Bias arising from missing data in predictive models
Bias arising from missing data in predictive models
复制标题
DOI：
10.1016/j.jclinepi.2004.11.029
复制
发表时间：
2006-10-01
影响因子：
7.2
通讯作者：
Gorelick, Marc H.
中科院分区：
文献类型：
--
作者：
Gorelick, Marc H.
目的：本研究的目的是确定三种常见的方法来处理缺失数据的预测model.Study设计和设置的结果的效果：使用模拟数据的Monte Carlo模拟研究。根据白色血细胞计数（WBC）（分为正常或高）、发热或进行的手术（PROC），使用完整数据进行基线logistic回归，以预测住院。然后进行一系列模拟，其中在各种缺失模式下删除不同比例（15-85%）的患者的WBC数据。三种分析方法：分析仅限于完整的数据，缺失的数据假设为正常（MAN），并使用插补values.Results：在基线分析，所有三个预测因子均与入院显着相关。使用MAN方法或插补，WBC的比值比（OR）根据缺失模式被大大高估或低估，发热的OR估计值存在相当大的偏倚。在CC分析中，WBC的OR始终偏向零值，PROC的OR偏向零值，发热的OR偏向或偏向零值。估计整体模型歧视大幅度偏置使用所有的分析approaches.Conclusions：所有三种方法处理大量的缺失数据可能会导致有偏估计的OR和预测模型的模型性能。测量不一致的预测变量可能会影响这些模型的有效性。(c)2006年爱思唯尔公司All rights reserved.
Objective: The purpose of this study is to determine the effect of three common approaches to handling missing data on the results of a predictive model.Study Design and Setting: Monte Carlo simulation study using simulated data was used. A baseline logistic regression using complete data was performed to predict hospital admission, based on the white blood cell count (WBC) (dichotomized as normal or high), presence of fever, or procedures performed (PROC). A series of simulations was then performed in which WBC data were deleted for varying proportions (15-85%) of patients under various patterns of missingness. Three analytic approaches were used: analysis restricted to cases with complete data, missing data assumed to be normal (MAN), and use of imputed values.Results: In the baseline analysis, all three predictors were all significantly associated with admission. Using either the MAN approach or imputation, the odds ratio (OR) for WBC was substantially over- or underestimated depending on the missingness pattern, and there was considerable bias toward the null in the OR estimates for fever. In the CC analyses, OR for WBC was consistently biased toward the null, OR for PROC was biased away from the null, and the OR for fever was biased toward or away from the null. Estimates for overall model discrimination were substantially biased using all analytic approaches.Conclusions: All three methods of handling large amounts of missing data can lead to biased estimates of the OR and of model performance in predictive models. Predictor variables that are measured inconsistently can affect the validity of such models. (c) 2006 Elsevier Inc. All rights reserved.