|
Abstract:
|
Bilateral correlated data often arise in ophthalmological studies, where each subject may contribute results of two eyes to the analysis. There is correlation between the two eyes of the same subject. In this article, MLE based Wald and profile likelihood confidence intervals for difference of proportions are adjusted for bilateral data, taking the intraclass correlation into account. The coverage probability and width of the resulting methods, together with a simple Wald CI and score method CI are compared in a Monte Carlo simulation study. Both have coverage probability close to nominal level with favorable interval width. A real working example is presented to illustrate the methodology. We develop a web calculator so the algorithm can be easily used in practice and results in the paper can be reproduced.
|
