Table 3

Association between characteristics of articles and statistically significant outcome or conclusions that favor the test drug: Multivariate logistic regression (full sample, n = 137)

Results Favor Test Drug

Conclusions Favor Test Drug


Characteristic

Category

Favorable n/Total n (%)

OR

(95% CI)

P

Favorable n/Total n (%)

OR

(95% CI)

P


Trial registration before publication

No

44/83 (53)

1.00

55/83 (66)

1.00


Yes

36/54 (67)

1.29 (0.54-3.08)

0.566

43/54 (80)

1.56 (0.60-4.05)

0.363



Blinding

Stringent

18/36 (50)

1.00

25/36 (69)

1.00


Not stringent

62/101 (61)

2.33 (0.95-5.76)

0.066

73/101 (72)

1.51 (0.59-3.91)

0.394



Sample Size

Natural log

-

2.28 (1.49-3.48)

<0.001

-

1.77 (1.18-2.66)

0.006



Comparison group

Active comparator

35/65 (54)

1.00

45/65 (69)

1.00


Placebo

45/72 (63)

2.06 (0.87-4.90)

0.101

53/72 (74)

1.47 (0.59-3.63)

0.410



Primary efficacy outcome

Survival

17/30 (57)

1.00

21/30 (70)

1.00


Surrogate

63/107 (59)

3.42 (1.15-10.14)

0.027

77/107 (72)

3.04 (0.95-9.68)

0.061



Sponsored

No

14/28 (50)

1.00

15/28 (54)

1.00


Yes

66/109 (61)

1.01 (0.37-2.67)

0.999

83/109 (76)

2.01 (0.76-5.31)

0.157


Rasmussen et al. Trials 2009 10:116   doi:10.1186/1745-6215-10-116

Open Data