Monitoring reproductive aging in a 5-year prospective study:: aggregate and individual changes in steroid hormones and menstrual cycle lengths with age

Objective: We describe a 5-year prospective study of reproductive aging, and present analyses of steroid hormone and menstrual cycle changes with age. Design: Participants were college-educated white women, primarily of northern European ancestry, recruited from the Tremin Research Program on Women's Health (n = 156, 25-58 years). In each of 5 consecutive years, they collected daily urine specimens for 6 months and recorded menstrual bleeds for all months. Urine specimens were assayed for estrone-3-glucuronide (E1G) and pregnanediol-3-glucuronide (PDG), urinary metabolites of estradiol and progesterone. Using multilevel models, we estimated hormone and cycle-length trajectories for individual women and within- and between-woman variance by age. Results: At the aggregate level, PDG declined beginning in the 30s, E1G increased into the 40s before declining, and cycle length became more variable with age. Individual-level models revealed substantial hormonal variation across women, in both absolute levels and rates of change. Most women showed declining E1G by the late 40s, declining PDG in the 30s, and increasing mean cycle length in the 40s. Hormonal variation decreased with age; cycle length variation decreased and then increased. Within individual women, cycle lengths Were highly variable while hormone levels were more stable. Women differed more from each other in hormone levels than for cycle lengths. Conclusions: Aggregate-level analyses show general changes in steroid hormones and cycle length but cannot show variation within and across women. Individuals' cycle lengths were too variable to predict hormone levels. Clinicians should obtain more data on individual women's hormonal patterns when determining fertility or menopause treatments.