Methods Inf Med 2002; 41(04): 342-348
DOI: 10.1055/s-0038-1634392
Original article
Schattauer GmbH

Implications of Nondifferential Misclassification on Estimates of Attributable Risk

C. Vogel
1   Department of Medical Informatics, Biometry and Epidemiology, Friedrich-Alexander-University of Erlangen-Nuremberg, Germany
,
O. Gefeller
1   Department of Medical Informatics, Biometry and Epidemiology, Friedrich-Alexander-University of Erlangen-Nuremberg, Germany
› Institutsangaben
Weitere Informationen

Publikationsverlauf

Received 30. November 2001

Accepted 31. Januar 2002

Publikationsdatum:
07. Februar 2018 (online)

Summary

Objectives: Only the effects of isolated nondifferential misclassification of exposure or disease on the estimates of attributable risk have been discussed in the literature. The aim of this paper is to broaden the spectrum of scenarios for which implications of misclassification are available.

Methods: For this purpose, a matrix-based approach allowing a comprehensive, unified analysis of various structures of misclassification is introduced. The relative bias or – in the situation of covariate misclassification – the relative adjustment are presented for the different misclassification scenarios.

Results: Under nondifferential misclassification of exposure or disease, the attributable risk is biased towards the null with the only exception of perfect sensitivity of exposure classification or perfect specificity of disease classification both leading to an unbiased attributable risk. From these two marginal effects, the consequences of simultaneous nondifferential independent misclassification of exposure and disease on the attributable risk are derived in the matrix-based approach. Misclassification of a dichotomous covariate leads to partial adjustment.

Conclusions: To a large extent, the results for the attributable risk are in accordance with the well-known results for the relative risk. The algebraic differences between the two risk measures, however, make it necessary to repeat the methodological considerations for the attributable risk.

 
  • References

  • 1 Barron BA. The effects of misclassification on the estimation of relative risk. Biometrics 1977; 33 (Suppl. 02) 414-8.
  • 2 Blettner M, Wahrendorf J. What does an observed relative risk convey about possible misclassification?. Methods Inf Med 1984; 23 (Suppl. 01) 37-40.
  • 3 Copeland KT, Checkoway H, McMichael AJ, Holbrook RH. Bias due to misclassification in the estimation of relative risk. Am J Epidemiol 1977; 105 (Suppl. 05) 488-95.
  • 4 Cox B, Elwood JM. The effect on the stratum-specific odds ratios of nondifferential misclassification of a dichotomous covariate. Am J Epidemiol 1991; 133 (Suppl. 02) 202-7.
  • 5 Flegal KM, Brownie C, Haas JD. The effects of exposure misclassification on estimates of relative risk. Am J Epidemiol 1986; 123 (Suppl. 04) 736-51.
  • 6 Greenland S. The effect of misclassification in the presence of covariates. Am J Epidemiol 1980; 112 (Suppl. 04) 564-9.
  • 7 Greenland S, Kleinbaum DG. Correcting for misclassification in two-way tables and matched-pair studies. Int J Epidemiol 1983; 12 (Suppl. 01) 93-7.
  • 8 Rippin G. Design issues and sample size when exposure measurement is inaccurate. Methods Inf Med 2001; 40 (Suppl. 02) 137-40.
  • 9 Savitz DA, Baron AE. Estimating and correcting for confounder misclassification. Am J Epidemiol 1989; 129 (Suppl. 05) 1062-71.
  • 10 Tzonou A, Kaldor J, Smith PG, Day NE, Trichopoulos D. Misclassification in case-control studies with two dichotomous risk factors. Rev Epidemiol Sante Publique 1986; 34 (Suppl. 01) 10-7.
  • 11 Armstrong BG. Effect of measurement error on epidemiological studies of environmental and occupational exposures. Occup Environ Med 1998; 55 (Suppl. 10) 651-6.
  • 12 Brenner H, Gefeller O. Use of the positive predictive value to correct for disease misclassification in epidemiologic studies. Am J Epidemiol 1993; 138 (Suppl. 11) 1007-15.
  • 13 Holcroft CA, Spiegelman D. Design of validation studies for estimating the odds ratio of exposure- disease relationships when exposure is misclassified. Biometrics 1999; 55 (Suppl. 04) 1193-201.
  • 14 Marshall RJ. Confounder prevalence and stratum-specific relative risks: implications for misclassified and missing confounders. Epidemiology 1994; 5 (Suppl. 04) 439-42.
  • 15 Weinkam JJ, Rosenbaum WL, Sterling TD. Recovering true risks when multilevel exposure and covariables are both misclassified. Am J Epidemiol 1999; 150 (Suppl. 08) 886-91.
  • 16 Thürigen D, Spiegelman D, Blettner M, Heuer C, Brenner H. Measurement error correction using validation data: a review of methods and their applicability in case-control studies. Stat Methods Med Res 2000; 9 (Suppl. 05) 447-74.
  • 17 Tinnerberg H, Bjork J, Welinder H. Evaluation of occupational and leisure time exposure assessment in a population-based case control study on leukaemia. Int Arch Occup Environ Health 2001; 74 (Suppl. 08) 533-40.
  • 18 Lash TL, Silliman RA. A sensitivity analysis to separate bias due to confounding from bias due to predicting misclassification by a variable that does both. Epidemiology 2000; 11 (Suppl. 05) 544-9.
  • 19 Emsley CL, Gao S, Hall KS, Hendrie HC. Estimating odds ratios adjusting for misclassification in Alzheimer’s disease risk factor assessment. Stat Med 2000; 19 11-12 1523-30. L (pii).
  • 20 Koster M, Alfredsson L, Michelsen H, Vingard E, Kilbom A. Retrospective versus original information on physical and psychosocial exposure at work. Scand J Work Environ Health 1999; 25 (Suppl. 05) 410-4.
  • 21 Hsieh CC. The effect of non-differential outcome misclassification on estimates of the attributable and prevented fraction. Stat Med 1991; 10 (Suppl. 03) 361-73.
  • 22 Hsieh CC, Walter SD. The effect of non-differential exposure misclassification on estimates of the attributable and prevented fraction. Stat Med 1988; 7 (Suppl. 10) 1073-85.
  • 23 Walter SD. Effects of interaction, confounding and observational error on attributable risk estimation. Am J Epidemiol 1983; 117 (Suppl. 05) 598-604.
  • 24 Levin ML. The occurrence of lung cancer in man. Acta Unio Int Contra Cancrum 1953; 9: 531-41.
  • 25 Walter SD. Prevention for multifactorial diseases. Am J Epidemiol 1980; 112 (Suppl. 03) 409-16.
  • 26 Whittemore AS. Statistical methods for estimating attributable risk from retrospective data. Stat Med 1982; 1 (Suppl. 03) 229-43.
  • 27 Benichou J. A review of adjusted estimators of attributable risk. Stat Methods Med Res 2001; 10 (Suppl. 03) 195-216.
  • 28 Vogel C, Land M, Gefeller O. Effects of independent non-differential misclassification on the attributable risk. In: Gaul W, Ritter G. eds. Classification, Automation, and New Media, Proceedings of the 24th Annual Conference of the GfKl. Berlin: Springer; 2002: 515-22.
  • 29 Greenland S. Basic methods of sensitivity analysis and external adjustment. In: Rothman KJ, Greenland S. editors. Modern Epidemiology. 2 ed. Philadelphia: Lippincott-Raven; 1998. p. 343-357.
  • 30 Vogel C, Gefeller O. The effects of simultaneous misclassification on the attributable risk. In: Opitz O, Schwaiger M. eds. Exploratory Data Analysis in Empirical Research, Proceedings of the 25th Annual Conference of the GfKl. Berlin: Springer; 2002: 392-400.