Skip to main content

Table 4 Performance for scenario III with true intervention effect of \({\beta }_{t}=-0.264\)

From: Comparison of statistical models for estimating intervention effects based on time-to-recurrent-event in stepped wedge cluster randomized trial using open cohort design

Event generation model

    

AG model

PWP-TT model

PWP-GT model

F

Bias

MSE

CP

Bias

MSE

CP

Bias

MSE

CP

Poisson

1

0.0307

0.0037

0.931

0.0015

0.0034

0.947

0.0408

0.0050

0.895

2

0.0338

0.0038

0.936

0.0013

0.0034

0.947

0.0435

0.0051

0.878

3

0.0358

0.0039

0.928

0.0013

0.0034

0.949

0.0447

0.0051

0.863

4

0.0369

0.0041

0.929

0.0010

0.0036

0.952

0.0445

0.0052

0.873

Mixed-Poisson

1

0.0343

0.0041

0.919

0.0126

0.0036

0.945

0.0601

0.0070

0.804

2

0.0366

0.0041

0.907

0.0125

0.0036

0.940

0.0612

0.0069

0.785

3

0.0390

0.0044

0.911

0.0135

0.0037

0.940

0.0615

0.0070

0.790

4

0.0405

0.0045

0.909

0.0139

0.0037

0.938

0.0624

0.0070

0.803

Weibull (constant)

1

0.0453

0.0052

0.874

0.0406

0.0049

0.894

0.0028

0.0030

0.953

2

0.0377

0.0046

0.916

0.0326

0.0044

0.918

0.0011

0.0029

0.959

3

0.0320

0.0042

0.932

0.0264

0.0041

0.934

0.0016

0.0030

0.940

4

0.0280

0.0042

0.938

0.0219

0.0042

0.935

0.0022

0.0030

0.937

Weibull (change)

1

0.1227

0.0161

0.119

0.0235

0.0027

0.929

-0.0002

0.0020

0.951

2

0.1206

0.0156

0.130

0.0173

0.0024

0.939

-0.0008

0.0020

0.944

3

0.1203

0.0155

0.139

0.0148

0.0023

0.949

0.0001

0.0019

0.950

4

0.1202

0.0155

0.133

0.0129

0.0024

0.953

0.0006

0.0020

0.948

  1. AG Andersen-Gill, PWP-TT Prentice-Williams-Peterson Total-Time, PWP-GT Prentice-Williams-Peterson Gap-Time, MSE Mean square error, CP Coverage probability