Once-Weekly Semaglutide in Adults with Overweight or Obesity
New England Journal of Medicine, 2021 · paper · Clinical trials
A double-blind trial of 1,961 adults with overweight or obesity without diabetes, given weekly semaglutide 2.4 mg or placebo plus lifestyle intervention for 68 weeks. Semaglutide produced a 14.9% mean weight loss versus 2.4% with placebo.
2 findings
Wording "reaching a nadir at week 60 (Fig. 1A and 1B)": in Fig. 1B (on-treatment data) the semaglutide curve is still falling at week 68, so only Fig. 1A shows a nadir at week 60.
- p. 5 Results, Change in Body Weight
reaching a nadir at week 60 (Fig. 1A and 1B)
- p. 6 Figure 1B, semaglutide curve, weeks 60 and 68
B Body Weight Change from Baseline by Week, Observed On-Treatment Data
- p. 11 Discussion, liraglutide comparison paragraph
did not reach the nadir until week 60
Only the authors can say what is right here.
Judge's note: Looked at Figure 1 on page 6. In panel A the semaglutide point at week 60 is the lowest (about -15.8) and week 68 sits slightly higher (about -15.6). In panel B the curve falls at every visit and week 68 is the lowest point (about -16.9, matching the trial product estimate in the text). The nadir statement therefore holds for panel A only, yet both panels are cited; the Discussion repeats "did not reach the nadir until week 60". Values are read from plotted points, since no numbers are printed for the curves.
Second check: another model agreed
Report a problem with this finding
Wording Table 3 footnote ‡: "In the placebo group, death due to glioblastoma, aspiration pneumonia, and severe sepsis occurred in one participant each who had discontinued placebo" reads as three participants, but Table 3 shows 1 (0.2) placebo participant with 3 fatal events and the text says "One death was reported in each group".
- p. 10 Table 3 footnote ‡
curred in one participant each who had discontinued placebo
- p. 10 Table 3, Fatal events row, placebo
1 (0.2)
- p. 9 Results, Safety and Side-Effect Profile
One death was reported in
- p. 9 Safety paragraph
One death was reported in each group
Judge's note: Page 10 image: the Fatal events row gives placebo 1 (0.2) participants and 3 events, and footnote ‡ reads as quoted. Page 9 text says one death in each group. The table and text agree that one placebo participant died, with three fatal event terms recorded, so "one participant each" misdescribes the data; "in one participant" is the evident intent. No statement elsewhere in the article supports three placebo deaths.
Second check: another model agreed
Report a problem with this finding
6 minor (typos and the like; most checked by the judge only)
Wording The Discussion gives the placebo proportion with at least 5% weight loss as "32% of those who received placebo" and later as "At week 68, 31% of participants who received placebo had lost at least 5% of baseline body weight"; the reported value is 31.5% (182 of 577).
- p. 11 Discussion, first paragraph
as compared with 32% of those who received placebo, lost 5% or more of baseline
- p. 11 Discussion, right column, paragraph beginning 'At week 68'
At week 68, 31% of participants who received placebo had lost at least 5% of baseline body
- p. 8 Table 2, body-weight reduction ≥5% row, placebo
31.5
- p. 11 Discussion, right column, paragraph beginning 'At week 68'
At week 68, 31% of participants who received placebo had lost at least 5% of baseline body weight
Judge's note: Page 11 image shows both sentences as quoted. Table 2 and Figure 1C give 31.5%, and the Results give 182 participants of 577, which is 31.54% and rounds to 32%. The 12% and 5% in the second sentence match the in-trial values 12.0% and 4.9%, so it is the same data set; the on-treatment value (33.1%, Fig. 1D) would not give 31% either. The 31% is therefore the inconsistent figure. The discrepancy is one percentage point of rounding and does not change any conclusion.
Second check: another model agreed
Report a problem with this finding
Typo Table 2 writes "Lipid levels, ratio of wk 68 value to baseline" but "C-reactive protein, ratio of wk-68 value to baseline": the same phrase is hyphenated in one row and not in the other.
- p. 8 Table 2, last row
ratio of wk-68 value to baseline
Judge's note: Both rows are visible on the upright image of page 8 and read as quoted. Purely cosmetic; either form could be the house style, but the two rows are inconsistent.
Second check: none (judge only)
Report a problem with this finding
Typo Table 2 footnote *: "see Table S2 for corresponding data for the estimand (which assessed treatment effect assuming all participants adhered to treatment and did not receive rescue intervention)"; the words "trial product" are missing before "estimand".
- p. 9 Table 2 footnote *
see Table S2 for corresponding data for the estimand (which
Judge's note: Read the sideways footnote on the page 9 image: the wording is as quoted with no gap or line-break loss. The footnote names the treatment policy estimand, then refers to an unnamed "the estimand" whose description matches the trial product estimand defined in the Methods; the Results text says the trial product estimand data are in Table S2. The meaning can be inferred from the parenthesis, so this is a missing-word slip.
Second check: none (judge only)
Report a problem with this finding
Typo Table 2 (continued) section header reads "Exploratory end-point assessed in the prediabetes subpopulation", hyphenated and singular, whereas the other headers read "end points" and two end points are listed beneath it.
- p. 9 Table 2 (continued)
Exploratory end-point assessed in the prediabetes subpopulation
Judge's note: Page 9 image shows the header as quoted. The other section headers on page 8 ("Coprimary end points", "Confirmatory secondary end points", "Supportive secondary end points") are unhyphenated and plural, and the term is unhyphenated throughout the article. Cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo Disclosures: "grant support, paid to Guys and St. Thomas’ Hospital" lacks the apostrophe that the affiliations give in "Guy’s and St. Thomas’ NHS Foundation Trust".
- p. 12 Disclosures, Dr. McGowan
grant support, paid to Guys and
- p. 13 Appendix, affiliations
Guy’s and St. Thomas’ NHS Foundation Trust
- p. 12 Disclosures, Dr. McGowan
paid to Guys and
Judge's note: The page 12 crop shows "Guys and St. Thomas’ Hospital" with no apostrophe in "Guys" while the apostrophe after "Thomas" is present, so this is not an extraction loss. The page 13 crop shows "Guy’s" in the affiliation list. The paper's own spelling supplies the fix.
Second check: none (judge only)
Report a problem with this finding
Typo Affiliations: "the National Institute of Health Research, UCLH Biomedical Research Centre" uses "of" where the body's name is given elsewhere as "for"; the same list also uses the abbreviation "NIHR Leicester Biomedical Research Centre" without introducing it.
- p. 13 Appendix, affiliations
the National Institute of Health Research, UCLH Biomedical Research Centre
- p. 13 Appendix, affiliations
the NIHR Leicester Biomedical Research Centre
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: the official name of the UK body abbreviated NIHR was the National Institute for Health Research at the time of publication. The page 13 crop shows the wording as quoted, and the same sentence uses "NIHR" for the Leicester centre. The correct name cannot be verified from the files, so this goes to the authors.
Second check: none (judge only)