About 1 in 10 American adults took a GLP-1 drug in the past year, and 44% of those users did not take it the way it was prescribed. Skipping doses was the most common change, reported by 28% of users in a CDC survey published November 12, 2026.1
Cost and availability were the reasons people gave most often. Roughly 1 in 5 users said they had taken a compounded version, and 62% of them departed from their prescription, against 36% of other users.1 The survey ran in July and August 2025, so it describes the year leading up to then, not today.
Who did the CDC survey ask?
The CDC survey asked 8,118 adults about any GLP-1 drug they had taken in the past 12 months, as part of Round 7 of its Rapid Surveys System. The adults came from two online panels built on probability samples of US households, and their answers were weighted to match the National Health Interview Survey so the results stand for the civilian adult population living outside institutions.1
The question named the brands people would recognize: Ozempic, Rybelsus, Wegovy, Mounjaro, Zepbound, Saxenda, Victoza, Trulicity, Byetta and Bydureon BCise. Those contain semaglutide, tirzepatide, liraglutide, dulaglutide or exenatide, and pills counted as well as injections, for diabetes or for weight loss.1 Of the 8,118, 976 said yes. Only those 976 were asked the follow-up questions about compounded drugs and about any changes to their use.1 The report's lead author is Samantha L. Pierce, writing with three co-authors, and it discloses no conflicts of interest.1
How many Americans take GLP-1 drugs?
An estimated 9.8% of US adults, about 1 in 10 Americans, took GLP-1 drugs at some point in the past year, with a confidence interval from 9.0% to 10.7%.1 Use was far from even across groups.
| Group | Took a GLP-1 in the past year |
|---|---|
| All adults | 9.8% |
| Women | 11.9% |
| Men | 7.6% |
| Ages 18 to 39 | 4.8% |
| Ages 40 to 64 | 13.7% |
| Ages 65 and older | 11.1% |
| Ever had diabetes | 47.4% |
| Never had diabetes | 5.7% |
| Healthy weight | 4.7% |
| Severe obesity (class 3) | 24.1% |
| Private insurance | 10.6% |
| Public insurance | 10.6% |
| Uninsured | 2.7% |
Nearly half of adults who had ever been diagnosed with diabetes took a GLP-1, compared with 5.7% of those who had not, the widest gap in the table. Use also ran high among people with sleep apnea, at 23.3%, and cardiovascular disease, at 21.4%.1 Women used them more than men, 11.9% to 7.6%. Uninsured adults reported the least use of any group.
Why were people at a healthy weight taking GLP-1s?
The CDC report does not say why people at a healthy weight were taking a GLP-1, though its authors suggest three possibilities. Some may have diabetes without excess weight, some may have reached a healthy weight on the drug, and some may be using it off-label.1 Most users, 80%, had overweight or obesity. BMI here comes from the height and weight people reported themselves.1
Among healthy-weight users, 26.9% had taken a compounded product, above the average for all users.1
How many GLP-1 users take compounded versions?
Of the people who took a GLP-1, 22.6% said they had taken a compounded version, with a confidence interval from 18.9% to 26.8%. Another 8.8% did not know whether theirs was compounded.1
| GLP-1 users | Took a compounded version |
|---|---|
| All users | 22.6% |
| Never had diabetes | 34.9% |
| Ever had diabetes | 9.4% |
| Women | 27.3% |
| Men | 14.8% |
| Ages 18 to 39 | 28.0% |
| Ages 65 and older | 18.6% |
| Private insurance | 25.9% |
| Public insurance | 12.5% |
About 1 in 3 users who had never had diabetes had taken a compounded version. Among users with diabetes it was fewer than 1 in 10, and women reported compounded use nearly twice as often as men.1 The CDC authors read the 8.8% who were unsure as "potentially reflecting challenges in distinguishing compounded from FDA-approved products."1
What does "compounded" mean in this survey?
Compounded meant whatever each person took it to mean, because the survey asked only whether they had taken "a generic, compounded version of this medication", with a note that compounded drugs contain the same active ingredients as brand-name ones but are not FDA-approved.1 It did not ask where the drug came from. Someone injecting semaglutide from a gray-market vial sold for research use could have answered yes, no or don't know, and the survey would count them alongside a patient filling a pharmacy prescription. Both the 22.6% and the 8.8% describe people who knew or suspected they were not on a brand-name product, whatever its source.
The report also cites earlier work on compounded GLP-1s. Analyses of FDA adverse event reports found "disproportionately higher reporting of preparation errors, contamination, and manufacturing issues for compounded products". In a review of websites selling compounded GLP-1s, about 40% stated or implied the drugs were FDA approved and about 80% used brand names in their advertising.1 Our compounded semaglutide guide covers what pharmacies are still allowed to make.
How did people change the way they took GLP-1s?
Of all users, 44% said they had not taken their GLP-1 the way it was prescribed, and skipping doses was the most common change, at 28%.1 The report also charts taking less medication and delaying a prescription fill, though its text gives no overall share for either. A smaller group, about 12% of users, reduced or stopped their GLP-1 "due to side effects or symptoms."1 Our GLP-1 side effects guide covers what the trials recorded, and the weight regain guide covers what studies measured after people stopped.
Why do people skip GLP-1 doses?
People who skipped or stretched their GLP-1 doses most often pointed to cost and supply. Among users who changed their use, 38% cited cost and 36% cited availability.1 Younger users felt the price most: 58% of those aged 18 to 39 who changed their use cited cost, and only 20% of those 65 and older did.1
An earlier study the report cites, covering 2022 to 2024, found that about half of adults with overweight or obesity but no diabetes who started a GLP-1 stopped because of cost or insurance problems. Another 12% could not fill their prescriptions because of shortages.1 The report's abstract ties the changes, "particularly among younger adults," in part to cost- and access-related challenges.1
Who was most likely to change how they took a GLP-1?
Younger adults were the most likely to change how they took a GLP-1, even though fewer of them took one at all. Only 4.8% of adults aged 18 to 39 used a GLP-1 in the past year. Those who did were the likeliest age group to be on a compounded version, at 28.0%, and 54% of them changed their use, against 30% of users 65 and older. They also delayed prescription fills far more often, 37% to 6%.1
Of users who had taken a compounded version, 62% changed their use, and 36% of other users did.1 The age and compounding findings overlap, and the report neither separates one from the other nor says why compounded users strayed more often.
Does skipping doses change how well a GLP-1 works?
The survey did not measure whether skipping doses changes how well a GLP-1 works. It asked what people did. It reported nothing about their weight, blood sugar or side effects afterward. The CDC writes that understanding these real-world changes matters because they "might have implications for safety and effectiveness."1
What the drug labels say about a missed dose is in our missed dose guide. Our microdosing guide covers what people mean when they take less on purpose.
How does this compare with Gallup's count?
Gallup's estimate lands close to the CDC's, though it measures something different. Its web survey of 5,065 adults, conducted May 28 to June 5, 2026, found 11% currently taking GLP-1 drugs for weight loss, up from 3% in 2024.2 The CDC counted any use for diabetes or weight loss over a full year, measured almost a year earlier.1
Gallup's text puts 18% of current users on "a compounded or custom-mixed version of the medicine", while its own table prints 19%, and the article does not say which is right.2 Gallup did not ask about skipping doses. The two surveys ask different questions a year apart, and both come out near 1 in 10.
How reliable is the CDC survey?
The CDC survey is large and weighted to represent US adults, and its authors list its limits themselves. Cumulative response rates were low, 13.3% for one panel and 3.8% for the other, and the authors say that could affect how well the results generalize. People without reliable internet access may be underrepresented. Answers about the past year depend on memory.1 Weight was self-reported too. Every figure about compounding and changing doses rests on the 976 users, so the subgroup estimates are loose, and the compounded share among users 65 and older has an interval running from 10.5% to 30.6%.1 Anyone who took a GLP-1 at any point in the 12 months counts the same as someone who took it all year.
Common questions
What percentage of Americans take GLP-1 drugs?
About 1 in 10 American adults took GLP-1 drugs in the past year, 9.8% in a CDC survey of 8,118 adults conducted in July and August 2025. A separate Gallup survey in mid-2026 found 11% currently taking one for weight loss.
How many GLP-1 users skip doses?
Skipping doses was reported by 28% of GLP-1 users in the CDC's 2025 survey, making it the most common change. In all, 44% changed how they took the drug from the way it was prescribed.
Why do people skip GLP-1 doses?
People who skip or stretch GLP-1 doses most often cite cost and availability, named by 38% and 36% of those who changed their use in the CDC survey. Among adults 18 to 39, 58% cited cost.
How many people use compounded GLP-1s?
About 1 in 5 GLP-1 users, 22.6%, said they had taken a compounded version in the CDC survey, and another 8.8% did not know. Among users who had never had diabetes, the share was 34.9%.
How many people take compounded semaglutide?
The 2025 CDC survey did not break compounded use out by drug, so it gives no figure for compounded semaglutide alone. Of all GLP-1 users, 22.6% said they had taken a compounded version of whichever drug they used.
Are compounded GLP-1 users more likely to skip or stretch doses?
Yes, 62% of users who had taken a compounded GLP-1 changed how they took it, against 36% of those who had not. The CDC survey does not say why.
Are younger adults more likely to skip GLP-1 doses?
Yes, 54% of GLP-1 users aged 18 to 39 changed how they took the drug in the CDC survey, compared with 30% of users 65 and older. Younger users also delayed prescription fills far more often, 37% to 6%.
How many people stop GLP-1s because of side effects?
About 12% of GLP-1 users in the CDC survey reduced or stopped their medication because of side effects or symptoms. The report gives one combined figure and does not separate those who reduced from those who stopped.
Do people without diabetes take GLP-1s?
Yes, 5.7% of US adults who had never had diabetes took a GLP-1 in the past year, compared with 47.4% of those who had. About 1 in 3 of the users without diabetes had taken a compounded version.
Are women more likely than men to take GLP-1s?
Yes, 11.9% of women and 7.6% of men took a GLP-1 in the past year in the CDC survey. Women who used one were also more likely to have taken a compounded version, 27.3% of women and 14.8% of men.
Does insurance affect GLP-1 use?
Uninsured adults reported the lowest use in the CDC survey, at 2.7%, compared with 10.6% among adults with either private or public insurance. Privately insured users were about twice as likely as publicly insured users to report a compounded version, 25.9% to 12.5%.
How many people take GLP-1s at a healthy weight?
In the CDC survey, 4.7% of adults at a healthy weight by self-reported BMI took a GLP-1 in the past year. The authors note that some may have diabetes, may have reached a healthy weight on the drug, or may be using it off-label.
Which drugs did the CDC survey count as GLP-1s?
The CDC survey question named Ozempic, Rybelsus, Wegovy, Mounjaro, Zepbound, Saxenda, Victoza, Trulicity, Byetta and Bydureon BCise. Those contain semaglutide, tirzepatide, liraglutide, dulaglutide or exenatide, and both pills and injections counted.
Did the CDC survey count gray-market peptides?
Not separately, because the 2025 CDC survey asked only whether someone had taken "a generic, compounded version" of the medication. A person using a gray-market vial could have answered yes, no or don't know.
How reliable is the CDC GLP-1 survey?
The CDC survey was weighted to represent US adults, but cumulative response rates were low, 13.3% and 3.8% for its two panels. Weight and use were self-reported, and the compounding and dose figures rest on 976 users.
How many Americans take Ozempic?
The CDC survey did not break out Ozempic on its own. About 1 in 10 adults, 9.8%, took a GLP-1 such as Ozempic, Wegovy, Mounjaro or Zepbound in the past year, and the report gives no figure for any single brand.
When was the CDC GLP-1 report published?
The CDC published it in Morbidity and Mortality Weekly Report on November 12, 2026, from survey data collected in July and August 2025.
Sources
- Pierce SL, Balasuriya L, Kompaniyets L, Goodman AB. Use of Glucagon-Like Peptide-1 Medications and Modified Medication Use Among Adults, United States, 2025. MMWR Morb Mortal Wkly Rep 2026;75(44):645-651. https://www.cdc.gov/mmwr/volumes/75/wr/mm7544a1.htm. Also at PubMed Central, https://pmc.ncbi.nlm.nih.gov/articles/PMC13759129.Read at source on cdc.gov and checked against the PubMed Central copy. Every figure on this page was confirmed against a verbatim fragment of the text or Table 1. The official title separates "Adults" and "United States" with a dash, replaced here with a comma under house style. Figures 1 and 2 hold values for taking less medication, delaying fills and reasons by age that neither page gave in readable form, so only the shares stated in the text appear here. The adverse event and website findings in the report's discussion come from studies it cites. This page has not read those studies. ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩ ↩
- Witters D. In U.S., GLP-1 Usage Reaches New High. Gallup, July 7, 2026, updated September 8, 2026. https://news.gallup.com/poll/712157/glp-usage-reaches-new-high.aspx.Read at source. Gallup National Health and Well-Being Index, a web survey of 5,065 US adults on the probability-based Gallup Panel, May 28 to June 5, 2026. Updated September 8, 2026 with revised weighting. The text gives 18% of current users on a compounded or custom-mixed version and the table gives 19%; both are printed on this page. ↩ ↩
Citing this page: Peptide Decoding. How Many Americans Take GLP-1s? 1 in 10, and 44% Skip or Stretch Them. Figures from the CDC's Morbidity and Mortality Weekly Report of November 12, 2026, with Gallup's 2026 survey for comparison. Reviewed November 14, 2026. https://peptidedecoding.com/news/cdc-glp1-use-survey

