B0291
Title: A modified log rank test for interval censored competing risk data
Authors: Yang-Jin Kim - Sookmyung Women University (Korea, South) [presenting]
Abstract: A competing risk analysis has been applied when a patient fails by one of several causes. Then available data include information about a failure time and a failure cause. Gray suggested a log rank test with a weighted risk process for right censored competing risk data. Our interest is to suggest a test statistics with an interval censored competing risk data. The available data is two sequential inspection times embracing a failure time and a cause indicator. We proposed a modified log rank test by combining Sun's test for interval censored data and Gray's test for competing risk. Therefore, the suggested test reduces to Sun's test statistic under one cause failure. The suggested method is applied to AIDS dataset and some simulation results are presented to evaluate the performance.
Journal