What does STEP 2 tell me about vomiting at the 1 mg dose?

Concretely: STEP 2 · vomiting · 1 mg.

I would like to know the limits of what can be inferred from this.

What I am trying to avoid is over-reading a single result, which I have done before.

What can I legitimately conclude from this figure?

Only what the 1 mg arm reported, and the denominator is that arm rather than the trial. Adverse-event tables are published per arm, so the 1 mg incidence of vomiting has its own numerator and its own denominator, and pooling it with the other arms produces a figure that describes nobody. Two further deductions before you use it. Subtract the placebo arm — vomiting occurs in people who received nothing, and the difference is the part attributable to the drug. And check whether STEP 2 counted events or counted participants: one participant with six episodes is one row in a participant count and six in an event count, and the two get quoted interchangeably. Nothing here is medical advice.

Look at the discontinuation rate alongside the efficacy figure. A large effect in the two thirds who stayed is a different result from a large effect in everyone.

Placebo arms in this class are not nothing. Lifestyle-intervention placebo arms in the major obesity trials commonly lose two to three per cent of body weight, so an active-arm figure quoted without its comparator overstates the drug effect by roughly that much.

Headline results, principal programmes

A composite endpoint is only as informative as its least serious component. Where a cardiovascular composite combines death, infarction and stroke, ask which component moved, because they are not interchangeable outcomes.

Meta-analyses in this area are dominated by whichever trial contributed the most participants, so read the forest plot rather than the summary estimate.

I am not a clinician and this is not medical advice; it is a reading of a published protocol.

The short version: check the endpoint, check the comparator, check who was excluded, then look at the number.

This is answerable from the published record, but only if you take the placebo arm seriously rather than reading the active arm alone.

Non-inferiority and superiority designs are not interchangeable. A non-inferiority result says the new agent is not meaningfully worse against a pre-specified margin — it does not say it is as good, and it certainly does not say it is better.

Open-label extensions are not the same evidence as the randomised phase. Once everyone knows what they are taking, the reported outcomes acquire a bias that no analysis fully removes.

Quote the interval alongside the estimate and half the disagreements on this site would not start.

edited 12 Nov 2025 by petra_hovland — added the citation requested in comments

On the detail: the trial answers a narrower question than the headline suggests, and the narrowing is where the useful information is.

Confidence intervals matter more than point estimates when two trials disagree. Two studies reporting fifteen and twenty per cent whose intervals overlap heavily have not disagreed about anything.

Put another way, intention-to-treat and per-protocol analyses answer different questions. ITT asks what happens if you offer the treatment; per-protocol asks what happens if it is taken as directed. The gap between the two is a measure of how tolerable the protocol was.

The cardiovascular outcome programme in this class runs to several large randomised trials — LEADER for liraglutide, SUSTAIN-6 and SELECT for semaglutide, REWIND for dulaglutide — and they are the reason the class is discussed as more than a weight intervention.

Be careful about generalising from a trial population to yourself. The exclusion criteria are usually the most informative page in the supplement.

When two sources disagree, the answer is almost always in the methods section of the one you have not read.