Mounjaro vs Ozempic: The Head-to-Head Trial, and What Ozempic Still Wins

Last reviewed 21 August 2026 · Educational, not medical advice · No sponsorships, no affiliate links

The head-to-head trial, and its one weakness

SURPASS-2 randomised adults with type 2 diabetes inadequately controlled on metformin to tirzepatide 5, 10, or 15 mg or semaglutide 1 mg, for 40 weeks. Tirzepatide beat semaglutide at every dose on both HbA1c and weight, and the margin widened with dose. On the stricter multi-target analysis, 29% of the tirzepatide 15 mg group met three or more intensive treatment targets against 8% on semaglutide.

The weakness, and Novo Nordisk is entitled to press it: semaglutide was capped at 1 mg. That was the highest approved Ozempic dose when the trial was designed, but the ceiling is now 2 mg. Nobody has run tirzepatide 15 mg against semaglutide 2 mg, so the true gap in current practice is probably narrower than SURPASS-2 suggests.

How much narrower is a guess, but not an unbounded one. Semaglutide’s dose-response between 1 mg and 2 mg is modest, and the one trial that tested both molecules at their full obesity ceilings - SURMOUNT-5, covered in Wegovy vs Zepbound - still favoured tirzepatide by 6.5 percentage points of body weight. The direction of the result is not in doubt; only its size is.

What real-world diabetes practice shows

A propensity-matched cohort of more than 18,000 US patients, published in JAMA Internal Medicine, tracked exactly this pair of products in ordinary care. Twelve-month weight change was -15.3% on tirzepatide against -8.3% on semaglutide, and tirzepatide patients were three times as likely to reach 15% weight loss.

The number in that paper that should shape your expectations more than either of those, though, is the discontinuation rate: 55.9% of tirzepatide patients and 52.5% of semaglutide patients had stopped within twelve months. A drug you are not taking in month thirteen has an HbA1c effect of zero, whichever molecule it was.

The cardiovascular and kidney asymmetry

This is where the comparison stops being about potency, and it is the section most likely to change a prescription.

Ozempic reduced major adverse cardiovascular events against placebo in SUSTAIN-6 and carries a cardiovascular risk-reduction indication in type 2 diabetes with established heart disease. It also carries a chronic kidney disease indication, granted on the FLOW trial, for reducing the risk of worsening kidney disease and cardiovascular death. In the EU it has additionally been recommended for peripheral artery disease following the STRIDE trial.

Mounjaro has SURPASS-CVOT, published in December 2025: more than 13,000 patients with type 2 diabetes and cardiovascular disease, over 4.5 years, tirzepatide against dulaglutide. Tirzepatide was non-inferior, with an 8% lower rate of major events. That is a genuinely large and reassuring trial. It is also an active-comparator design, which answers "is it at least as good as a drug we already trust" rather than "by how much does it beat nothing." Tirzepatide has no kidney indication and no placebo-controlled cardiovascular outcome trial.

For a patient with type 2 diabetes and albuminuria or declining eGFR, that difference is decisive in a way no amount of extra HbA1c reduction compensates for.

Cost, and Lilly’s awkward internal arithmetic

At US manufacturer self-pay rates in August 2026, Mounjaro runs $499 a month at every dosethrough LillyDirect. Ozempic runs $349 through 1 mg and $499 at 2 mg through NovoCare. At maintenance, then, the two cost the same.

The uncomfortable comparison is not between the brands but inside one of them. Zepbound is the identical tirzepatide molecule at identical doses, and it sells for $299 to $449 a month. A patient with type 2 diabetes paying cash for Mounjaro is paying up to $200 a month more than a patient with obesity paying cash for the same drug at the same strength. Nothing pharmacological explains that; it is indication pricing, and it is worth raising with a prescriber if both labels are clinically defensible for you.

Insurance inverts the picture entirely. Diabetes-labelled products are the covered ones. With commercial coverage and a savings card, either drug can land near $25 a month, at which point every figure in this section stops mattering.

Practical differences at the point of use

• Neither causes hypoglycaemia on its own, but both do when combined with insulin or a sulfonylurea. Those doses usually need to come down when either drug is started or escalated.
• Six dose steps against four.Mounjaro’s longer ladder gives more room to find a tolerable maintenance dose, and more months of titration before you get there.
• Gastrointestinal side effects dominate both, cluster after dose increases rather than at steady state, and are managed the same way: slow down.
• Only semaglutide has an oral form for diabetes, Rybelsus, which must be taken on an empty stomach with a 30-minute wait.

What to track when the endpoint is HbA1c, not the scale

HbA1c is a ninety-day rolling average. That makes it an excellent measure of control and a terrible feedback signal: if a dose is not working, the blood test tells you a quarter after the fact. Everything useful happens between the labs, and only if you wrote it down.

For this pair specifically, three things belong in the log. Glucose readings with times, so patterns separate from noise. Every dose change with its date, on both the GLP-1 and anything else. And - the one people skip - every adjustment to insulin or a sulfonylurea, because those are the changes that make the difference between a smooth escalation and a hypoglycaemic episode. A low that happened at some point last week is an anecdote. A low with a timestamp, a dose, and a meal next to it is something your prescriber can act on.

Zenday App is our top-ranked GLP-1 companion at 4.55/5 and the one we would hand to someone starting either drug tomorrow: it keeps doses, food, activity, and side-effect days in one place, works from first injection through maintenance, and treats side-effect management as a core pillar. For the quarterly endocrinology appointment,Shotsy exports a clinician-ready PDF, which turns twelve weeks of scattered memory into one page a doctor can read in thirty seconds. Both sit at the top of ourGLP-1 app ranking.

What the evidence does not show

• No head-to-head at current ceilings. SURPASS-2 compared 15 mg against 1 mg, not against 2 mg.
• No placebo-controlled cardiovascular outcome trial for tirzepatide, and no kidney outcome trial at all.
• No head-to-head on hard outcomes. SURPASS-2 measured HbA1c and weight over 40 weeks. It did not measure heart attacks, kidney failure, or death.

How to choose

• Glycaemic control is not where you want it, and weight is part of the problem: Mounjaro. The head-to-head is clear and the composite endpoint gap is large.
• You have chronic kidney disease: Ozempic. It is the only one of the two with the indication and the trial behind it.
• You have established cardiovascular disease: Ozempic has the stronger evidence class, though SURPASS-CVOT makes tirzepatide a defensible choice.
• You are paying cash: the same at maintenance. Ask whether the obesity-labelled version of tirzepatide is appropriate for you, because it is materially cheaper.
• You are already on Ozempic and want to move: see switching from Ozempic to Mounjaro, which covers the restart dose and the insulin adjustments.

Method and sources

Efficacy figures come from SURPASS-2, SURPASS-CVOT, SUSTAIN-6, FLOW, and the JAMA Internal Medicine propensity-matched cohort as published. Prices are US manufacturer self-pay rates published by LillyDirect and NovoCare and read on 21 August 2026; they change frequently and are not what you will pay with insurance. We take no money from either manufacturer and run no affiliate links. For the molecule-level comparison behind this page, see semaglutide vs tirzepatide.