Obstetric fistula is gaining visibility within national and international public health agencies. With increased attention to this dire condition comes the demand for better epidemiologic data. This article assesses the current state of knowledge regarding population-based estimates of the incidence and prevalence of obstetric fistulas; proposes a method for better estimating these rates; and discusses the feasibility of this method. We show that there are no solid population-based estimates of the numbers of obstetric fistulas anywhere (the estimate from the most rigorous study being based on only 2 reported fistula cases); and for advocacy purposes, we recommend using the estimates from the 1990 Global Burden of Disease report. To estimate the incidence and prevalence of obstetric fistula, we propose an adaptation of the sibling-based method for direct estimation of maternal mortality. A series of questions are proposed for this use and sample size calculations are provided. The questions may require refinement, however, and we invite research groups to consider testing them.