Cook, Richard J.Boruvka, Audrey2016-06-062016-06-062016-02-29http://dx.doi.org/10.1093/biostatistics/kxv042http://hdl.handle.net/10012/10528This is a pre-copyedited, author-produced PDF of an article accepted for publication in Bioinformatics following peer review. The version of record Boruvka, Audrey and Cook, Richard J. (2016). Biostatistics, 17(2): 350-363. DOI: 10.1093/biostatistics/kxv042 is available online at: http://dx.doi.org/10.1093/biostatistics/kxv042Semiparametric methods are well-established for the analysis of a progressive Markov illness-death process observed up to a noninformative right censoring time. However often the intermediate and terminal events are censored in different ways, leading to a dual censoring scheme. In such settings unbiased estimation of the cumulative transition intensity functions cannot be achieved without some degree of smoothing. To overcome this problem we develop a sieve maximum likelihood approach for inference on the hazard ratio. A simulation study shows that the sieve estimator offers improved finite-sample performance over common imputation-based alternatives and is robust to some forms of dependent censoring. The proposed method is illustrated using data from cancer trials.enCox modelInterval censoringMethod of sievesProfile likelihoodProgression-free survivalArticle