Skip to main content

Table 3 Adjusted odds ratios for reporting any exploration of subgroup analysis and for reporting subgroup analysis using appropriate methods

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

Condition

Predict any subgroup analysis OR (95 % CI)

Predict subgroup analysis using appropriate methods OR (95 % CI)

Journal of publication

p = 0.60

p = 0.15

 British Medical Journal

1.00

1.00

 Annals

1.28 (0.43, 3.82)

7.12 (0.92, 55.31)

 Journal of the American Medical Association

1.34 (0.53, 3.35)

2.40 (0.63, 8.37)

 Lancet

1.35 (0.60, 3.03)

3.58 (1.18, 10.88)

 New England Journal of Medicine

1.90 (0.80, 4.48)

2.28 (0.73, 7.11)

Year of publication

p = 0.39

p = 0.05

 Year

1.04 (0.95, 1.15)

0.88 (0.76, 1.00)

Biostatistician as coauthor

p = 0.90

p = 0.11

 No biostatistician as coauthor

1.00

1.00

 Biostatistician as coauthor

1.03 (0.63, 1.68)

1.71 (0.89, 3.30)

Medical condition under study

p = 0.10

p = 0.02

 Obstetrics/gynecological

1.00

1.00

 Cardiovascular

2.47 (0.83, 7.35)

2.30 (0.38, 14.02)

 Infectious disease

2.03 (0.61, 6.77)

0.51 (0.08, 3.07)

 Cancer

3.60 (1.09, 11.84)

0.46 (0.08, 2.67)

 Psychiatry/neurology

3.03 (0.87, 10.63)

0.47 (0.07, 3.08)

 Autoimmune, including DM

3.53 (1.03, 12.16)

0.71 (0.10, 4.98)

 Pulmonary/critical care

1.37 (0.36, 5.22)

1.88 (0.22, 15.85)

 Other chronic disease

4.80 (1.41, 16.35)

1.27 (0.18, 8.85)

 Other, uncategorized

1.29 (0.37, 4.45)

3.07 (0.22, 43.87)

First author’s region

p = 0.65

p = 0.47

 Other

1.00

1.00

 North America

1.48 (0.65, 3.40)

1.51 (0.56, 4.05)

 Europe

1.38 (0.63, 2.98)

1.78 (0.71, 4.44)

Funding

p = 0.01

p = 0.007

 No industry funding

1.00

1.00

 Industry funding

1.94 (1.17, 3.21)

0.39 (0.20, 0.77)

Significance of the primary outcome

p = 0.022

p = 0.19

 Not significant

1.00

1.00

 Significant

0.55 (0.33, 0.92)

0.64 (0.33, 1.24)

Sample size

p < 0.001

p = 0.18

 Quintiles

1.98 (1.64, 2.40)

1.21 (0.91, 1.61)

  1. Model included all variables and robust error terms; DM Diabetes Mellitus