Why do the Phase 2 GLP-1 trials titrate amounts up so slowly?

Your read is basically right. In the trial write-ups the stated reasons are almost always (1) GI adverse events cluster at the start and around each step up, and slow escalation blunts that peak, and (2) it improves retention, which protects statistical power. A few also note it mirrors how the drug would be labeled if approved, so the trial reflects real-world use.

One nuance worth reading for: some trials report that a slower schedule doesn't change the final efficacy much, only the dropout rate. That's a recurring finding across the class, and it's a nice example of dose context that has nothing to do with any individual's plan.

That's really helpful, thanks. The retention/power point is one I hadn't connected — makes sense that dropouts in the aggressive arm would quietly wreck the comparison. I'll look for the review you mentioned.

Adding one caution for anyone reading along: the escalation schedules in these papers are tied to specific formulations, populations, and monitoring setups. They're describing what a supervised trial did, not a template. Easy to forget when the schedule looks so tidy in a table.