Table 2:

Expertise associated with independent comments and journal editorials

Relevant research/expertiseNo. (%) of comments*p value
Independent comments
n = 104		Editorial comments
n = 21
Relevant academic research†
Papers published on disease/outcome in source research article, median (95% CI)2 (1–5)27 (6–55)< 0.001
Papers published on treatment/exposure in source research article, median (95% CI)0 (0–1)2 (0–21)0.02
No published papers on disease/outcome or treatment/exposure in source research article29 (28)1 (5)0.02
Research conducted of similar design to source research article48 (46)19 (90)< 0.001
Expertise‡
Academic expertise§55 (53)18 (86)0.007
Clinical expertise58 (56)10 (48)0.6
Neither academic nor clinical expertise26 (25)0 (0)0.007

  • Note: CI = confidence interval.

  • ↵* Unless stated otherwise.

  • ↵† In the 5 years before publication of the source article.

  • ↵‡ Thirty-nine independent commenters and 7 editorialists had both academic and clinical expertise.

  • ↵§ Published ≥ 5 papers relevant to the source research article during the 5 years before its publication.