Skip to main content

Table 5 Summaries of posterior probabilities of the absolute risk difference of predischarge mortality

From: Advantages of Bayesian monitoring methods in deciding whether and when to stop a clinical trial: an example of a neonatal cooling trial

  RD posterior mean (95 % credible interval) Futility monitoring
Pr(RD > 0.01)a
Safety monitoring
Pr(RD < −0.05)b
Neutral Enthusiastic Neutral Enthusiastic Neutral Enthusiastic
32.0 °C for 72 h −0.02 (−0.08, 0.03) −0.02 (−0.08, 0.04) 11 % 15 % 19 % 16 %
33.5 °C for 120 h −0.03 (−0.09, 0.02) −0.03 (−0.09, 0.03) 8 % 9 % 28 % 25 %
32.0 °C for 120 h −0.06 (−0.15, 0.03) −0.06 (−0.15, 0.03) 5 % 7 % 61 % 54 %
  1. aRD > 0.01 indicates 1 % or more reduced mortality
  2. bRD < −0.05 indicates a 5 % or more absolute increase in mortality
  3. The three experimental groups are compared standard cooling (33.5 °C for 72 h) under a neutral and enthusiastic prior. Positive values of RD favor the experimental groups
  4. Pr probability, RD risk difference