Case-cohort sampling is a commonly used and efficient method for studying large cohorts. In many situations, some covariates are easily measured on all cohort subjects, and surrogate measurements of the expensive cova...Case-cohort sampling is a commonly used and efficient method for studying large cohorts. In many situations, some covariates are easily measured on all cohort subjects, and surrogate measurements of the expensive covariates also may be observed. In this paper, to make full use of the covariate data collected outside the case-cohort sample, we propose＇a class of weighted estimators with general time-varying weights for the additive hazards model, and the estimators are shown to be consistent and asymptotically normal. We also identify the estimator within this class that maximizes efficiency, and simulation studies show that the efficiency gains of the proposed estimator over the existing ones can be substantial in practical situations. A real example is provided.展开更多
文摘目的将共享随机效应模型(shared random-effect model,SREM)应用于轻度认知障碍(mild cognitive impairment,MCI)向认知正常(normal cognition,NC)逆转的研究,比较不同纵向认知标志物对MCI逆转的预测性能,并评价影响因素的协变量效应。方法SREM模型包括两个子模型,其中纵向子模型采用线性混合效应模型对纵向认知标志物的变化轨迹建模,生存子模型采用比例风险模型对生存过程建模。基于对数似然函数值和信息准则进行模型拟合优度检验,采用ROC曲线下面积(area under the curve,AUC)评价不同纵向认知标志物(MMSE、CDRSB、FAQ、ADAS11、ADAS13和ADASQ4)对MCI逆转的预测性能;同时进行纵向子模型和生存子模型的影响因素分析。结果843名MCI患者中72名(8.54%)在随访结束后逆转为NC。以spline-PH-GH参数分布为基准风险函数的SREM模型对数似然函数值最大,AIC和BIC最小;以CDRSB为纵向认知标志物建立的SREM模型拟合最好,在不同时间的AUC值均表现良好,范围为0.797~0.852,且预测误差最小,范围为0.0427~0.0429;年龄、性别、受教育程度、婚姻状况和APOEε4基因均会影响MCI患者的认知功能和日常活动功能,六种纵向认知标志物均会影响MCI患者的逆转。结论CDR评分对MCI患者的认知功能和逆转预测性能最佳;认知功能和日常活动功能可作为MCI逆转的动态监测指标。
基金partly supported by the National Natural Science Foundation of China Grants(No.11231010,11171330 and 11101314)Key Laboratory of RCSDS,CAS(No.2008DP173182)and BCMIIS
文摘Case-cohort sampling is a commonly used and efficient method for studying large cohorts. In many situations, some covariates are easily measured on all cohort subjects, and surrogate measurements of the expensive covariates also may be observed. In this paper, to make full use of the covariate data collected outside the case-cohort sample, we propose＇a class of weighted estimators with general time-varying weights for the additive hazards model, and the estimators are shown to be consistent and asymptotically normal. We also identify the estimator within this class that maximizes efficiency, and simulation studies show that the efficiency gains of the proposed estimator over the existing ones can be substantial in practical situations. A real example is provided.
