Table 4

Quality aspects related to statistical analysis

Epidemiologist/biostatistician
Yes % (n=178)No % (n=289)Crude OR (95% CI)Adjusted OR4-150 (95% CI)
Mentioning statistical procedures
  Yes98.395.92.5 (0.8, 8.4)1.9 (0.5, 7.1)
Reporting p value and/or CI
  Yes98.997.91.9 (0.4, 9.3)0.9 (0.2, 4.8)
Analysis according to intention to treat
  Yes57.332.22.8 (1.9, 4.2)2.0 (1.3, 3.0)
Use of multivariate analysis if needed
  Yes/not applicable93.590.41.6 (0.8, 3.2)1.3 (0.6, 2.9)
Estimation of statistical power if results are non-significant
  Yes/not applicable87.684.81.3 (0.7, 2.2)1.1 (0.6, 2.0)
  • 4-150 Adjusted for number of participating centres, funding, and study sample size.