Skip to main content

Table 3 Type I error and power of GLMM and GEE at aggregated level for testing treatment-by-period interactions with AR1 correlation structure (\(N_h\) denotes the number of clusters allocated to switch pattern h)

From: Comparison of analysis methods and design choices for treatment-by-period interaction in unidirectional switch designs: a simulation study

Design

\(\sigma _c^2\)

Cluster allocation

Type I error

Power

\(N_0\)

\(N_1\)

\(N_2\)

\(N_3\)

\(N_4\)

\(N_5\)

GLMM Agg

GEE Agg

GLMM Agg

GEE Agg

PGD

0.2

24

0

0

0

0

24

0.077

0.046

0.595

0.406

PGD

0.5

24

0

0

0

0

24

0.078

0.050

0.323

0.181

PGD

0.8

24

0

0

0

0

24

0.076

0.045

0.228

0.128

DSDu

0.2

16

0

0

16

0

16

0.088

0.044

0.510

0.288

DSDu

0.5

16

0

0

16

0

16

0.084

0.037

0.299

0.140

DSDu

0.8

16

0

0

16

0

16

0.076

0.038

0.222

0.101

DSDo

0.2

14

0

0

24

0

10

0.090

0.035

0.431

0.249

DSDo

0.5

14

0

0

24

0

10

0.086

0.040

0.246

0.114

DSDo

0.8

14

0

0

24

0

10

0.007

0.043

0.187

0.086

USDu

0.2

8

8

8

8

8

8

0.073

0.039

0.407

0.195

USDu

0.5

8

8

8

8

8

8

0.083

0.039

0.244

0.101

USDu

0.8

8

8

8

8

8

8

0.078

0.036

0.180

0.074

USDo(1)

0.2

6

9

9

9

9

6

0.090

0.031

0.372

0.157

USDo(1)

0.5

6

9

9

9

9

6

0.081

0.029

0.228

0.085

USDo(1)

0.8

6

9

9

9

9

6

0.082

0.029

0.185

0.068

USDo(2)

0.2

5

10

9

9

10

5

0.085

0.033

0.341

0.144

USDo(2)

0.5

5

10

9

9

10

5

0.083

0.026

0.214

0.068

USDo(2)

0.8

5

10

9

9

10

5

0.071

0.024

0.166

0.061