Background: Syncope is a common reason for emergency department visits. Guidelines suggest minimal initial workup, including history, physical exam, and ECG testing. Additional cardiac testing is recommended in only high risk patients and neuroimaging is rarely indicated. Because of the low yield of neuroimaging, in 2012, the Choosing Wisely campaign recommended against its use for syncope. Our objective was to examine trends in cardiac testing and neuroimaging of patients presenting to the ED with syncope, before and after the Choosing Wisely recommendations and to describe hospital variation in testing rates.
Methods: We linked State Inpatient and Emergency Department Databases to conduct a retrospective study of all ED visits in 2009 and 2013 for 8 states reporting procedure utilization. We calculated rates of ECG, advanced cardiac testing (echocardiogram, stress testing, diagnostic catheterization) and neuroimaging (head CT, brain MRI, carotid ultrasound) for all adults with a discharge diagnosis of syncope. Differences between years were estimated using mixed effect regression modeling adjusted for patient characteristics, comorbidities, and hospital random effects.
Results: We identified 287,261 ED visits for syncope in 2009 and 315,221 ED visits in 2013 from 676 hospitals. Between 2009 and 2013, adjusted rates of ECG testing increased from 81.1% of discharges to 84.3% (p<.0001). Rates of advanced cardiac testing increased from 10.3% to 12.4% (p<.0001), driven primarily by a substantial increase in the use of echocardiograms (8.3% vs 11.3%, p<.0001). Rates of neuroimaging increased from 36.0% to 42.0% (p<.0001) with increased utilization of all tests. Rates of ECG, advanced cardiac testing and neuroimaging varied significantly between hospitals in both years (Figure). The median hospital-level change in testing between years was 1.6% (IQR -3.5 to 11.2) for ECG, 1.2% (IQR -1.8 to 5.9) for advanced cardiac imaging and 4.2% (IQR -5.3 to 18.4) for neuroimaging.
Conclusions: Among patients presenting to the ED with syncope, rates of both high- and low-value diagnostic testing increased between 2009 and 2013, with substantial variation between hospitals. Thus, the 2012 Choosing Wisely recommendations do not appear to have had a significant effect on testing for patients presenting with syncope.