Denosumab for Prevention of Fractures in Postmenopausal Women with Osteoporosis
New England Journal of Medicine, 2009 · paper · Clinical trials
A three-year placebo-controlled trial of twice-yearly denosumab injections in 7,868 postmenopausal women with osteoporosis. Denosumab reduced new vertebral fractures by 68%, hip fractures by 40% and nonvertebral fractures by 20%, without increased cancer or infection risk.
2 findings
Reference or label Figure 2 legend says "Changes in mean values for serum C-telopeptide of type I collagen (CTX) (Panel C) and serum procollagen type I N-terminal propeptide (PINP) (Panel D) are shown", but the y-axes of Panels C and D are labelled "Median Percent Change".
- p. 6 Figure 2 legend
mean values for serum C-telopeptide of type I collagen (CTX) (Panel C)
- p. 6 Figure 2, Panels C and D, y-axis label
Median Percent Change
- p. 8 Discussion, fourth paragraph
dian of 86% at 1 month
- p. 8 Discussion, left column, fourth paragraph
Denosumab reduced bone resorption by a me-
Judge's note: Page 6 image: both Panel C and Panel D y-axes read 'Median Percent Change'; the legend says 'Changes in mean values' for the same panels. The Discussion (page 8 image) says "Denosumab reduced bone resorption by a median of 86% at 1 month" and the legend itself says the marker comparisons used the Wilcoxon rank-sum test, so median is the statistic plotted and the legend's 'mean' (apparently carried over from the BMD sentence) is the mislabel. No passage reconciles the two.
Second check: another model agreed
Report a problem with this finding
Wording Abstract says there were "no adverse reactions to the injection of denosumab", but Results report "Local reactions after injection of a study drug occurred in 33 subjects (0.8%) in the denosumab group and 26 subjects (0.7%) in the placebo group".
- p. 1 Abstract, Results, last sentence
and no adverse reactions to the injection of denosumab
- p. 7 Results, Adverse Events, second paragraph
study drug occurred in 33 subjects (0.8%) in the
- p. 1 Abstract, Results, last sentence
no adverse reactions to the injection of denosumab
- p. 7 Results, Adverse Events, second paragraph
Local reactions after injection of a study drug occurred in 33 subjects (0.8%) in the denosumab group and 26 subjects (0.7%) in the placebo group
Only the authors can say what is right here.
Judge's note: Read the abstract on the page 1 image and the Adverse Events paragraph on the page 7 image. The abstract clause is coordinated with "there were no cases of osteonecrosis of the jaw and no adverse reactions to the injection", i.e. it reads as zero events, not as "no increase". The Results give 33 local injection reactions on denosumab (33/3886 = 0.85%, printed 0.8%; 26/3876 = 0.67%, printed 0.7%). The paper defines no distinction between an 'adverse reaction to the injection' and a 'local reaction after injection', and nothing else in the article explains the difference. The abstract wording is broader than the data; the authors probably meant no excess over placebo. Rated S3 (wording overstates) since the counts themselves are reported correctly in the Results.
Second check: another model agreed
Report a problem with this finding
3 minor (typos and the like; most checked by the judge only)
Wording Figure 2 legend: "P<0.001 for all between-group comparisons at all time points on the basis of analysis-of-covariance (ANCOVA) models" is followed by "for CTX and PINP, the comparisons were calculated with the use of the Wilcoxon rank-sum test".
- p. 6 Figure 2 legend
parisons at all time points on the basis of analysis-of-covariance (ANCOVA) models
- p. 6 Figure 2 legend, last sentence
for CTX and PINP, the comparisons were calculated with the use of the Wil
Judge's note: Read the legend on the page 6 image. The first sentence attributes all comparisons in the figure to ANCOVA; the next says the CTX and PINP comparisons used the Wilcoxon rank-sum test. The intended meaning (ANCOVA for bone mineral density, Wilcoxon for the markers) is recoverable from the second sentence, so this is loose wording and not a wrong number.
Second check: another model agreed
Report a problem with this finding
Typo References 14 and 15 lack punctuation between title and place of publication: "Agresti A. Categorical data analysis New York: John Wiley, 2002." and "Kalbfleisch JD, Prentice RL. The statistical analysis of failure time data New York: John Wiley, 1980."
- p. 10 Reference 14
Categorical data analysis
- p. 10 Reference 15
tistical analysis of failure time data New
Judge's note: Checked on the page 10 image: no full stop after 'analysis' or after 'data' in the two book entries, whereas every other entry closes its title with a full stop. Present in print, not an extraction artefact. Cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo Reference 16 title printed as "Mechanisms of action of bisphosphonates: similarities and differences and their potential influences of clinical efficacy".
- p. 10 Reference 16
ces and their potential influences of clini
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: that the cited Osteoporosis International article's title ends 'potential influence on clinical efficacy'. The page 10 image shows the printed wording 'potential influences of clinical efficacy' exactly as quoted, which is ungrammatical on its face; the correct title cannot be verified from the files.
Second check: none (judge only)