Rank Estimation of Treatment Differences Based on Repeated Measurements Subject to Dependent Censoring

• Published in: Journal of the American Statistical Association Vol. 94; no. 447; pp. 888 - 895
• Main Authors: Glidden, D. V., Wei, L. J.
• Format: Journal Article
• Language: English
• Published: Alexandria, VA Taylor & Francis Group 01.09.1999; American Statistical Association; Taylor & Francis Ltd
• Subjects: AIDS; Applications; Censorship; Clinical trials; Covariance matrices; Cumulative hazard function; Estimating function; Estimation; Estimation methods; Exact sciences and technology; Informative censoring; Longitudinal data; Mathematics; Measurement; Medical sciences; Medical treatment; Missing data; Outcome variables; Parametric inference; Parametric models; Patients; Point estimators; Probability and statistics; Resampling method; Sampling; Sciences and techniques of general use; Statistics; Theory and Methods; Time dependence; Wilcoxon statistic; Infection; Rank statistic; Missing data; Selection problem; Data analysis; Viral disease; Treatment efficiency; Repeated measurement; Clinical trial; AIDS; Medical application
• ISSN: 0162-1459; 1537-274X
• DOI: 10.1080/01621459.1999.10474194

In comparing the effectiveness of two treatments, suppose that for each patient repeated measurements of an outcome variable are taken at prespecified time points, but some observations may be missing due to the patient's dependent right censoring. In this article, a simple rank estimation procedure, constructed based on an artificial censoring technique, is proposed for the treatment differences over time without imposing a parametric structure on the dependence between the outcome measures and the censoring variable. Our method can be easily implemented and is illustrated by a dataset from an AIDS clinical trial.