Immortal time bias in a claims-database study — one year on posts 31–47

Helpful, and easy to find again, which is half of what a good reply is.

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.

Answering the question post #32 raises rather than the one it answers.

Checked the Immortal time bias claim against the primary source this morning. It survives, with a narrower scope than the version quoted here. Posting the narrower scope.

The arithmetic in post #32 is right; the assumption feeding it is the part to check.

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.

That has been true for the cases I have seen and I have not seen many.

Placebo-controlled versus active-controlled changes what a result means entirely, and comparing across the two is one of the most common errors in this subcategory.

A modest claim, modestly supported.

Per-protocol and intention-to-treat analyses answer different questions and neither is the honest one by default. Reporting both is the practice worth insisting on.

Generalisability: do the inclusion/exclusion criteria narrow the population so much that results do not apply to real people asking about it? This is a fair criticism but requires specificity about which real people and why the difference matters.

Narrowing post #35, because the general version has more than one answer.

Since Immortal time bias keeps coming up, it should probably be a maintained page rather than a recurring thread. I am happy to draft it if someone with more direct experience will review it.

Coming back to post #35, because the follow-up matters more than the original answer.

Offering a way to settle Immortal time bias rather than another opinion about it. Two measurements, taken the same way, a fortnight apart. If the difference is within the noise, the question was not answerable at this precision.

That is clearer than the version I had in my head. Thank you.

This follows post #39 rather than contradicting it.

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.

Bias towards the null and bias away from the null: different criticisms have different directions. Differential dropout might bias away from null; conservative statistical analysis might bias toward null.

Defending a design that is being criticised unfairly is as useful as criticising one that deserves it, and it happens far less often.

I am aware this is the third time this month I have made this point.

Coming back to post #42, because the follow-up matters more than the original answer.

On Immortal time bias, the part that usually goes wrong is that the question is asked as though it has one answer. It has a range, and the width of the range is the interesting bit.

If you can post the two or three numbers you are working from, several people here will check the arithmetic rather than argue about the conclusion.