Skip to main content
. 2008 Oct 31;337:a1864. doi: 10.1136/bmj.a1864

Table 3.

 Prevalence of preferences and characteristics of patients in trials. Values are numbers (percentages) unless stated otherwise

Authors Sample size Female Mean (SD) age (years) Indifferent With preference With preference for “experimental” treatment With preference for control: usual care or alternative treatment Missing preference data
Adamson et al, 2005w3 122* 52 (42) 35.7 (35.7) 16 (13) 102 (84) 90 (88) 12 (12) 4 (3.3)
Carr et al, 2005w4 237 133 (56) 42.2 (10.9) 152 (64) 70 (30) 10 (14) 60 (86)‡ 15 (6)
Hay et al, 2003w5; Thomas et al, 2004w6 207 110 (53) 57.4 (13.4) 81 (39.1) 125 (60.4) 83 (66) 42 (34)‡ 1 (0.5)
Johnson et al, 2007w7 234 140 (60) 47.9 (11.2) 100 (42.7) 134 (57.3) 114 (85) 20 (15) 0 (0)
Kitchener et al, 2006w8 291 291 (100) 50.3 (10.5) 141 (48.5) 144 (49.5) 134 (93) 10 (7) 6 (2.1)
Klaber Moffett et al, 1999w1 187 106 (57) 41.9 (8.9) 69 (36.9) 118 (63.1) 118 (100) (0) 0 (0)
Klaber Moffett et al, 2005w2 268 168 (63) 48.3 (14.6) 142 (53) 126 (47) 43 (34) 83 (66)‡ 0 (0)
Klaber Moffett et al, 2006w9 315 188 (60) 45.0 (15.1) 261 (82.9) 51 (16.2) 37 (73) 14 (27) 3 (1.0)
McLean et al, publication pendingw10 151 90 (60) 54.0 (14.3) 66 (43.7) 82 (54.3)§ 34 (45) 41 (55)‡ 3 (2.0)
Salisbury et al, 2005w11 556 327 (59) 48.0 (19.1) 63 (11.3) 474 (85.3) 328 (69) 146 (31) 19 (3.4)
Salter et al, 2006w12 24 18 (75) 47.7 (16.5) 9 (37.5) 15 (62.5) 15 (100) (0) 0 (0)

*Two missing people in dataset.

†Trials included in patient level data meta-analysis.

‡Physiotherapy was control treatment in these trials.

§Seven participants removed from dataset for meta-analysis as they had preference for treatment not available in trial.