Skip to main content

Advertisement

Table 1 Articles and randomized controlled trials (RCTs) included in the final sample

From: No improvement in the reporting of clinical trial subgroup effects in high-impact general medical journals

Characteristic Articles represented N = 416 RCTs included N = 437
Journal of publication   
Annals 36 (9) 37 (8)
British Medical Journal 47 (11) 48 (11)
Journal of the American Medical Association 72 (17) 76 (17)
Lancet 115 (28) 123 (28)
New England Journal of Medicine 146 (35) 153 (35)
Year of publication   
 2007 113 (27) 119 (27)
 2010 140 (34) 144 (33)
 2013–2014 163 (39) 174 (40)
Biostatistician as coauthor 239 (57) 254 (58)
Medical condition under study --  
 Cardiovascular   101 (23)
 Infectious disease   82 (19)
 Cancer   59 (14)
 Psychiatry/neurology   40 (9)
 Autoimmune, including diabetes   37 (8)
 Pulmonary/critical care   29 (7)
 Obstetrics/gynecological   21 (5)
 Other chronic disease   41 (9)
 Other, uncategorized   27 (6)
First author’s region --  
 North America   185 (42)
 Europe   188 (43)
 Other   64 (15)
Funding --  
 Industry funding   188 (43)
 No industry funding   249 (57)
Significance of the primary outcomea --  
 Not significant   153 (36)
 Significant   277 (64)
Study design --  
 Parallel group   425 (97)
 Crossover   12 (3)
Analysis reported --  
 Subgroup analysis with appropriate methods   185 (42)
 Subgroup analysis without appropriate methods   85 (19)
 No subgroup analysis   167 (38)
Sample size -- 506 (7–170,432)
  1. n (%) or median (range)
  2. an = 7 trials were excluded for not reporting a statistical test for the primary outcome