Most People Never Reach the Dose

There is a simple explanation for why real-world results look worse than trial results, and it is not that patients are different. A UK population study found that only 58% of people starting semaglutide reached the recommended maintenance dose after four weeks, and 21% were still on the starting dose — the one that exists to be tolerated rather than to work.^[1]

Where people actually end up

By the tenth prescription, the commonest subcutaneous semaglutide doses were 0.5 mg and 1 mg.^[1] Those are the middle of the ladder, not the top of it — and this is after ten prescriptions, which is most of a year.

Escalation was faster in people with higher BMI, and the authors found no further variation by other patient characteristics, including established cardiovascular disease. Delayed escalation was, in their words, frequent across all patient stratifications.^[1]

A trial escalates on protocol. Ordinary care escalates when someone remembers to.

And half of them stop

More than half of the people who started the oral version stopped.^[1] That is the formulation sold on convenience — and it is the one with the strictest daily conditions, requiring an empty stomach on waking with a wait before anything else, which we covered in travel and time zones. Convenience of format and convenience in practice turn out to be different things.

It also has the worst nausea profile of the class, which we covered in the side-effect ranking — for orforglipron, a different oral drug, but the pattern is worth noticing.

Why this reframes so much else

We wrote recently that real-world data shows outcome differences by race and ethnicity where randomized trials showed none, and argued the gap lies in access, continuity and titration rather than in patients — see who responds in the real world. This study puts numbers on the titration half of that argument.

If a fifth of people never leave the starting dose and nearly half stop altogether, then a large part of “the drug worked less well than in trials” is really “the drug was never taken at the dose it was tested at, for the duration it was tested for”. Those are different problems with different solutions.

What this does not establish

• It is a type 2 diabetes population in UK primary care. Dosing targets, prescribing incentives and follow-up patterns differ from US weight-management practice.
• It does not say why people stopped. Side effects, cost, supply and simply not returning are all in there, undifferentiated.
• Slow escalation is sometimes correct. Holding a dose because someone is nauseated is good medicine, not a failure — and the labels build that judgment in.
• Higher is not automatically better. The point is that most people never get the chance to find out.