Association between blood pressure, race, ethnicity and likelihood of admission to the hospital from United States emergency departments – A cross sectional study

Abstract

Background: The emergency department (ED) has emerged as the primary portal for entry to the hospital for most patients with health care problems, including hypertension. Hypertension is the most important risk factor for heart disease. Disparities may exist in access to hospitalization across race/ethnicity. Our objective was to estimate how the likelihood of hospital admission based on blood pressure (BP) was modified by race/ethnicity. Methods: We used data from the 2014 National Hospital Ambulatory Medical Care Survey, a representative sample of non-federal, U.S. emergency department visits. We plotted probability of admission by blood pressure stratified by race/ethnicity to assess for a linear relationship. We then fit logistic regression models that adjusted for other potential confounders including patient-, visit-, and hospital-level factors. All analyses were conducted with relevant SURVEY functions in SAS to account for design. Results: Just over 21,000 visits were included in the study, representing approximately 1.4 million U.S. ED visits. We included the range of systolic blood pressure from 110 to 180 mmHg based on the linear relationship with probability of admission. We found the odds ratio for admission was 1.11 [95% CI: (1.06, 1.18)] for each 10 mmHg rise in systolic blood pressure in the unadjusted analysis. In the final adjusted model accounting for confounders, we found that the relationship between BP and admission was no longer significant 0.96 [0.91 to 1.01]. Whites were substantially more likely to be admitted compared to Blacks and Hispanics at odds ratio 1.5 [1.2 to 2]. Conclusions: The relationship between BP and hospital admission is complicated. Blacks and Hispanics appear less likely to be admitted to the hospital from the ED at a given level of blood pressure even after accounting for triage severity, and other individual and hospital level factors. Further research is needed to better understand this disparity.