KFF poll finds 12% of US adults currently use GLP-1 drugs such as Ozempic

A KFF Health Tracking Poll conducted February 24 to March 2, 2026 found that 12% of U.S. adults say they are currently using a GLP-1 medication such as semaglutide or tirzepatide, and 6% say they have used one in the past but are not using one now ^[1].

The survey reached a nationally representative sample of 1,343 U.S. adults online and by telephone, in English (n=1,268) and Spanish (n=75). The margin of sampling error for the full sample is plus or minus 3 percentage points ^[1]. The March 2026 poll is the most recent KFF measurement of GLP-1 use; KFF also asked the question in November 2025 and May 2024, and the topline notes that a footnote covers wording changes where they apply ^[1].

KFF's question describes GLP-1 medications as prescription drugs "used for weight loss or for treating diabetes and other chronic conditions," and lists brand names including Ozempic, Wegovy, Mounjaro, Zepbound, Rybelsus and Trulicity, along with "others" ^[1]. Respondents were asked whether they are currently using or have ever used one of these drugs "to lose weight or treat a chronic condition such as diabetes or heart disease" ^[1].

Because the question groups all of those products together, and because the published topline does not break the results out by brand or by molecule, the poll does not show how many adults use semaglutide (sold as Ozempic, Wegovy and Rybelsus) compared with tirzepatide (sold as Mounjaro and Zepbound). It also does not separate people using a GLP-1 for weight loss from people using one for diabetes or another chronic condition, even though the question covers both. Those figures have not been disclosed in the topline ^[1].

The same poll measured cost pressure on prescription drugs more broadly. It found 59% of adults are very or somewhat worried about affording prescription drug costs for themselves and their families ^[1]. In the past 12 months, 43% reported at least one cost-related action: 27% said they did not fill a prescription, 19% said they cut pills in half or skipped doses, and 31% said they took an over-the-counter drug instead of filling a prescription ^[1]. Separately, 35% said they compared prescription drug prices online, 36% said they used a discount card or coupon, 16% said they bought a lower-cost prescription from an online pharmacy without using insurance, and 8% said they bought a prescription directly from a manufacturer's website ^[1].

Those cost figures cover all adults in the survey. KFF does not report them separately for GLP-1 users, so it is not known from this poll whether people taking these drugs face cost barriers at higher or lower rates than other adults ^[1].

Why it matters for patients

The 12% figure gives a national sense of scale: roughly one in eight U.S. adults reports current GLP-1 use, which is relevant context for anyone weighing whether these drugs are common in their community or at their workplace. The separate 6% who report past use means about 18% of adults have some personal experience with the drugs, either now or previously ^[1].

For patients trying to compare products, the poll has a clear limit. Readers often want to know how many people take semaglutide versus tirzepatide, or how many use a GLP-1 for weight loss versus diabetes. This survey does not answer either question, and no brand-level or reason-for-use split appears in the published topline ^[1].

Several practical details about GLP-1 use are also absent. The poll does not report insurance coverage, out-of-pocket spending, dose, how long people have taken the drugs, or whether they obtained a prescription from a primary care doctor or another source. None of that has been disclosed ^[1]. Because the question asks people to identify their own medications, the totals reflect what respondents reported about themselves rather than pharmacy records or claims data ^[1].

The survey's margin of sampling error of plus or minus 3 percentage points applies to the full sample; KFF notes that sampling error is only one source of error in any poll and that other unmeasured error is possible ^[1].