
Sample Size Determination for Correlated Outcomes Using GEE 123
4.5.3 Application Example
Suppose researchers would like to design a clinical trial to assess the efficacy
of an experimental drug in preventing the occurrence of pulmonary fibrosis in
subjects with scleroderma. Patients will be randomly assigned to either receive
the drug or placebo with a randomization ratio of 1:1. For each subject, the
presence or absence of pulmonary fibrosis will be assessed m = 6 times, at
baseline, and at months 6, 12, 18, 24, and 30. The primary interest is to detect
the difference in the rate of change between the two groups. The targeted
power and two-sided type