Background: Syncope accounts for 1% of US emergency department (ED) visits. Because the etiology of syncope is often benign, diagnostic testing for syncope can be low yield and routine hospitalization has not been shown to improve outcomes. Thus, professional society guidelines from cardiology and emergency medicine have advocated shifting priorities in the decision to hospitalize patients from a focus on determining the etiology of syncope to stratifying the need for hospitalization by patients’ cardiac risk. Changes in utilization and clinical outcomes as a result of this paradigm shifts are not known. Thus, our objective was to characterize longitudinal changes in hospitalization and readmission rates of patients presenting to the ED with syncope.
Methods: Using linked State Inpatient and Emergency Department Databases from the Agency for Healthcare Research and Quality, we conducted a retrospective study of all ED visits in 12 geographically diverse states in 2009 and 2013. First, we calculated annual incidence rates of syncope ED visits and subsequent hospitalizations. We then calculated 30-day ED revisit and hospital readmission rates for the subset of 3 states with readmission identifiers. Differences between years were estimated using mixed effect regression modeling adjusted for patient clinical and demographic characteristics and hospital random effects.
Results: We identified 471,379 ED visits for syncope in 2009 and 535,521 ED visits in 2013. The annual incidence of syncope ED visits increased by 11.4% from 697 to 777 visits per 100,000 persons. Rates of subsequent hospitalization decreased from 41.2% in 2009 to 29.0% in 2013 (Adjusted Odds Ratio [AOR] 0.50; 95% CI [0.47-0.52]). Though age was the strongest predictor of hospitalization, hospitalization rates decreased across all age groups, (7.2% to 4.7% for patients ages 18-25, p<.0001, and 66.6% to 52.3% for patients age ≥85, p<.0001). The 30-day ED revisit rate following an index syncope ED visit decreased for patients discharged from the ED during the index visit (14.2% in 2009 to 13.7% in 2013; AOR 0.96 [95% CI 0.93-0.98]) and did not significantly change for patients admitted during the index visit (16.0% in 2009 to 16.5% in 2013; AOR 1.01 [95% CI 0.98-1.05]). The 30-day readmission rate decreased for patients discharged from the ED during the index visit (4.3% in 2009 to 3.9% in 2013, p<.0001) and did not significantly change for patients admitted during the index visit (9.7% in 2009 to 9.5% in 2013, p=.19).
Conclusions: Although the incidence of ED visits for syncope increased from 2009 to 2013, hospitalization rates for declined by 25% without an adverse effect on ED revisits or hospital readmissions.