Two in Five GLP-1 Users Skipped, Cut or Delayed Doses in 2025, a CDC Survey Finds
A Nov. 12 CDC report finds 9.8 percent of U.S. adults used a GLP-1. Of users, 22.6 percent took a compounded version and 44 percent changed their dosing.
About 9.8 percent of U.S. adults took a glucagon-like peptide-1 receptor agonist in the 12 months before they were surveyed in July and August 2025, and 44 percent of those users changed how they took it, most often by skipping doses. The figures come from a Centers for Disease Control and Prevention report published Nov. 12, 2026, in Morbidity and Mortality Weekly Report.
Among users, 22.6 percent said they had taken a compounded version and another 8.8 percent did not know. Compounded users were more likely to change how they took the drug: 62 percent did, against 36 percent of users of non-compounded products. Among those who changed their use, cost was the most common reason, cited by 38 percent, followed by availability at 36 percent.
Who answered, and who uses these drugs?
The estimates come from round 7 of CDC's Rapid Surveys System, 8,118 completed online surveys drawn from two probability panels, NORC's AmeriSpeak and Ipsos's KnowledgePanel, weighted to National Health Interview Survey totals. Of those respondents, 976 reported GLP-1 use and answered follow-up questions. Cumulative response rates were 13.3 percent and 3.8 percent for the two panels, and use was self-reported, including whether a product was compounded.
Use was highest at ages 40 to 64 (13.7 percent) and lowest at 18 to 39 (4.8 percent). It was 11.9 percent among women and 7.6 percent among men, 47.4 percent among adults ever told they had diabetes and 24.1 percent among those reporting class 3 obesity. Only 2.7 percent of uninsured adults reported use.
Which patients reported compounded products?
Compounded use was concentrated among younger users, women, the privately insured and people without diabetes. The table pulls the subgroup estimates from the report.
| Group of GLP-1 users | Used a compounded version |
|---|---|
| All users | 22.6% (18.9 to 26.8) |
| Women | 27.3% (21.9 to 33.5) |
| Men | 14.8% (11.1 to 19.4) |
| Ages 18 to 39 | 28.0% (19.9 to 37.8) |
| Ages 65 and older | 18.6% (10.5 to 30.6) |
| Private insurance | 25.9% (21.3 to 31.1) |
| Public insurance | 12.5% (7.8 to 19.6) |
| Ever told they had diabetes | 9.4% (6.6 to 13.2) |
| No diabetes | 34.9% (28.7 to 41.6) |
One in three users without diabetes reported a compounded product, against fewer than one in ten users with diabetes. The group without diabetes includes patients taking these drugs for weight management rather than glycemic control.
How does the survey window line up with the shortages?
The report does not say when respondents took compounded products, and the timing matters. Compounders could make copies of semaglutide and tirzepatide while the drugs were on the Food and Drug Administration's shortage list, and FDA then gave them grace periods to stop. We lined the 12-month recall window up against FDA's own dates.
| Date | Event |
|---|---|
| Dec. 19, 2024 | FDA finds the tirzepatide shortage resolved |
| Feb. 21, 2025 | FDA finds the semaglutide shortage resolved |
| March 19, 2025 | Grace period ends for outsourcing facilities compounding tirzepatide |
| May 22, 2025 | Grace period ends for outsourcing facilities compounding semaglutide |
| July to August 2025 | Survey fielded, 40 to 101 days after May 22 |
| April 30, 2026 | FDA proposes excluding semaglutide, tirzepatide and liraglutide from the 503B bulks list |
| Nov. 12, 2026 | MMWR report published, about 14 months after fieldwork ended |
By our arithmetic, a respondent surveyed July 1, 2025, was asked about a window in which 235 of 365 days fell before FDA declared the semaglutide shortage over. For a respondent surveyed Aug. 31, it was 174 of 365 days. So the 22.6 percent figure mixes compounding during the shortage with compounding after it, and the survey cannot separate the two. What it does show is that, one to three months after the last grace period closed, more than one in five users reported a compounded product within the past year.
The survey cannot say whether a patient's compounded semaglutide came from the shortage months or after them. A medication history can, if someone asks.
Why did patients change their dosing?
Skipping doses was the most common change, reported by 28 percent of users. About 12 percent reduced or stopped use because of side effects or symptoms. The age gradient was steep: 54 percent of users aged 18 to 39 modified their use against 30 percent of those 65 and older, and 37 percent of the younger group delayed filling a prescription against 6 percent of the older group. Cost drove much of the difference, cited by 58 percent of younger users who modified their use and 20 percent of older ones. The authors summarize the finding plainly: "Approximately two in five GLP-1 medication users reported modifying their medication use."
What it means in clinic
A medication list that reads "semaglutide 1 mg weekly" may not describe what the patient is injecting or how often. The report supports asking two direct questions at each visit: where the product comes from, and what dose the patient actually took in the past month. That history matters for interpreting weight and glycemic response and for counseling on newer label warnings, including the optic nerve warning FDA added to all GLP-1 labels this month.
Cost-driven skipping also bears on how practices weigh newer options, such as the oral agent orforglipron, once prices are known. On compounding, the open item is FDA's April proposal to keep the three molecules off the 503B bulks list, whose comment period closed June 29, 2026; we found no final decision as of Nov. 13. HHS took a more permissive line on six compounded peptides, including BPC-157, in an Oct. 28 policy whose comment period closes Nov. 13, so the two compounding tracks are moving in opposite directions.
Reported, fact-checked and edited by The Script Pad's editorial staff. See How we use AI.
This article is for professional education and does not replace clinical judgment. Treatment decisions should be based on the individual patient and current guidelines.
