Table 1

Characteristics of randomized controlled trials with multiple arms by trial design
All trials, n = 298 Trial design
Parallel group, n = 221 Parallel group with factorial, n = 37 Crossovera, n = 40
No. of study groups
 3 172 (57.7) 152 (68.8) 0 (0.0) 20 (50.0)
 4 84 (28.2) 39 (17.7) 34 (91.9) 11 (27.5)
 > 4 42 (14.1) 30 (13.6) 3 (8.1) 9 (22.5)
Common specialties
 Anesthesia 54 (18.1) 47 (21.3) 3 (8.1) 4 (10.0
 Endocrinology 32 (10.7) 20 (9.1) 5 (13.5) 7 (17.5)
 Cardiology 31 (10.4) 25 (11.3) 6 (16.2) 0 (0.0)
 Infectious disease 22 (7.4) 20 (9.1) 1 (2.7) 1 (2.5)
 Rheumatology 19 (6.4) 18 (8.1) 1 (2.7) 0 (0.0)
Intervention
 Drug 192 (64.4) 148 (68.0) 22 (59.5) 22 (55.0)
 Surgery or procedure 17 (5.7) 12 (5.4) 1 (2.7) 4 (10.0)
 Counseling or lifestyle interventions 47 (15.8) 33 (14.9) 10 (27.0) 4 (10.0)
 Equipment or device 26 (8.7) 19 (8.6) 2 (5.4) 5 (12.5)
 Others 16 (6.7) 5 (1.7) 2 (0.7) 9 (3.0)
Study centers
 Single 80 (26.9) 54 (24.4) 8 (21.6) 18 (5.0)
 Multiple 141 (47.3) 110 (49.8) 24 (64.9) 7 (13.5)
 Not reported 77 (25.8) 57 (25.8) 5 (13.5) 15 (37.5)
No. of randomized patients per:
 Trial, median (10th to 90th percentile) 136 (25 to 800) 148 (45 to 650) 468 (120 to 1653) 21 (12 to 61)
 Arm, median (10th to 90th percentile) 39 (7 to 228) 43 (12 to 204) 117 (30 to 413) 5 (3 to 16)
Cluster trial 13 (4.4) 8 (3.6) 3 (8.1) 2 (5.0)
Trial type
 To show superiority 260 (87.3) 195 (88.8) 33 (89.2) 32 (80.0)
 To show non-inferiority or equivalence 12 (4.0) 8 (3.6) 4 (10.8) 0 (0.0)
 Pharmacokinetic/pharmacodynamic objective 26 (8.7) 18 (8.1) 0 (0.0) 8 (20.0)
Randomization design
 Balanced 263 (88.3) 190 (86.0) 33 (89.2) 40 (100.0)
 Unbalanced 30 (10.1) 28 (12.7) 2 (5.4) 0 (0.0)
 Unclear 5 (1.7) 3 (1.3) 2 (5.4) 0 (0.0)
Funding source
 Solely or partially industry 101 (33.9) 77 (34.8) 11 (29.7) 13 (32.5)
 Public 118 (39.6) 87 (39.4) 14 (37.8) 17 (42.5)
 None 8 (2.7) 6 (2.7) 1 (2.7) 1 (2.5)
 Unknown 71 (23.8) 51 (23.1) 11 (29.8) 9 (22.5.0)
Trial registration reported 144 (48.3) 109 (49.3) 27 (73.0) 8 (20.0)

a There were 4 of the 40 crossover trials with a factorial design.

Baron et al.

Baron et al. BMC Medicine 2013 11:84   doi:10.1186/1741-7015-11-84

Open Data