Skip to main content

Table 2 Results from simulations based on MIST2 dataset (OR = 1)

From: Accounting for centre-effects in multicentre trials with a binary outcome – when, why, and how?

 

Mean treatment effect

Type I error rate (%)

Convergence (%)

Unadjusted for centre-effects

1.00

4.7

100

Fixed-effects

1.01

5.1

99.8

Random-effects

1.00

4.8

99.4

GEE

1.00

5.1

100

Mantel-Haenszel

1.01

4.0

100