Bias in relative odds estimation owing to imprecise measurement of correlated exposures.

A series of graphs is presented that show the estimated degree of bias in logistic coefficient estimates for two correlated continuous exposures measured with imprecision. These graphs indicate that even when the correlation coefficient between the exposure of interest and a correlated exposure is as low as 0.2, imprecision in the measurement of the latter exposure can result in at least as serious bias in the logistic coefficient estimate for the exposure of interest as measurement imprecision in the exposure of interest itself. The implications for the design and interpretation of epidemiological studies are discussed.

[1]  G R Howe,et al.  Methodological issues in case-control studies. III--The effect of joint misclassification of risk factors and confounding factors upon estimation and power. , 1984, International journal of epidemiology.

[2]  D R Jacobs,et al.  High density lipoprotein cholesterol as a predictor of cardiovascular disease mortality in men and women: the follow-up study of the Lipid Research Clinics Prevalence Study. , 1990, American journal of epidemiology.

[3]  A. Phillips,et al.  Declaring independence: why we should be cautious. , 1990, Journal of epidemiology and community health.

[4]  N. Day,et al.  Misclassification in more than one factor in a case-control study: a combination of Mantel-Haenszel and maximum likelihood approaches. , 1989, Statistics in medicine.

[5]  B Rosner,et al.  Correction of logistic regression relative risk estimates and confidence intervals for systematic within-person measurement error. , 2006, Statistics in medicine.

[6]  W. Winkelstein,et al.  Smoking and cervical cancer--current status: a review. , 1990, American journal of epidemiology.

[7]  J M Robins,et al.  Confounding and misclassification. , 1985, American journal of epidemiology.

[8]  A A Tsiatis,et al.  The impact that group sequential tests would have made on ECOG clinical trials. , 1989, Statistics in medicine.

[9]  D A Savitz,et al.  Estimating and correcting for confounder misclassification. , 1989, American journal of epidemiology.

[10]  J Kaldor,et al.  Latent class analysis in chronic disease epidemiology. , 1985, Statistics in medicine.

[11]  M. Gail,et al.  Introduction. Errors‐in‐variables workshop , 1989 .

[12]  A. Phillips,et al.  How independent are "independent" effects? Relative risk estimation when correlated exposures are measured imprecisely. , 1991, Journal of clinical epidemiology.

[13]  D. Gordon,et al.  High-density lipoprotein--the clinical implications of recent studies. , 1989, The New England journal of medicine.

[14]  B Rosner,et al.  Correction of logistic regression relative risk estimates and confidence intervals for measurement error: the case of multiple covariates measured with error. , 1990, American journal of epidemiology.

[15]  T. Gordon,et al.  The diet-heart idea. Outline of a history. , 1988, American journal of epidemiology.

[16]  L L Kupper,et al.  Effects of the use of unreliable surrogate variables on the validity of epidemiologic research studies. , 1984, American journal of epidemiology.

[17]  S Greenland,et al.  The effect of misclassification in the presence of covariates. , 1980, American journal of epidemiology.

[18]  D. English,et al.  A review of the effects of random measurement error on relative risk estimates in epidemiological studies. , 1989, International journal of epidemiology.

[19]  R. Bawol,et al.  Epidemiology as a guide to clinical decisions. The association between triglyceride and coronary heart disease. , 1980, The New England journal of medicine.