Abstract
INTRODUCTION: Recent reports suggest rising intensity of emergency department (ED) billing practices, sparking concerns that this may represent up-coding. However, it may reflect increasing severity and complexity of care in the ED population. We hypothesize that this in part may be reflected in more severe manifestations of illness as indicated by vital sign abnormalities. METHODS: Using 18 years of data from the National Hospital Ambulatory Medical Care Survey, we conducted a retrospective secondary analysis of adults (>18 years). We assessed standard vital signs using weighted descriptive statistics (heart rate, oxygen saturation, temperature, and systolic blood pressure [SBP]), as well as hypotension and tachycardia. Finally, we evaluated for differing effects stratifying by subpopulations of interest, including age (<65 vs ≥65), payer type, arrival by ambulance, and high-risk diagnoses. RESULTS: In total there were 418,849 observations representing 1,745,368,303 ED visits. We found only minimal variations in vital signs over the study period: heart rate (median 85, interquartile range [IQR] 74-97); oxygen saturation (median 98, IQR 97-99); temperature (median 98.1, IQR 97.6-98.6); and SBP (median 134, IQR 120-149). Similar results were found among the subpopulations tested. The proportion of visits with hypotension decreased (first/last year difference 0.5% [95% CI 0.2%-0.7%]) while there was no difference in the proportion of patients with tachycardia. CONCLUSIONS: Arrival vital signs in the ED have largely remained unchanged or improved over the most recent 18 years of nationally representative data, even for key subpopulations. Greater intensity in ED billing practices is not explained by changes in arrival vital signs.