can close that gap before the coverage changes land is the question the next set of numbers will answer.
Written by
MTNYC Editorial TeamThe MTNYC Editorial Team is a group of healthcare writers, researchers, and addiction specialists dedicated to providing accurate, compassionate, and evidence-based information about addiction treatment and recovery resources in New York State.
