Each week, we will be publishing layman’s abstracts of new articles from our prestigious portfolio of journals in statistics. The aim is to highlight the latest research to a broader audience in an accessible format.
Guo, L., Xiong, Y. and Joan Hu, X. (2020), Estimation in the Cox cure model with covariates missing not at random, with application to disease screening/prediction. Can J Statistics. doi:10.1002/cjs.11550
It is a well-received intervention in screening of infectious diseases tracing people who have been in (physical) contacts with infected persons. It can help to stop the spread of many diﬀerent infections in the community involving an infection. Quantitative analysis is in great demand in the related public health studies to evaluate the disease development and address eﬀectiveness and eﬃciency concerns. However, many issues have been encountered in the required analysis, including missing entries of the potential covariates and censored (incompletely observed) ﬁnal clinical outcomes in the study data. The existence of a substantial portion of subjects who are nonsusceptible to the infection adds another layer of complexity.
This article presents a development of statistical methodology for dealing with the statistical challenges. We employ supplementary data to account appropriately for missing covariates. Strategies are provided for identifying risk factors, estimating probability of developing the disease, and predicting the time to the disease onset for those who are deemed to develop the disease. The data from the TB (tuberculosis) contact study at the BC Centre for Disease Control (Cook et al, 2005) are used throughout the paper to motivate and illustrate the research.