Most of them are stomach trouble, and most are mild. Pooled across 55 placebo-controlled trials covering 106,395 people, gallstones rose to a risk ratio of 1.46 and reflux to 2.19 [2]. In absolute terms that is about two extra gallstone cases per thousand and four extra reflux cases. Pancreatitis, bowel obstruction, ileus, perforation, bleeding and gastroparesis came back null. The frightening items on the list are the ones with nothing behind them.
Everyday nausea is common and varies by drug and by dose. In one phase 2b trial, gastrointestinal events affected 51.9% on weekly semaglutide 1 mg. On a longer-acting comparator the range ran 81.8% to 87.3%, mostly grade 1 or 2 [3]. Dosing less often did not mean feeling better. One matched analysis found coded nausea and vomiting at half the rate in men as in women on tirzepatide. The hazard ratio was 0.50 [4]. Whether that is half the symptom or half the coding is not settled.
At the severe end the picture is flatter. Tolerability is the second reason people give for choosing one of these drugs over another, after price. The two get traded off constantly. This study is the largest attempt to check whether that trade-off exists at the severe end. It does not. That is a purchasing-relevant negative, which is the kind this market produces least often. It sits alongside the price comparison in what the tirzepatide premium buys.
What was compared
A new-user, active-comparator cohort study identified adults with type 2 diabetes starting dulaglutide, semaglutide or tirzepatide. The window ran from January 2019 to August 2024. It built three pairwise comparisons, propensity-matched 1:1 within each [1].
That produced 65,238 matched pairs for semaglutide against dulaglutide, 20,893 for tirzepatide against dulaglutide, and 46,620 for tirzepatide against semaglutide. The primary outcome was a composite of acute pancreatitis, biliary disease, bowel obstruction, gastroparesis and severe constipation.
What it found, which is nothing
Semaglutide against dulaglutide gave 0.96 (95% CI 0.87 to 1.06). Tirzepatide against dulaglutide gave 0.96 (95% CI 0.77 to 1.20). Tirzepatide against semaglutide gave 1.07 (95% CI 0.90 to 1.26).
All three intervals contain 1.0, and contain it with room to spare. The last of those, 1.07, is the one most likely to be misread. A point estimate inside a null interval is not a trend toward harm. It is noise around no difference. The authors’ own conclusion is that the three drugs have similar gastrointestinal safety profiles.
Set against what the two cost
Across the 407 injected semaglutide figures and 384 injected tirzepatide figures recorded here, the median billed rates are $179 and $249. That is a difference of $70, or $840 over a year. Roster read September 2026.
There are real arguments for paying it. The weight evidence favors tirzepatide. A seller publishing its whole dose ladder is worth more than one that does not. What this study removes is the argument that the dearer molecule is easier on you. At the severe end, across 132,751 matched pairs, it is not.
The comparison a buyer can actually make
The safety difference is not there to shop on. That leaves the variables this site measures. What a month costs once the fees are in it. Whether the figure survives a dose increase. Whether the seller states either. Those are in what a GLP-1 actually costs and nobody says what happens at a higher dose. The spread between the cheapest and dearest figure for one molecule is wider than the gap between the molecules.