Skip to main content
. 2013 Jan 29;346:f457. doi: 10.1136/bmj.f457

Table 1.

 Characteristics of trials using surrogate primary outcomes and final patient relevant primary outcomes. Values are numbers (percentages) unless stated otherwise

Characteristics Surrogate outcomes (n=84) Patient relevant outcomes (n=101)
Intervention clinical area:
 Cardiovascular 20 (24) 25 (25)
 Endocrinology 4 (5) 3 (3)
 Gastrology and hepatology 9 (11) 10 (10)
 Infectious disease 18 (21) 21 (21)
 Nephrology and urology 1 (1) 4 (4)
 Neurology 0 (0) 2 (2)
 Obstetrics 5 (6) 4 (4)
 Oncology 5 (6) 2 (2)
 Other 17 (20) 24 (23)
 Pulmonology 5 (6) 6 (6)
Population clinical area:
 Cardiovascular 25 (30) 25 (25)
 Endocrinology 4 (5) 7 (7)
 Gastrology and hepatology 8 (10) 9 (9)
 Infectious disease 13 (15) 15 (14)
 Nephrology and urology 0 (0) 5 (5)
 Neurology 0 (0) 1 (1)
 Obstetrics 7 (8) 7 (7)
 Oncology 5 (6) 2 (2)
 Other 16 (19) 25 (25)
 Pulmonology 6 (7) 5 (5)
Journal:
Annals of Internal Medicine 8 (10) 6 (6)
BMJ 7 (8) 11 (11)
Journal of the American Medical Association 20 (24) 22 (22)
Lancet 21 (25) 19 (19)
New England Journal of Medicine 27 (32) 42 (41)
PLoS Medicine 1 (1) 1 (1)
Publication year:
 2005 40 (48) 49 (49)
 2006 44 (52) 52 (51)
Centre status:
 Single centre 19 (23) 9 (9)*
 Multicentre 65 (77) 92 (91)*
Intervention:
 Drugs 49 (59) 61 (60)
 Medical devices 7 (8) 7 (7)
 Surgical procedures 4 (5) 8 (8)
 Health promotion activities 7 (8) 2 (2)
 Other therapeutic technologies 17 (20) 23 (23)
Sponsor:
 Profit 24 (29) 29 (29)
 Not for profit 49 (59) 56 (55)
 Mixed 11 (12) 16 (16)
Median (interquartile range) sample size 371 (162-787) 741 (300-4731)†
Median (interquartile range) follow-up (days) 255 (133-540) 180 (35-730)

2 test, P=0.01.

†Mann-Whitney U test, P<0.001.