(Credit: © zimmytws - stock.adobe.com)
GLP-1 Spending Hit $65 Billion, and the Financial Burden Is Not Shared Equally
In A Nutshell
- U.S. spending on GLP-1 drugs rose nearly 15-fold in a decade, from $4.4 billion in 2013 to $64.9 billion in 2023.
- Nearly one in five users (18.2%) spent at least 10% of household income on the drugs.
- Lower-income and publicly insured users paid fewer dollars than wealthier, privately insured users but spent a bigger share of their income.
- Burden declined over the decade, yet income gaps persisted.
Total U.S. spending on GLP-1 drugs, the diabetes and weight-loss medications that include semaglutide and tirzepatide, jumped nearly 15-fold in a decade, from $4.4 billion in 2013 to $64.9 billion in 2023, according to a study in the Journal of General Internal Medicine. Behind that eye-catching number is a less obvious finding about who feels the cost most, and it has little to do with who pays the biggest pharmacy bills.
Dollars alone can mislead. Publicly insured and lower-income users typically paid less out of pocket each year than privately insured and higher-income users. But $1,500 a year takes a far bigger bite out of a $30,000 household income than a $200,000 one. Once the study measured costs against income, the pattern flipped.
Nearly one in five users, 18.2%, spent at least 10% of household income on the drugs, a level the researchers labeled a high burden. Uninsured, publicly insured, and lower-income adults were among the most likely to cross it.
Survey of 2,797 Adults Shows GLP-1 Spending Grew 31% a Year on Average
Data come from the Medical Expenditure Panel Survey, a federal survey that follows health spending among a nationally representative group of American adults. Participants list the medications they take, and with permission, the survey checks with pharmacies to confirm what was filled and what was paid. From 2013 through 2023, researchers found 2,797 adults who used GLP-1 drugs. Weighted to reflect the whole country, that group stands in for about 33 million U.S. adults over the decade.
Researchers tracked total national spending, what the typical user paid out of pocket in a year (copays, deductibles, and other payments straight from the patient), and what share of household income that bill represented. The survey records prices at the register, so totals do not reflect rebates or discounts negotiated behind the scenes.
Use exploded along with the cost. Annual users grew about tenfold, from roughly 891,000 in 2013 to 9.1 million in 2023, lifting the share of American adults on the drugs from under 1% to about 5%. Users were most often ages 50 to 64 (43%), female (53%), non-Hispanic White (69%), and privately insured (68%). Spending grew faster than use, averaging about 31% more each year.
Newer Drugs Fueled GLP-1 Spending as Semaglutide Reached Over Half of the 2023 Total
Older medications set the early pace. Liraglutide climbed to $9.1 billion in 2022 before falling to $3.4 billion in 2023 as newer options gained ground. Semaglutide went from $0.7 billion in 2018 to $35.5 billion in 2023, more than half of that year’s total. Tirzepatide appeared in the data only in 2023 and drew $9.1 billion right away. Overall spending nearly doubled between 2021 and 2023, and authors pointed to a shift toward newer, more clinically effective, but pricier drugs.
Uninsured and Lower-Income Adults Devoted the Largest Shares of Income to GLP-1 Drugs
Typical users paid $1,528 out of pocket per year. Earlier insurance-claims research put the figure near $863, but authors said the gap likely comes from measurement: the survey captures cash payments and uninsured adults that claims data leave out. Income groups in the study were based on the federal poverty line, with lower income meaning under twice that level and higher income meaning at least four times it.
By that yardstick, privately insured users paid a typical $1,850 and higher-income users $2,160, while publicly insured users paid $763 and lower-income users $640. Uninsured adults, at $1,704, landed near the privately insured group. Authors said the higher amounts among better-insured and higher-earning users may reflect greater access to care and the ability to afford costlier drugs, or coverage exclusions, since many employer plans do not cover GLP-1 drugs for obesity.
Measured against household income, uninsured adults spent an average of 14.4% on the drugs, lower-income adults 11.0%, and publicly insured adults 8.5%, compared with 4.7% for higher-income households.
Nearly three in ten uninsured adults (29.2%) crossed the 10% line, as did more than one in four lower-income adults (27.9%) and about one in four adults 65 and older (24.3%). Among higher-income adults, roughly one in eight did (11.9%). Both uninsured figures carry wide margins of error, with the average burden anywhere from 3.1% to 25.8%, so they are less certain than the others.
GLP-1 Out-of-Pocket Burden Declined Over Time, but Income Gaps Persisted
One encouraging trend emerged. The share of users crossing the 10% line dropped from about one in three (32.7%) in 2013 through 2016 to about one in seven (14.8%) in 2020 through 2023. Gaps between income groups persisted, though. Lower-income adults averaged a 10.2% burden in 2020 through 2023, against 3.7% for higher-income adults. The study was not designed to determine why burden declined.
One limit matters here. The survey counts only filled prescriptions, so anyone who abandoned a prescription because of cost would not appear in the data, and the true affordability problem may be larger than what was measured.
Lower dollar bills did not mean lighter burdens in this data. As GLP-1 use continues to grow, the open question is whether affordability can keep pace, especially for patients with the least financial room to absorb the cost.
Disclaimer: This article summarizes a peer-reviewed study and is intended for general informational purposes only. It is not medical, financial, or insurance advice. Anyone with questions about medications or their costs should consult a qualified health care professional or benefits adviser.
Paper Notes
Limitations
Several caveats apply. Medication records rely partly on what respondents report, which can introduce recall errors or underreporting, though pharmacy verification and government imputation procedures help reduce that risk. The survey captures filled prescriptions only, so it cannot assess prescriptions written but never filled because of cost, coverage denials, or other access barriers. The analysis was descriptive and did not adjust for factors such as other health conditions, disease severity, or plan generosity, so it cannot explain why costs or burdens changed. Spending figures reflect gross costs and exclude rebates and other post-sale price concessions. The survey also may not fully capture manufacturer coupons or patient assistance programs. Twenty-six respondents (1%) were identified only through a broader drug class code that may include non-GLP-1 incretin mimetics, and the uninsured burden estimate has a wide statistical range because of greater uncertainty in that subgroup. As a cross-check, cumulative spending for 2018 through 2023 landed within about 2% of an earlier claims-based study ($200.1 billion versus $204.5 billion), though year-by-year and drug-by-drug estimates sometimes differed more.
Funding and Disclosures
Authors reported no funding for the study and declared no competing interests. The study used publicly available, deidentified data and was deemed exempt from institutional review by Virginia Commonwealth University.
Publication Details
Study title: “Spending and Affordability of Glucagon-Like Peptide-1 Receptor Agonists and Other Incretin Mimetics in the United States, 2013–2023.” Authors: David T. Zhu (Medical Scientist Training Program and School of Medicine, Virginia Commonwealth University, Richmond, Virginia), Chloe Gao (Department of Medicine, University of British Columbia, Vancouver), Ye In Christopher Kwon (School of Medicine, Virginia Commonwealth University), and Alan Lai (School of Medicine, Virginia Commonwealth University). Journal: Journal of General Internal Medicine. Received December 11, 2025; accepted July 13, 2026; published online July 30, 2026. DOI: https://doi.org/10.1007/s11606-026-10658-2. Corresponding author: David T. Zhu ([email protected]). Data source: Medical Expenditure Panel Survey, Agency for Healthcare Research and Quality (https://meps.ahrq.gov/mepsweb/). The article is open access under a Creative Commons Attribution 4.0 International License.







