Abstract
OBJECTIVES: Consumption of high potency alcohol is associated with greater healthcare burden, yet little attention has been placed on the change in types of alcohol consumed during the COVID-19 pandemic. We estimate the change in alcohol consumption by beverage type attributable to the COVID-19 pandemic. METHODS: The National Institute on Alcohol Abuse and Alcoholism provided apparent alcohol consumption ("consumption") by beverage type for 10 states for January 2017 through November 2020 based on sales and tax data. The 38-month period to February 2020 was used to train quasi-Poisson regression models. The models then predicted the monthly consumption based on the historical trends in the absence of the COVID-19 pandemic from March through November 2020. The difference between the observed and predicted is the change in consumption attributable to the COVID-19 pandemic. RESULTS: Beyond what was expected based on historical trends, spirits consumption increased significantly for 6 states (Colorado, Massachusetts, Missouri, North Dakota, Minnesota, and Tennessee) ranging from 4% (95% confidence interval [CI] 1%-6%) to 17% (95% CI 6%-28%) which is equivalent to 7 (95% CI 2-18) to 32 95% CI 12-48) excess standard spirits drinks per-capita; Alaska, Florida, Illinois, and Kentucky had no significant change. Wine consumption increased 10% (95% CI 3%-18%) in Colorado and 8% (95% CI 3%-12%) in Tennessee. Wine consumption in Alaska decreased 6% (95% CI, 3%-10%) and beer consumption decreased 8% (95% CI 4%-11%). CONCLUSIONS: During the COVID-19 pandemic, spirits consumption increased relative to wine and beer. Increased consumption of higher potency alcohol beverages could lead to higher alcohol-related healthcare and societal burden.