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

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.

[1]  R. Prentice.  Covariate measurement errors and parameter estimation in a failure time regression model , 1982 .

[2]  R. Prentice,et al.  Further results on covariate measurement errors in cohort studies with time to response data. , 1989, Statistics in medicine.

[3]  D. Ruppert,et al.  Random-Effect Models in Nonlinear Regression with Applications to Bioassay , 1989 .

[4]  R. Burnett,et al.  Bias due to aggregation of individual covariates in the Cox regression model. , 2004 .

[5]  H. Friedmann.  Health Effects of Exposure to Radon (BEIR VI Report), National Research Council, National Academic Press, Washington DC, ISBN 0-309-05645-4 , 2000 .

[6]  L Kreienbrock,et al.  Two dimensions of measurement error: Classical and Berkson error in residential radon exposure assessment , 2004, Journal of Exposure Analysis and Environmental Epidemiology.

[7]  H. Zeeb,et al.  Mortality from cancer and other causes among male airline cockpit crew in Europe , 2003, International journal of cancer.

[8]  R. Carroll,et al.  Segmented regression with errors in predictors: semi-parametric and parametric methods. , 1997, Statistics in medicine.

[9]  H. Wichmann,et al.  CHARACTERISTICS OF THE GERMAN URANIUM MINERS COHORT STUDY , 2002, Health physics.

[10]  D. Hall.  Measurement Error in Nonlinear Models: A Modern Perspective , 2008 .

[11]  Ralf Bender,et al.  Generating survival times to simulate Cox proportional hazards models , 2005, Statistics in medicine.

[12]  N. L. Johnson,et al.  Survival Models and Data Analysis , 1982 .

[13]  Thomas Augustin,et al.  Cox’s Proportional Hazards Model under Covariate Measurement Error , 2002 .

[14]  P. Guénel,et al.  Job exposure matrices in industry. , 1993, International journal of epidemiology.

[15]  John D. Kalbfleisch,et al.  Misspecified proportional hazard models , 1986 .

[16]  K. Liestøl,et al.  Attenuation caused by infrequently updated covariates in survival analysis. , 2003, Biostatistics.

[17]  H. Wichmann,et al.  The German uranium miners cohort study: feasibility and first results. , 1999, Radiation research.

[18]  Thomas Augustin,et al.  Cox's Proportional Hazards Model under Covariate Measurement Error - A Review and Comparison of Methods , 2001 .

[19]  D. Ruppert,et al.  Measurement Error in Nonlinear Models , 1995 .

[20]  Raymond J. Carroll,et al.  Measurement error in nonlinear models: a modern perspective , 2006 .