Abstract Placental abruption is an obstetrical complication defined as premature placental separation. Despite its elusive etiology, it is believed to be the consequence of acute stimuli – ischemia, inflammation, and oxidative stress at the maternal‐fetal interface – associated with rupture of the decidual artery, resulting in (premature) placental separation. The study of abruption and maternal and newborn long‐term cardiovascular and hemorrhagic/thrombotic cerebrovascular events is the central focus of the proposed project. We will examine the long‐term impact of abruption on rates of cardiovascular and cerebrovascular morbidity and mortality in women and in their children. We will also investigate the relationship of clinical classification (mild and severe forms) of abruption, risks based on abruption across successive pregnancies, and abruption in twin pregnancies, on rates of cardiovascular and cerebrovascular events. We will undertake a causal mediation analysis to evaluate the extent to which these associations may be mediated through (i) preterm delivery and (ii) small for gestational age births. We will perform this analysis by clinical classification of abruption and estimate the extent of mediation following corrections for both measured (socio‐demographic characteristics, including maternal smoking, comorbid medical conditions, and obstetrical events), and unmeasured confounding. A unique aspect of this project will be to identify, through applications of Support Vector Machines and Deep Learning algorithms, subsets of women at high risk for abruption and cardiovascular and cerebrovascular mortality and morbidity. Finally, we will examine whether maternal race/ethnicity and socioeconomic status are effect measure modifiers of the association between abruption and risks of cardiovascular and cerebrovascular events. We propose to address these aims through a large population‐based epidemiologic study, utilizing data from the Myocardial Infarction Data Acquisition System (MIDAS), a New Jersey statewide database of all patients admitted to all non‐federal acute care hospitals in NJ with a CVD diagnosis, with longitudinal follow‐up of up to 30 years. The MIDAS data will be linked to the NJ fetal death and linked live birth‐ infant death data with associated maternal and newborn hospitalization data between 1980‐2017 to create one of the largest and most comprehensive databases in the US to evaluate the extent to which sentinel events in pregnancy impart lasting risk for women's and children's health later in life. This project will provide unprecedented opportunities to address public health, policy implications and clinical screening recommendations of women during the period following delivery regarding risk susceptibility to cardiovascular and cerebrovascular disease.