Trends in emergency department visits related to acute alcohol consumption before and during the COVID-19 pandemic in the United States, 2018–2020

Background Excessive drinking accounts for more than 95,000 deaths annually in the United States. Coronavirus disease 2019 (COVID-19) pandemic-related factors (e.g., social, economic, policy) may have affected alcohol consumption. Emergency department (ED) visits involving acute alcohol consumption (referred to as “alcohol-related”) are a useful indicator for assessing changes in alcohol-related harms. Methods The 2018–2020 National Syndromic Surveillance Program data, which include nonfatal ED visits from facilities in 49 states and Washington, DC, were analyzed. Trends in the number of alcohol-related ED visits among people ≥15 years, and weekly alcohol-related ED visit rates (per 10,000 total visits) overall, by demographic characteristics, and quarter (Q) were assessed. Quarterly rates for 2018 and 2019 were averaged to increase baseline data stability. Results Alcohol-related visits accounted for 1.6% of 60,474,770 total visits (2018), 1.7% of 61,564,380 total visits (2019), and 1.8% of 52,174,507 total visits (2020). The number of alcohol-related ED visits generally increased during the first eight months of 2018 and 2019. However, it sharply declined at the onset of the COVID-19 pandemic in mid-March–mid-April 2020, before resuming pre-pandemic patterns. Alcohol-related ED visits per 10,000 were higher during quarters in 2020 than corresponding quarters in 2018–2019 (Q1: +7.3%, Q2: +23.8%, Q3: +9.7%, Q4: +6.5%). Conclusions Alcohol-related ED visit rates per 10,000 total visits increased during 2020 versus 2018–2019, with the greatest relative difference in the second quarter. Fewer people sought ED care in 2020 than 2018–2019 but alcohol-related visits declined to a lesser extent than total visits.


Introduction
Excessive alcohol use is responsible for more than 95,000 deaths each year in the US, shortening the lives of those who die by an average of 29 years ( Esser et al., 2020 ). About half of these deaths are associated with the effects of acute drinking (e.g., alcohol poisoning, injuries) ( Esser et al., 2020 ).
Some evidence suggests alcohol consumption may have increased during the COVID-19 pandemic. For example, a small survey of US adults showed a greater prevalence of binge drinking in April compared with February 2020, and found that respondents consumed a greater number of drinks per day ( Barbosa et al., 2020 ). Fluctuations in alcohol 2. Methods

Data source
The NSSP evolved from BioSense, which originally launched to provide surveillance for potential bioterrorism-related illnesses. The NSSP now facilitates monitoring changes over time for various public health problems (e.g., substance use ( Nolan et al., 2017 ) or drug overdose ( Holland et al., 2021 )). The NSSP receives deidentified ED visits from hospitals in 49 states (excluding Hawaii) and Washington, DC, with approximately 71% of all US EDs participating ( Centers for Disease Control and Prevention, 2021 ). Data in this study include December 31, 2017 through January 2, 2021 to align with the surveillance weeks to capture each day of 2018 and 2020. This activity was reviewed by the CDC and was conducted consistent with applicable federal law and CDC policy (see, e.g., 45 CFR part 46; 21 CFR part 56; 42 USC §241(d), 5 USC §552a, 44 USC §3501 et seq).

Query of ED visits related to acute alcohol consumption
A query was designed for monitoring ED visits associated with acute alcohol consumption based on discharge diagnosis codes and chief complaint text (see Supplemental Tables 1 and 2). The query includes International Classification of Diseases, Tenth Revision, Clinical Modification (ICD) − 10-CM diagnosis codes with evidence of acute alcohol use. This query does not include codes that only indicate chronic alcohol use (e.g., alcoholic liver cirrhosis) without an indication of acute drinking.

Analyses
We assessed trends in the number of alcohol-related ED visits among people aged ≥ 15 years and weekly alcohol-related ED visit rates (per 10,000 ED visits) overall, by population characteristics, and by quarter (Q) during 2020 compared with corresponding quarters during 2018-2019. Because of fluctuations in facilities reporting to the NSSP over time, data were included from the facilities that were consistently reporting throughout the study period so that only facilities with more complete data were included ( Radhakrishnan et al., 2022 ) (where the facility-level average percentages for reporting diagnosis codes to the NSSP was ≥ 70%). Analyses focused on rates per 10,000 ED visits because of the substantial decline in total ED visits during the onset of the COVID-19 pandemic Hartnett et al., 2020 ).
We calculated weekly rates and then calculated relative percent differences and standard deviations in quarterly rates, based on average weekly rates, of alcohol-related ED visits per 10,000 visits during 2020 compared with corresponding quarters during 2018-2019. We averaged the rates for corresponding quarters in 2018 and 2019 to increase the stability of the baseline data for comparisons with 2020 data. Analyses were conducted using RStudio version 1.3.1093-1.

Trends in the number of alcohol-related ED visits
The number of alcohol-related ED visits generally increased for the first eight months of 2018 and 2019. In 2020, the number sharply declined at the onset of the COVID-19 pandemic in mid-March-mid-April, coinciding with the US declaration of the COVID-19 national emergency on March 13, 2020 ( Fig. 1 A). The number of alcohol-related ED visits then sharply increased from mid-April to late-June, and then returned to levels similar to corresponding weeks in 2018 and 2019, before generally declining slowly until mid-December 2020, and increasing at the end of the year.

Trends in rates of alcohol-related ED visits
The average of weekly rates of alcohol-related ED visits per 10,000 total visits increased from 163.4 in 2018 and 166.7 in 2019 to 184.8 in 2020. In 2020, rates of alcohol-related ED visits per 10,000 total visits increased consistently from late-January to a peak in late-April through early-June, more sharply than in 2018 or 2019, and the peak rates were higher than any rate during 2018-2019 ( Fig. 1 B). Alcohol-related ED visit rates then declined until mid-December 2020, where rates were comparable to corresponding weekly rates during 2019. Towards late-December, alcohol-related ED visit rates increased and were higher in 2020 than 2018 or 2019.
Average quarterly alcohol-related ED visit rates per 10,000 visits were higher during 2020 than corresponding quarters in 2018-2019. Rates were 7.3% higher in Q1, 23.8% higher in Q2, 9.7% higher in Q3, and 6.5% higher in Q4 ( Table 1 ). For both males and females, alcoholrelated ED visit rates were higher in each quarter of 2020 than corresponding 2018-2019 quarters. Rates ranged from 3.5% (Q4)-16.7% (Q2) higher among males and from 6.0% (Q4)-25.0% (Q2) higher among females. Among adults aged ≥ 25 years, in each age group, rates were higher in all quarters of 2020 than 2018-2019. Among people aged 15-24 years, only the Q1 and Q2 rates were higher in 2020 than 2018-2019. By geographic region and within the 10 HHS regions, changes in alcohol-related ED visit rates during 2020 versus 2018-2019 varied by location.
Trends in the number and rates of alcohol-related ED visits by population groups are shown in the online supplement . Alcohol-related ED visit rates among both males and females were generally higher than during 2020 than 2018-2019. By age group, alcohol-related ED visit rates during 2020 were higher, on average, than 2018-2019. However, among people aged 15-24 years, rates were slightly lower in the second half of 2020 than 2018-2019. By geographic region and HHS region, alcohol-related ED visit rates peaked during mid-April to mid-June, and the rates remained somewhat elevated but followed more typical seasonal fluctuations during the second half of 2020.

Discussion
The study assessed more than 174 million total ED visits from facilities that contributed consistently to the NSSP during 2018-2020. We observed that the overall average rate of alcohol-related ED visits per 10,000 total visits increased to 184.8 in 2020, up from 163.4 in 2018 and 166.7 in 2019. The quarterly alcohol-related ED visit rates per 10,000 total visits were 7-24% higher during quarters in 2020 than corresponding quarters in 2018-2019. The greatest relative difference in alcoholrelated ED visit rates occurred during the second quarter, which is consistent with another study that showed increases in drinking frequency during May-June 2020 relative to 2019 ( Pollard et al., 2020 ). However, the number of alcohol-related ED visits declined at the beginning of the pandemic, unlike corresponding dates in earlier years. This decline is consistent with the substantial decreases in total ED visits and ED visits for life threatening conditions and coincides with when stay-at-home orders went into effect ( Hartnett et al., 2020 ;Lange et al., 2020 ). Taken together, these study findings suggest that ED visits for alcohol-related emergencies declined to a lesser extent than ED visits for other causes.
People may have tried to avoid EDs, unless their condition was severe enough to warrant emergency medical attention, including complications associated with acute alcohol consumption, as well as mental health, drug overdose, and violence related outcomes ( Holland et al., 2021 ). Changes in the trends of alcohol-related ED visits during the COVID-19 pandemic may be related to people drinking to cope with Fig. 1. a. Number of weekly alcohol-related emergency department visits, 2018-2020. Fig. 1 b. Weekly alcohol-related emergency department visit rates per 10,000 emergency department visits, 2018-2020.

Table 1
Alcohol-related emergency department visit rates a per 10,000 emergency department visits and relative difference b in alcohol-related emergency department visit rates by quarter and by characteristics, 2018-2019 and 2020.  social isolation and other stressors such as increases in unemployment ( Czeisler et al., 2020a ;Wardell et al., 2020 ).
Several other factors might have influenced alcohol-related ED visit rates in 2020, including changes in the availability of and access to alcohol during the COVID-19 pandemic. The initial closure of bars and restaurants may have temporarily reduced access to alcohol for consumption on-premises, but many states deemed establishments that sold alcohol for consumption off-premises (e.g., liquor stores) as essential businesses, permitting them to remain open ( National Institute on Alcohol Abuse and Alcoholism, 2021 ). In addition, many states implemented legislation that may have increased access to alcohol for offpremise consumption, including allowing direct to consumer shipping and home delivery of alcohol ( National Institute on Alcohol Abuse and Alcoholism, 2021 ). Several states have converted new alcohol carryout and delivery policies into permanent laws ( National Institute on Alcohol Abuse and Alcoholism, 2021 ), increasing the availability of and access to alcohol compared to pre-pandemic times.
This study has limitations. First, the NSSP data are not weighted to represent the US population so the findings may not be generalizable to EDs that do not participate in the NSSP and the use of alcohol-related diagnosis codes may not be uniform across EDs. Second, changes in the demographic characteristics of the patients who visited the participating EDs over time for all causes could also affect the rates reported in this study, as it was not possible to assess patient characteristics such as race and ethnicity or income level. Third, these findings are likely conservative estimates of alcohol-related emergency events compared to non-pandemic times because avoidance of medical care has been reported during the pandemic ( Czeisler et al., 2020b ). Fourth, these data may underestimate alcohol-related ED visits because discharge diagnoses or chief compliant text relating to alcohol use may be incomplete ( Uong et al., 2022 ). Lastly, it is not possible to attribute changes in alcohol-related ED visits to the pandemic; we were not able to control for factors such as changes in alcohol availability, ED capacity for patients with COVID-19, and the implementation and enforcement of COVID-19 community mitigation measures.
Nevertheless, the general patterns of the rates of alcohol-related ED visits in this study are consistent with the epidemiology of alcoholattributable deaths ( Esser et al., 2020 ), suggesting the usefulness of this surveillance to inform the prevention of adverse alcohol-related outcomes. In addition, per capita alcohol sales data are consistent with the region-specific study findings on alcohol-related visit rates (e.g., highest in the West and Northeast, and lowest in the South) ( Slater and Alpert, 2019 ).

Conclusions
Alcohol-related ED visit rates per 10,000 total visits increased during 2020 versus 2018-2019, with the greatest relative difference in the second quarter. While the number of all-cause ED visits declined substantially during the early stage of the COVID-19 pandemic and remained lower during 2020 than 2019 , the number of alcohol-related ED visits did not decrease as much as total ED visits. Strong alcohol policies, including those that increase the price and reduce the availability of and access to alcohol (e.g., increasing alcohol taxes, regulating alcohol outlet density), are associated with decreased excessive drinking and related harms ( Community Guide, 2016 ). A comprehensive approach including more widespread implementation of these effective population-level alcohol policies ( Community Guide, 2016 ), messaging to promote help-seeking for behavioral or mental health support ( Holland et al., 2021 ), and alcohol screening and brief intervention (face-to-face or via electronic modes for providing telehealth services) could help to reduce excessive drinking and alcohol-related ED visits ( O'Connor et al., 2018 ), and the burden of alcohol use on US healthcare systems.