復發事件過程與終止事件時間之獨立性檢定

Testing the Independence of Recurrent Event Process and Terminal Event Time

陳力新(Li-Hsin Chen)

指導教授 ： 張淑惠

國立臺灣大學/公共衛生學院/流行病學研究所/碩士(2008年)

https://doi.org/10.6342/NTU.2008.01728

摘要

在長期追蹤的研究中，針對某一感興趣之事件收集觀察個體重複發生此事件的資料稱為復發事件資料，而此事件有順序地重複發生的過程稱為復發事件過程。ㄧ般而言，在復發事件資料中，迫使復發停止的事件，且此事件和復發事件有關，稱為終止事件。以精神分裂症病人住院治療為例：自第一次進行精神分裂症住院治療後，仍可能再回到醫院進行住院治療，直到和精神分裂症有關之因素所引發之死亡發生為止，故此種死亡事件可視為是復發事件過程的ㄧ種相依設限。過去文獻在探討復發事件過程對終止事件的影響時，常利用隨機變數模式化復發事件過程與終止事件的相關性，然而本文則嘗試以Day et al.(1997)利用預測風險比測度單一信號事件時間與單一目標事件時間相關性的概念為基礎，推廣至包括單一目標事件的多個有序警訊標記與單一目標事件的資料，由於多個警訊標記有時序地發生對應復發事件資料中的復發事件過程，而單一目標事件則對應復發事件資料中的單一終止事件，故可將其視為復發事件資料處理。本文以對數排序(Log-rank)型式之檢定方法檢定復發事件過程與終止事件時間的獨立性，最後藉由模擬呈現此方法的檢定能力。

關鍵字

復發事件過程 ； 終止事件 ； 相依設限 ； 預測風險比 ； 對數排序檢定

並列摘要

Recurrent event data arise commonly in longitudinal follow-up studies. During the follow-up period, the recurrent event process could be precluded by a terminal event, such as death. For example, schizophrenia patients may experience repeated hospitalizations related to their schizophrenic symptoms until death or end of study. In the analysis of the recurrent event process, the terminal event (e.g. death) can be regarded as the dependent censoring of the recurrent event process. In literature, frailty models are often applied to tackling the correlation between the recurrent event process and the terminal event. Alternatively, when the terminal event time is the major outcome in the analysis, recurrent events can be considered as serial biological markers for predicting the occurrence of the terminal event. Therefore, the concept of the predictive hazard ratio (Day et al., 1997) can be employed to form the test of the independence of the recurrent event process and the terminal event time. The properties of the proposed test statistic are studied by simulation.

並列關鍵字

dependent censoring ； frailty model ； predictive hazard ratio ； recurrent event process ； terminal event

參考文獻

Clayton, D. G.. (1978).A model for association in bivariate life tables and its application in epidemiological studies of familial tendency in chronic disease incidence. Biometrika, 65, 141-151

Day, R., Bryant, J., and Lefkopoulou, M. (1997). Adaptation of bivariate frailty models for prediction, with application to biological markers as prognostic indicators. Biometrika, 84, 45-56

Kalbfleish, J. D., and Prentice, R. L. (1981).Estimation of the average hazard ratio. Biometrika, 68,105-112

Kalbfleish, J. D., and Prentice, R. L. (2002).The Statistical Analysis of FailureTtime Data. Wiley :New York.

Lin, D. Y., Wei, L. J., Yang, I., and Ying, Z. (2000).Semiparametric regression for the mean and rate functions of recurrent events.Journal of the Royal Statistical Society, Series B, 62, 711-730