Skip to main content

Table 3 Sensitivity analyses

From: Impact of peer review on discussion of study limitations and strength of claims in randomized trial reports: a before and after study

  Manuscript Publication Crude difference (95%CI) Adjusted difference (95%CI)
Number of limitation-acknowledging sentences (mean, SD) 2.61 (3.81) 3.96 (4.51) NA 1.74 (0.21–3.28)
Unweighted hedges (%) 2.06 (0.76) 2.13 (0.74) 0.06 (0.03–0.10) − 0.01 (− 0.18–0.15)
Unweighted hedges, limitation-acknowledging sentences excluded (%) 2.01 (0.77) 2.05 (0.76) 0.04 (0.01–0.08) 0.06 (− 0.03–0.16)
Weighted hedges (%) 7.07 (2.91) 7.30 (2.82) 0.20 (0.07–0.34) 0.03 (− 0.62–0.26)
Weighted hedges, limitation-acknowledging sentences excluded (%) 6.92 (2.95) 7.05 (2.89) 0.13 (0.01–0.26) 0.08 (−0.58–0.74)
  1. Results obtained via the same calculations as in Table 2, but excluding BMJ Open and BMC Medicine whose editorial team sizes were extremely large compared to the other 26 journals, namely, 84 and 182, respectively. After omitting these two journals, the median team size was 8 (IQR interquartile range; IQR 6–14). N = 357 because we were unable to find the impact factor of BMC Dermatology (contributing six manuscript-publication pairs); hedges were counted (and weighted), divided by the total number of words in the discussion section and multiplied by 100. NA not available, since that model did not converge and no coefficients were calculated
  2. SD standard deviation
  3. Crude difference estimated using a mixed regression model without covariables (N = 363);
  4. Adjusted for journal impact factor (continuous), editorial team size (continuous) and composition of authors in terms of English proficiency (three dummy variables)