Correction of Risk Estimates for Measurement Error in Epidemiology

S. A. Bashir; S. W. Duffy

doi:10.1055/s-0038-1634621

Methods of Information in Medicine, Table of Contents

Methods Inf Med 1995; 34(05): 503-510
DOI: 10.1055/s-0038-1634621

Original Article

Schattauer GmbH

Correction of Risk Estimates for Measurement Error in Epidemiology

Authors

S. A. Bashir

¹MRC Biostatistics Unit, Institute of Public Health, University Forvie Site, Cambridge U. K.
S. W. Duffy

¹MRC Biostatistics Unit, Institute of Public Health, University Forvie Site, Cambridge U. K.

Abstract

Abstract:

Epidemiologists are under considerable pressure to acknowledge the presence of measurement error in the determination of risk factors. Repeatability and validation studies are often prescribed in conjunction with epidemiological studies. We describe some practical uses for repeatability and validation study data, in terms of correcting risk estimates for measurement error. Commonly available methods are described, with their advantages and shortcomings. A user-friendly computer program to carry out the analyses described accompanies the paper.

Keywords:

Epidemiology - Risk Factors - Measurement Errors

Full Text

References

References
1 Armstrong BG, Whittemore AS, Howe GR. Analysis of case-control data with covariate measurement error: Application to diet and colon cancer. Statist Med 1989; 1151-63.
2 Berkson J. Are there two regressions?. J Amer Statist Ass 1950; 45: 164-80.
3 Blettner M, Wahrendorf J. What does an observed relative risk convey about possible misclassification?. Meth Inform Med 1984; 23: 37-40.
4 Clayton D. Using test-retest reliability data to improve estimates of relative risk; an application of latent class analysis. Statist Med 1985; 4: 445-55.
5 Thomas DC, Stram D, Dwyer J. Exposure measurement error: Influence in exposuredisease relationships and methods of correction. Ann Rev Public Health 1992; 14: 69-93.
6 Walker AM, Lanes SF. Misclassification of covariates. Statist Med 1991; 10: 1181-96.
7 Duffy SW, Rohan TE, Day NE. Misclassification in more than one factor in a case-control study: A combination of Mantel-Haenszeel and maximum likelihood approaches. Statist Med 1989; 8: 1529-36.
8 MacMahon S, Peto R, Cutler J. et al. Blood pressure, stroke, and coronary heart disease: Part 1. The Lancet 1990; 335: 765-74.
9 Bross I. Misclassification in 2 x 2 tables. Biometrics 1954; 10: 478-86.
10 Mertens TE. Estimating the effects of misclassification. The Lancet 1994; 342: 418-21.
11 Qizilbash N, Iones L, Warlow C, Mann J. Fibrinogen and lipid concentration as risk factors for transient ischaemic attacks and minor ischaemic strokes. BMJ 1991; 303: 605-9.
12 Qizilbash N, Duffy SW, Prentice CRM, Boothby M, Lopez Arrieta J, Warlow C. Von Willebrand factor and risk of ischaemic stroke. Stroke. 1995 (to be publ.).
13 Cox B, Elwood JM. The effect on stratum specific odds ratios of nondifferential misclassification of a dichotomous covariate. Am J Epidem 1991; 28: 202-7.
14 Rosner B, Willet WC, Spiegelmann D. Correction of logistic regression relative risk estimates and confidence intervals for systematic within-person measurement error. Statist Med 1989; 8: 1051-69.
15 Kuha J. Corrections for exposure measurement error in logistic regression models with an application to nutritional data. Statist Med 1994; 13: 1135-48.
16 Duffy SW, Maximovitch DM, Day NE. External validation, repeat determination, and precision of risk estimation in misclassified exposure data in epidemiology. J Epidem Commun Health 1992; 46: 620-4.
17 Greenland S. On correcting for misclassification in twin studies and other matched-pair studies. Statist Med 1989; 8: 825-9.
18 Greenland S. The effect of misclassification in matched-pair case-control studies. Amer J Epidem 1982; 116: 402-6.
19 Greenland S, Kleinbaum DG. Correcting for misclassification in two-way tables and matched studies. Intern J Epidem 1983; 12: 93-7.
20 Kaldor J, Clayton D. Latent class analysis in chronic disease epidemiology. Statist Med 1985; 4: 327-35.
21 Richardson S, Gilks WR. A Bayesian approach to measurement error problems in epidemiology using conditional independence models. Amer J Epidem 1993; 138: 430-42.
22 Thomas DC, Gauderman J, Kerber R. A non-parametric Monte Carlo approach to adjustment for covariate measurement errors in regression analysis. Biometrics. 1995 (to be publ.).
23 Dempster AP, Laird NM, Rubin DB. Maximum likelihood from incomplete data via the EM algorithm. J Royal Statist Soc, Series B 1977; 39: 1-38.
24 Liu X, Liang K-Y. Adjustment for non-differential misclassification error in the generalized linear model. Statist Med 1991; 10: 1197-211.
25 Rosner B, Spiegelman D, Willet WC. Correction of logistic regression relative risk estimates and confidence intervals for measurement error: The case of multiple covariates measured with error. Amer J Epidem 1990; 132: 734-45.