Josiah Brown Poster Abstract


Alexander S. Chen
Joseph A. Ladapo, John Mafi
John N. Mafi, MD, MPH, Peter Smulowitz, MD, Joseph A. Ladapo, MD, PhD, Robert H. Brook, MD, ScD, Bruce E. Landon, MD, MBA
Comparing the Value and Efficiency of Emergency Care among U.S. Nurse Practitioners, Physician Assistants, and Physicians
CTSI Summer Program

Many physicians believe that nurse practitioners and physician assistants (NPs and PAs) provide inferior value of emergency care compared with physicians. Because this perception lacks empirical evidence, we used nationally representative data on emergency department visits to test this belief. We analyzed nationally representative data from the National Hospital Ambulatory Medical Care Survey on visits to ED providers from 2009–2013. This analysis includes 100,535 physician visits and 20,387 NP/PA visits, representing around 515 million ED visits in the study period. We used logistic regression models weighted to reflect national estimates and adjusted for age, sex, ethnicity, triage severity, comorbidities, insurance status, and geographic location. We looked at 5 Choosing WiselyTM overuse measures and 2 misuse measures: use of (1) CT/MRI for back pain, headache, or syncope, (2) plain x-rays for back pain, (3) any antibiotics for upper respiratory infection (URI) or skin abscess after drainage, (4) opioid medications for back pain or headache, (5) low-value hospitalizations for uncomplicated back pain, headache, syncope, URI, and skin abscess, (6) inappropriate antibiotics for simple urinary tract infection, and (7) Beers Criteria inappropriate medications in older adults (e.g., benzodiazapines). We found that NP/PAs prescribed more medications and low value antibiotics when compared to physicians after controlling for patient characteristics but practice patterns for other services remained similar, suggesting a similar value and efficiency of care among the measures we studied. Some future improvements include using propensity weighting to adjust for providers seeing different populations of patients.