Using a data-fusion technique to increase detection of maternal fatalities that are not discovered using regular monitoring tactics was the goal of this study. From 2011 to 2018, researchers conducted a retrospective cohort research using electronic health data from a tertiary medical hospital. Maternal death cases were identified in two ways: 1) using a standard medical informatics service query of hospital data, and 2) using the TriNetX discovery tool as patients with a vital status of “deceased” and evidence of antecedent pregnancy exposure based on factors such as obstetric diagnostic codes or obstetric-related procedures. Potential causes of maternal death identified by the latter technique were subjected to chart review in order to validate the timing of death in comparison to the time of the last significant pregnancy and to characterize the characteristics of these fatalities. The primary outcome in the discovery cohort was pregnancy-associated death during pregnancy or within the first postpartum year compared to the hospital-identified group. Causes of mortality and comorbidities were secondary outcomes. The normal service inquiry revealed 23 maternal fatalities throughout the research period. The discovery method discovered 18 more patients who were verified on later record review to be pregnancy-related fatalities, representing a 78 percent increase in ascertainment, with a higher proportion representing postpartum deaths. The majority of newly discovered deaths were due to heart causes or other medical comorbidities. Although many hospital-ascertained cases were related to fatalities following the birth of an alive infant, the discovery tool revealed more deaths following early pregnancy loss or termination.Modern data analytics can help improve the detection of pregnancy-related mortality.
