GLP-1s Are Reshaping Consumer Spending and Employer Health Benefits

PwC put GLP-1 adoption at 21% of U.S. households, up 57% over 18 months. Its chart showed lower spending in grocery, fast food, pizza, and chicken, alongside higher apparel spending. The graphic attributed the analysis to PwC’s GLP-1 Consumer Survey and Numerator’s GLP-1 transaction panel, May 2024 refresh, covering roughly 110,000 households.

Ali Furman said PwC has been studying the consumer-spending implications of weight-loss drugs for about three years. Even when the population taking the drugs was smaller, PwC began seeing material changes in spending. Furman’s broader claim is that GLP-1s are disrupting an economy that, for decades, has been built around “more”: more volume, more impulse purchases, more convenience, more calories, more flavor.

The GLP-1 consumer, in her account, is not simply buying a smaller version of the old basket. They are interested in less of some things and more of others. That reallocation is showing up across grocery, restaurants, apparel, travel, fitness, and wellness.

The grocery decline is not inconsistent with more spending on fresh produce. Furman put the average GLP-1 user’s calorie consumption about 40% lower. Fresh produce can become a higher share of the grocery basket while total volume falls. That is why PwC sees grocery spending down roughly 5.5%. During the Q&A, Furman put the dollar value at about an $18-per-month decrease.

The shift is not isolated to the individual taking the medication. Furman said GLP-1 users are spending differently for their entire households, “including their pets.” PwC’s work with companies, she said, is increasingly about how to support a growing demographic that is not only pursuing weight loss but rethinking sleep, nutrition, exercise, and mental health.

Restaurants face a more uneven version of the same pressure. Furman described overall restaurant spend among GLP-1 users as down, with quick-service and fast food falling more than casual dining. She put casual dining down about 3% to 4%, compared with fast food down about 9%. The explanation, in her view, is not that indulgence disappears. It becomes more intentional.

“Permissioned indulgence” is the phrase PwC uses. GLP-1 users are not necessarily eliminating treats; they are changing the conditions under which indulgence makes sense. That helps explain why premium mini packaging and resealable packaging are doing well, while king-size and value-size formats are less aligned with the consumer’s new behavior.

The same logic is beginning to reshape product formulation and merchandising. Furman said food and beverage companies are responding with ingredients and attributes that matter to GLP-1 users and are often advised by doctors: protein, fiber, hydration, and micronutrients such as zinc, copper, and iron. Protein’s rise has causes beyond GLP-1s, she said, but the drugs are a catalyst.

Furman also said 80% of people on GLP-1s are taking some kind of vitamin or dietary supplement alongside the drug. The most common are protein supplements, fiber supplements, stool softeners, and magnesium. PwC is seeing a 1% increase in drugstore-channel spending that Furman believes is correlated with this supplement behavior.

Ali Furman resisted treating GLP-1 users as a single average consumer. PwC sees different cohorts and motivations, and those differences matter for businesses trying to interpret the data.

Early cohorts who started taking the drugs when they became available for weight loss appeared, based on weight lost over time, to be clinically obese or significantly overweight. More recent cohorts, over the same period of time, are losing less weight. Furman said one could infer that some of those newer users may be taking the drugs for different reasons, including aesthetic or lifestyle reasons, as the drugs become more mainstream.

She described at least two broad behavior patterns. One group treats the prescription as the main intervention: the doctor is in charge of health, the medicine is prescribed, and the patient takes it without making many other lifestyle changes. Instead of eating a whole bag of chips, that person may eat half a bag.

Another group uses the drug as the catalyst for a wider health transformation. That is where PwC sees larger changes in spending. These users join gyms more often. Their travel changes. Some move away from food-and-wine-centered travel — Furman gave Las Vegas and Napa Valley as examples — toward med spas, yoga retreats, and outdoor adventure vacations that align with greater physical agility.

The apparel data, in Furman’s telling, is not just about needing a different size. Jewelry spending among GLP-1 users is up almost 40%, she said, and jewelry is not tied to clothing size. PwC also sees a shift in spending allocation from mass retail toward more premium brands, including increases in women’s swimwear and women’s jeans. Furman interpreted this as an aspiration and identity shift.

That identity shift is why Furman compared GLP-1s to the smartphone: not because the mechanisms are similar, but because the downstream economic effects may be hard to predict at the moment of adoption. When smartphones first appeared, she said, no one anticipated Uber, TikTok, or Netflix streaming. GLP-1s, in her view, are a “physiological disruption” that will generate new brands, products, and industries around the ecosystem the drugs are creating.

Her advice to companies follows from that segmentation. Do not panic about what consumers want less of; identify which GLP-1 demographic aligns with a company’s products and services, and support that group’s new priorities.

The pressure to get GLP-1s has already changed how at least one manufacturer reaches patients. Laura Steele said Eli Lilly would not have expected, 10 years ago, to create a direct-to-consumer platform for patients. LillyDirect emerged, in her account, because consumer demand and access were not matching.

Steele described LillyDirect as a digital health platform designed around patient barriers. Past shortages were one problem. Another was that 50% of patients with obesity do not have a primary healthcare provider they see consistently. LillyDirect was meant to offer a simpler and more transparent way for patients to get supportive care and, if a healthcare provider determined it was appropriate, a prescription medicine.

For Zepbound, one of Lilly’s GLP-1 agents, Steele said LillyDirect has become a significant channel: 50% of all new patients flow through LillyDirect, and about a third of total prescriptions go through it.

In Steele’s account, the platform was born after a pharmacy benefit manager decided not to cover Zepbound, forcing Lilly to innovate on access. She said the company now sees it as part of a broader access ripple effect: commercial coverage, employer coverage, PBM coverage, and Medicare access. She also said Lilly would be on all three major PBMs again starting October 1.

Bertha Coombs noted that Reuters had reported one of the big three PBMs had decided not to cover the drug for its own employees for weight loss. That set up the central tension: patients and employers are demanding access while coverage decisions remain contested and expensive.

Employer access is changing quickly, according to Steele. She said that last year about 20% of employers with 200 or more employees covered GLP-1s. In 2026, Lilly estimates that about 67% are covering them. Employers, she argued, are beginning to see coverage as an investment in their workforce, not simply a pharmacy-cost line item.

Her economic rationale was that employers are also bearing the costs of not treating obesity, including productivity loss and increases across short-term disability, long-term disability, workers’ compensation, and lost working hours. She cited 9% to 23% growth across those kinds of cost categories. Employers still have real benefit-design and budget constraints, but the market is offering more ways to manage them.

Those options include PBM designs, Lilly’s Employer Connect, and employer-direct arrangements. Steele said she has never seen so many options on the table through PBMs in her 23 years at Lilly. Employer Connect adds roughly 15 independent companies that help tailor solutions, including whether an employer wants to focus only on drug cost or also provide wraparound care. In an employer-direct model, a company that does not want an intermediary could negotiate directly with Lilly or another manufacturer.

Steele did not say every employer will do this. But she said those that try it may value transparency: what budget they set, whether they came in on budget, what costs were expected, what employees actually paid out of pocket, and how access affects retention and productivity. In her view, that could become a significant movement.

Kosali Simon described GLP-1s as moving through several economic layers before their effects show up in households or health outcomes. The pipeline is expanding through new indications, molecules, and formulations. Policy questions include Medicare, Medicaid, and FDA decisions about remaining compounding. Markets determine how people actually buy the drugs. Outcomes include not only changes in shopping baskets but health, healthcare use, and whether people live longer, healthier, happier lives.

Simon’s own research on Medicaid suggests policy is not a marginal variable. She and colleagues reviewed state Medicaid policy documents, including provider manuals, to see which states added weight-loss indications and which formulations were covered year by year. She said 17 states had added a weight-loss indication, while some had begun taking it off. When the researchers looked at filled prescriptions across Medicaid programs and states through the end of 2025, state policy had a strong effect on whether low-income patients received the medication.

Simon also said the study found no evidence that people were simply switching from cash payment to insurance coverage in those states. In her interpretation, when Medicaid covers the drugs, it brings in new people. Nor did the study find evidence of off-label prescribing in the setting they examined. She connected that evidence to Medicare: if Medicare adds or expands a weight-loss indication, the Medicaid findings suggest new people may come in rather than coverage merely formalizing use that was already happening off-label.

Access remains uneven, but Furman said PwC’s May 2024 data did not show adoption concentrated only among high-income households. Adoption was “pretty flat” across low-, middle-, and high-income cohorts, in a range of 17% to 23%. She attributed that in part to affordability mechanisms including Medicaid, manufacturer couponing strategies, and promotions.

PwC’s demographic data also showed the highest percentage of users among Black and Hispanic consumers, with white consumers third. Adoption was at parity between men and women. By generation, Gen X had the most people on the drugs, followed by Baby Boomers, Millennials, and Gen Z. Furman said a Gen Xer is 90% more likely to be on a GLP-1 than a Gen Zer.

Steele added that only about one in 10 patients who could be eligible for these products are taking them. In her view, stigma remains a major barrier even where access exists. Lilly’s task, as she described it, is partly educational: to frame obesity as a chronic illness that warrants treatment and medical intervention.

The introduction of oral agents may widen access and reduce stigma. Steele said Lilly had launched what the transcript renders as “Foundeo” or “fondaio,” describing it as Lilly’s first oral agent, in April. Because the source provides only the spoken name, the exact product name should be treated cautiously. Steele’s substantive point was that the oral format has a lower price point and manufacturing and scalability advantages, but the psychological access point matters too: it gives some patients permission to engage. Some have been hesitant because of price, stigma, dislike of needles, or discomfort with what feels like a more intense medical intervention. A daily oral agent can feel less severe and easier to try with a healthcare provider.

Coombs described a Medicare Bridge Program as a potentially large access shift, saying the administration had negotiated it with Lilly and Novo and that it would allow Medicare patients to access the drugs through their benefits for a $50 copay. Steele said Lilly anticipates a large Medicare surge. There could be 20 million eligible Medicare patients, she said. She described eligibility criteria as BMI of 35 without a comorbidity, or BMI of 27 to 34 with another comorbidity.

According to Steele, the Bridge Program launches July 1 and runs through December 31, 2027. Its goal, she said, is to prove real-world outcomes, improved cost, and savings across the system, with the hope of expanding access further.

An audience member raised the adherence problem directly, citing a prior statement that within about a year, roughly half of people taking GLP-1s stop using them. The question was whether that makes the consumer and economic effects look more like boom-and-bust than durable transformation.

Ali Furman said PwC’s survey identifies three leading reasons people stop or lapse: cost, insurance coverage, and side effects. The percentages reporting each reason are fairly even, she said. Her answer treated discontinuation as a function of current barriers rather than a fixed feature of the category. Hundreds of trials are underway to mitigate side effects, she said, and cost and access are also being worked on. If those barriers fall, more people may stay on therapy. If they do not, discontinuation may remain similar.

Her main rebuttal to the boom-and-bust interpretation was adoption growth despite those barriers. In PwC’s first year studying the category, many clients argued adoption would remain flat because people go on and off the drugs. Furman said the data has since shown significant adoption growth over 18 months while cost, access, and side effects remain substantial barriers.

She also introduced a longer-term scenario based on changes in format. Using flu-shot uptake as a rough proxy, she noted that 47% of people accept a flu shot from their doctor during an annual physical. If GLP-1s eventually became a once-a-year injection for weight management, she said, a doctor could offer it in the same annual context. With 75% of Americans considered overweight and 40% considered clinically obese, Furman said there is a world in which 50% to 75% of U.S. adults could be on these drugs in 10 years.

Laura Steele agreed that access matters for adherence. Lilly sees improved adherence when patients have formulary coverage, she said, and also through LillyDirect. But she argued that adherence is not only a coverage problem. It is also bound up with whether obesity is understood as a chronic illness.

Some patients and providers treat the medication as a temporary weight-loss tool: once the goal weight is reached, the patient stops. Steele contrasted that with asthma or blood-pressure medication, which patients generally would not stop simply because the condition is controlled. Obesity care, she said, still carries the stigma that patients should be able to manage the condition with diet and exercise alone.

That stigma affects clinical timing. Steele said oral agents are activating patients in earlier stages of obesity, with about 60% of new patients in stage one or two. Those patients historically might be sent home with diet-and-exercise advice repeatedly. She said it takes 11 rounds of diet and exercise before a healthcare professional says, in effect, “we need to do something differently.” If those rounds correspond to annual visits, a patient with early obesity could wait 11 years before medical intervention.

Coombs asked whether the stigma could invert: if someone is visibly struggling with obesity, will people begin to ask why they are not on a GLP-1? Steele answered from the provider side. When Lilly asks healthcare providers why it takes so many rounds of diet and exercise before action, she said, they often say they do not want to hurt the patient’s feelings. Steele argued for a more empathetic bridge: the conversation should not be about appearance, but long-term health.

A central question for employers, payers, and policymakers is whether GLP-1 spending is offset by lower spending elsewhere in healthcare. Kosali Simon said the early evidence does not show a simple medical-cost offset.

Simon described one study by her and colleagues, plus two other independently done studies, that examined millions of healthcare records. She said the evidence spans private health insurance records, Veterans Administration records, and a general study. In her summary, the studies found the same broad pattern: after people start taking GLP-1s, they do not have fewer healthcare visits or lower spending on other healthcare. Spending on other healthcare may go up a little.

That pattern did not surprise her. The medication can require more monitoring and more visits, and medical technologies often increase healthcare use rather than reduce it. The payoffs, in Simon’s framing, may be found outside narrow healthcare utilization: employee productivity, the ability to do things in life that people value, and broader measures of health and well-being. That does not make the budget question disappear. It shifts where the benefits must be evaluated.

Bertha Coombs compared the employer and payer pressure to the “Sovaldi moment” roughly 12 years earlier: an expensive drug that is not confined to a tiny specialized population but has broad use. Employers are grappling with access because the eligible population is large, and the spending impact is immediate.

Simon said the question is part of a larger unresolved problem in healthcare: how to cover powerful new medications, including drugs that may cost very large amounts for small populations, while also considering population health spending. GLP-1s intensify the issue because demand and eligible populations are both large.

Inflation adds another layer. Asked how a rising-cost environment might affect GLP-1 growth, Simon said forecasts depend heavily on the price points of new formulations compared with older formulations becoming generic. She said her research finds that rebates matter greatly when calculating what is actually charged, and that new formulations can have a lower “real price” when measured by price per weight-loss outcome. In other words, a newer drug may have a higher nominal price but a lower cost per unit of outcome.

Furman added a consumer-spending perspective. Across income cohorts, PwC sees a more value-conscious and budget-conscious consumer than before, due to inflation and rising energy and gas costs. But value-conscious does not mean uniformly frugal. Consumers are still willing to splurge on things they believe are valuable, especially categories related to self-improvement. Even high-income consumers are trading down in substitutable, commoditized items and reallocating toward premium travel. Furman expects similar trade-off behavior around GLP-1s over time.

An audience member asked whether U.S. GLP-1 prices reflect the broader asymmetry in pharmaceutical pricing between the United States and other developed countries. The answer did not reduce the issue to a single number.

Laura Steele said the Medicare Bridge negotiation brought the cost of medicine down. She also said Lilly agrees that the U.S. should not bear all the cost of innovation. Medicare pricing and cash-pay pricing reflect that discussion, in her view. But she pointed to a “middle piece”: the healthcare system’s intermediaries, rebates, and the true cost to the system.

Bertha Coombs pressed on whether government and cash-pay pricing could lead to higher costs elsewhere, as lower government reimbursement can sometimes be offset by higher private-market charges in provider markets. She asked whether commercial and employer-market discussions were different this year and whether there was more cooperation to get better pricing.

Steele answered that as access expands, there is price flexibility in discussions that unlocks broader access. She pointed to self-pay, PBMs, new commercial plans, Medicare, Medicaid, and employer channels as different routes for innovation to reach patients. When Coombs asked directly whether prices went down this year, Steele said that across the board, the expansion of access reflects those discussions.

Kosali Simon added that pricing is difficult to interpret because “the drug” is not a single static product. It is a bundle of characteristics: side effects, efficacy, and ease of use. She compared it to pricing computers, where the top model may cost more but also contains more capability. Healthcare pricing already lacks transparency, and drugs are no exception. Rebates make it still harder to know what price movement means.

The result is a pricing fight distributed across public programs, cash-pay models, commercial insurance, employer direct purchasing, PBM arrangements, and rebates. What different actors pay depends on which channel they use and how much of the nominal price is altered by negotiation, rebate, or benefit design.

Mani Keita asked the closing question: GLP-1s seem to have created an unusually empowered consumer. What conditions made that possible, and can they be applied elsewhere in healthcare, including care management and doctor-patient communication?

Laura Steele answered by distinguishing patient-centricity from a generic consumer-access playbook. The barriers differ by therapeutic area. The patient with obesity faces a different journey from the patient with Alzheimer’s.

She used LillyDirect’s neuroscience and Alzheimer’s work as the example. In Alzheimer’s, the access bottleneck is less about demand for a drug and more about diagnostic availability. There is a backlog of specialists who can diagnose and treat patients in time. For newer Alzheimer’s agents, Steele said timing is critical: the patient must be in a certain disease stage for the medicines to work. If that window is missed, the medicines should not be prescribed or infused.

Steele said that when she worked in neuroscience, patients often waited 17 months from first noticing memory and thinking issues to reaching the right office with the right diagnostic test. The vision for LillyDirect in that area is to unlock diagnostics at scale, including through blood-based biomarkers, and make access less dependent on waiting six to nine months for a neurologist in a city — a particularly acute issue for rural communities.

Her broader point was that companies need to map the patient journey, identify “leakage points,” and put solutions where barriers occur. That approach may expand to other therapeutic areas, but the solution must fit the disease, not simply copy the GLP-1 channel.

Kosali Simon emphasized the novelty of new engagement channels. GLP-1s have shown the force of TikTok, word of mouth, and 800 numbers in bringing patients into care. Other stakeholders — employers, managed care, and others — are likely watching how patient engagement through GLP-1s changes the relationship between patient and provider. Instead of waiting for the healthcare provider to offer an intervention, the medication itself may be what brings someone into the doctor’s office asking what is available.

Ali Furman closed the consumer thread by calling the GLP-1 situation a cultural movement that became consumer-led. America had not previously had a scalable solution to the overweight problem, she said, and consumers rallied around the first intervention that appeared to address it at scale. Around that, they are building ecosystems.

The business implication, in her view, is not to focus only on the decline in categories consumers now want less of. It is to follow what they value. Many users report feeling happier and more self-confident. If that happens across a large adult population, Furman said, the implications are material for society — not necessarily all positive, but significant.

Coombs added a generational possibility: if parents with obesity use the drugs to improve healthspan and develop healthier habits, they may help children build more sustainable ways of eating and managing health, potentially affecting childhood obesity patterns in future generations. It was offered as a possibility, not as evidence already established.