Table 2 Parameter setting for binomial outcomes (log odds scale)
From: Power and sample size calculation for stepped-wedge designs with discrete outcomes
Simulation | EPT Trial | LIRE Trial | |
|---|---|---|---|
# of sequence | 3 | 4 | 5 |
# of time period (Nt) | 4 | 5 | 6 |
# of cluster per sequence | 4, 8, 12 | 6 | 20 |
# of pcp per cluster | NA | NA | 35 |
Cluster size | 20, 50, 100 | ? | ? |
Prevalence (roughly) | 0.03, 0.12, 0.43 | 0.08 | 0.19 |
Fixed time effect βtime | (0.1, 0.1, 0.1) | (−0.008, −0.08, -0.17, −0.11) | −0.124 |
Effect size for the intervention βp | 0.2 | −0.3 | −0.055 |
SD of random cluster effect | 0.05 | 0.2 | 0.011 |
SD of random treatment effect | 0.05, 0.1 | NA | 0.0054 |
SD of random time effect | 0.05, 0.1 | 0.12 | NA |
SD of random cluster (pcp) effect | NA | NA | 0.0015 |