Adults whose first sleeping pill prescription covered about a month were far more likely to still be filling sleeping pills years later than adults who started with a two-week supply, according to a study of 129,287 US patients published September 3, 2026 in JAMA Network Open. In the year after that first prescription, 38.1% of the month-supply group filled another one, compared with 15.6% of the shortest-supply group.
The gap was still there seven years on. The researchers, led by Cloughesy and colleagues, used insurance claims covering January 1, 2007 through February 29, 2024, and tracked patients starting zolpidem, eszopiclone or zaleplon, the drugs known as Z-drugs.
Key takeaways
- A first sleeping pill prescription of 25 to 34 days was linked to a refill rate 22.4 percentage points higher one year later than a first prescription of 1 to 14 days.
- Seven years after that first prescription, patients who started on 25 to 34 days were still 7.6 percentage points more likely to be filling a Z-drug.
- This is claims data, not a trial, so it cannot prove the longer prescription caused the longer use.
What the study found
The researchers sorted patients by how many days their very first Z-drug prescription covered: 1 to 14 days, 15 to 24 days, 25 to 34 days, or 35 days and up.
In year one, 15.6% of the 1-to-14-day group filled another Z-drug prescription. That rose to 38.1% for the 25-to-34-day group and 49.5% for those given 35 days or more. Comparing the 25-to-34-day group with the shortest group, that is an adjusted difference of 22.4 percentage points, or about 22 more people out of every 100 still filling pills a year later. The math is very likely to land between 21.8 and 23.0 percentage points (95% CI, 21.8 to 23.0). Put another way, roughly 2.4 times as many patients were still refilling.
By year seven, the rates had fallen but the ranking held: 7.9% for the shortest group, 15.6% for the 25-to-34-day group and 18.9% for the longest. That leaves the 25-to-34-day group 7.6 percentage points higher, very likely between 7.2 and 8.1 (95% CI, 7.2 to 8.1).
Volume followed the same pattern. Patients who started with 25 to 34 days filled 42.8 additional days’ supply in the first year (95% CI, 41.7 to 44.0) and 17.0 additional days in the seventh year (95% CI, 16.1 to 18.0).
Dr. Kumar’s take
The honest caveat comes from the authors themselves: these associations cannot separate an effect of the prescription length from confounding by indication. Doctors may hand a 30-day supply to the patients whose insomnia is worse to begin with, and those are the patients who would refill anyway.
That caveat does not explain the whole shape of the finding. A seven-year tail is a long time for a drug that is meant for short courses, and the same story has played out with other short-course drugs that became permanent, as in the tripling of PPI use in US adults over twenty years.
Here is the mechanism the numbers point at without proving it. A longer first prescription means more nights on the drug before the bottle runs out. When the bottle runs out, sleep gets worse for a few nights. The patient reads those bad nights as confirmation that the insomnia was real and untreated. What they may actually be reading is the drug leaving. The study cannot tell those two readings apart, and the authors say so plainly: longer prescriptions may have gone to patients with more severe or chronic insomnia, who were also more likely to keep taking Z-drugs. What the numbers do show is that the shortest first prescriptions in the study, 1 to 14 days, ended with the fewest patients still refilling.
The useful part is that acting on this costs nothing and risks nothing.
What it means for you
If you are being started on a sleeping pill for the first time, ask for the smallest supply that makes sense, roughly a week, with a scheduled visit to decide what happens next. A shorter bottle does not deny you treatment. It just forces a second conversation.
Ask at the same visit for a referral to cognitive behavioral therapy for insomnia, or CBT-I. This is not alternative medicine. The American College of Physicians guideline states that “ACP recommends that all adult patients receive cognitive behavioral therapy for insomnia (CBT-I) as the initial treatment for chronic insomnia disorder,” a strong recommendation. Medication is the add-on for when CBT-I alone has not worked.
If you are already years into refills, none of this means stop abruptly. Sleeping pills should be tapered with your prescriber, not dropped. The study is about how these prescriptions begin, and its practical value is aimed at people who have not started one yet.