This health network-based retrospective study will be conducted using data from University of Pittsburgh Medical Center (UPMC), a semi-closed health network in Western Pennsylvania.
Two UPMC hospitals will be utilized to identify the source population in this study – UPMC Magee-Women’s Hospital, which serves Pittsburgh, Pennsylvania and the surrounding area and has over 9,000 deliveries per year, and UPMC Hamot Hospital, which serves Erie, Pennsylvania and has over 3,000 deliveries per year. The integrated electronic health record at these hospitals allows for >95% of mothers and their infants to be linked within the EHR allowing for accurate characterization of maternal immunization. Furthermore, both hospitals participate in the Magee Obstetric Maternal & Infant (MOMI) Database, which collects over 300 maternal and infant variables from pregnancy, delivery, and postpartum for every pregnancy, allowing for standardized data collection for >95% of deliveries occurring in these hospitals.
The University of Pittsburgh and UPMC’s bioinformatics group (R3) can routinely obtain structured data elements from the electronic health record including maternal and infant immunization records, laboratory results, medications, diagnosis codes, demographic and birth characteristics. Additional details of specific medical encounters, including details on reported clinical symptoms, vital signs, and physical exam findings, can be obtained using unstructured data abstraction with trained data abstractors, who enter data directly into electronic data collection tool. This detailed abstraction will allow for accurate characterization of viral infection that includes LRTD.