Skip to main content

Table 5 Performance for scenario IV 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

0.0320

0.0037

0.919

0.0020

0.0033

0.948

0.0429

0.0050

0.876

2

0.0350

0.0036

0.911

0.0015

0.0031

0.944

0.0470

0.0052

0.840

3

0.0361

0.0037

0.910

0.0012

0.0031

0.943

0.0497

0.0054

0.825

4

0.0361

0.0037

0.909

0.0009

0.0031

0.944

0.0512

0.0055

0.810

Mixed-Poisson

1

0.0344

0.0038

0.921

0.0121

0.0033

0.941

0.0604

0.0066

0.799

2

0.0377

0.0038

0.919

0.0128

0.0031

0.945

0.0639

0.0068

0.763

3

0.0388

0.0038

0.911

0.0130

0.0030

0.945

0.0647

0.0068

0.754

4

0.0388

0.0038

0.910

0.0127

0.0030

0.943

0.0643

0.0067

0.745

Weibull (constant)

1

0.0475

0.0051

0.866

0.0429

0.0048

0.882

0.0028

0.0028

0.948

2

0.0418

0.0044

0.886

0.0367

0.0042

0.907

0.0024

0.0025

0.942

3

0.0391

0.0042

0.890

0.0339

0.0039

0.912

0.0026

0.0023

0.945

4

0.0381

0.0041

0.899

0.0329

0.0039

0.913

0.0022

0.0023

0.947

Weibull (change)

1

0.1282

0.0174

0.062

0.0264

0.0026

0.908

0.0001

0.0018

0.950

2

0.1264

0.0169

0.065

0.0233

0.0024

0.922

-0.0002

0.0017

0.946

3

0.1249

0.0165

0.068

0.0222

0.0023

0.930

-0.0004

0.0017

0.949

4

0.1246

0.0165

0.069

0.0219

0.0023

0.930

-0.0004

0.0017

0.949

  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