PROJECT SUMMARY-ABSTRACT It is believed that women who become pregnant again shortly after giving birth are at elevated risk of having a preterm infant in the subsequent pregnancy. The World Health Organization states: “After a live birth, the recommended interval before attempting the next pregnancy is at least 24 months in order to reduce the risk of adverse maternal, perinatal and infant outcomes.” The American Congress of Obstetricians and Gynecologists states “The optimal interval between delivery and subsequent pregnancy is 18 months to 5 years; the greatest risk of low birth weight and preterm birth occurs when the interconception interval is less than 6 months.” Recently, novel evidence has questioned short intervals’ causality for preterm birth. Studies in Australia, California, Canada and Sweden used case- crossover designs, comparing preterm birth following short versus longer intervals within women who had 3 livebirths, and therefore 2 intervals, answering the question “is risk of preterm birth after a short interval greater than that of the same woman after a longer interval?” Two studies found intervals <6 months not to carry increased risk, and two found slight increases for short intervals, although the risk was less than was observed in a corresponding conventional analysis. However, experts have raised concerns regarding the case-crossover design for this question. Because of these concerns, a critical need exists to assess the causality the association of short interpregnancy interval and preterm birth, so that existing recommendations can be either reaffirmed or modified. Our long term goal is to understand the etiology of preterm birth, and learn how it may be prevented. Our objective in this application is to establish the most appropriate way to evaluate the impact of interpregnancy interval on preterm birth; and explore the impact of study design on this question utilizing detailed data collected by the National Survey of Family Growth. We will employ modern causal inference methods to compare results of a conventional analysis; an analysis utilizing the enhanced NSFG data, including the degree of pregnancy intendedness and how long the first child was breastfed; and several within-woman crossover analyses. We propose 2 aims: 1)To determine whether after adjustment for enhanced time- varying confounders, women who become pregnant shortly after a birth are at increased risk of preterm delivery. 2) To describe the effects of study design, compare the properties of different designs, and determine the optimal design for the association of interval and preterm birth. Upon completion of this project we expect to have provided important data to inform official recommendations regarding pregnancy spacing, thereby influencing clinical care and policy, and improving maternal and neonatal health.