Mazroui Yassin, Mathoulin-Pélissier Simone, Macgrogan Gaetan, Brouste Véronique, Rondeau Virginie
INSERM, ISPED, Centre INSERM U-897-Epidemiologie-Biostatistique, 146 rue Léo Saignat, 33076, Bordeaux Cedex, France; Université Bordeaux Segalen, ISPED, Centre INSERM U-897-Epidemiologie-Biostatistique, Bordeaux, F-33000, France.
Biom J. 2013 Nov;55(6):866-84. doi: 10.1002/bimj.201200196. Epub 2013 Aug 9.
Individuals may experience more than one type of recurrent event and a terminal event during the life course of a disease. Follow-up may be interrupted for several reasons, including the end of a study, or patients lost to follow-up, which are non informative censoring events. Death could also stop the follow-up, hence, it is considered as a dependent terminal event. We propose a multivariate frailty model that jointly analyzes two types of recurrent events with a dependent terminal event. Two estimation methods are proposed: a semiparametrical approach using penalized likelihood estimation where baseline hazard functions are approximated by M-s plines, and another one with piecewise constant baseline hazard functions. Finally, we derived martingale residuals to check the goodness-of-fit. We illustrate our proposals with a real dataset on breast cancer. The main objective was to model the dependency between the two types of recurrent events (locoregional and metastatic) and the terminal event (death) after a breast cancer.
在疾病的病程中，个体可能会经历不止一种类型的复发事件和一个终末事件。随访可能会因多种原因中断，包括研究结束，或患者失访，这些都是无信息删失事件。死亡也可能导致随访终止，因此，它被视为一个相关的终末事件。我们提出了一种多变量脆弱模型，该模型联合分析两种类型的复发事件和一个相关的终末事件。提出了两种估计方法：一种是使用惩罚似然估计的半参数方法，其中基线危险函数由M样条近似，另一种是具有分段常数基线危险函数的方法。最后，我们推导了鞅残差以检验拟合优度。我们用一个关于乳腺癌的真实数据集来说明我们的提议。主要目的是对乳腺癌后两种类型的复发事件（局部区域复发和转移）与终末事件（死亡）之间的相关性进行建模。
