A woman I counseled some years ago had been on buprenorphine for two years and was doing well by every measure that mattered to her. She had her job back, she had her daughter back, and she was sleeping through the night for the first time since her twenties, partly because her psychiatrist had put her on a low dose of clonazepam for the panic attacks that had followed her out of active addiction. Then a new doctor looked at her chart, saw an opioid and a benzodiazepine on the same medication list, and told her one of them had to go. She chose to taper the buprenorphine, because that was the one that carried the stigma. She was dead within the year.
I have thought about her often, and a large Scottish study published this summer in the International Journal of Drug Policy speaks directly to what happened to her. Its finding is more specific than the usual warning about mixing sedatives with opioids.
What the Study Found
Megan Glancy, Andrew McAuley, and colleagues at Glasgow Caledonian University and partner institutions linked prescribing records, specialist drug treatment records, and death registrations for 40,753 people who were prescribed opioid agonist treatment in Scotland between 2015 and 2020. Opioid agonist treatment, or OAT, is the medical term for methadone or buprenorphine maintenance. During follow-up, 40 percent of the cohort was also prescribed a benzodiazepine at some point, and over 189,202 person-years of observation there were 3,159 drug-related deaths.
The design feature that matters most is this. In addition to asking whether a benzodiazepine prescription raised the death rate overall, the researchers split each person's time into periods on OAT and periods off it, out to twenty-four months after treatment ended, and compared death rates within each.
Taken across all time, a prescribed benzodiazepine was not associated with a statistically meaningful rise in drug-related death, with an adjusted incidence rate ratio of 1.13 and a confidence interval that crossed one. During the months a person was actually on methadone or buprenorphine, the ratio was 0.89, meaning there was no detectable excess risk at all. During the months after a person left treatment, the picture changed. In those periods a prescribed benzodiazepine was associated with a 55 percent higher rate of drug-related death, with a confidence interval that did not cross one.
Illicit benzodiazepines behaved differently. People who reported using street benzodiazepines at the start of the study had a 60 percent higher rate of drug-related death, and that excess was present whether they were on OAT or off it. The study period coincided with a rise in drug deaths across Scotland in which illicit benzodiazepines were increasingly found at post-mortem, so this is not an abstract concern there.
Why the Timing Is the Whole Story
Picture the same medication list in two different months. In the first, the patient is taking her buprenorphine every morning and her clonazepam at night, and in this dataset that combination did not carry a measurable excess of death. In the second month, she has stopped the buprenorphine, her opioid tolerance has been falling week by week, and the clonazepam is still in the cabinet. That is the month in which the death rate climbed.
Benzodiazepines and opioids both suppress breathing, and the danger of combining them has been understood for decades. What this study adds is that the association was concentrated in the periods off treatment. My interpretation, offered as a clinician rather than as a claim from the paper, is that this is the point where opioid exposure has become unstable, because a person who leaves treatment and returns, even occasionally, to a drug supply of unknown strength has lost the steady tolerance that treatment provided.
This fits what many of the same researchers reported in 2023 in The Lancet Public Health. In a cohort of 46,453 people prescribed OAT in Scotland between 2011 and 2020, drug-related death rates were almost three and a half times higher during periods off treatment than during periods on it, after adjusting for other factors. Time off OAT has carried a far higher death rate than time on it for years. The new study suggests that a benzodiazepine prescription tends to amplify that particular danger rather than create a new one.
What This Study Cannot Tell Us
This was a retrospective cohort, which means the researchers observed what happened rather than assigning anyone to a treatment. A prescription record shows that a drug was dispensed, not that it was taken as directed. Illicit benzodiazepine use was self-reported at the start of the study and was not tracked over time, so a person who started or stopped using street pills midway through would have been misclassified. People who are prescribed benzodiazepines may also differ from those who are not in ways that no statistical adjustment fully removes, since they tend to carry more anxiety, more trauma, and more psychiatric complexity to begin with.
The setting matters as well. Scottish overdose deaths in those years increasingly involved street benzodiazepines, and the drug supply there differs from the fentanyl-driven American one, so the direction of the finding is likely to travel while the exact numbers may not.
What This Means for You or Someone You Love
If you are on methadone or buprenorphine and someone has told you that your anxiety medication means you need to come off your opioid treatment, this study is worth bringing to that conversation. The evidence here points the other way. The risk in the data sat with leaving OAT, not with staying on it while taking a prescribed benzodiazepine, and the earlier work from many of the same researchers found death rates more than three times higher during time off treatment than on it. Ask your physician to address the benzodiazepine question inside your treatment rather than by ending it.
If you are a family member, understand that the medication your loved one takes every morning is not a crutch or a substitute addiction. In these data, being on it was associated with a lower death rate, even when a second sedating drug was in the picture. The pressure families sometimes apply to "get off the methadone" comes from love and from stigma in roughly equal parts, and it can cost the life it is meant to protect.
If you are using street benzodiazepines, which in the United States often means counterfeit pills of unknown content, that risk was present on and off treatment alike, and it is the one thing in this study that OAT did not offset. That is an argument for telling your treatment team the truth about what you are taking, not for hiding it.
Our Rescue From Rehab program was built around transitions like the one that killed my former client. We do not treat the end of a medication as a finish line, and we treat the anxiety, sleep disturbance, and trauma that drive benzodiazepine use as clinical problems to be worked on within the recovery plan. That work belongs alongside what we know about how long brain recovery actually takes, and alongside the growing evidence that addiction is a disorder of identifiable brain networks rather than a failure of character. The woman I described did not die because she took two medications. She died because a system that should have protected her told her to give up the one that was keeping her alive.
Illustration created with AI image tools.
This article was drafted with the assistance of AI writing tools, then reviewed, edited, and approved by Dr. Drew W. Edwards, Ed.D., who holds full editorial responsibility for its content.
Sources: Glancy M, Palmateer N, Curnow E, Hickman M, Macleod J, Yeung A, Bishop J, Priyadarshi S, Barnsdale L, Trayner KMA, Hutchinson S, McAuley A. Drug-related death rates associated with prescribed and illicit benzodiazepine exposure among people receiving opioid-agonist treatment in Scotland, 2015 to 2020. International Journal of Drug Policy. 2026;155:105387. DOI: 10.1016/j.drugpo.2026.105387 · McAuley A, Fraser R, Glancy M, et al. Mortality among individuals prescribed opioid-agonist therapy in Scotland, UK, 2011–20: a national retrospective cohort study. The Lancet Public Health. 2023;8(7):e484-e493. DOI: 10.1016/S2468-2667(23)00082-800082-8)