Evidence · Journal club

[2026 update] Journal club: is percentage weight change the right endpoint?

Evidence › Journal club effect size risk of bias randomised trial

MSaarinenTL3Regular7 Aug 2025#1

Journal club: is percentage weight change the right endpoint? — that is the question, and I have not found it answered plainly anywhere I have looked.

A question about what the numbers mean rather than what they are.

SURPASS-2 (N Engl J Med, 2021) reports its primary result in relative terms, and the absolute figures are in a table two pages later. The two produce very different impressions of the same finding.

I would like to work through which one belongs in a summary, and why the answer is not simply "both".

45 likes 12mo

l.sarkissianTL2Member8 Aug 2025#2

Where I part company with the opening post, and it is a narrow parting.

Where the group cannot agree, record the disagreement rather than resolving it by seniority. The recorded disagreement is more honest and more useful later.

Anyone with a larger sample, please post it.

0 likes 12mo

c.nybergTL28 Aug 2025#3

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.

The right answer here may simply be that it has not been measured.

3 likes 12mo

h.hutchingsTL1Member8 Aug 2025#4

l.sarkissian, post #2: Where I part company with the opening post, and it is a narrow parting. Where the group cannot agree, record the disagreement rather than resolving it by seniority. The recorded disagreement is more honest and more useful later. Anyone with a larger sample, please post it. Go to post

Sessions on negative or null results are consistently the most instructive and the hardest to get anyone to attend.

11 likes in reply to #2 12mo

r.sobczakTL28 Aug 2025#5

h.hutchings, post #4: Sessions on negative or null results are consistently the most instructive and the hardest to get anyone to attend. Go to post

Building on post #4 rather than restating it.

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.

That is all the detail I have. Someone else will have more.

32 likes in reply to #4 12mo

e.almeidaTL2Member9 Aug 2025#6

Second this, and I would have said it less carefully.

0 likes 12mo

n.ramosTL29 Aug 2025 · edited#7

Reading order that works for these sessions: registry entry, methods, baseline table, primary result, then abstract last. Reading the abstract first anchors everything that follows.

6 likes 12mo

IHollingworthTL2Member9 Aug 2025#8

Where the group cannot agree, record the disagreement rather than resolving it by seniority. The recorded disagreement is more honest and more useful later.

16 likes 12mo

h.brandtTL29 Aug 2025#9

Taking post #8 at face value and following it one step further.

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.

0 likes 12mo

isotonic_sheetTL3Regular10 Aug 2025#10

The baseline characteristics table is the single most useful page for the questions that get asked here, because it tells you who the result applies to.

1 like 12mo

two_year_lineTL3Regular10 Aug 2025 · edited#11

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.

That has held every time I have looked, which is not the same as always.

3 likes 12mo

c.chowdhuryTL210 Aug 2025#12

h.hutchings, post #4: Sessions on negative or null results are consistently the most instructive and the hardest to get anyone to attend. Go to post

Sessions on negative or null results are consistently the most instructive and the hardest to get anyone to attend.

The honest answer is that it depends, and here is what it depends on.

0 likes in reply to #4 12mo

batchlogTL3Regular10 Aug 2025#13

Reading back through, this was answered upthread and I missed it. My fault.

31 likes 12mo

d.ferreiraTL211 Aug 2025#14

Collapsed as off-topic by two members at trust level 3 or above

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.

bias_varianceTL4Biostatistician11 Aug 2025#15

isotonic_sheet, post #10: The baseline characteristics table is the single most useful page for the questions that get asked here, because it tells you who the result applies to. Go to post

Sessions on negative or null results are consistently the most instructive and the hardest to get anyone to attend.

That is the version I use. It may not be the version that is correct.

6 likes in reply to #10 12mo

s.ostergaardTL211 Aug 2025#16

n.ramos, post #7: Reading order that works for these sessions: registry entry, methods, baseline table, primary result, then abstract last. Reading the abstract first anchors everything that follows. Go to post

Post #15 answers the question as asked. The question underneath it is different.

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.

That holds for the case as described. Change the assumptions and it may not.

1 like in reply to #7 12mo

impurity_tableTL3Analytical chemist11 Aug 2025#17

Reading order that works for these sessions: registry entry, methods, baseline table, primary result, then abstract last. Reading the abstract first anchors everything that follows.

I would be interested in a counterexample if anyone has one.

0 likes 12mo

i.norgaardTL211 Aug 2025#18

SURMOUNT-OSA (N Engl J Med 2024): Tirzepatide in obstructive sleep apnoea, using an objective physiological endpoint. Notable because soft endpoints are avoided. Two parallel trials addressed the confounder directly.

Anyone who has looked at this more carefully, please correct the record.

22 likes 12mo

RodriguesTL311 Aug 2025#19

Collapsed as off-topic by two members at trust level 3 or above

Coming back to post #17, because the follow-up matters more than the original answer.

The baseline characteristics table is the single most useful page for the questions that get asked here, because it tells you who the result applies to.

e.bakkenTL212 Aug 2025#20

Where the group cannot agree, record the disagreement rather than resolving it by seniority. The recorded disagreement is more honest and more useful later.

3 likes 12mo

BBramleyTL3Regular12 Aug 2025#21

h.brandt, post #9: Taking post #8 at face value and following it one step further. 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. Go to post

The baseline characteristics table is the single most useful page for the questions that get asked here, because it tells you who the result applies to.

Someone will know this better than I do and I hope they say so.

0 likes in reply to #9 12mo

t.tullochTL212 Aug 2025#22

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.

2 likes 12mo

t.nardoneTL3Regular12 Aug 2025#23

Post #22 is the version of this I will quote in future. One addition.

Sessions on negative or null results are consistently the most instructive and the hardest to get anyone to attend.

13 likes 12mo

c.serranoTL212 Aug 2025#24

Where I part company with post #20, and it is a narrow parting.

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.

The claim is narrower than it sounds, and deliberately so.

27 likes 11mo

HRouhaniTL1Member13 Aug 2025#25

d.ferreira, post #14: 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. Go to post

Adding a data point of agreement rather than a data point.

0 likes in reply to #14 11mo

t.verhoevenTL213 Aug 2025#26

Everything in post #24 holds. The case it does not cover is the one I have.

Reading order that works for these sessions: registry entry, methods, baseline table, primary result, then abstract last. Reading the abstract first anchors everything that follows.

5 likes 11mo

vial_deskTL3Regular13 Aug 2025#27

Building on post #26 rather than restating it.

The baseline characteristics table is the single most useful page for the questions that get asked here, because it tells you who the result applies to.

That is my reading. Someone else read the same page differently and was reasonable.

19 likes 11mo

e.mensaTL213 Aug 2025#28

Where the group cannot agree, record the disagreement rather than resolving it by seniority. The recorded disagreement is more honest and more useful later.

Worth checking against a second source before it gets quoted onward.

0 likes 11mo

HHidalgoTL2Member13 Aug 2025#29

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 reads as pedantry until the day it does not.

29 likes 11mo

st.dialloTL213 Aug 2025#30

On post #28 — agreed on the reasoning, with one qualification.

Sessions on negative or null results are consistently the most instructive and the hardest to get anyone to attend.

I have left out the parts I could not verify.

0 likes 11mo

Sponsored Sigma-Aldrich — certified reference materialsReference standards, analytical reagents and certified materials for method calibration. If you are validating a method, this is where the standard comes from.Reference standards →