Skip to main content

Table 3 The performance of BPP in comparison with CP when the null hypothesis is false

From: Seamless phase 2/3 design for trials with multiple co-primary endpoints using Bayesian predictive power

Items

ρ

Mean#

n1 = 50

n1 = 100

n1 = 150

n1 = 200

Stop percentage difference* (%)

0

-8.07

1.58

2.53

3.20

0.3

-6.29

1.27

2.22

2.89

0.6

-4.92

0.96

1.83

2.24

total

-6.43

1.27

2.19

2.78

Sample size difference*+

0

143.97

23.85

31.86

39.74

0.3

155.63

63.61

64.55

61.12

0.6

184.22

102.65

93.04

83.75

total

161.28

63.37

63.15

61.54

Correct dose selection (%)

0

90.94

96.70

98.71

99.54

0.3

89.19

95.37

98.03

99.26

0.6

87.48

94.42

97.20

98.86

total

89.20

95.50

97.98

99.22

  1. *represents the difference of BPP and CP, Value = BPP – CP; + : Sample size difference represents the difference of sample size used for each group (eg. selected dose group) during the whole phase 2/3 trial. #: represents the average value of the nine different seroresponse scenarios
Back to article page

BMC Medical Research Methodology

ISSN: 1471-2288

Contact us