Background: Payment reforms and other policy initiatives are accelerating the shift of risk from payers to providers. As a result, population health management is playing an increasing role in decision making by providers, guided by the Medicare Triple Aim. Reducing avoidable hospitalizations is an important tool for achieving this aim, by increasing quality of care and containing hospital costs. Previous studies have examined the trends over time of emergency department (ED) visits for major diseases. However, there is very little data assessing ED visits related to the symptoms of these major diseases. This study examined the trends of ED visits for chest pain (CP), a symptom suggestive of coronary artery disease (CAD), and of shortness of breath (SOB), a symptom suggestive of congestive heart failure (CHF).
Methods: We conducted a population-based cross-sectional study to estimate ED visits for CP suggestive of CAD and for SOB suggestive of CHF in the US for the years 2006 through 2013 at encounter level, using the Nationwide Emergency Department Sample (NEDS) database. We defined CP suggestive of CAD as a principal diagnosis of CP for the ED visit (ICD-9-CM code: 786.5), with a CAD code (410-414) as a secondary diagnosis but no diagnosis for other serious conditions that may trigger CP (e.g., aortic dissection). We defined SOB suggestive of CHF as a principal diagnosis of SOB (ICD-9-CM code: 786.05) for the ED visit with a CHF code (428.0-428.4) as a secondary diagnosis but no diagnosis for other conditions that may trigger SOB (e.g., pneumothorax). Outcome measures included annual number of ED visits and subsequent admissions, weighted for national estimates (2006-2013). We performed a trend analysis in rates over time, which accounted for US census population, for ED visits and subsequent admissions, using a generalized linear regression model with a Poisson distribution and a Wald test.
Results: The number of ED visits for CP suggestive of CAD per 100,000 population increased 24.3% from 197 in 2006 to 245 in 2013 (p<0.01), while subsequent admissions for CP suggestive of CAD decreased by 36.1% from 90 in 2006 to 58 in 2013 (p<0.01). However, we found a consistently small number of ED visits for SOB suggestive of CHF over time, from 4 ED visits in 2006 to 5 ED visits in 2013 (p>0.1). Similarly, subsequent admissions for SOB suggestive of CHF were relatively low and stable, from 0.61 admissions per 100,000 in 2006 to 0.72 admissions in 2013 (p>0.1).
Conclusions: Our results showed an increasing trend for ED visits and a decreasing trend for subsequent admissions over time for CP suggestive of CAD. However, there appeared no change for ED visits and subsequent admissions over time for SOB suggestive of CHF. Future research is warranted to examine possible reasons for the different ED visit rates for symptoms associated with major diseases such as CAD and CHF.