Trends in use of echocardiography in hospitalized patients with syncope

National Cost Savings From Use of Artificial Intelligence Guided Echocardiography in the Assessment of Intermediate-Risk Patients With Syncope in the Emergency Department

Objectives

Our primary objective was to estimate the realistic impact of an artificial intelligence (AI)-based trans-thoracic echocardiogram (TTE)-first strategy on the annual national cost savings among eligible adult emergency department (ED) patients presenting with syncope in the United States. Our secondary outcomes were the estimated reduction in avoidable ED bed hours and comprehensive TTE studies.

Methods

Using publicly available estimates for inputs such as the size of the adult ED syncope population, typical disposition and risk stratification proportions, and frequency of comprehensive TTE studies, we created a model and ran 1000 trials of a Monte Carlo simulation. Using this simulation, we modeled the national annual cost savings and potential bed hours averted through the impact of avoiding comprehensive TTE studies. We report the descriptive statistics modeling the distribution of all endpoints.

Results

An AI-assisted TTE-first strategy was estimated to save a mean (±SD) of $815 million (±$260 million) by avoiding 468,000 (±141,000) comprehensive TTE studies resulting in 12,500,000 (±4,600,000) bed hours saved.

Conclusion

If adopted widely, an AI-based TTE-first strategy applied to eligible ED patients presenting with syncope could yield substantial benefits by averting avoidable comprehensive TTE studies and saving bed hours.

Regional variation in outcomes and healthcare resources utilization in, emergency department visits for syncope

BACKGROUND

Management of patients with syncope lacks standardization. We sought to assess regional variation in hospitalization rates and resource utilization of patients with syncope.

METHODS

We identified adults with syncope using the Nationwide Emergency Department Sample from years 2006 to 2014. Demographics and comorbidity characteristics were compared across geographic regions in the US. Multiple regression was conducted to compare outcomes.

RESULTS

9,132,176 adults presented with syncope. Syncope in the Northeast (n = 1,831,889) accounted for 20.1% of visits; 22.6% in the Midwest (n = 2,060,940), 38.5% in the South (n = 3,527,814) and 18.7% in the West (n = 1,711,533). Mean age was 56 years with 57.7% being female. The Northeast had the highest risk-adjusted hospitalization rate (24.5%) followed by the South (18.6%, ORadj 0.58; 95% CI 0.52-0.65, p < 0.001), the Midwest (17.2%, ORadj 0.51; 95% CI 0.46-0.58, p < 0.001) and West (15.8%, ORadj 0.45; 95% CI 0.39-0.51, p < 0.001). Risk-adjusted rates of syncope hospitalizations significantly declined from 25.8% (95% CI 24.8%-26.7%) in 2006 to 11.7% (95% CI 11.0%-12.5%) in 2014 (Ptrend < 0.001). The Northeast had the lowest risk-adjusted ED (Emergency Department) service charges per visit ($3320) followed by the Midwest ($4675, IRRadj 1.41; 95% CI 1.30-1.52, p < 0.001), the West ($4814, IRRadj 1.45; 95% CI 1.31-1.60, p < 0.001) and South ($4969, IRRadj 1.50; 95% CI 1.38-1.62, p < 0.001). Service charges increased from $3047/visit (95% CI $2912-$3182) in 2006 to $6267/visit (95% CI $5947-$6586) in 2014 (Ptrend < 0.001).

CONCLUSIONS

Significant regional variability in hospitalization rates and ED service charges exist among patients with syncope. Standardizing practices may be needed to reduce variability.