| The goal of the study is to evaluate anovulation, menstrual cycle characteristics, and urinary estrogen and progesterone metabolite levels as well as their relationships to demographic and lifestyle factors, using hormone assay results from prospectively collected daily urine samples for multiple consecutive menstrual cycles, from a large group of working women. Assessment of metabolites of sex steroid hormones in daily urine samples often involves considerable complexity. Statistical theory and methods for reproductive epidemiology are consequently somewhat limited.; Algorithms to assess ovulatory status and the day of ovulation using daily urinary levels of estrogen (EIC) and progesterone (PdG) metabolites have been previously developed and applied. Several Bayesian models were used to estimate the probability of occurrence of anovulation, the sensitivity and specificity of two algorithms for assessing ovulatory status, and association between anovulation and risk factors, respectively. The Bayesian approach was taken because it allows for the determination of ovulatory status to be imperfect.; The linear mixed model (LMM) and the marginal logistic regression model (MLR) were used to examine menstrual outcomes, to account for the intra-woman correlation for the repeated outcome measures. A semiparametric smoothing spline mixed-effects (SPSM) model was used to examine covariate effects on the excretion profiles of urinary steroid hormone metabolites. Population-averaged curves were modeled for EIC and PdG, respectively.; The present study is one of the largest to date to include use of daily hormonal measures to assess risk factors for ovarian function. Our results show that a woman's individual demographic and lifestyle factors affect her ovarian function. We also found that the means of the continuous menstrual characteristics were also highly predicted by the length of the prior luteal phase.; In addition, this research explored the methodology for investigating the influence of host and environmental factors on the ovarian function of women, based on daily dairies and urine collection, and provided guidelines for the future studies. |
