How Ozempic, Wegovy and Mounjaro Change the Grocery Bill: 5% Less in Six Months in the US, £389 a Year Less in Britain
US and UK data show GLP-1 users' households buying fewer snacks and less fast food, then partly rebounding. The savings are small next to the drug bill.
Key points
- Worldpanel by Numerator estimated on November 19, 2026, that British households with a GLP-1 user spent more than £1 billion less on groceries than expected in the year after someone started, about £389 per household, compared with households without a user
- The share of British households with a GLP-1 user rose from 6.3% in February 2026 to 8.5% in August; pills made up 11.2% of users and men 32%
- In UK user households, alcohol and crisps spending fell 15% more than in non-user households, while frozen ready meal volume rose 30%, according to Worldpanel
- A peer-reviewed Journal of Marketing Research study of a roughly 150,000-household U.S. panel found grocery spending fell 5.3% within six months of a household member starting a GLP-1 drug (8.2% in higher-income households), with savory snacks down 10.1% and fast-food and coffee-shop spending down 8.0%; yogurt rose the most
- In that study, households that stopped the drug drifted back toward their old spending and bought slightly less healthy baskets than before they started
- Circana reported that by the second year of treatment, the restaurant spending index of U.S. user households had returned to within 0.5 points of its starting level and the candy index was 5.7 points above it; these are index points, not percentage changes in spending
- A Danish receipt study of 293 GLP-1 starters found small shifts toward lower-sugar, higher-protein and less ultraprocessed purchases compared with 884 matched non-starters
- Worldpanel's average UK saving of about £32 a month is a fraction of private Mounjaro prices, which started at £133.19 a month in a November 2026 comparison of 28 UK providers
British households with a GLP-1 user spent more than £1 billion less on groceries than expected
On November 19, Worldpanel by Numerator, a company that tracks what a sample of British households buy, put a number on how weight-loss drugs are changing the weekly shop. Households with at least one person taking a GLP-1 drug, such as Mounjaro (tirzepatide) or Wegovy (semaglutide), spent more than £1 billion (about $1.32 billion) less on groceries in the year after someone started than Worldpanel expected, judged against households without a user. Reuters reported that this works out to £389 a year for each user household.
The group keeps growing. Worldpanel said 8.5% of British households had a GLP-1 user in August, up from 6.3% in February, and that user numbers rose almost as much in those six months as in the whole previous year. Pills already accounted for 11.2% of users, even though, as Reuters put it, regulatory approval came only in June. Men made up 32% of users, up from 23% in February. Worldpanel expects more than 10% of British households to include a user in 2027. Its survey sample totaled 9,629 households.
In June, the same company had reported a smaller total: £780 million, or £418 per household compared with non-user households, from a survey of more than 11,500 households. The two releases used different samples and time windows, so the move from £418 to £389 per household is not a trend anyone should read much into. The total rose along with the number of households that have a user. Our November 20 digest carried the Reuters report.
Fewer crisps, chocolate and fast-food meals; more yogurt and frozen ready meals
Worldpanel's November release said spending on alcohol and crisps in user households fell 15% more than in non-user households. Frozen ready meals moved the other way, with volume sales up 30% among users. Worldpanel described them as easy, portion-controlled options. In June the company reported that chocolate spending among users had fallen 18 percentage points more than among non-users since treatment began. It also reported increases of 20 points for mouthwash and 24 points for chewing gum, which it linked to dry mouth and bad breath, sometimes called "Ozempic mouth." In the June survey, 54% of users said they had fewer cravings and less "food noise," the constant background thinking about food (our food noise explainer covers how researchers measure it).
The best-documented American numbers come from a peer-reviewed study in the Journal of Marketing Research by Sylvia Hristakeva and Jūra Liaukonytė of Cornell University and Leo Feler. Within six months of someone starting a GLP-1 drug, households cut grocery spending by 5.3% on average. Savory snacks fell 10.1%. Cornell's summary of the paper says sweets, baked goods and cookies fell by similar amounts, and that even bread, meat and eggs declined. A few categories rose. Yogurt rose the most, followed by fresh fruit, nutrition bars and meat snacks, though Hristakeva noted those increases were "modest relative to the overall decline." Spending at fast-food chains, coffee shops and other limited-service restaurants fell 8.0%.
A third data set, from Denmark, looked at what was in the basket rather than what it cost. Researchers at Copenhagen University Hospital used digital supermarket receipts donated by members of the SMIL cohort and compared 293 people who started a GLP-1 drug between 2019 and 2022 with 884 matched people who did not. In the year after starting, the average energy density of what initiators bought dipped from 209.4 to 207.3 calories per 100 grams, sugar from 15.7 to 15.1 grams per 100 grams, and protein rose from 6.6 to 6.9 grams per 100 grams. The share of ultraprocessed items fell. The comparison group mostly moved the opposite way. The authors, writing in JAMA Network Open, called the changes modest for any one person.
The alcohol figure fits a wider pattern of interest in whether these drugs reduce drinking. That question is being tested in clinical trials, which our alcohol and addiction brief follows. Grocery purchase data cannot settle it, because a receipt does not show how much anyone drank.
Matching 150,000 households' receipts to survey answers
The Cornell-led study is worth a closer look because its design shows both the strength and the weak spots of this kind of research. Numerator runs a panel of about 150,000 U.S. households that record their grocery and restaurant purchases. The researchers linked those records to repeated surveys asking whether anyone in the household took a GLP-1 drug, when they started and why. They then compared households that started with similar households that did not, which lets them estimate what spending would have looked like without the drug.
Four details change how the 5.3% average reads:
- Higher-income households cut grocery spending by 8.2% within six months, more than the 5.3% average.
- The reductions lasted through the first year among households that kept taking the drug, but they shrank after the first six months.
- About one-third of users stopped during the study period, according to Cornell. Those households drifted back toward their old spending, and their baskets became slightly less healthy than before they started, partly because of more candy and chocolate.
- People taking the drugs for weight loss skewed younger and wealthier, while those taking them for diabetes were older and more evenly spread across income groups.
The authors write that they cannot fully separate the drug's biological effect from other changes people make when they start treatment. They point to the clinical trial evidence on appetite, and to the rebound after stopping, as signs that appetite suppression is likely a main driver. The share of U.S. households reporting at least one user rose from about 11% in late 2023 to more than 16% by mid-2024 in their data. The paper appeared online in December 2025 and in the June 2026 print issue.
By the second year, Circana's candy index sits above where it started
Circana, another U.S. market research company, followed households for longer. Its November 10 report, GLP-1 Unlocked, said 37% of surveyed U.S. households have had someone use a GLP-1 drug, up from 17% in 2024, and 30% have a current user. Our November 13 digest covered the release.
Circana reports its spending results as a "Consumption Index," and the numbers are easy to misread. The index compares average monthly spending in households with a current user against households that have never used the drugs, over the same period. An index of 100 means equal spending. Before treatment, user households had a restaurant index of 121, meaning they spent about 21% more at restaurants than never-user households. That index fell 4.3 points during the first year of treatment and returned to within 0.5 points of its starting level in the second year. Retail food and beverage spending stayed 1.9 points below its starting level in year two. The candy index climbed 5.7 points above its pre-treatment level by year two, and frozen desserts 5.3 points.
Circana itself warns that these are index points, "rather than percentage increases" in what the households actually spent. A 5.7-point rise in the candy index does not mean a household bought 5.7% more candy. It means user households spent more on candy relative to non-users than they did before treatment. The release does not say whether the second-year households were the same people who started the drug, or whether some had lowered their dose or stopped. Circana's analysis covers purchases from August 2023 through July 2026 and identifies users through a July 2026 survey of its receipt panel. Of households where someone had stopped, 48% said they were at least somewhat likely to restart within a year.
Circana's candy numbers and the Cornell finding on people who stop both point to the snack cutback fading for some households. The two data sets measure different things, though, and neither tracks what any person ate.
Reading "less than expected" and other panel numbers
Consumer panels are good at spotting broad shifts in what households buy. They are a weaker guide to what happens to one person's diet, for several reasons.
They are samples. Worldpanel's November figures come from 9,629 households and Circana's from its receipt panel, and totals such as £1 billion are estimates scaled up to the whole country.
"Expected" is a model. Worldpanel's £389 compares user households with what similar non-user households did. If people who start these drugs differ in ways the comparison misses, such as income, age, health goals or a diet begun at the same time, part of the gap reflects those differences. The Cornell team found that weight-loss users skew younger and wealthier. Drug use is also self-reported: the panels learn who takes a GLP-1 drug from surveys, not prescriptions. The Danish study is the exception, since it used Denmark's prescription registry.
Purchases also belong to the whole household. A user household can include people who take nothing, and a receipt shows what was bought, not who ate it, how much was thrown away, or what was eaten at work or with friends.
The company figures have also not been peer-reviewed. Worldpanel and Circana publish summaries for retailers and food companies and sell the full methods to clients. The Cornell and Danish studies did go through peer review, but the Danish one was small (293 starters), drew on people willing to share receipts, and had no body weight data.
Roughly £32 a month off groceries, against £133 or more a month for Mounjaro
Divide Worldpanel's £389 a year by 12 and the average grocery saving is about £32 a month. A private prescription costs several times that. Medino, an online pharmacy that publishes a comparison of 28 UK providers, found the cheapest Mounjaro at £133.19 a month for the 2.5 mg starting dose (its own price) when it last checked prices on November 2, 2026. Boots charged £179 for that dose and £339 for the 15 mg top dose. People who get Mounjaro or Wegovy through the NHS in England pay the standard prescription charge of £9.90 per item, unless they are exempt.
In the U.S., Novo Nordisk's NovoCare pharmacy sells Wegovy pens at doses up to 2.4 mg to people paying cash for $349 a month, after an introductory $199 for each of the first two monthly fills. At $349, that is $4,188 a year. Wegovy's list price is $1,349.02 per package. Insured patients pay whatever their plan sets. The Cornell paper reports its results in percentages, so it does not give a dollar figure for what a typical household saved.
Price is also a common reason British users give for stopping, Worldpanel said in June. In that survey, 41% of people who had been taking a GLP-1 drug and stopped in 2026 said it was too expensive. For more on prices, see our international GLP-1 price comparison.
A smaller basket that still covers protein, fiber and vitamins
Shopping data show people buying less, and a separate line of research shows some of them eating too little. In a study of 14 patients at one English general practice published November 17 in BMJ Nutrition, Prevention & Health, average intake over two recalled days was 1,231 calories. More than three-quarters fell below two-thirds of the UK reference intake for fiber and for vitamin D, and none ate the recommended 400 grams of fruit and vegetables a day. Protein averaged 153% of the reference. The study was tiny, and the BMJ summary describes no comparison group; our brief on undereating covers it in detail.
The main U.S. guidance on eating during GLP-1 treatment is a 2025 joint advisory from the American College of Lifestyle Medicine, the American Society for Nutrition, the Obesity Medicine Association and The Obesity Society, published in the American Journal of Clinical Nutrition. Its advice lines up in part with how user households are already shopping. It tells clinicians to emphasize "nutrient-dense, minimally processed foods" such as fruits, vegetables, whole grains, legumes, lean proteins, nuts and seeds, and to counsel people to avoid refined carbohydrates, sugar-sweetened drinks, red and processed meats, "most fast foods, ultraprocessed sweets, and savory snacks." The fall in snack and fast-food spending moves in that direction. The drops in bread, meat and eggs that Cornell reported, and the rise in frozen ready meals in Britain, are less clear-cut. The advisory lists frozen meals and take-out among the habits clinicians should ask patients about.
On protein, the advisory says an absolute target of 80 to 120 grams a day "may enhance adherence while ensuring adequate intake," as an alternative to calculating needs from lean body mass. Because appetite is low, it suggests eating protein-rich foods first in a meal and choosing lower-volume, nutrient-dense options such as fish, eggs, Greek yogurt, cottage cheese, nuts and nut butters. Some people meet targets with shakes or bars. The authors add that more protein alone is likely not enough to preserve muscle without strength training, which our protein and resistance training guide covers.
For constipation, the advisory recommends fluids and fiber from foods, added gradually. Its table of habits to encourage includes regular small meals at consistent times, enough fluids and minimal alcohol. It says supplements such as vitamin D, calcium, B12 or a multivitamin "can be proactively considered" for nutrients at risk, tailored to each person. Cost comes up too: in a survey the advisory cites on barriers to healthy eating, 81% of respondents listed cost. For people facing those barriers, the authors suggest that medically tailored groceries or meals may help. A registered dietitian nutritionist, the advisory says, can help work out an eating pattern that meets nutrition goals while fitting a person's needs and preferences.
Updates and corrections
No corrections yet.
Limitations
- Worldpanel and Circana figures come from company press releases and news reports; their full methods are not public and have not been peer-reviewed
- "Less than expected" and index figures depend on comparison households; users differ from non-users in age, income and health goals, and the comparisons may not fully account for that
- Panel data record household purchases, not what any person ate; use of GLP-1 drugs is self-reported in the Worldpanel, Circana and Cornell data
- The June and November 2026 Worldpanel releases used different samples and time windows, so their per-household figures are not directly comparable
- The Danish receipt study was small, drew on volunteers who share receipts, lacked body weight data, covered GLP-1 starts from 2019 to 2022, and did not break results down by drug or reason for use
- We could not open Worldpanel's own release pages; UK figures come from the Reuters report and trade-press reproductions of the June release
- Drug prices change often; the UK prices cited come from one pharmacy's comparison and the U.S. price from Novo Nordisk's self-pay pharmacy as of November 2026
Citations
- 1. The No-Hunger Games: How GLP-1 Medication Adoption Is Changing Consumer Food Demandpeer-reviewed Journal of Marketing Research 2026
- 2. Consumer Food Purchases After Glucagon-Like Peptide-1 Receptor Agonist Initiationpeer-reviewed JAMA Network Open 2026
- 3. Uncovering nutritional deficiencies and the adequacy of wrap-around care for patients on GLP-1 and dual GIP/GLP-1 receptor agonistspeer-reviewed BMJ Nutrition, Prevention & Health 2026
- 4.
- 5.
- 6.
- 7.
- 8.
- 9.
- 10. NHS prescription charges (NHS)government 2026
- 11. Price your Wegovy (NovoCare)manufacturer 2026
Peptides in this article
Full peptide profiles with evidence levels, dosing data, and safety notes live on peptidelist.org.
Related insights
Thanksgiving on Ozempic, Wegovy, Mounjaro or Zepbound: Big Meals, Alcohol, Nausea and When to Get Help
What the Wegovy, Ozempic, Zepbound, Mounjaro and Foundayo labels say about nausea, missed doses and warning signs, and dietitians' advice on plates and drinks.
Buying Wegovy or Mounjaro Abroad: Price Gaps, Import Rules, Counterfeits and Cold Storage
South Korea now lets travelers bring home three months of weight-loss drugs, then cracked down on resale. Rules elsewhere, storage limits and fake pens.
Eating Too Little on Ozempic, Wegovy, Mounjaro or Zepbound: Calorie, Fiber, Vitamin D and Iron Gaps
A 14-patient English study found average intake near 1,230 calories a day and wide shortfalls in fiber, vitamin D, iron and produce. Records data on anemia, a small lean mass trial and the 2025 nutrition advisory help show who should watch for it.