Evidence · Trials

Subgroup analyses: pre-specified versus discovered

Evidence › Trials randomised trial intention to treat surrogate endpoints estimand

j.habermannTL3Regular16 Apr 2026#1

Subgroup analyses: pre-specified versus discovered — setting out what I have, and where I think it stops being reliable.

Comparing SURMOUNT-2 (Lancet, 2023) with SURPASS-4 (Lancet, 2021) and finding the comparison harder than it looks.

Different populations, different durations, different endpoints defined slightly differently, and in one case a different estimand. People compare the headline percentages anyway, including me until recently.

Is there a defensible way to put these side by side, or is the honest answer that there is not and we should stop?

0 likes 3mo

a.eriksenTL2 Moderator14 Jun 2026#2

This follows the opening post rather than contradicting it.

Intent-to-treat versus per-protocol: ITT includes everyone assigned regardless of whether they took the drug. Per-protocol includes only those who completed it as intended. The two can give substantially different results.

22 likes 1mo

IsaksenTL3Regular27 Jul 2026#3

j.habermann, post #1: Subgroup analyses: pre-specified versus discovered — setting out what I have, and where I think it stops being reliable. Comparing SURMOUNT-2 ( Lancet , 2023) with SURPASS-4 ( Lancet , 2021) and finding the comparison harder than it looks. Different populations, different durations, different endpoints defined slightly differently, and… Go to post

Worth separating two things that the opening post runs together.

Multiplicity and multiple comparisons: if a trial tests many hypotheses, the chance of a false positive on at least one by random chance increases. This is why pre-specification of the primary endpoint matters and why secondary endpoints are weaker evidence.

10 likes in reply to #1 1d

Promoted into the documentation commons. The content of this topic is maintained at SUSTAIN 6 — trial digest, with named maintainers and a review date. The promotion was discussed in doc review. Corrections are best raised against the document, which is the version that gets kept current.

Suggested topics

Topic	Participants	Replies	Views	Activity
Revisiting: Primary endpoint hierarchies and why order matters Evidence › Trials estimand surrogate endpoints discontinuation & dropout intention to treat Posting this under the heading it deserves: Revisiting: Primary endpoint hierarchies and why order matters Everything below is what sits behind that. Comparing SURMOUNT-4 ( JAMA , 2024) with SURPASS-2 ( N…	+21	25	46k	3mo
Trial registration and comparing the protocol with the paper Evidence › Trials randomised trial open label intention to treat Trial registration and comparing the protocol with the paper — setting out what I have, and where I think it stops being reliable. I have seen LEADER ( N Engl J Med , 2016) cited in support of a claim I do…	+38	44	4.8k	10mo
Run-in periods and the population they select Evidence › Trials surrogate endpoints discontinuation & dropout estimand randomised trial On the subject in the title: Run-in periods and the population they select Working notes rather than a conclusion. Session topic: PIONEER 6 ( N Engl J Med , 2019). Please read it before posting; the…	+130	142	7.3k	1mo
Interim analyses and stopping rules Evidence › Trials intention to treat surrogate endpoints open label discontinuation & dropout Interim analyses and stopping rules Writing it up because I had to work it out twice and would rather nobody else did. I have seen SUSTAIN 6 ( N Engl J Med , 2016) cited in support of a claim I do not think…	+25	29	35k	16d
About the Trials category Evidence › Trials randomised trial estimand intention to treat Pinned Wiki Individual randomised trials: design, population, endpoints, results, and limitations. This post is a community wiki: any member at trust level 3 or above can edit it, and every edit is recorded with its…	+6	10	27k	12mo

Related topics — sharing the tags randomised trial, intention to treat, surrogate endpoints

Topic	Participants	Replies	Views	Activity
Why a glucagon receptor agonist in a weight-loss compound is not a contradiction — one year on Compounds › Retatrutide preprint retatrutide TRIUMPH programme randomised trial Why a glucagon receptor agonist in a weight-loss compound is not a contradiction — one year on — that is the question, and I have not found it answered plainly anywhere I have looked. I have seen SURPASS-2 (…	+3	7	1k	2mo
Journal club: STEP 1 and the treatment-policy estimand — does this still hold? Evidence › Journal club randomised trial risk of bias effect size journal club Asking directly, because I could not find a straight answer: Journal club: STEP 1 and the treatment-policy estimand — does this still hold? Comparing STEP 2 ( Lancet , 2021) with SURPASS-2 ( N Engl J Med ,…	+70	76	4.8k	15h
Cagrilintide molecular characteristics and analytical considerations Compounds › Cagrilintide & amylin analogues cagrilintide randomised trial amylin analogues cagrisema Cagrilintide molecular characteristics and analytical considerations — setting out what I have, and where I think it stops being reliable. I have seen SURMOUNT-2 ( Lancet , 2023) cited in support of a claim I…		4	41k	12mo
Interim analyses and stopping rules — does this still hold? Evidence › Trials estimand discontinuation & dropout open label surrogate endpoints Solved Interim analyses and stopping rules — does this still hold? — that is the question, and I have not found it answered plainly anywhere I have looked. Comparing SCALE ( N Engl J Med , 2015) with SURMOUNT-4 (…	+30	34	29k	13mo
Second pass at: Cagrilintide's dosing interval and the pharmacokinetics behind it Compounds › Cagrilintide & amylin analogues cagrisema effect size randomised trial Wiki On the subject in the title: Second pass at: Cagrilintide's dosing interval and the pharmacokinetics behind it Working notes rather than a conclusion. Session topic: SURMOUNT-1 ( N Engl J Med , 2022). Please…	+2	6	2.4k	7mo

Sponsored Sigma-Aldrich — certified reference materialsReference standards, analytical reagents and certified materials for method calibration. If you are validating a method, this is where the standard comes from.Reference standards →