Effect of Berkson measurement error on parameter estimates in Cox regression models

Abstract

We study the effect of additive and multiplicative Berkson measurement error in Cox proportional hazard model. By plotting the true and the observed survivor function and the true and the observed hazard function dependent on the exposure one can get ideas about the effect of this type of error on the estimation of the slope parameter corresponding to the variable measured with error. As an example, we analyze the measurement error in the situation of the German Uranium Miners Cohort Study both with graphical methods and with a simulation study. We do not see a substantial bias in the presence of small measurement error and in the rare disease case. Even the effect of a Berkson measurement error with high variance, which is not unrealistic in our example, is a negligible attenuation of the observed effect. However, this effect is more pronounced for multiplicative measurement error.

This is a preview of subscription content, access via your institution.

References

  1. Abrahamowicz MR, du Berger D, Krewski R, Burnett G, Bartlett R,Tamblyn M, Leffondre K (2004) Bias due to aggregation of individual covariates in the cox regression model. Am J Epidemiol 160:696–706

    Article  Google Scholar 

  2. Andersen PK, Liestol K (2003) Attentuation caused by infrequently updated covariates in survival analysis. Biostatistics 4:633–649

    MATH  Article  Google Scholar 

  3. Augustin T (2005) An exact corrected log-likelihood function for cox’s proportional hazards model. Scand J Stat 70:381–387

    Google Scholar 

  4. Augustin T, Schwarz R (2002) Cox’s proportional hazards model under covariate measurement error, A review and comparison of methods. In: Van Huffel S, Lemmerling P (eds). Total least squares and error-in-variables modelling: analysis, algorithms and applications. Kulwer, Dortrecht, pp. 175–184

    Google Scholar 

  5. Bender R, Augustin T, Blettner M (2005) Generating survival times to simulate Cox proportional hazards models. Stat Med 24:1713–1723

    Article  MathSciNet  Google Scholar 

  6. Blettner M, Zeeb H, Auvinen A, Ballard TJ, Caldora M, Eliasch H, Gundestrup M, Haldorsen T, Hammar N, Hammer GP, Irvine D, Langner I, Paridou A, Pukkala E, Rafnsson V, Storm H, Tulinius H, Tveten U, Tzonou A (2003) Mortality from cancer and other causes of male airline cockpit crew in europe. Int J Cancer 106:946–952

    Article  Google Scholar 

  7. Carroll RJ, Ruppert D, Stefanski LA, Crainiceanu CM (2006) Measurement error in nonlinear models. A modern perspective 2nd edn. Chapman and Hall, Boca Raton

    Google Scholar 

  8. Elandt-Johnson RC, Johnson NL (1980) Survival models and data analysis. Wiley, New York

    Google Scholar 

  9. Goldberg M, Kromhout H, Gunel PP, Fletcher AC, Grin M, Glass DC, Heederik D, Kauppinen T, Ponti A (1993) Job exposure matrices in industry. Int J Epidemiol 22:10–15

    Google Scholar 

  10. Heid I, Küchenhoff H, Miles J, Kreinbrock L, Wichmann HE (2004) Two dimensions of measurement error: classical and berkson error in residential radon exposure assessment. J Expos Anal Environ Epidemiol 14:365–377

    Article  Google Scholar 

  11. Kreuzer M, Brachner A, Lehmann F, Martignoni K, Wichmann H-E, Grosche B (2002) Characteristics of the german uranium miners cohort study. Health Phys 83:26–34

    Article  Google Scholar 

  12. Kreuzer M, Grosche B, Brachner A, Martignoni K, Schnelzer M, Schopka H-J, Brüske-Hohlfeld I, Wichmann H-E, Burkart W (1999) The german uranium miners cohort study: feasibility and first results. Radiat Res 153:56–58

    Article  Google Scholar 

  13. Küchenhoff H, Carroll RJ (1997) Segmented regression with errors in predictors: semi-parametric and parametric methods. Stat Med 16:169–188

    Article  Google Scholar 

  14. National Research Council (1999) Committee on health risks of exposure to radon (BEIR VI): Health effects of exposure to radon. National Academy, Washington, DC

    Google Scholar 

  15. Pepe MS, Self MS, Prentice RL (1989) Further results in covariate measurement errors in cohort with time to respones. Stat Med 8:829–838

    Google Scholar 

  16. Prentice RL (1982) Covariate measurement errors and parameter estimation in a failure time regression model. Biometrika 69:331–342

    MATH  Article  MathSciNet  Google Scholar 

  17. Rudemo M, Ruppert D, Streibig JC (1989) Random effect models in nonlinear regression with applications to bioassay. Biometrics 45:349–362

    MATH  Article  MathSciNet  Google Scholar 

  18. Struthers CA, Kalbfleisch J (1986) Misspecified proportional hazard models. Biometrika 73:363–369

    MATH  Article  MathSciNet  Google Scholar 

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Helmut Küchenhoff.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Küchenhoff, H., Bender, R. & Langner, I. Effect of Berkson measurement error on parameter estimates in Cox regression models. Lifetime Data Anal 13, 261–272 (2007). https://doi.org/10.1007/s10985-007-9036-2

Download citation

Keywords

  • Error in variables
  • Berkson error
  • Cox regression