Abstract
Murrayand Tsiatis (1996) described a weighted survival estimate thatincorporates prognostic time-dependent covariate informationto increase the efficiency of estimation. We propose a test statisticbased on the statistic of Pepe and Fleming (1989, 1991) thatincorporates these weighted survival estimates. As in Pepe andFleming, the test is an integrated weighted difference of twoestimated survival curves. This test has been shown to be effectiveat detecting survival differences in crossing hazards settingswhere the logrank test performs poorly. This method uses stratifiedlongitudinal covariate information to get more precise estimatesof the underlying survival curves when there is censored informationand this leads to more powerful tests. Another important featureof the test is that it remains valid when informative censoringis captured by the incorporated covariate. In this case, thePepe-Fleming statistic is known to be biased and should not beused. These methods could be useful in clinical trials with heavycensoring that include collection over time of covariates, suchas laboratory measurements, that are prognostic of subsequentsurvival or capture information related to censoring.
Similar content being viewed by others
References
D.R. Cox, “Regression models and life-tables (with discussion),” Journal of the Royal Statistical Society, B vol. 34 pp. 1038–1041, 1972.
D. M. Finkelstein and D. A. Schoenfeld, “Analyzing survival in the presence of an auxiliary variable,” Statistics in Medicine vol. 13 pp. 1747–1754, 1994.
R. J. Gray, “A kernel method for incorporating information on disease progression in the analysis of survival,” Biometrika vol. 81 pp. 527–539, 1994.
E. L. Kaplan and P. Meier, “Nonparametric estimation from incomplete observations,” J. Am. Statist. Assoc. vol. 53 pp. 457–81, 1958.
M. R. Kosorok and T. R. Fleming, “Using surrogate failure time data to increase cost effectiveness in clinical trials,” Biometrika vol. 80 pp. 823–33, 1993.
H. M. Malani, “A modification of the redistribution to the right algorithm using disease markers,” Biometrika vol.82 pp. 515–526, 1995.
S. Murray, “Nonparametric estimation and testing for survival data in the two sample censored data problem incorporating longitudinal covariates,” Sc.D. dissertation, Department of Biostatistics, Harvard University, 1994.
S. Murray and A. A. Tsiatis, “A nonparametric approach to incorporating prognostic longitudinal covariate information in survival estimation,” Biometrics vol. 52 pp. 137–151, 1996.
M. S. Pepe and T. R. Fleming, “Weighted Kaplan-Meier statistics-a class of distance tests for censored survival data,” Biometrics vol. 45 pp. 497–507, 1989.
M. S. Pepe and T. R. Fleming, “Weighted Kaplan-Meier statistics: large sample theory and optimality considerations,” J. R. Statist. Soc. B vol. 53 pp. 341–52, 1991.
J. M. Robins and A. Rotnitzky, “Recovery of information and adjustment for dependent censoring using surrogate markers,” in AIDS Epidemiology: Methodological Issues, (N. Jewell, K. Dietz and V. Farewell, eds.), Boston: Birkhäauser-Boston, 1992.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Murray, S., Tsiatis, A.A. Using Auxiliary Time-Dependent Covariates to Recover Information in Nonparametric Testing with Censored Data. Lifetime Data Anal 7, 125–141 (2001). https://doi.org/10.1023/A:1011392622173
Issue Date:
DOI: https://doi.org/10.1023/A:1011392622173