Skip to main content

Table 2 Simulation results for powers and sample sizes

From: Power and sample sizes estimation in clinical trials with treatment switching in intention-to-treat analysis: a simulation study

p

 

no switching

switching

s.dist = “beta”

switching

s.dist = “gamma”

switching

s.dist = “unif”

switching

s.dist = “indepExp”

m2/m1=1.5

  0.2

n

130

166

165

168

169

 

E1

104.0

130.8

130.1

132.5

132.8

 

E2

89.2

113.8

113.2

115.1

115.8

 

Powera

0.80

0.70

0.70

0.70

0.69

 

ns/nno − s

1.28

1.27

1.29

1.30

  0.4

n

130

215

219

217

235

 

E1

104.0

167.0

169.9

168.5

180.9

 

E2

89.2

147.6

150.2

148.9

161.1

 

Powera

0.80

0.57

0.58

0.57

0.54

 

ns/nno − s

1.65

1.68

1.67

1.81

  0.6

n

130

294

301

297

344

 

E1

104.0

225.0

230.0

227.1

259.4

 

E2

89.2

201.7

206.4

203.9

236.0

 

Powera

0.80

0.47

0.45

0.48

0.40

 

ns/nno − s

2.26

2.32

2.28

2.65

  0.8

n

130

418

437

416

534

 

E1

104.0

314.7

328.5

313.3

394.8

 

E2

89.2

286.4

299.6

285.2

366.4

 

Powera

0.80

0.35

0.33

0.35

0.29

 

ns/nno − s

3.22

3.36

3.20

4.11

  1.0

n

130

627

782

657

937

 

E1

104.0

464.9

578.1

487.2

678.4

 

E2

89.2

430.3

536.3

450.7

642.2

 

Powera

0.80

0.24

0.23

0.24

0.18

 

ns/nno − s

4.82

6.02

5.05

7.21

m2/m1=2

  0.2

n

48

64

63

62

64

 

E1

38.4

49.9

48.9

48.2

49.4

 

E2

28.4

38.1

37.4

36.9

38.0

 

Powera

0.80

0.69

0.69

0.69

0.68

 

ns/nno − s

1.33

1.31

1.29

1.33

  0.4

n

48

87

84

84

88

 

E1

38.4

65.8

63.5

63.7

65.7

 

E2

28.4

51.8

49.9

49.9

52.2

 

Powera

0.80

0.56

0.57

0.56

0.53

 

ns/nno − s

1.81

1.75

1.75

1.83

  0.6

n

48

114

124

116

127

 

E1

38.4

83.8

90.7

85.1

92.1

 

E2

28.4

67.5

73.5

69.0

76.0

 

Powera

0.80

0.44

0.42

0.45

0.40

 

ns/nno − s

2.38

2.58

2.42

2.65

  0.8

n

48

167

182

167

202

 

E1

38.4

119.0

129.0

118.9

140.0

 

E2

28.4

99.1

108.1

99.0

119.8

 

Powera

0.80

0.32

0.31

0.31

0.27

 

ns/nno − s

3.48

3.79

3.48

4.21

  1.0

n

48

265

296

269

355

 

E1

38.4

182.7

203.3

185.7

236.2

 

E2

28.4

157.4

175.7

159.4

210.7

 

Powera

0.80

0.23

0.22

0.23

0.19

 

ns/nno − s

5.52

6.17

5.60

7.40

  1. aPower at the sample size nno − s. E1 and E2: the expected numbers of events in the control group and the experimental group. ns and nno − s denote the required sample size with and without treatment switching, respectively at the power of 0.8 and the significance level of 0.05