Skip to main content

Table 1 Reporting of aspects of trial design, conduct and analysis: Stratified by diseases and types of Intervention

From: High prevalence of potential biases threatens the interpretation of trials in patients with chronic disease

Interventions All Drug   COPD Heart failure Diabetes Stroke
    Non-drug Drug Non-drug Drug Non-drug Drug Non-drug Drug Non-drug
Number of trials 161 70 91 20 29 12 19 26 22 12 21
Reporting of between- group comparisons, % (n) 90% (145) 90% (63) 89% (81) 100% (20) 93% (27) 67% (8) 95% (18) 92% (24) 82% (18) 92% (11) 86% (18)
Primary outcome, % (n) 33% (53) 50% (35) 20% (18) 70% (14) 7% (2) 17% (2) 21% (4) 42% (11) 36% (8) 67% (8) 19% (4)
Primary outcome is only one measure, % (n) 21% (34) 31% (22) 13% (12) 35% (7) 0% (0) 8% (1) 16% (3) 27% (7) 32% (7) 58% (7) 10% (2)
No statistical comparison of baseline characteristics, % (n) 68% (110) 79% (55) 60% (55) 85% (17) 59% (17) 83% (10) 68% (13) 73% (19) 73% (16) 75% (9) 43% (9)
Reporting of handling of missing data, % (n) 17% (28) 21% (15) 13% (12) 20% (4) 10% (3) 8% (1) 5% (1) 27% (7) 9% (2) 25% (3) 33% (7)
Intention-to-treat analysis reported, % (n) 42% (67) 53% (37) 33% (30) 70% (14)% 24% (7) 33% (4) 26% (5) 46% (12) 27% (6) 58% (7) 57% (12)
Reporting of P values and 95% CI % (n) 24% (38) 31% (22) 20% (18) 45% (9) 24% (7) 8% (1) 16% (3) 19% (5) 32% (7) 58% (7) 5% (1)