Tirzepatide vs Dulaglutide: A Real Head-to-Head

Almost every GLP-1 result you read compares a drug against placebo, which is why comparing two of them usually means comparing two separate trials — something this site refuses to do. This one is different. A randomized trial put 13,299 people on either tirzepatide or dulaglutide and followed them a median of four years, so the comparison is inside the randomization rather than across studies.^[1]

The result

The primary composite kidney outcome occurred in 396 people on tirzepatide (6.0%) and 498 on dulaglutide (7.6%) — a 23% lower risk, with a hazard ratio of 0.77 (95% CI 0.68 to 0.88, p=0.0002).^[1]

The benefit came from two different places depending on how much kidney disease people started with. In those at low-to-moderate risk, it was driven by fewer new cases of persistent macroalbuminuria — protein appearing in the urine. In those at high risk, it was driven by slower loss of filtration: the annual decline in eGFR was lower by 0.29 mL/min per 1.73m² overall, and by 0.93 in the high-risk group.^[1]

What a head-to-head does and does not buy you

The value is that it removes the usual objection. When people set tirzepatide’s weight results beside semaglutide’s, they are reading across trials with different populations and different placebo arms, which is not a comparison — we make that point in the tirzepatide timeline. Here, both groups were enrolled together and assigned at random.

1. Both arms were treated. Nobody received placebo, so this tells you which drug did better and not how much either beats no treatment at all.
2. It is a pre-specified analysis of a cardiovascular trial, not a study designed around kidneys. Pre-specified is far better than picking through results afterward, and it is still not the same as being the primary question.
3. Dulaglutide is not a placebo. It is an established GLP-1 with its own kidney data, so tirzepatide is beating an active, effective comparator rather than nothing.
4. The population was specific: people with type 2 diabetes and atherosclerotic cardiovascular disease, roughly a third of whom already had microalbuminuria at baseline.

The cost side is in the same abstract

Nausea, vomiting and diarrhea were all more common with tirzepatide than with dulaglutide.^[1] That is not a footnote — it is the trade-off, stated by the trial itself, and it is the reason a better kidney outcome does not automatically make it the better drug for a given person.

A drug that protects kidneys slightly better and is tolerated slightly worse is a genuine choice, not an obvious upgrade.

For anyone weighing this, our kidneys article covers the separate and much larger question of what these drugs do for kidney disease against no treatment, including the acute risk that runs the other way when someone becomes dehydrated.