联合脆弱模型在含有终止事件的复发事件数据分析中的应用

Application of Joint Frailty Model for Analysis of Recurrent Event Data with Terminal Event

目的探讨联合脆弱模型在含有终止事件的临床复发事件数据分析中的应用及R软件的实现。方法收集肺癌患者复发数据(多次住院),构建联合脆弱模型,拟采用最大惩罚似然估计(MPnLE)进行模型的参数估计,并评估肺癌患者个体内多次复发间的相关性以及复发事件与终止事件(死亡)间的相关性。结果联合脆弱模型分别评估了协变量对疾病复发进程与死亡进程的效应,同时也考虑了多次复发与死亡的相关性,结果解释合理,软件实现方便。结论联合脆弱模型可以充分挖掘含终止事件的肺癌患者复发数据所蕴含的信息,也可用于其他肿瘤患者预后因素的分析,为临床诊断和治疗提供统计学支持。

Objective To explore the application of joint frailty model in the analysis of clinical recurrent event data with a terminal event.Methods The recurrence data of lung cancer patients was collected to establish joint frailty model.Parameters were estimated by using maximum penalized likelihood estimation(MPnLE),and the correlation of intra-subjects as well as the correlation between recurrent event and terminal event(death)was evaluated.Results Joint frailty model respectively evaluated the effects of covariates on the process of recurrent event and death,also consider the correlation between recurrent event and terminal event.The interpretation of results is reasonable and software implementation is convenient.Conclusion Joint frailty models could fully extract the information contained in the recurrence data of lung cancer patients with terminal event and can also be used for the analysis of prognostic factors in other cancer patients,providing statistical support for clinical diagnosis and treatment.