What Patients Want in an Obesity Drug

Eight hundred US adults with overweight or obesity were handed a stack of hypothetical drug choices and told to pick. Route of administration and weight-loss efficacy won, over and over, ahead of nearly everything else on the list. Dosing complexity barely moved the needle. That is the headline finding from OPTIC, a discrete-choice experiment published September 17, 2026 in Diabetes, Obesity and Metabolism^[1], a survey design built specifically to force people to trade one desirable thing off against another rather than just rate everything highly. One of the two hypothetical drug profiles it tested was built to resemble an oral form of semaglutide, the GLP-1 receptor agonist already sold as Wegovy and Rybelsus.

How the survey worked

US adults with overweight or obesity and at least one obesity-related complication completed an online survey in October and November 2025^[1]. The sample was split on purpose. Four hundred participants had never taken an obesity medication before; 400 had. A separate 400 had never injected a medication; 400 had. Each person then worked through a discrete-choice experiment, repeatedly picking between two hypothetical medication profiles that varied on attributes like average weight loss, cardiovascular risk reduction, side effects, route and frequency of dosing, and dosing instructions. A random-parameters logit model, a statistical technique for estimating how much weight people implicitly assign to each attribute, turned those repeated choices into relative-importance scores^[1].

What patients said mattered most

Route of administration ranked as the single most important attribute, ahead of average weight-loss percentage, cardiovascular risk reduction, and the share of people who reach 20% or more total weight loss^[1]. Dosing instructions ranked dead last. Whether a pill had to be taken on an empty stomach barely swayed anyone. Convenience of taking the drug outweighed the fine print of how to take it, though the efficacy attributes still carried real weight in every trade-off respondents made.

It helps to anchor what an 'average weight-loss percentage' means in practice, because a survey attribute is an abstraction until you attach real numbers to it. In the STEP-1 trial, weekly injectable semaglutide produced a mean body-weight reduction of 14.9% at 68 weeks, against 2.4% on placebo^[2]. In SELECT, semaglutide also cut major cardiovascular events, meaning heart attack, stroke, and cardiovascular death, by 20% in adults with overweight or obesity and existing cardiovascular disease but no diabetes^[3]. Those are the numbers behind a line like '20% weight loss' on a survey card.

Why the pill question mattered so much

Nearly every obesity medication on the market today is an injection. An oral option removes the needle entirely, and the survey found real appetite for exactly that: 73.3% of participants who had never taken an obesity medication said they were open to trying an oral one^[1]. Dosing instructions ranking last does not mean instructions are irrelevant to everyone. Only 23.4% of participants said taking a medication on an empty stomach and waiting 30 minutes before eating would be disruptive to their daily life^[1], which is a real restriction for some patients, just not most of them. Semaglutide's existing oral tablet, Rybelsus, is dosed once daily on an empty stomach with up to 120 ml of water, with the dose titrated upward over a series of monthly steps^[4].

A hypothetical pill beat a hypothetical pill, not a verdict on real drugs

The survey also ran a head-to-head comparison between two constructed profiles: Treatment A, built to resemble an oral form of semaglutide, and Treatment B, built to resemble oral orforglipron, an investigational oral GLP-1 receptor agonist from Eli Lilly. Respondents chose Treatment A over Treatment B by 84.2% to 15.8% in the discrete-choice questions and by 90.0% to 10.0% in a separate fixed-choice question^[1]. That is a stated preference for one modeled set of attributes over another, built from published data available when the survey was designed, not a clinical head-to-head. It does not show which drug performs better in the body. It shows which attribute package survey respondents found more appealing on paper.

Who ran the survey, and who paid for it

RTI Health Solutions, an independent research contractor, designed and fielded the survey^[1]. Four of the eight listed authors are Novo Nordisk employees, the company that manufactures semaglutide under the Ozempic, Wegovy and Rybelsus brand names^[1]. That funding relationship does not make the attribute-importance findings wrong. A random-parameters logit model is a standard, peer-reviewed method for this kind of research, and the underlying methodology has been used in dozens of unrelated drug-preference studies. Even so, the specific comparison between a semaglutide-like profile and a competitor's investigational drug reads best as sponsor-funded research, not an independent verdict.

What this means

Stated preference in a survey is not the same as what a person actually does in a pharmacy. Cost, insurance coverage, and a clinician's recommendation all weigh in too, and none of those showed up as a survey attribute here. The sample was US adults only, surveyed online across two months in late 2025, so it cannot tell us how preferences look in other countries or in people without an obesity-related complication. What the OPTIC results add is a clearer picture of the trade-offs patients say they are willing to make: an oral option is worth a lot to people who currently inject, efficacy and heart-protection numbers still carry real weight, and dosing rules are a smaller sticking point than drug developers sometimes assume. None of this is a reason to start, switch or stop an obesity medication without talking to a prescribing clinician who knows your full history.