[2026 update] Composite endpoints and the component doing the work posts 61–88
This is a continuation of a long topic, addressed by post number rather than by page. Start at post 1.
Post #58 describes the usual case. This is about the unusual one.
I would rather this thread reach "we do not know" about Composite endpoints than reach a confident answer that nobody can support when asked.
Risk of bias: structured appraisal of internal validity. Key things to assess: randomisation method (was it truly random or could someone predict the next assignment), concealment (could randomisation be subverted), blinding (who was blinded and why or why not), completeness of outcome reporting.
Composite endpoints has been discussed here with more heat than it deserves, mostly because two definitions have been in play the whole time.
On post #62 — agreed on the reasoning, with one qualification.
A treatment-policy estimand asks what happens to people assigned to a strategy, including those who abandon it. A hypothetical estimand asks what would have happened had everyone continued. Both are legitimate and they give different numbers.
Composite endpoints should be read component by component. A composite driven entirely by its softest component is a different finding from one where the components move together.
Someone should write this up properly, and it should probably not be me.
Two claims get bundled together under Composite endpoints and they need separating. The descriptive one — this is what was observed — is usually well supported. The causal one — this is why — usually is not.
Almost every disagreement in threads like this one dissolves once you say which of the two you are making.
Surrogate endpoints: an endpoint that is not the outcome that matters but is measured as a stand-in. HbA1c is a surrogate for long-term glucose control and the short-term complications it prevents. Weight loss is a surrogate for metabolic health and long-term outcomes. Surrogates are useful but not identical to the endpoint that matters.
Reading this Composite endpoints thread as someone who came in with a fixed view: the third and seventh replies moved me and the confident ones did not.
The number people quote for Composite endpoints is a central estimate presented without its interval, and the interval is wide enough that the estimate is nearly uninformative on its own.
I read post #69 twice before replying, because I had assumed the opposite.
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.
A single observation, in a thread that deserves better than single observations.
Reading rather than answering, but this is the post I would point somebody at.
On post #73 — agreed on the reasoning, with one qualification.
A treatment-policy estimand asks what happens to people assigned to a strategy, including those who abandon it. A hypothetical estimand asks what would have happened had everyone continued. Both are legitimate and they give different numbers.
The confident version of this sentence would be wrong, so here is the hedged one.
The version of Composite endpoints that I was taught turned out to be a teaching simplification. Useful, and not true in the way I had assumed it was.
Post #77 is right about the mechanism and I think understates the practical bit.
Composite endpoints should be read component by component. A composite driven entirely by its softest component is a different finding from one where the components move together.
The answer changed when I changed how I was measuring, which was informative.
Post #77 put the caveat in the right place and I want to underline it.
Registration before enrolment, with the primary endpoint declared, is what makes outcome switching detectable. Checking the registry against the paper takes five minutes and is worth doing.
Building on post #77 rather than restating it.
Risk of bias: structured appraisal of internal validity. Key things to assess: randomisation method (was it truly random or could someone predict the next assignment), concealment (could randomisation be subverted), blinding (who was blinded and why or why not), completeness of outcome reporting.
I am not the right person to answer the follow-up to this.
Careful with the language on Composite endpoints. "Not detected" and "not present" are different findings and the first is a statement about the method.
I would keep Composite endpoints and the decision it usually gets used for separate in this thread. They are related and they are not the same question, and merging them is why the last one went badly.
Post #82 is the version of this I will quote in future. One addition.
Absolute and relative effects answer different questions. Write down the event rate in each arm and the difference between them; everything quotable is derived from those two numbers.
That is consistent with mine, for whatever one more account is worth.
Confirming post #85 from a second method, which matters more than confirming it from a second person.
Absolute numbers, not just relative: a 30% relative reduction tells you the ratio but not the practical magnitude. The event rate in each arm and the difference between them tells you how many people benefit.
It cost nothing to check and would have cost something not to.
Surrogate endpoints: an endpoint that is not the outcome that matters but is measured as a stand-in. HbA1c is a surrogate for long-term glucose control and the short-term complications it prevents. Weight loss is a surrogate for metabolic health and long-term outcomes. Surrogates are useful but not identical to the endpoint that matters.
This is the sort of thing that ought to be settled and apparently is not.
This topic was referenced in
- [2026 update] Reading a supplementary appendix and finding the interesting partEvidence › Trials · 44 replies
Suggested topics
| Topic | Participants | Replies | Views | Activity |
|---|---|---|---|---|
|
Comparators chosen for regulatory reasons rather than clinical ones — one year on
Comparators chosen for regulatory reasons rather than clinical ones — one year on — setting out what I have, and where I think it stops being reliable. Posting a small dataset on Comparators chosen for…
|
+82 | 87 | 1.1k | 12mo |
|
Adjudicated events and why the definition matters — the long version
Adjudicated events and why the definition matters — the long version Writing it up because I had to work it out twice and would rather nobody else did. A narrow question about Adjudicated events, deliberately…
|
+125 | 146 | 13k | 15mo |
|
How to read a forest plot, properly, from scratch
The question in the title: How to read a forest plot, properly, from scratch I will give what I have already checked below so nobody repeats it. Reading STEP 4 ( JAMA , 2021) for the population rather than…
|
+73 | 81 | 17k | 7mo |
|
Open-label extensions: what survives and what does not
Posting this under the heading it deserves: Open-label extensions: what survives and what does not Everything below is what sits behind that. I have spent a fortnight trying to pin open-label extensions down…
|
4 | 22k | 3mo | |
|
Composite endpoints and the component doing the work
Composite endpoints and the component doing the work Writing it up because I had to work it out twice and would rather nobody else did. Collecting what is known about composite endpoints in one place, because…
|
+3 | 7 | 33k | 13h |
Related topics — sharing the tags surrogate endpoints, discontinuation & dropout, estimand
| Topic | Participants | Replies | Views | Activity |
|---|---|---|---|---|
|
Journal club: FLOW and the renal composite, component by component
On the subject in the title: Journal club: FLOW and the renal composite, component by component Working notes rather than a conclusion. Reading FLOW ( N Engl J Med , 2024) for the population rather than the…
|
+84 | 91 | 18k | 2y |
|
Follow-up: Open-label extensions: what survives and what does not
Open-label extensions: what survives and what does not — setting out what I have, and where I think it stops being reliable. I have spent a fortnight trying to pin Open-label extensions down and I want to set…
|
+22 | 26 | 397 | 16mo |
|
Journal club: indirect comparison between two programmes, defended and attacked
Journal club: indirect comparison between two programmes, defended and attacked Writing it up because I had to work it out twice and would rather nobody else did. Trying to work out what would count as…
|
+122 | 131 | 5.2k | just now |
|
Run-in periods and the population they select — the long version
Run-in periods and the population they select — the long version Writing it up because I had to work it out twice and would rather nobody else did. A narrow question about Run-in periods, deliberately narrow,…
|
2 | 8.9k | 4mo | |
|
Journal club: SUSTAIN 6 and its retinopathy signal — does this still hold?
The question in the title: Journal club: SUSTAIN 6 and its retinopathy signal — does this still hold? I will give what I have already checked below so nobody repeats it. Reading SUSTAIN 6 ( N Engl J Med ,…
|
+60 | 66 | 12k | 16mo |