The types of exposures studied in cancer epidemiology pose special challenges from a data analytic standpoint. For example, nutritional exposures form the basis for many etiologic hypotheses concerning cancer. However, nutrient intake is difficult to measure precisely. The degree of measurement error may mask true underlying relationships due to the regression dilution problem. It is the role of measurement error correction methods to estimate the relationship between cancer incidence and """"""""true"""""""" nutrient intake. To accomplish this requires data from both a main study where disease and the surrogate exposure are measured, and a validation study where both the surrogate measure and the gold standard for nutrient intake are assessed. In this proposal, we seek to extend the previous work on measurement error correction which is based on intake reported at a single survey to the situation where diet is reported at multiple surveys over time. Another focus of this proposal is to extend previous measurement error models which were specified at the nutrient level to models specified at the food level, which is the level at which people actually report their intake. The issue is that different foods have different degrees of measurement error, which should be taken into account when considering measurement error both at the food and nutrient level. Another issue is that many nutrients have contributions from both foods and supplements which are likely to have differing degrees of measurement error. We also consider measurement error issues for non-nutritional exposures in cancer epidemiology. For example, proband studies using family registers for a specific type of cancer collect data from a cancer case and other nonaffected people in the same family. Special analytic methods are required to take account of the familial nature of the data. We propose to extend measurement error correction to be applicable to this type of data structure. Second, some exposure-disease relationships are inherently non-linear, and are best captured using splines (e.g., the relationship of skin cancer to low levels of arsenic in drinking water). We propose to extend measurement error correction methods to curves fitted with splines. Also, ROC curves are used in imaging studies for breast cancer detection but are based in imperfect continuous measures. We propose to assess the impact of measurement error on the estimation of the ROC curve. Finally, there is inevitably misclassification in the pathological classification of disease stage in some types of cancer (e.g., pancreatic cancer). We propose to investigate the impact of this misclassification on estimated racial differences in survival for persons with pancreatic cancer.
| Aschard, Hugues; Spiegelman, Donna; Laville, Vincent et al. (2018) A test for gene-environment interaction in the presence of measurement error in the environmental variable. Genet Epidemiol 42:250-264 |
| Kim, Sehee; Li, Yi; Spiegelman, Donna (2016) A semiparametric copula method for Cox models with covariate measurement error. Lifetime Data Anal 22:1-16 |
| Khudyakov, Polyna; Gorfine, Malka; Zucker, David et al. (2015) The impact of covariate measurement error on risk prediction. Stat Med 34:2353-67 |
| Rosner, Bernard; Tworoger, Shelley; Qiu, Weiliang (2015) Correcting AUC for Measurement Error. J Biom Biostat 6: |
| Yi, Grace Y; Ma, Yanyuan; Spiegelman, Donna et al. (2015) Functional and Structural Methods with Mixed Measurement Error and Misclassification in Covariates. J Am Stat Assoc 110:681-696 |
| Rosner, Bernard; Hendrickson, Sara; Willett, Walter (2015) Optimal allocation of resources in a biomarker setting. Stat Med 34:297-306 |
| Wang, Molin; Liao, Xiaomei; Spiegelman, Donna (2013) Can efficiency be gained by correcting for misclassification? J Stat Plan Inference 143: |
| Zucker, David M; Gorfine, Malka; Li, Yi et al. (2013) A regularization corrected score method for nonlinear regression models with covariate error. Biometrics 69:80-90 |
| Liao, Xiaomei; Spiegelman, Donna; Carroll, Raymond J (2013) Regression calibration is valid when properly applied. Epidemiology 24:466-7 |
| Spiegelman, Donna; Logan, Roger; Grove, Douglas (2011) Regression calibration with heteroscedastic error variance. Int J Biostat 7:4 |
Showing the most recent 10 out of 51 publications
