What does REDEFINE-1 tell me about dizziness at the 2.4 mg dose?

The case in front of me: REDEFINE-1 · dizziness · 2.4 mg.

The figures are clear enough; the question is what they mean and what they do not.

I can supply the numbers if the specifics change the answer.

What can I legitimately conclude from this figure?

Answer first: read the primary endpoint, the comparator and the population before you read the effect size. Almost every argument on this site about a trial is really an argument about one of those three.

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.

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.

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.

Read the protocol and the statistical analysis plan if the result matters to you. Both are usually published alongside.

Before comparing two trials, check whether they share an endpoint definition. Frequently they do not, and the numbers then are not comparable in any sense.

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.

It helps to be literal here: duration decides what can be seen. A 68-week trial can measure weight and glycaemia; it cannot measure anything whose event rate is one per cent per year without enrolling tens of thousands.

Where a result is quoted from a conference abstract rather than a peer-reviewed publication, the numbers routinely move between the two. It is worth checking which one you are reading.

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

edited 21 Feb 2026 by lyoph_cake — added the placebo-arm figures

A trial establishes what happened to a defined group under a defined protocol. Extending it beyond that group is inference, and inference is allowed as long as it is labelled.

Trial populations are selected. Exclusion criteria in this class routinely remove people with significant renal impairment, prior pancreatitis and unstable psychiatric illness, which is exactly the population the results are then quoted for.

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

If a claim cannot be traced to a named trial with a named endpoint, treat it as a claim rather than as evidence.

More usefully, the trial answers a narrower question than the headline suggests, and the narrowing is where the useful information is.

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.

Registry entries at ClinicalTrials.gov carry the pre-specified primary endpoint with a timestamp, which is the cheapest available check on whether an endpoint was changed after the data were seen.

One qualification: absence of a signal in a trial of this size is not evidence of absence for a rare event. It is evidence that the event is rarer than the trial could detect.

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