Antenatal Betamethasone for Women at Risk for Late Preterm Delivery
New England Journal of Medicine, 2016 · paper · Clinical trials
A randomized trial of betamethasone versus placebo in women at high risk of delivery at 34–36 weeks. Serious neonatal respiratory complications occurred in 11.6% versus 14.4% of infants, while newborn low blood sugar was more common with betamethasone.
1 finding
Substance Table 2 (p.7) gives severe respiratory complications in the betamethasone group as "113 (8.0)", but 113/1427 = 7.9%, which is the figure the Results text gives ("7.9% vs. 12.1%").
- p. 7 Table 2, Severe respiratory complication row
113 (8.0)
- p. 5 Results, Neonatal Outcomes, second paragraph
(7.9% vs. 12.1%; relative risk, 0.66;
- p. 5 Results, Neonatal Outcomes paragraph
(7.9% vs. 12.1%; relative risk, 0.66
Judge's note: Checked Table 2 on the p.7 image and the text on p.5. The Table 2 column N is 1427, and the only footnote that changes a denominator (§) applies to the resuscitation row. 7.92% cannot round to 8.0. The placebo value 169/1400 = 12.1% and the RR of 0.66 are consistent with 113 events.
Second check: another model agreed
Report a problem with this finding
4 minor (typos and the like; most checked by the judge only)
Wording Methods (p.2) make a woman ineligible if she "was expected to deliver in less than 12 hours for any reason", but Results (p.4) give the most common exclusion as "the expectation that delivery would occur within 24 hours" (6203 women), and Figure 1 (p.5) says "6203 Were expecting delivery in <24 hr".
- p. 2 Methods, Screening and Recruitment
less than 12 hours for any reason
- p. 4 Results, Characteristics of the Participants
ery would occur within 24 hours, which was
- p. 5 Figure 1, exclusion box
6203 Were expecting delivery in <24 hr
- p. 2 Screening and Recruitment
expected to deliver in less than 12 hours for any reason
Only the authors can say what is right here.
Judge's note: Checked the page images for p.2, p.4 and p.5. The Methods give only a 12-hour exclusion threshold. The only 24-hour rule is the window for indicated deliveries (planned induction or cesarean "between 24 hours and 7 days"), and the discussion repeats the 12-hour oxytocin delay. The '<24 hr' exclusion category may combine the 12-hour rule with the indicated-delivery window, but the paper never says so. The authors should clarify which criterion the 6203 exclusions reflect.
Second check: another model agreed
Report a problem with this finding
Wording Table 3 (p.8) gives hyperbilirubinemia as 167 (11.7) vs 140 (10.0), RR "1.17 (0.95–1.40)". The upper limit cannot be reproduced from the counts with the log-RR method, which gives about 0.95–1.45.
- p. 8 Table 3, Hyperbilirubinemia row
1.17 (0.95–1.40)
Only the authors can say what is right here.
Judge's note: I recomputed this myself. RR = 1.170 and SE(ln RR) = sqrt(1/167-1/1427+1/140-1/1400) = 0.1082, which gives exp(0.1573±0.2121) = 0.947–1.447. The same method reproduces the hypothermia row (132/1427 vs 112/1400 gives 0.91–1.47, as printed). The printed interval is not centred on 1.17 on the log scale. No footnote marks a different method for this row (only bronchopulmonary dysplasia is footnoted as exact). The printed P = 0.15 matches z = 1.45. The authors should confirm the upper limit.
Second check: another model agreed
Report a problem with this finding
Typo Study Outcomes (p.3): "at least 0.30 for at least at least 24 continuous hours": the words 'at least' are duplicated.
- p. 3 Study Outcomes, neonatal secondary outcomes
at least 0.30 for at least at least 24 continuous
- p. 3 Study Outcomes, secondary outcomes
for at least at least 24 continuous
- p. 3
at least 0.30 for at least at least 24 continuous hours
Judge's note: The page-image crop shows 'at least at least' on a single line, so this is not an extraction artefact.
Second check: none (judge only)
Report a problem with this finding
Typo Figure 1 prints "1,535 Declined to participate" with a thousands comma, while the other four-digit counts in the same box (6074, 6203, 3269, 1056, 1142) have none.
- p. 5 Figure 1
1,535 Declined to participate
Judge's note: Checked Figure 1 on the p.5 image. Only the five-digit counts (24,133, 21,302, 19,587) and 1,535 carry commas. The exclusion counts add up correctly.
Second check: none (judge only)