GLP-1 Drugs at Two Years: A Public Health Audit

Two years ago the GLP-1 receptor agonist class was a diabetes medication with a growing off-label footprint. Today it is a public health story, a consumer trend, a stock-price mover, and a small but meaningful factor in national prescription-drug spending. This audit looks at what the public data — CDC prescribing trends, NEJM outcomes trials, and the disclosed financials of the three dominant manufacturers — actually shows.

We are limiting the audit to the two drugs that dominate use: semaglutide (marketed as Ozempic for diabetes and Wegovy for weight loss) from Novo Nordisk, and tirzepatide (marketed as Mounjaro for diabetes and Zepbound for weight loss) from Eli Lilly. Both are peptide analogs; both are administered by injection; both work in part by modulating gut-hormone signaling that governs appetite and glucose homeostasis.

What the outcome trials show

The efficacy data on both drugs is unusually strong by pharmaceutical standards. The STEP trials for semaglutide and the SURMOUNT trials for tirzepatide both produced average weight loss in the 15–22% of body-weight range over 68–72 weeks, sustained on continued therapy. These are results that historically only bariatric surgery achieved.

The cardiovascular outcome trials are the more consequential story. SELECT, the large semaglutide cardiovascular outcomes trial, showed a statistically significant reduction in major adverse cardiovascular events in patients with cardiovascular disease but not diabetes. That result — a weight-loss drug reducing cardiovascular events on its own — pulled GLP-1s out of the diabetes-and-obesity conversation and into cardiology.

The kidney outcome data (FLOW trial) and the MASH liver-disease data have added further indications. What began as a diabetes drug is looking increasingly like a broad cardiometabolic agent.

Side effects: real, common, mostly tolerable

The gastrointestinal side-effect profile is well-established: nausea, vomiting, diarrhea, and constipation are common, particularly during dose escalation. Most patients tolerate them; a meaningful minority discontinue.

The more concerning signals — pancreatitis, medullary thyroid cancer (based on rodent studies), gastroparesis, and the loss of lean muscle mass — remain areas of active surveillance. The FDA's adverse event reporting system shows elevated but not alarming reporting rates. The consensus in the endocrinology literature as of mid-2026 is that the safety profile is favorable for the indicated populations, with real but manageable trade-offs.

The economic story

The economic footprint is enormous. Novo Nordisk and Eli Lilly's public earnings show GLP-1 revenues that have collectively passed sixty billion dollars annually and continue to grow. Both companies have executed capacity expansions of a scale that pharmaceutical manufacturing rarely sees, and both continue to face supply constraints for the weight-loss indications.

The second-order effects are showing up in unexpected places. Retail analysts have tracked declines in snack food and alcohol sales in high-GLP-1-penetration markets. Airlines have reported weight-shift data. Health insurers are re-modeling long-term liabilities in ways that reflect both the direct cost of the drugs and their potential to reduce downstream cardiometabolic care.

The most-watched question — whether payers will broadly cover GLP-1s for weight loss — has moved from a hard no to a partial yes. Medicare's 2024 coverage expansion for semaglutide in cardiovascular indications was a landmark, and commercial payers are following selectively.

The generic cliff

The competitive story is that both dominant drugs face patent-cliff pressure over the next several years. Generic semaglutide is projected to enter multiple markets between 2026 and 2031 depending on jurisdiction. Compounded semaglutide, sold widely through US telehealth channels during the FDA shortage, remains a legally contested product. The next-generation entrants — dual and triple receptor agonists, oral GLP-1s from Lilly and Roche's CT-388 — are in late-stage trials.

The likely 2030s picture is a broader class of drugs, more oral options, meaningfully lower unit costs, and a much wider treated population.

What the audit does not settle

The audit does not settle the long-term safety question — most GLP-1 users are still measured in years of exposure, not decades. It does not settle the psychosocial and cultural question of what a widely available appetite-suppressing drug does to a society's relationship with food. And it does not settle the equity question of whether these drugs will remain concentrated in wealthier populations or will reach the patients who most need them.

The data through mid-2026 supports one clear reading: the drugs work, the safety profile is manageable, the economic effects are real, and the class is not going away. Everything else remains open.

Sources

• NEJM STEP semaglutide weight loss trials — nejm.org/doi/full/10.1056/NEJMoa2032183
• NEJM SURMOUNT tirzepatide trials — nejm.org/doi/full/10.1056/NEJMoa2206038
• NEJM SELECT cardiovascular outcomes trial — nejm.org/doi/full/10.1056/NEJMoa2307563
• FDA Adverse Event Reporting System — fda.gov/drugs/questions-and-answers-fdas-adverse-event-reporting-system-faers

From the editorial archive

• How the $370B IRA Bet Is Actually Performing — Another audit of a policy or program in mid-play.
• After the AI Bubble: An Audit of 2025's Promises — The other big "what actually happened?" audit.
• The Rule of 40 Is Dead: What SaaS Boards Actually Look At in 2026 — How public and private capital is reading the story.