Prevention of Recurrent Preterm Delivery by 17 Alpha-Hydroxyprogesterone Caproate
Obstetrics & Gynecology, 2003 · paper · Clinical trials
A double-blind trial of weekly 17P hormone injections versus placebo in 463 women with a prior spontaneous preterm delivery. Delivery before 37 weeks occurred in 36.3% with 17P versus 54.9% with placebo, and infants had fewer certain complications.
1 finding
Substance Table 3, intraventricular hemorrhage of any grade: '0.25 (0.8–0.82)'. The lower bound 0.8 lies above the point estimate 0.25, so the interval is impossible as printed (from the counts it is about 0.08).
- p. 6 Table 3, Intraventricular hemorrhage, Any grade row
0.25 (0.8–0.82)
- p. 6 Discussion, neonatal outcomes paragraph
There were significant reductions in the rates of necrotizing enterocolitis, any intraventricular hemorrhage
Judge's note: Page 6 image and crop both show '0.25 (0.8–0.82)'. The lower limit is above the estimate and almost equal to the upper limit. From 4/305 vs 8/153: RR = 0.251, SE(log RR) = 0.604, 95% CI = 0.077 to 0.82, so 0.08 was evidently intended. The Discussion's statement that the reduction in 'any intraventricular hemorrhage' was significant fits 0.08–0.82.
Second check: another model agreed, code check
Report a problem with this finding
9 minor (typos and the like; most checked by the judge only)
Wording Table 2: the Black/nonblack percentages '64 (57.5)', '47 (42.5)', '47 (55.9)', '37 (44.1)' cannot be reproduced from the counts with any stated denominator, and they are not the within-race rates that the relative risks compare.
- p. 5 Table 2, Black women / Nonblack women rows
64 (57.5)
- p. 5 Table 2, Nonblack women row
47 (42.5)
- p. 5 Table 2, placebo column, Black women row
47 (55.9) 0.68 (0.51–0.90)
- p. 5 Table 2, Black women / Nonblack women rows
64 (57.5) 47 (55.9)
Only the authors can say what is right here.
Judge's note: Checked on the page 5 image and crop. In each arm the counts add to the preterm totals (111 and 84) and the percentages add to 100, so they look like shares of preterm deliveries. But 64/111 = 57.7 and 47/111 = 42.3, not 57.5 and 42.5; 47/84 = 56.0 and 37/84 = 44.0, not 55.9 and 44.1. Within-race rates using the Table 1 race counts (183 and 90 Black women) give 35.0% and 52.2%. So the '%' in these rows means something different from the other rows under 'no. (%)', and the printed values match neither reading. Only the authors can say which denominator was intended.
Second check: another model agreed
Report a problem with this finding
Wording Table 2, corticosteroids for fetal lung maturity: '52 (17.8)', '30 (19.7)', '0.91 (0.60–1.35)'. The crude ratio from these counts is 0.90.
- p. 5 Table 2, Corticosteroids for fetal lung maturity row
0.91 (0.60–1.35)
Only the authors can say what is right here.
Judge's note: Page 5 crop. The footnote gives denominators 292 and 152. The crude RR is 0.902, and a crude log-scale interval from the same counts is 0.60–1.35, matching the printed CI exactly. The methods do not say how the RRs were estimated (only chi-square/Fisher tests are named). Another row is also slightly off the crude value (hospital visit 49/306 vs 21/152 = 1.159, printed 1.15), so a different estimator cannot be ruled out. Query for the authors.
Second check: another model agreed
Report a problem with this finding
Typo Table 2, chorioamnionitis: '11 (3.6)', '5 (3.3)', '1.09 (0.39–3.09)'. The crude ratio is exactly 1.10.
- p. 5 Table 2, Chorioamnionitis row
1.09 (0.39–3.09)
Judge's note: Page 5 crop. (11/306)/(5/153) = 1.10 exactly. The printed 1.09 matches the ratio of the rounded percentages, 3.6/3.3. The printed upper limit of 3.09 also departs from a crude Wald interval (about 3.11). The estimator is not stated in the methods, so this is a minor consistency point only.
Second check: none (judge only)
Report a problem with this finding
Typo The funding statement lists grant 'HD34116' twice ('HD34208, HD34116, HD21410' and 'HD40544, HD34116, HD40560').
- p. 6 Funding statement
HD34208, HD34116, HD21410
- p. 6 Funding statement
HD40544, HD34116, HD40560
- p. 6 Funding statement at end of Discussion
HD34116, HD40560, HD40512
- p. 6 Funding footnote
HD34116, HD21410
Only the authors can say what is right here.
Judge's note: Page 6 image shows the grant list with HD34116 in both line 2 and line 4. Only the authors can say whether this is a duplicate or a mistyped second grant number.
Second check: none (judge only)
Report a problem with this finding
Typo Reference 4 title reads 'a controlled clinical study of pregestational therapy', where 'progestational' appears intended.
- p. 7 Reference 4
clinical study of pregestational therapy
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: the original title of LeVine 1964 uses 'progestational'. Page 7 image confirms 'pregestational'. The introduction cites refs 4-7 for trials of 'progestational compounds', which supports the misspelling reading.
Second check: none (judge only)
Report a problem with this finding
Typo Reference 28 gives the volume as '2003;1884:419-24'.
- p. 7 Reference 28
2003;1884:419-24
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: Am J Obstet Gynecol 2003 was volume 188. Page 7 image confirms '1884'. Within the paper, ref 12 cites Am J Obstet Gynecol 1998 as volume 178, which fits volume 188 in 2003. A four-digit volume is malformed.
Second check: none (judge only)
Report a problem with this finding
Typo Reference 4 page range is printed '1964;72:3036' with no hyphen (presumably 30-6).
- p. 7 Reference 4
1964;72:3036
Judge's note: Page 7 crop shows '72:3036' with no dash. Every other reference prints its pages as a hyphenated range.
Second check: none (judge only)
Report a problem with this finding
Typo Reference 27 page range is printed '1994;170: 5462.' with no hyphen (presumably 54-62).
- p. 7 Reference 27
5462.
- p. 7 Reference 27 (Hobel et al.)
1994;170: 5462.
Judge's note: Page 7 crop shows '170:' followed by '5462.' on the next line, with no dash. This is the same defect as ref 4 (C010) but in a different reference.
Second check: none (judge only)
Report a problem with this finding
Typo The same journal is abbreviated 'Br J Obstet Gynecol' in ref 10 and 'Br J Obstet Gynaecol' in ref 11.
- p. 7 Reference 10
Br J Obstet Gynecol
- p. 7 Reference 11
Br J Obstet Gynaecol
- p. 7 Reference 10 (Goldstein et al.)
Br J Obstet Gynecol 1989;96: 265-74.
- p. 7 Reference 11 (Keirse)
Br J Obstet Gynaecol 1990;97:149-
Judge's note: Page 7 crop shows both spellings in consecutive references. The ref 11 form matches the journal's British spelling, so ref 10 should be made consistent with it.
Second check: none (judge only)