This is a continuation of a long topic, addressed by post number rather than by page. Start at post 1 · go to the accepted answer.
Surrogate endpoints are not automatically bad and their validity is compound-specific and population-specific. The question is whether this surrogate has been validated for this use.
One more caveat and then I will stop qualifying: the sample selected itself.
Reporting rather than recommending, on measurement error. What happened is above. Whether it should have is a different question and not one I am qualified to answer.
Checked the measurement error claim against the primary source this morning. It survives, with a narrower scope than the version quoted here. Posting the narrower scope.
Worth separating two things that post #60 runs together.
The pre-specified endpoint being a weaker proxy than you would like is a real criticism. It is a smaller one than saying the result was chosen after the fact.
Take it as a starting point and not as a specification.
I came in to disagree and I am leaving without a disagreement.
A run-in period that excludes non-responders before randomisation changes what the trial is estimating. It is legitimate design and it must be stated in any summary.
I am reporting what happened, not recommending it.
A note on how measurement error gets discussed rather than on measurement error itself: the confident posts get the replies and the careful ones get ignored, and the careful ones have been right more often.
Worth stating the null on measurement error before we explain it: the observation may be nothing. That possibility deserves a sentence and usually does not get one.
Post #66 is the version of this I will quote in future. One addition.
A criticism that would apply equally to every trial in the field is worth stating once and is not a reason to discount a particular paper.
That matches what I was told, which is not the same as knowing it.
Post #71 is right about the mechanism and I think understates the practical bit.
Before the thread moves on from measurement error — what is the sample size behind the claim? I am not being difficult; I have seen the same figure quoted from an n of four and from an n of four hundred.
That is consistent with mine, for whatever one more account is worth.
I keep a log for measurement error specifically because my memory of it turned out to be systematically wrong in one direction. Six weeks of notes cost nothing and settled it.
I read post #71 twice before replying, because I had assumed the opposite.
Hold a trial to the standard something could actually have met. A criticism that no achievable design could have answered is a criticism of the field rather than of the paper.
A modest claim, modestly supported.
Post #75 answers the question as asked. The question underneath it is different.
Worth separating measurement error as a question about the compound from measurement error as a question about the documentation. They get answered by different people and only one of them is answerable here.
Where I have landed on measurement error, having got it wrong once in public: the direction is clear, the magnitude is not, and anyone quoting a precise magnitude has borrowed it from somewhere that did not measure it.
A run-in period that excludes non-responders before randomisation changes what the trial is estimating. It is legitimate design and it must be stated in any summary.
Where I part company with post #75, and it is a narrow parting.
Source for the measurement error figure, since it was asked for. It is in the discussion rather than the abstract, which is why the version circulating is stronger than the paper is.
Reading the surrounding paragraph is worth the two minutes. The authors are more careful than their summarisers.
Quietly grateful for the plain phrasing. Not every thread gets that.
Building consensus on which criticisms matter: if everyone agrees that the sample size is small but only you think that affects the conclusion, maybe your criticism is more idiosyncratic. That does not make it wrong but it is worth noticing.
I would want the raw data before agreeing with my own summary of it.
The honest answer on measurement error is that it depends, and the useful part is the list of what it depends on. Four items, in rough order of how much they matter.
Most people get the first two right and then argue about the fourth.
Good question, well framed, and I would like to see it answered properly.
On measurement error I would separate what is worth knowing from what is worth acting on. The first list is long and the second is short, and conflating them is how threads get heated.
Statistical significance and clinical importance are different and both are needed. A significant difference below the minimal important difference is a real finding of no practical consequence.
Stating my assumptions rather than smuggling them in.
Generalisability and validity are separate axes. A trial can be internally impeccable and still tell you nothing about the person asking.
I have seen it go both ways, which is why I hedge.
The arithmetic in post #88 is right; the assumption feeding it is the part to check.
The arithmetic on measurement error is the easy part and it is where the errors are, which is an uncomfortable combination. Show your working and someone will catch it.
Answering the question post #86 raises rather than the one it answers.
Where the measurement error discussion usually stalls is that nobody wants to say "I do not know" and everyone is willing to say "it varies". Those are the same sentence with different clothes on.