COVID-19 pandemic-associated changes in overall emergency department visits by age group, race, and ethnicity - United States, January 2019-April 2022

Abstract

Background: ED data are an important source of surveillance data for monitoring many conditions of public health concern and are especially useful in describing trends related to new, or unusual public health events. The COVID-19 pandemic led to significant changes in emergency care seeking behavior. We described the trends in all-cause emergency department (ED) visit volumes by race, ethnicity, and age using ED data from the National Syndromic Surveillance Program (NSSP) during December 30, 2018-April 2, 2022.

Methods: We described total and race, ethnicity, and age group-specific ED visit volumes during the COVID-19 pandemic by comparing quarterly visit volumes during the pandemic period to the relevant quarters in 2019. We quantified the variability of ED visits volumes by calculating the coefficient of variation in mean weekly ED visit volume for each quarter during Q1 2019-Q1 2022.

Results: Overall ED visits dropped by 32% during Q2 2020, when the COVID-19 pandemic began, then rebounded to 2019 baseline by Q2 2021. ED visits for all race, ethnicity, and age groups similarly dropped in Q2 2020 and adults of all race and ethnicity groups rebounded to at or above pre-pandemic levels while children remained at or below the pre-pandemic baseline except during Q3 2021. There was larger variation in mean weekly ED visits compared to the respective quarter in 2019 for 6 of 9 quarters during Q1 2020-Q1 2022.

Conclusions: ED utilization fluctuated considerably during the COVID-19 pandemic. Overall ED visits returned to within 5% of 2019 baseline during Q2 2021, however, ED visits among children did not return to the 2019 baseline until Q3 2021, then again dropped below the 2019 baseline in Q4 2021. Trends in ED visit volumes were similar among race and ethnicity groups but differed by age group. Monitoring ED data stratified by race, ethnicity and age can help understand healthcare utilization trends and overall burden on the healthcare system as well as facilitate rapid identification and response to public health threats that may disproportionately affect certain populations.

Keywords: COVID-19; Emergency department; Race and ethnicity; Surveillance.

Fig. 1

Weekly trends of total all-cause ED visits (A) and ED visits by age group, race, and ethnicity (B)— National Syndromic Surveillance Program (NSSP)*, United States, December 30, 2018–March 27, 2022. Abbreviations: ED = emergency department; NSSP = National Syndromic Surveillance Program. * NSSP is a collaboration among CDC, local, and state health departments, and federal, academic, and private sector partners. To reduce artifactual impact from changes in reporting patterns, analyses were restricted to facilities with a coefficient of variation ≤40 and ≥ 75% complete information on race and ethnicity throughout 2019–2022. Visit data from 1595 facilities were included in this analysis. All facilities from Connecticut; District of Columbia; Georgia; Linn County, IA; Maryland; New Jersey; New York City; North Carolina; Ohio; Pennsylvania; San Mateo, CA; and Texas Region 6/5 were not included in this analysis because they did not meet the inclusion criteria.

Fig. 2

Weekly Visit Volume Mean, Standard Deviation, and Coefficient of Variation (CoV) by Quarter — National Syndromic Surveillance Program (NSSP)*, United States, December 30, 2018–March 27, 2022. Abbreviations: ED = emergency department; NSSP = National Syndromic Surveillance Program; CoV = coefficient of variation. * NSSP is a collaboration among CDC, local, and state health departments, and federal, academic, and private sector partners. To reduce artifactual impact from changes in reporting patterns, analyses were restricted to facilities with a coefficient of variation ≤40 and ≥ 75% complete information on race and ethnicity throughout 2019–2022. Visit data from 1595 facilities were included in this analysis. All facilities from Connecticut; District of Colombia; Georgia; Linn County, IA; Maryland; New Jersey; New York City; North Carolina; Ohio; Pennsylvania; San Mateo, CA; and Texas Region 6/5 were not included in this analysis because they did not meet the inclusion criteria. Coefficient of variation was calculated by dividing the mean weekly ED visit volume by the standard deviation for each quarter. Higher CoV corresponds to larger variation in weekly ED visit volumes.