Abstract
Existing methods for the analysis of data involving assay data subject to measurement error are deficient. In particular, classical calibration methods have been shown to yield invalid inferences unless the measurement error is small. Regression calibration, a form of conditional mean imputation, has better properties, but is not well suited to adjusting for heteroscedastic measurement error. Bayesian multiple imputation is less common for measurement error problems than for missing data, but we argue that it represents an attractive option for measurement error, providing superior inferences to existing methods and a convenient way of adjusting for measurement error using simple complete-data methods and multiple imputation combining rules. It also provides a convenient approach to limit of quantification issues, another area where current approaches are in our view deficient. We review some recent work that develops multiple imputation methods for assay data, focusing particularly on three key aspects: internal versus external calibration designs, the role of the nondifferential measurement error assumption in these designs, and heteroscedastic measurement error. Future research topics are outlined.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Buonaccorsi, J. P. 2010. Measurement error: Models, methods and applications. Boca Raton, FL, Chapman and Hall/CRC.
Carroll, R. J., K. Roeder, and L. Wasserman. 1999. Flexible parametric measurement error models. Biometrics, 55, 44–54.
Carroll, R. J., D. Ruppert, L. A. Stefanski, and C. M. Crainiceanu. 2006. Measurement error in nonlinear models: A modern perspective, 2nd ed. Boca Raton, Chapman and Hall/CRC.
Carroll, R. J., and L. A. Stefanski. 1990. Approximate quasi-likelihood estimation in problems with surrogate predictors. J. Am. Stat. Ass., 85(411), 652–663.
Chu, H., L. Nie, and T. W. Kensler. 2008. A Bayesian approach estimating treatment effects on biomarkers containing zeros with detection limits. Stat. Med., 27, 2497–2508.
Cole, S. R., H. Chu, and S. Greenland. 2006. Multiple-imputation for measurement-error correction. Int. J. Epidemiol., 35, 1074–1081.
Cooper, G. S., D. A. Savitz, R. Millikan, and T. Chiu Kit. 2002. Organochlorine exposure and age at natural menopause. Epidemiology, 13(1), 729–1733.
Currie, L. A., 1968. Limits for qualitative detection and quantitative determination—Application to radiochemistry. Anal. Chem., 40, 586–592.
Delaigle, A., and A. Meister. 2008. Density estimation with heteroscedastic error. Bernoulli, 14, 562–579.
Freedman, L. S., D. Midthune, R. J. Carroll, and V. Kipnis. 2008. A comparison of regression calibration, moment reconstruction and imputation for adjusting for covariate measurement error in regression. Stat. Med., 27, 5195–5216.
Fuller, W. A. 1987. Measurement error models. New York, NY, Wiley.
Gelman, A., J. B. Carlin, H. S. Stern, and D. B. Rubin. 2004. Bayesian data analysis, 2nd ed. Boca Raton, FL, Chapman and Hall/CRC.
Guo, Y., O. Harel, and R. J. A. Little. 2010. How well quantified is the limit of quantification? Epidemiology, 21, S10–S16.
Guo, Y., and R. J. A. Little, 2011. Regression analysis with covariates that have heteroscedastic measurement error. Stat. Med., 30, 2278–2294.
Guo, Y., R. J. A. Little, and D. McConnell. 2012. On using summary statistics from an external calibration sample to correct for covariate measurement error. Epidemiology, 23(1), 165–174.
He, Y., and A. M. Zaslavsky. 2009. Combining information from cancer registry and medical records data to improve analyses of adjuvant cancer therapies. Biometrics, 65, 946–952.
Helsel, D. R. 2004. Nondetects and data analysis: Statistics for censored environmental data. New York, NY, Wiley
Higgins, K. M., M. Davidian, G. Chew, and H. Burge. 1998. The effect of serial dilution error on calibration inference in immunoassay. Biometrics, 54, 19–32.
Hossian, S., and P. Gustafson. 2009. Bayesian adjustment for covariate measurement errors: a flexible parametric approach. Stat. Med., 28, 1580–1600.
Little, R. J. A., and D. B. Rubin. 2002. Statistical analysis with missing data, 2nd ed. New York, NY, Wiley.
May, R. C., J. G. Ibrahim, and H. Chu. 2011. Maximum likelihood estimation in generalized linear models with multiple covariates subject to dection limits. Stat. Med., 30, 2551–2561.
Messer, K., and L. Natarajan. 2008. Maximum likelihood, multiple imputation and regression calibration for measurement error adjustment. Stat. Med., 27, 6332–6350.
Raghunathan, T., P. Solenberger, and J. van Hoewyk. 2011. IVEware: Imputation and variance estimation software: Installation instructions and user guide. Survey Research Center, Institute of Social Research, University of Michigan, Ann Arbor.
Raghunathan, T. E. 2006. Combining information from multiple surveys for assessing health disparities. Allgemeines Statistisches Archiv., 90(4), 515–526.
Reiter, J.R. 2008. Multiple imputation when records used for imputation are not used or disseminated for analysis. Biometrika, 95(4), 933–946.
Richardson, D. B., and A. Ciampi. 2003. Effects of exposure measurement error when an exposure variable is constrained by a lower limit. Am. J. Epidemiol., 157, 355–363.
Richardson, S., W. R. Gilks. 1993. Conditional independence models for epidemiological studies with covariate measurement error. Stat. Med., 12, 1703–1722.
Richardson, S., L. Leblond, I. Jaussent, and P. J. Green. 2002. Mixture models in measurement error problems, with reference to epidemiological studies. J. R. Stat. Soc. Ser. A Stat. Society, 165, 549–566.
Robins J. M., and N. Wang. 2000. Inference in imputation estimators. Biometrika, 87, 113–124.
Rosner B., D. Spiegelman, and W. C. Willett. 1990. Correction of logistic regression relative risk estimates and confidence intervals for measurement error: The case of multiple covariates measured with error. Am. J. Epidemiol., 132, 734–745.
Rubin, D. B. 1987. Multiple imputation for nonresponse in surveys. New York, NY, Wiley.
Sadler, W. A., and M. H. Smith. 1985. Estimation of the response-error relationship in immunoassay. Clinical Chemistry, 31, 1802–1805.
SAS 2011. Proc MI and Proc MIANALYZE in statistical analysis software, Version 9.3. Cary, NC, SAS Institute, Inc.
Singh, A., and I. Nocerino. 2002. Robust estimation of mean and variance using environmental data sets with below detection limit observations. Chemometrics Intelligent Lab. Systems, 60, 69–86.
Spiegelman, D. 1990. Validation Studies. In Colton, T. and Armitage, P., eds., Encyclopedia of Biostatistics, 4694–1700. Sussex, UK, Wiley.
Spiegelman, D., R. J. Carroll, and V. Kipnis. 2001. Efficient regression calibration for logistic regression in main study/internal validation study designs with an imperfect reference instrument. Stat. Med., 20, 139–160.
Spiegelman, D., R. Logan, and D. Grove. 2011. Regression calibration with heteroscedastic error variance. Lnte. J. Biostat., 7, 1–34.
Stata. 2011. Stata: Data analysis and statistics software, Version 12. College Station, TX, StataCorp LP.
Van Buuren, S., and K. Groothuis-Oudshoorn. 2011. MICE: Multivariate imputation by chained equations. R package, version 2.9. http://CRAN.R-project.org/package=mice.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Guo, Y., Little, R.J. Bayesian Multiple Imputation for Assay Data Subject to Measurement Error. J Stat Theory Pract 7, 219–232 (2013). https://doi.org/10.1080/15598608.2013.772018
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1080/15598608.2013.772018