ACCOUNTING FOR PREGNANCY DEPENDENCE IN EPIDEMIOLOGIC STUDIES OF REPRODUCTIVE OUTCOMES

The aim of this paper is to evaluate the contribution of hierarchical mixed models to the analysis of epidemiologic studies of environmental exposure and reproductive outcomes. We have re-analyzed, with a logis tic-normal mixed model, four studies investigating the relation betwee n the frequency of spontaneous abortions and paternal or maternal envi ronmental exposures. The data include multiple pregnancies for some wo men. The fitted models allow for between-woman variation of the propen sity for spontaneous abortion, by including a random intercept in the logistic model to adjust for within-woman correlations on pregnancy ou tcomes. We have discussed and implemented two estimation methods, maxi mum likelihood and Bayesian inference We found similar values in the v arious epidemiologic studies of the between-woman variance of the intr insic risk of spontaneous abortion. The size of this variance correspo nds to a substantial variability in risk between women. Indeed, the ri sk of spontaneous abortion calculated for ''nonexposed'' pregnancies, that is, with mother's age, birth order, tobacco consumption, and mate rnal environmental exposure equal to the referent class, can vary, acc ording to this model, from 2% to 17%.