On the Nondifferential Misclassification of a Binary Confounder

Consider a study with binary exposure, outcome, and confounder, where the confounder is nondifferentially misclassified. Epidemiologists have long accepted the unproven but oft-cited result that, if the confounder is binary, then odds ratios, risk ratios, and risk differences that control for the mismeasured confounder will lie between the crude and the true measures. In this paper, we provide an analytic proof of the result in the absence of a qualitative interaction between treatment and confounder, and we demonstrate via counterexample that the result need not hold when there is such a qualitative interaction. We also present an analytic proof of the result for the effect of treatment among the treated and describe extensions to measures conditional on or standardized over other covariates.

[1]  Sander Greenland,et al.  Interval Estimation for Messy Observational Data , 2009, 1010.0306.

[2]  R. Cooper,et al.  Socioeconomic Status and Health in Blacks and Whites: The Problem of Residual Confounding and the Resiliency of Race , 1997, Epidemiology.

[3]  G. Pershagen,et al.  Exposure to Air Pollution from Traffic and Childhood Asthma Until 12 Years of Age , 2013, Epidemiology.

[4]  S Greenland,et al.  The effect of misclassification in the presence of covariates. , 1980, American journal of epidemiology.

[5]  W. Willett,et al.  An overview of issues related to the correction of non-differential exposure measurement error in epidemiologic studies. , 1989, Statistics in medicine.

[6]  M. Szklo,et al.  Epidemiology: Beyond the Basics , 1999 .

[7]  T T Chen.  A review of methods for misclassified categorical data in epidemiology. , 1989, Statistics in medicine.

[8]  G Maldonado,et al.  Adjusting a relative-risk estimate for study imperfections , 2008, Journal of Epidemiology and Community Health.

[9]  Jonathan A C Sterne,et al.  The impact of residual and unmeasured confounding in epidemiologic studies: a simulation study. , 2007, American journal of epidemiology.

[10]  T. VanderWeele.  The Sign of the Bias of Unmeasured Confounding , 2008, Biometrics.

[11]  S Greenland,et al.  Basic methods for sensitivity analysis of biases. , 1996, International journal of epidemiology.

[12]  A. Ahlbom,et al.  Aspects of misclassification of confounding factors. , 1992, American journal of industrial medicine.

[13]  Manuel J. Rodríguez,et al.  Maternal Exposure to Drinking-water Chlorination By-products and Small-for-gestational-age Neonates , 2012, Epidemiology.

[14]  G R Howe,et al.  Methodological issues in case-control studies. III--The effect of joint misclassification of risk factors and confounding factors upon estimation and power. , 1984, International journal of epidemiology.

[15]  W. Youden,et al.  Index for rating diagnostic tests , 1950, Cancer.

[16]  Sander Greenland,et al.  Multiple‐bias modelling for analysis of observational data , 2005 .

[17]  Sander Greenland,et al.  Interval estimation by simulation as an alternative to and extension of confidence intervals. , 2004, International journal of epidemiology.

[18]  M. Goldner,et al.  Hyperglycemia and glycosuria due to thiazide derivatives administered in diabetes mellitus. , 1960, The New England journal of medicine.

[19]  Tjeerd P van Staa,et al.  Long-Term Exposure to Outdoor Air Pollution and Incidence of Cardiovascular Diseases , 2013, Epidemiology.

[20]  S. Greenland,et al.  Risk factors, confounding, and the illusion of statistical control. , 2004, Psychosomatic medicine.

[21]  M. Dufour,et al.  Epidemiology. Introduction. , 1995, Recent developments in alcoholism : an official publication of the American Medical Society on Alcoholism, the Research Society on Alcoholism, and the National Council on Alcoholism.

[22]  D A Savitz,et al.  Estimating and correcting for confounder misclassification. , 1989, American journal of epidemiology.

[23]  Diane C. Tsai.  Methods in Observational Epidemiology , 1988, The Yale Journal of Biology and Medicine.

[24]  H Brenner,et al.  Bias due to non-differential misclassification of polytomous confounders. , 1993, Journal of clinical epidemiology.

[25]  Sander Greenland,et al.  Modern Epidemiology 3rd edition , 1986 .

[26]  B G Armstrong,et al.  Analysis of case-control data with covariate measurement error: application to diet and colon cancer. , 1989, Statistics in medicine.

[27]  Timothy L Lash,et al.  Semi-Automated Sensitivity Analysis to Assess Systematic Errors in Observational Data , 2003, Epidemiology.

[28]  Timothy L Lash,et al.  A method to automate probabilistic sensitivity analyses of misclassified binary variables. , 2005, International journal of epidemiology.

[29]  Mario Recker,et al.  Negative epistasis between the malaria-protective effects of α+-thalassemia and the sickle cell trait , 2005, Nature Genetics.

[30]  Carl V Phillips,et al.  Quantifying and Reporting Uncertainty from Systematic Errors , 2003, Epidemiology.