BBB Research: Time-to-Event Data Analysis

BBB has a long history of methodological contributions to the analysis of time-to-event data, also called survival analysis. Although much of the existing literature is on methods for a single time-to-event variable, increasingly, in many applications, researchers are interested in studies with more than one time-to-event variable. Such data, called multivariate survival data, are frequently encountered in many longitudinal studies. Examples of such data abound in studying reproductive, perinatal health outcomes as well as adolescent behavioral studies. For instance, time to pregnancy loss is an example of competing risks. Progression of labor through its various stages provides examples of recurrent event, multistage models. However, due to the unique nature of reproductive/obstetric events, continuous monitoring is not possible, leading to issues of unknown time “zero” as well as issues of interval censoring and length bias. Research efforts are focused on developing robust statistical inference in multistage models with various issues of incompleteness. Additional focus includes issues of modeling bivariate survival data, subject to length-bias, truncation, and various types of censoring including current status data, interval censored data, and so forth.

Principal Investigator

Raji Sundaram, M.Stat., Ph.D.

Selected Publications

Lee, Y., Wang, M.-C., Grantz, K. L., & Sundaram, R. (2019). Joint modeling of competing risks data and current status data: an application to spontaneous labour study. Journal of the Royal Statistical Society: Series-C, 68, 1167-1182. DOI: 10.1111/rssc.12351 external link

Ma, L., & Sundaram, R. (2018). Analysis of gap times based on panel count data with informative observation times and unknown start time. Journal of the American Statistical Association, 113(521), 294-305.

Sundaram, R., Ma, L., & Ghoshal, S. (2017). Median cost analysis associated with recurrent episodic illnesses in the presence of terminal event. International Journal of Biostatistics, 13(1). PMID: 28453440

Lum, K. J., Sundaram, R., & Louis, T. A. (2015). Accounting for length-bias and selection effects in estimating the effects of menstrual cycle length. Biostatistics, 16(1), 113-128. PMID: 25027273. PMCID: PMC4263226

Sundaram, R., McLain, A. C., & Buck, L. G. M. (2012). A survival analysis approach to modeling human fecundity. Biostatistics, 13(1), 4-17. PMID: 21697247. PMCID: PMC3276273

Sundaram, R. (2006). Semi-parametric inference in proportional odds model with time-dependent covariates. Journal of Statistical Planning and Inference, 136, 320-334.

Datta, S., & Sundaram, R. (2006). Nonparametric estimation of stage occupation probabilities in a multistage model using current status data. Biometrics, 62(3), 829-837. PMID: 16984326

top of pageBACK TO TOP