This poster presents progress made on a biostatistics project under the mentorship of Dr. Mary R. Boland from the Perelman School of Medicine’s Department of Biostatistics, Epidemiology, and Informatics. The study focused on geographic analysis of adverse pregnancy outcomes in Philadelphia using data from Electronic Health Records (EHR), the U.S. Census Bureau, and other public databases.

Over the course of the project, census data was utilized to add yearly variation within current and newly added covariates. The yearly variation for each neighborhood-level covariate was used to track and visualize changes within Philadelphia census tracts between 2010-2017.

Adverse pregnancy outcomes remain a major public health concern despite considerable advancements in maternal health over the recent decades. Stillbirth is the delivery of a baby who has passed away, after the 20th week of pregnancy. Although it is rare (1/100 pregnancies), stillbirth can be devastating. Preterm birth is when a baby is born before 37 weeks of pregnancy have been completed. This birth outcome is more common (1/10 of infants born in the U.S) and accounted for 17% of infant deaths in 2017. In the preliminary results, it was discovered that several significantly correlated covariates for both stillbirth and preterm birth. Additionally, it was discovered that the neighborhood-level covariate data has a possible application to COVID-19 outcomes.