Correcting for Bias due to Misclassification when Error-prone Continuous Exposures Are Misclassified

Abstract

To investigate the association between a continuous exposure and an outcome it is common to categorize the exposure and estimate the relative associations between categories. Error in measurement of the continuous exposure results in misclassification when the exposure is categorized. In this paper we investigate methods for correcting for this misclassification. We consider applications of methods for continuous exposures and for fundamentally categorical exposures. A particular challenge is that even nondifferential error in the underlying continuous exposure can result in differential misclassification in the categorized exposure, i.e. misclassification dependent on the outcome. For continuous exposures, there exist a range of methods for correcting for the effects of exposure measurement error on the exposure-outcome association, including regression calibration (RC), multiple imputation (MI), moment reconstruction (MR) and simulation extrapolation (SIMEX). There are also correction methods for use with genuinely categorical exposures, using estimated misclassification probabilities. Alongside simple methods using estimated misclassification probabilities, we also consider two RC based methods, MI and MR of the continuous exposure followed by categorization, and a new SIMEX method. Simulation studies are used to compare the methods when the true exposure is available in a validation study and the more common situation in which replicate or additional error-prone exposure measurements are available in a subsample. We restrict attention to the case where the underlying association between the continuous exposure and the outcome is linear on the appropriate scale. RC and SIMEX methods fail to correct adequately for bias. However, MI and MR perform well. Methods using estimated misclassification probabilities also perform well, provided differential misclassification is assumed, however these methods are restricted to estimation of odds ratios and have other practical drawbacks. MI and MR have the benefit of being flexible for use with different analysis models, with quantile-based cutpoints, and more easily accommodate covariate adjustment. In summary, we found that MI and MR can be applied to correct exposureoutcome associations for the effects of misclassification error when the association is linear. Extending MI and MR for use with categorized continuous exposures under nonlinear exposureoutcome associations is now an important area for further research.

Higher Dispositional Optimism Predicts Better Health-Related Quality of Life After Esophageal Cancer Surgery: A Nationwide Population-Based Longitudinal Study

“…Misclassification might happen due to the measurement error in the LOT-R, which can cause potential bias in either direction. 47 Third, patients with lower dispositional optimism and poor HRQL might be more likely to decline to participate in the study and not answer the follow-up questionnaires, which could make the observed associations underestimated. Last but not least, prediction does not equal causation.…”

A toolkit for measurement error correction, with a focus on nutritional epidemiology

“…Classical error in a multivariate exposure setting can result in bias in any direction even in a linear regression model [4]. Where there are non-linear terms, for example, a quadratic term, in the exposureoutcome model, classical measurement error has the effect of making the association appear more linear [32]. Other types of error, such as systematic error, which depends on the true exposure; heteroscedastic error; and differential error, which depends on the outcome, may in general result in biases of any form, for example, bias either away from or towards the null [4,33].…”

STRATOS guidance document on measurement error and misclassification of variables in observational epidemiology: Part 2—More complex methods of adjustment and advanced topics

“…Recently Keogh et al 128 investigated several methods of adjusting for misclassification due to dichotomizing a continuous error-prone predictor. In contrast to previous work, a measured continuous exposure was specified to follow a linear measurement error model, thereby allowing for systematic error.…”