In this study, the primary goal is to examine the relationship between DDT (dichlorodiphenyltrichloroethane) levels and the odds of loss of clinically-recognized pregnancies amongst women in Limpopo, South Africa. The initial protocol (which will be referred to as Part 1) planned to enroll 2,400 non-pregnant women, draw blood for measurement of DDT exposure, and follow 1,200 pregnant participants to ascertain occurrence of miscarriage. An initial two-year pilot of 850 non-pregnant women was proposed to evaluate field procedures, recruitment strategies and the reproducibility of DDT levels. Of the 444 women enrolled in the first 10 months of the pilot of Part 1, only 27 have become pregnant. Although the field procedures have been well executed, the number of pregnancies has been low. Based upon the experience to date, the approach taken so far is unlikely to generate enough pregnancies to monitor the occurrence of miscarriages. Therefore, the revised protocol (referred to as Part 2) will modify enrollment criteria to include only pregnant women who are in the early stages of pregnancy, confirmed by rising human chorionic gonadotropin (hCG) blood levels taken 1 week apart or pregnant women who have a pregnancy loss within 4 weeks of enrollment. Participants who were enrolled in Part 1 of the pilot, who are still not pregnant at implementation of Part 2 will be withdrawn from the study, while pregnant participants from Part 1 will continue to be followed. In Part 2, a total of 2,400 pregnant women will be enrolled. Only those who meet the follow-up criteria (n=1,200) will be followed to determine the outcome of pregnancy. The reproducibility study of DDT levels amongst a subset of 200 women will also be dropped and replaced with an analysis using specimens that were collected in Part 1.