Approaches to addressing missing values, measurement error, and confounding in epidemiologic studies

• Published in: Journal of clinical epidemiology Vol. 131; pp. 89 - 100
• Main Authors: van Smeden, Maarten, Penning de Vries, Bas B.L., Nab, Linda, Groenwold, Rolf H.H.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.03.2021; Elsevier Limited
• Subjects: Bayesian analysis; Bias; Blood pressure; Body mass index; Confidence intervals; Confounding; Data analysis; Epidemiology; Imputation; Internal Medicine; Measurement error; Missing data; Multivariate statistical analysis; Oils & fats; Performance assessment; Performance measurement; Recall; Regression; Regression analysis; Regression calibration; Simulation; Statistical analysis; Statistics; Missing data; Data analysis; Simulation; Confounding; Regression; Regression calibration; Imputation; Measurement error
• ISSN: 0895-4356; 1878-5921; 1878-5921
• DOI: 10.1016/j.jclinepi.2020.11.006

Epidemiologic studies often suffer from incomplete data, measurement error (or misclassification), and confounding. Each of these can cause bias and imprecision in estimates of exposure–outcome relations. We describe and compare statistical approaches that aim to control all three sources of bias simultaneously. We illustrate four statistical approaches that address all three sources of bias, namely, multiple imputation for missing data and measurement error, multiple imputation combined with regression calibration, full information maximum likelihood within a structural equation modeling framework, and a Bayesian model. In a simulation study, we assess the performance of the four approaches compared with more commonly used approaches that do not account for measurement error, missing values, or confounding. The results demonstrate that the four approaches consistently outperform the alternative approaches on all performance metrics (bias, mean squared error, and confidence interval coverage). Even in simulated data of 100 subjects, these approaches perform well. There can be a large benefit of addressing measurement error, missing values, and confounding to improve the estimation of exposure–outcome relations, even when the available sample size is relatively small.