Skip to main content

Advertisement

Table 3 Frequency of different judgments for the same supporting explanation related to percent of attrition in RCT groups and comments about statistics

From: Assessments of attrition bias in Cochrane systematic reviews are highly inconsistent and thus hindering trial comparability

Supporting explanation n = total number of Cochrane reviews that had this supporting explanation N = number of analyzed Cochrane reviews Risk of bias judgment
Low, N (%) Unclear, N (%) High, N (%)
Percent of attrition in the RCT groups with higher attrition
Attrition between the groups was under 10%, n = 264, N = 122 101 (82.8) 16 (13.1) 5 (4.1)
Attrition between the groups that was 10–20%, n = 354, N = 143 91 (63.6) 28 (19.6) 24 (16.8)
Attrition between the groups that was 21–30%, n = 215, N = 60 34 (56.7) 5 (8.3) 21 (35)
Attrition between the groups that was above 30%, n = 276, N = 70 18 (25.7) 9 (12.9) 43 (61.4)
Supporting explanations about statistics
ITT analysis used, n = 825, N = 193 140 (72.5) 21 (10.9) 32 (16.6)
ITT analysis was not used, n = 238, N = 35 20 (57.1) 9 (25.7) 6 (17.1)
PP analysis used, n = 81, N = 8 7 (87.5) 1 (12.5) 0 (0)
LOCF analysis used, n = 66, N = 25 13 (52) 3 (12) 9 (36)
  1. Abbreviations: ITT intention-to-treat, LOCF last observation carried forward, PP per protocol, RCT randomized controlled trial,