Objectives: The main objectives were to estimate the prevalence of predicted and unpredicted last births using a prospective approach and to estimate the prevalence of violence during the last pregnancy. In addition, the relationship between birth predictedness and violence during pregnancy was examined. Methods: The target population for this study was women who had participated in the 1994 Demographic and Health Survey (DHS) and lived in El Alto and La Paz Bolivia (n = 1308). In 1997, 816 women were located and re-interviewed. During this three-year interval, 127/816 women had given birth to their last child. Results: Of the last births that occurred during the three-year interval, 82% were unpredicted (18% were to women who stated in 1994 that they wanted to postpone childbirth for more than three years and 64% were to women who stated they wanted to wanted to forego childbearing entirely). Twenty-eight percent of women reported that they had experienced violence during their last pregnancy. No statistically significant relationship was found between birth predictedness and violence during their pregnancy. Conclusions: The majority of births that occurred in the three-year study interval were unpredicted. The prevalence of violence during pregnancy was alarmingly high among this sample of women. Further investigation on violence during pregnancy is needed and should be expanded to examine how violence during pregnancy impacts maternal and infant outcomes, which have remained poor in this country. In addition, the high rates of unpredicted births illustrate that work remains to be done in addressing women’s ability to control their fertility.