Unpredictable bias when using the missing indicator method or complete case analysis for missing confounder values: an empirical example
Unpredictable bias when using the missing indicator method or complete case analysis for missing confounder values: an empirical example
复制标题
DOI：
10.1016/j.jclinepi.2009.08.028
复制
发表时间：
2010-07-01
影响因子：
7.2
通讯作者：
Geerlings, Mirjam I.
中科院分区：
文献类型：
--
作者：
Knol, Mirjam J.;Janssen, Kristel J. M.;Geerlings, Mirjam I.
目的：遗漏指标法(MIM)和完全病例分析(CC)是处理遗漏混杂资料的常用方法。研究设计与背景：从一项队列研究中，我们选择了暴露(婚姻状况)、结果(抑郁)和混杂因素(年龄、性别和收入)。“收入”中的缺失值是根据不同的缺失值模式创建的：缺失值完全是随机创建的，并取决于暴露和结果值。缺失值的百分比从2.5%到30%不等。结果：当缺失值完全随机时，MIM给出了高估的优势比，而CC和MI给出了公正的结果。当遗漏的值取决于观测值时，MIM和CC给出了低估或高估。偏差的大小和方向取决于缺失值与暴露和结果的关系。随着缺失值百分比的增加，偏倚增加。结论：MIM不应用于处理丢失的混杂数据，因为它即使在缺失值百分比很小的情况下也会给出不可预测的比数比偏差。当缺失值完全随机时，可以使用CC，但它会造成统计能力的损失。(C)2010 Elsevier Inc.保留所有权利。
Objective: Missing indicator method (MIM) and complete case analysis (CC) are frequently used to handle missing confounder data. Using empirical data, we demonstrated the degree and direction of bias in the effect estimate when using these methods compared with multiple imputation (MI).Study Design and Setting: From a cohort study, we selected an exposure (marital status), outcome (depression), and confounders (age, sex, and income). Missing values in "income" were created according to different patterns of missingness: missing values were created completely at random and depending on exposure and outcome values. Percentages of missing values ranged from 2.5% to 30%.Results: When missing values were completely random, MIM gave an overestimation of the odds ratio, whereas CC and MI gave unbiased results. MIM and CC gave under- or overestimations when missing values depended on observed values. Magnitude and direction of bias depended on how the missing values were related to exposure and outcome. Bias increased with increasing percentage of missing values.Conclusion: MIM should not be used in handling missing confounder data because it gives unpredictable bias of the odds ratio even with small percentages of missing values. CC can be used when missing values are completely random, but it gives loss of statistical power. (C) 2010 Elsevier Inc. All rights reserved.