Interhospital variability in hospital admissions for patients with low-risk syncope presenting to the emergency department

BACKGROUND Guidelines and risk scores have sought to standardize the management of syncope in the emergency department (ED), but variation in practice remains. OBJECTIVE The purpose of this study was to explore factors associated with admission for patients presenting to the ED with low- risk syncope. METHODS Our study population included adult patients in the Nationwide Emergency Department Sample between 2006 and 2019 who presented to an ED with a primary diagnosis of syncope. Multivariable hierarchical logistic regression analyses determined the association of patient or hospital factors with admission. Reference effect measures methodology assessed the relative contributions of patient, hospital, and unmeasured hospital factors. RESULTS Of the 3,206,739 qualifying encounters during the study period, 804,398 (25.1%) met low-risk criteria. Of these patients, 20,260 were admitted to the hospital (2.5%). Factors associated with increased odds of admission included increasing age and weekend presentation to the hospital, while female sex, lack of medical insurance, hospital region, teaching status, and higher ED volume decile were associated with lower odds of admission. Reference effect measures methodology demonstrated that unmeasured site variability contributed the widest range of odds for admission (odds ratio [OR] 5th percentile vs 95th percentile 0.23-4.38) compared with the composite patient (OR 0.33-3.68) or hospital (OR 0.65-1.30) factors. CONCLUSION Admission patterns for low-risk syncope varies widely across institutions. Unmeasured site variation contributes significantly fi cantly to the variability in admission rates, suggesting which hospital a patient presents to plays a disproportionate role in admission decisions. Further guidance to reduce practice variation in syncope care in the ED is needed.