Dichlorodiphenyldichloroethane (DDE) and polychlorinated biphenyls (PCBs) have been suggested to affect the timing of menopause. However, results from studies on DDE/PCBs and age at menopause have been inconsistent and may be limited by exposure assessment at widely divergent ages, remote from the time of etiologic relevance. Attempts to estimate measures at a common and etiologically appropriate age have been limited by the paucity of data on the predictors of long-term changes in organochlorines. Understanding the dynamics of individual body burdens of these organochlorines and the factors that predict them would improve exposure assessment, and determining whether DDE/PCBs affect age at menopause would contribute to the knowledge of their public health significance.

Data were collected from 512 participants from a 1978-1982 baseline study and a 2003-2004 follow-up study. Baseline and follow-up organochlorine measures from 123 participants were used to characterize the intra-individual changes and predictors of change in DDE/PCB levels, in order to provide a model for exposure estimation. Interview and ovarian hormone data at follow-up were used to classify menopausal status and define age at menopause, and the effect on timing of menopause from DDE/PCB exposures, estimated at age 40, was evaluated.

Serum DDE and PCBs dramatically declined (median drop of 84% and 55%, respectively) over the follow-up period. Baseline levels were strongly correlated with and predictive of follow-up levels (Spearman correlations (r_s): 0.72 for DDE; 0.43 for PCBs). Prediction of follow-up PCBs substantially improved with data on initial concentration, lactation duration, baseline body mass index, and percent change in body fat (r_s = 0.75 between predicted and actual follow-up levels), whereas DDE prediction improved slightly (r _s = 0.83). Effect estimates for age at menopause were inconsistent across organchlorine categories (compared to the lowest exposure category, menopause occurred -0.02, 0.4, 0.7, -0.3 years later across ascending categories of DDE, and -0.4, 0.4, 0.8, and 0.6 years later across PCB categories).

Results suggest a single organochlorine measure provides considerable information on relative ranking at distant times and prediction is improved with data on a few easily collected variables. However, the data provide no evidence for an effect of DDE/PCBs on age at menopause.