Utilizing Serial Measures of Breast Cancer Risk Factors
Annual rept. 31 Dec 96-30 Dec 97
NEW YORK UNIV MEDICAL CENTER NY
Pagination or Media Count:
A model for adjusting hormone measurements for systematic fluctuations over the menstrual cycle, which will allow more valid comparisons of hormone levels between pre-menopausal cases and controls, was developed. The model was utilized to adjust hormone levels for day of cycle in an analysis of the association between total estradiol and risk of pre-menopausal breast cancer. A bootstrap method for obtaining variances of the logistic regression parameters which takes into account the error associated with estimation of the adjusted hormone levels has also been devised and will be implemented in the final year of the study. In addition, a method for analyzing correlated panel data using a continuous time Markov model is proposed.
- Medicine and Medical Research
- Statistics and Probability