134 Value of TriNetX national database to inform characteristics of observation patients assessed for acute syncope

Abstract

Objectives/Goals: Syncope is a common diagnosis in observation medicine, and characterization of observation patients is often limited to single unit, single center, or single payer systems. TriNetX, a federated deidentified multicenter national public database, provides an opportunity to study these patients from across the USA. Methods/Study Population: This retrospective cohort study queried data from 56 health care organizations (HCO) in TriNetX to examine differences between observation patients with syncope who required admission vs. those who were discharged. All observation stays with a diagnosis of syncope, defined by ICD-10, CPD, and SNOWMED codes, were queried. A total of 281,162 observation encounters were included in analysis, of which 46.4% (n = 130,357) were admitted and 53.6% (n = 150,805) were discharged. Data on age, gender, race, ethnicity, presence of congestive heart failure, EKG, and serum labs were collected for comparative analysis. T-test and Chi-square analyses were deployed with significance  =  p Results/Anticipated Results: The cohorts demonstrated statistically significant differences across all demographic factors, however, they were not clinically meaningful. Clinically significant differences include that only 72.8% of admitted patients and 68.3% of discharged patients had an EKG recorded in TriNetX during the period of observation (p Discussion/Significance of Impact: Patients admitted from observation status were more likely to have CHF, higher BNP, and pro-BNP values. TriNetX is a powerful tool to study patients across multiple hospital systems and payer types. Limitations, however, include incomplete data and inaccuracies among claims records.