Utilizing Serial Measures of Breast Cancer Risk Factors.
Abstract
Failure to account for errors-in-measurement of exposure and confounder variables can result in biased estimates of relative risk. We have developed a technique for correcting for measurement error when subjects have a variable number of repeated measurements and the average of the measurements is used as the subject's measure of exposure in the analysis. A bootstrap method for obtaining confidence intervals of the relative risk estimates, which takes into account the variability in the estimates of the correction factors, has also been devised. In addition, we have considered a model for adjusting hormone measurements for systematic fluctuations over the menstrual cycle, which will allow more valid comparisons of hormone levels between premenopausal cases and controls. Finally, most studies on the half-lives of environmental contaminents such as PCBs, which have been linked with an increased risk of breast cancer, have been based on small sample sizes and a limited number of repeated measurements. We have developed guidelines for choosing the number of subjects, repeated measurements, and time interval between measurements in the design of half-life studies.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 1996
- Accession Number
- ADA305839
Entities
People
- Mimi Y. Kim
Organizations
- NYU Langone Health