Prescribed and Trapped: Medicare's Benzodiazepine Numbers Reveal a Crisis Built by Doctors

In April 2026, a person writing for the Benzodiazepine Information Coalition described what it has been like to spend the past three years tapering off 4 milligrams of alprazolam — the generic form of Xanax, the most prescribed benzodiazepine in the United States. They had been switched to diazepam, which is supposed to be easier to taper because of its longer half-life, a strategy first described in the Ashton Manual, a self-help tapering guide written by a British physician and pharmacologist in 1999 because so few clinicians knew how to help patients stop. Six months into the taper, still experiencing cognitive fog, waves of anxiety, physical symptoms that arrive without warning, the person wrote: “It’s impossible to work when I’m in withdrawal.”

This person did not start taking Xanax because they were looking for a high. They started because a doctor prescribed it. That distinction — between the popular image of drug dependence and the pharmacological reality of prescription benzodiazepine use — sits at the center of a treatment crisis that new data is making harder to ignore.

Fifty-one percent higher, and rising

A study published in February 2026 in Frontiers in Medicine by Silvernail, Ritvo, Silvernail, and Martin examined Medicare Part D benzodiazepine prescribing patterns across the United States from 2017 to 2023. The finding is striking: while overall U.S. benzodiazepine prescriptions declined approximately 26 percent over that period — falling from roughly 110 million to 81 million annually — Medicare prescriptions for seniors went in the opposite direction, rising approximately 51 percent.

The Medicare beneficiaries in the study received an average of 108 days of benzodiazepine medication per year. The FDA recommends a maximum of 30 days. Every year in the study period exceeded that guideline by more than threefold.

The drugs doing the most prescribing work were alprazolam, lorazepam, and clonazepam — the same three drugs that dominate the clinical landscape and the same three that the patient advocacy community specifically flags as particularly difficult to discontinue. Alprazolam alone accounted for 38 to 47 percent of all Medicare benzodiazepine claims across the study period. It is a high-potency, short-acting drug; those pharmacological properties make it both the most effective anxiety suppressant and among the most difficult benzodiazepines to stop taking.

Most striking in the Frontiers study is the prescriber breakdown. Psychiatrists — the specialists ostensibly most qualified to manage these medications — showed the highest prescribing rates and increased prescribing by approximately 250 percent over the six-year period. Family practice and internal medicine physicians doubled their prescribing. Nurse practitioners and physician assistants prescribed shorter durations, though their overall prescribing volumes also rose. The specialists charged with understanding dependence are the ones prescribing the most, for the longest.

Geographically, 13 states showed significantly higher odds of extended prescribing compared to California. New Jersey had 2.5 times higher odds. This isn’t a regional variation around the margins — it describes materially different standards of practice operating under the same FDA label guidance.

What dependence actually feels like

Dependence on benzodiazepines does not develop because a patient lacks willpower. It develops because benzodiazepines work on GABA-A receptors — the primary inhibitory neurotransmitter system in the brain — and chronic exposure causes the brain to physiologically adapt. The receptors downregulate. The natural calming system becomes less responsive. What started as a medication becomes a maintenance requirement.

The psychological experience of this process is, in the words of a 2024 qualitative study published in Drug and Alcohol Dependence Reports, a form of “tipping point” — a moment when patients realize that the drug that was helping them is now something they cannot function without. The dominant finding from that study’s interviews with people who had developed benzodiazepine dependence was not recklessness or recreational use: it was prescriber-initiated dependence developed within prescribed parameters. People followed their doctors’ instructions, exactly, and became dependent anyway.

The dominant finding from that study’s interviews with people who had developed benzodiazepine dependence was not recklessness or recreational use: it was prescriber-initiated dependence developed within prescribed parameters.

The brain science explains why. Tolerance builds gradually — the calming effect diminishes while the dose that established that calm becomes, neurologically, the new baseline. Anxiety returns between doses, not at its original intensity but amplified by rebound hyperactivity in the GABA system. That rebound anxiety — more intense than what prompted the prescription in the first place — reinforces continued use. From the outside, this looks like the person needing more medication. From the inside, the person cannot distinguish between their original anxiety, the drug’s effect, and withdrawal.

Psychologically, long-term benzodiazepine use can produce what the literature describes as a progressive loss of self-efficacy: the belief that one cannot sleep, cannot manage stress, cannot go to a social event without the medication. This is not a personality trait. It is an acquired neurological state that can be attributed to the drug’s effects on cognitive confidence and emotional regulation. The research on benzodiazepine-associated cognitive impairment describes measurable deficits in memory, processing speed, and executive function — deficits that persist for months and sometimes years into discontinuation.

In a subset of patients estimated at 10 to 15 percent, withdrawal produces a protracted course of symptoms — described in the literature as “waves and windows,” periods of acute symptoms alternating with periods of relative stability — that may continue for a year or more after the medication is fully stopped. The person tapering from 4 milligrams of alprazolam, six months in and still not through, is not unusual. They are, by the research’s own accounting, in a reasonably common experience for which the medical system has historically provided very little support.

A guideline fifty years in the making

In May 2025, ten professional medical societies — including the American Society of Addiction Medicine, the American Academy of Family Physicians, the American Psychiatric Association, the Society of General Internal Medicine, and five others — published a joint clinical practice guideline on benzodiazepine tapering. Authored by Aleksandra E. Zgierska of the University of Pennsylvania, Maureen P. Boyle of Rutgers, and Joseph Conigliaro of the Icahn School of Medicine at Mount Sinai, the guideline represents the most comprehensive clinical consensus on benzodiazepine discontinuation ever assembled.

Its core recommendation: taper no faster than 5 to 10 percent of the total daily dose, every two to four weeks. Adjust based on patient response. Do not discontinue abruptly. Monitor for withdrawal symptoms at every step. Offer adjunctive psychosocial support. Respect that the process may take more than a year.

This is the Ashton Manual’s core methodology, codified in a peer-reviewed guideline from 10 specialty societies, 26 years after a retired pharmacologist in Newcastle wrote it herself because the clinical establishment hadn’t gotten around to it. The patient advocacy community has been saying this for decades. They can now point to a CPG.

The guideline also makes an important epidemiological distinction: approximately 1.5 percent of patients treated with benzodiazepines long-term develop a benzodiazepine use disorder that requires specialty addiction treatment. The other 98.5 percent have physiological dependence — a different clinical entity, one that requires a slow taper and monitoring rather than a 28-day residential program. These populations get conflated in clinical practice, in policy, and in insurance coverage decisions in ways that fail both groups.

The treatment gap that the numbers describe

The Frontiers study’s 51 percent rise in Medicare benzo prescribing, set against the 30-day guideline, describes an accumulated stock of people who have been on these medications for years and will need help stopping — most of whom will encounter a medical system that doesn’t know the guideline well, doesn’t know how to do a slow taper, and in many cases will simply tell them to cut their dose in half and call in two weeks.

The patient communities documenting this experience are large. The Reddit forums for benzo recovery and benzo tapering — combined membership over 180,000 — are full of people describing the same experience: a doctor who prescribed for anxiety, a months-long gradual dependence they didn’t notice until they tried to stop, a follow-up appointment where the doctor was unfamiliar with the Ashton Method or the new joint guideline, a sense of being on their own with the internet as their only reliable information source.

That treatment gap — between the scale of the prescribing data and the clinical infrastructure to manage discontinuation — is not a niche problem. It is happening to seniors, to people who started on benzodiazepines for insomnia after a job loss or a divorce or a health scare, to people who have been following their doctor’s instructions for eight years and are now trying to stop and finding that the drug has reorganized their nervous system around its presence.

That treatment gap — between the scale of the prescribing data and the clinical infrastructure to manage discontinuation — is not a niche problem.

The joint CPG from 10 societies is a meaningful step. The 51 percent rise in Medicare prescribing, in the same period that the rest of the country was prescribing less, suggests that the population that most needs that guideline implemented — elderly patients in long-term care, patients on fixed incomes with limited access to specialists who actually know how to taper — is the one most likely to be getting the care that looks nothing like it.

Getting people off benzodiazepines they no longer need, safely and slowly, is possible. The guideline exists. The Ashton Manual has always existed. The people who wrote in April on the Benzodiazepine Information Coalition’s blog have always existed, and will keep existing. What’s been missing is a clinical establishment willing to match the scale of what it prescribed with the scale of what it takes to stop.