Evaluation of recruitment strategies for prospective studies of spontaneous abortion.

The examination of spontaneous abortions may offer a sensitive index to potential adverse health effects due to environmental exposures. However, difficulties in ascertaining spontaneous abortions on a population basis, as well as problems in recruiting unselected women into studies of this nature, have severely limited the use of this end point in environmental epidemiology studies. The current research assesses the feasibility of recruiting a representative sample of women into a prospective pregnancy study. Four protocols varying in intensity of data collection were offered to the participants. Fifty-two percent of the eligible women agreed to participate. Seventy-four percent chose the most intensive protocol level. Overall compliance for urine collection and questionnaire completion were 80% and 81%, respectively. Four pregnancies were observed, which was the expected number for the study area during this time frame, suggesting that the majority of pregnancies occurring within a population could be ascertained with this approach.