KF Statistical Consulting Company.
Stat Methods Med Res. 2020 Feb;29(2):603-616. doi: 10.1177/0962280219842352. Epub 2019 Apr 16.
In this paper, we consider joint modeling of repeated measurements and competing risks failure time data to allow for more than one distinct failure type in the survival endpoint. Hence, we can fit a cause-specific hazards submodel to allow for competing risks, with a separate latent association between longitudinal measurements and each cause of failure. We also consider the possible masked causes of failure in joint modeling of repeated measurements and competing risks failure time data. We also derive a score test to identify longitudinal biomarkers or surrogates for a time-to-event outcome in competing risks data which contain masked causes of failure. With a carefully chosen definition of complete data, the maximum likelihood estimation of the cause-specific hazard functions and of the masking probabilities is performed via an expectation maximization algorithm. The simulations are used to explore how the number of individuals, the number of time points per individual, and the functional form of the random effects from the longitudinal biomarkers considering heterogeneous baseline hazards in individuals influence the power to detect the association of a longitudinal biomarker and the survival time.
在本文中，我们考虑联合建模重复测量和竞争风险失效时间数据，以允许生存终点有多种不同的失效类型。因此，我们可以拟合一个特定原因的危险子模型，以允许竞争风险，并允许纵向测量与每种失效原因之间存在单独的潜在关联。我们还考虑了在重复测量和竞争风险失效时间数据的联合建模中可能存在的失效原因的隐蔽性。我们还推导出了一个得分检验，用于识别竞争风险数据中存在隐蔽失效原因的时间事件结果的纵向生物标志物或替代物。通过仔细选择完整数据的定义，通过期望最大化算法对特定原因的危险函数和掩蔽概率进行最大似然估计。模拟用于探索在个体的基线危险异质性下，个体的数量、每个个体的时间点数量以及考虑纵向生物标志物的随机效应的函数形式如何影响检测纵向生物标志物与生存时间关联的能力。
