Semaglutide vs. Tirzepatide, Read Honestly

For three years the comparison everyone wanted was one nobody had run. People held up the semaglutide trial next to the tirzepatide trial, noted that the second number was bigger, and declared a winner. That is not how trials work. Different populations, different sites, different eras, different everything.

Then SURMOUNT-5 ran the actual comparison — the two drugs, head to head, in the same study, in adults with obesity and without diabetes, both titrated to their maximum tolerated doses over 72 weeks.

So we finally have an answer to the narrow question. I want to give it to you accurately, and then explain why it settles less than you’d think.

The result

Tirzepatide produced a mean weight reduction of roughly 20%. Semaglutide produced roughly 14%. Waist circumference followed the same pattern, with a difference of several centimetres favouring tirzepatide.

That is a real, statistically robust, clinically meaningful difference in the average. On the question “which of these two drugs produces more weight loss across a population?” — the answer is tirzepatide, and I do not think that is seriously contestable anymore.

Why “on average” is doing enormous work

Here is what a mean of 20% versus 14% looks like when you open it up.

Both trials report responder distributions, and they are wide. In the tirzepatide arm, a substantial group lost more than 25% of their body weight. Another group lost under 5%. The same shape holds for semaglutide, shifted left. The two distributions overlap across most of their range.

Which means: there are people for whom semaglutide will outperform tirzepatide, and there is currently no test, biomarker, or genotype that tells us in advance who they are. I have patients who plateaued at 6% on tirzepatide and lost 18% after switching to semaglutide. That is not supposed to happen if you believe the drugs are ranked. It happens regularly.

A trial tells you what happens to a thousand people. It does not tell you what happens to you. It narrows the bet; it does not place it.

The evidence bases are not the same shape

This is the part that gets flattened in every social media comparison, and it is the part that actually changes my prescribing.

Semaglutide has the cardiovascular outcome data. SELECT enrolled adults with established cardiovascular disease and overweight or obesity, without diabetes, and followed them for major adverse cardiovascular events. Semaglutide reduced them by about 20% relative to placebo. That is an outcomes trial — hard endpoints, heart attack and stroke and cardiovascular death, not a surrogate marker. For a patient with a prior myocardial infarction, that evidence is not decorative. It is the reason to choose the drug.

Tirzepatide has the deeper obesity and sleep apnea data. Larger weight effect, and dedicated trials in obstructive sleep apnea showing meaningful reduction in apnea-hypopnea index. For the patient whose dominant problem is severe obesity with OSA, that is the more relevant literature.

These are not the same claim. “More weight loss” and “fewer heart attacks” are different endpoints, established in different populations, and one does not automatically imply the other. Tirzepatide’s own cardiovascular outcomes trial is ongoing. When it reads out, this section will need rewriting. I will rewrite it.

What usually decides it in my office

Effect size is maybe the fourth thing on the list. Ahead of it:

Coverage and cost. These are extraordinarily expensive drugs and formulary coverage is capricious. The best drug is frequently the one your plan will actually pay for over 24 months, because a superior drug you stop at month five is worse than an adequate drug you stay on.

Tolerability. Some people are miserable on one and comfortable on the other, and it is not predictable. Nausea, reflux, constipation, sulfurous burping, fatigue — the side-effect profiles are broadly similar but individual experience is not.

Comorbidity. Established cardiovascular disease pushes me toward semaglutide. Sleep apnea, or a prior inadequate response to a GLP-1 alone, pushes me toward tirzepatide. Type 2 diabetes changes the whole conversation and the appropriate branded product with it.

Response, once you can measure it. Since there is no way to predict in advance who does better on which, the only real evidence about you is what happens after you start — and that requires having recorded it. Patients who switch drugs on an impression tend to switch on a bad month; patients who switch on six months of dose, side-effect and measurement data switch for a reason. I keep mine in the Zenday App, which matters here because the question at a switch decision is not “how much do I weigh” but “what did an adequate trial at an adequate dose actually produce, and how did I tolerate it” — and that is a record, not a memory.

Injection experience. The devices differ. This sounds trivial until you meet the patient who will not use the pen they find intimidating.

Supply. We have all lived through the shortages. A prescription you cannot fill is not a treatment plan.

Two things I want to say about the alternatives

Compounded versions. I understand the economics completely, and I am not going to lecture anyone about the impossible position that drug pricing puts patients in. What I will say plainly: compounded semaglutide and tirzepatide are not FDA-approved products, potency and purity vary by facility, and the salt forms sometimes sold are not the same molecule studied in these trials. Dosing errors from vials and syringes have caused real harm. If you are using one, please at least do it under the supervision of a clinician who knows you are.

Oral options. Oral semaglutide exists, and higher-dose oral formulations for weight management have been studied. The pill is not automatically gentler; the GI effects are mechanistic, not route-dependent. It solves a needle problem, not a nausea problem.

The answer I actually give

If you do decide to move between them, the practical mechanics matter more than the comparison does — there is no validated dose-equivalence table, and you re-titrate from near the bottom. Switching between GLP-1 medications sets out how.

When someone asks me which one is better, what I say is:

Tirzepatide produces more weight loss on average, by a margin that is real. Semaglutide has the better-established cardiovascular benefit. Your insurance, your stomach, and your other diagnoses will probably decide this before either of those facts does. And if the first one doesn’t work well for you after a fair trial at an adequate dose, switching is a completely reasonable next move — not a failure.

That is less satisfying than a winner. It is closer to true.

One practical difference that does not show up in any comparison table: the two molecules have different missed-dose windows, and the two semaglutide brands differ from each other too. What to do if you miss a dose has the numbers.