Journal club: SURMOUNT-1, read without the press release Writing it up because I had to work it out twice and would rather nobody else did.
Posting a small dataset on SURMOUNT-1. It is mine, it is uncontrolled, and the method is stated so it can be discounted appropriately.
What I would like is not agreement but a second dataset collected by someone with no stake in mine. If one exists I would rather read it than argue for this one.
Offering a way to settle SURMOUNT-1 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.
Confirming the opening post from a second method, which matters more than confirming it from a second person.
An hour is enough for one paper read properly and not enough for two read partly. The temptation is always to add the second.
The arithmetic on SURMOUNT-1 is the easy part and it is where the errors are, which is an uncomfortable combination. Show your working and someone will catch it.
Coming back to post #5, because the follow-up matters more than the original answer.
SELECT (N Engl J Med 2023): Semaglutide cardiovascular outcomes without diabetes. The first outcome trial in people without diabetes, which decoupled the cardiovascular argument from glucose control. Read the absolute numbers, not just the relative reduction.
Post #5 is right about the mechanism and I think understates the practical bit.
Where the SURMOUNT-1 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.
SURMOUNT-1 came up in a thread eighteen months ago and was answered well. I cannot find it, which is itself the problem, so here is the reconstruction.
The arithmetic in post #9 is right; the assumption feeding it is the part to check.
Session format: read the paper before posting. The discussion is much better when everyone has. Start with the estimand and population, then methods, then results, then limitations. That order makes critique coherent.
Building on post #10 rather than restating it.
The most useful reply I ever got about SURMOUNT-1 was a request to state my units. It sounds like pedantry and it has saved me twice.
The figures usually contain the finding and the text usually contains the interpretation. Separating them for the first twenty minutes is a discipline worth keeping.
It cost nothing to check and would have cost something not to.
No notes. Posting so the count is not one.
Post #12 and I disagree about the size of the effect, not about the direction.
Summarising the SURMOUNT-1 thread so far, since it is long and the answer is buried: the first reply has the method, the fourth has the correction to it, and the rest is people agreeing at length.
What I can speak to on SURMOUNT-1 is narrow, so I will keep it narrow rather than generalising from it. Beyond that boundary I do not know.
Where the group cannot agree, record the disagreement rather than resolving it by seniority. The recorded disagreement is more honest and more useful later.
A partial answer, offered because a partial answer beats none.
I came in to disagree and I am leaving without a disagreement.
SURMOUNT-1 (N Engl J Med 2022): Tirzepatide obesity trial. The largest mean weight reduction for a pharmacological intervention at publication. Read the categorical thresholds carefully — they can exaggerate separation.
I would put this at better than even and not much better.
Adding what did not work for me on SURMOUNT-1, since the failures never get written up and they are half the useful information.
An observation about SURMOUNT-1 that I cannot explain and am posting anyway, on the principle that unexplained observations are more useful public than private.
Picking up post #21: that is the part I would want checked first.
Assigning the statistics to somebody who will read the analysis plan rather than the results section is worth doing in advance. It is the part that gets skipped by default.
Posted with less confidence than the sentence structure implies.
Reframing SURMOUNT-1 slightly, because I think the disagreement is about the question rather than the answer. If the question is "does it happen", yes. If it is "how often", nobody here knows.
SURMOUNT-1 (N Engl J Med 2022): Tirzepatide obesity trial. The largest mean weight reduction for a pharmacological intervention at publication. Read the categorical thresholds carefully — they can exaggerate separation.
This follows post #25 rather than contradicting it.
The documentation on SURMOUNT-1 is better than this thread and I say that as someone who has posted in the thread.
The bit of SURMOUNT-1 that nobody enjoys is that the answer changes depending on what you are trying to decide with it. Say what the decision is and the thread will converge.
Post #29 is the version of this I will quote in future. One addition.
The most common failure mode is spending fifty minutes on the effect size and ten on the population. Reversing that ratio would improve most sessions.
A single observation, in a thread that deserves better than single observations.