Reading a supplementary appendix and finding the interesting part posts 31–60
This is a continuation of a long topic, addressed by post number rather than by page. Start at post 1.
Grateful for the specificity. Vague answers to this question are what sent me looking.
A definition problem is doing most of the work in this reading a supplementary appendix discussion. Once the term is pinned down I suspect the disagreement mostly goes away and what is left is small.
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.
The arithmetic in post #34 is right; the assumption feeding it is the part to check.
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.
Worth one more sentence than it usually gets.
What I would check first on reading a supplementary appendix is whether the thing being measured moved or whether the way of measuring it moved. Those look identical in a graph.
Reading a supplementary appendix was covered in the wiki last year and the page has a review date on it, which is a better starting point than my memory of a thread.
Post #36 put the caveat in the right place and I want to underline it.
Discontinuation handling is the methodological detail that most changes a result and gets the least attention. Read how missing data was imputed before reading the effect size.
Small point, but it is the one that usually catches people.
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.
I think the reading a supplementary appendix question is answerable and has not been answered, which is a more optimistic position than most of this thread.
Post #41 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.
A partial answer, offered because a partial answer beats none.
What I would tell a new member reading about reading a supplementary appendix for the first time: the confident posts are not the reliable ones, and the reliable ones are longer.
Clear enough that I do not think I have a follow-up, which is unusual.
Building on post #45 rather than restating it.
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.
This is the sort of thing that ought to be settled and apparently is not.
Narrowing post #49, because the general version has more than one answer.
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.
Two sources, same conclusion, and I could not rule out that one copied the other.
Answering the reading a supplementary appendix question as asked, then the question I think is meant. As asked: yes, with the qualification below. As meant: it depends on how the first measurement was taken.
Building on post #52 rather than restating it.
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.
On reflection I would soften that slightly.
Post #50 put the caveat in the right place and I want to underline it.
On reading a supplementary appendix: the maintained page in the documentation commons covers the general case with citations and a review date, which is more reliable than any reply here including this one.
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.
Thank you for taking the time. That was more work than a reply usually is.
Post #54 and I disagree about the size of the effect, not about the direction.
Reading the supplementary appendix is where most of the real information is, and it is where almost nobody goes. The baseline table alone answers half the generalisability questions asked here.
Written from notes rather than memory, which is why the numbers are specific.
Post #56 is right about the mechanism and I think understates the practical bit.
Discontinuation handling is the methodological detail that most changes a result and gets the least attention. Read how missing data was imputed before reading the effect size.
The part I am sure of is shorter than the part I have written.
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.
If this contradicts something upthread, the upthread version may well be the better one.