Research Reveals The Divergent Drinking Patterns of US Adults During the Pandemic

Newswise — A study of adult drinking in the US over time has identified distinct demographic subpopulations with diverging drinking trajectories during the first ten months of the COVID-19 pandemic. Although population-wide average drinking increases in the pandemic were modest and time-limited, the aggregated data mask escalating drinking in certain subpopulations, raising public health concerns. The study, reported in Alcoholism: Clinical and Experimental Research, provides insights into the sociodemographic communities that might benefit from intervention to counteract drinking increases − or reinforce drinking reductions − that were acquired during the pandemic.

The data were from ~8000 U.S. adult respondents in the nationally representative Understanding America Study Coronavirus in America survey. Participants completed a baseline web-based survey on March 10, 2020 − the day before COVID-19 was declared a global pandemic by the WHO − and biweekly follow-up surveys through to mid-January 2021. Each survey included questions on both drinking frequency (the number of days on which alcohol was consumed during the past week) and drinking intensity (the number of drinks consumed on a typical drinking day in the past week). Drinking intensity data provided an estimate of binge drinking, defined as at least 4 drinks (for women) or 5 drinks (men) on a drinking occasion.

When the data were collapsed across all survey participants, the average number of past-week drinking days increased slightly from 1.2 days in March to ~1.5 days in April, before gradually declining to an average of 1.2−1.3 days. The probability of past-week binge-drinking in the overall sample was around 6% across all timepoints, and fairly stable throughout the period of analysis. However, using a statistical modeling method known as growth mixture modeling, the research team was also able to identify multiple subpopulations with distinct drinking trajectories over time. In all, four different drinking frequency trajectories, and four different drinking intensity trajectories, were revealed.

For drinking frequency, most participants (73%) showed a “minimal/stable” trajectory, with an average of less than 1 past-week drinking day throughout the analysis period. One group (6.7%) had a “moderate/late decreasing” trajectory, with an average of 3.1 drinking days per week in March reducing to 2.1 drinking days by January 2021. The remaining participants (one-fifth of the total) increased their drinking frequency. The “moderate/early increasing” group (12.9%) began with an average 2.1 drinking days in March, which increased through April to an average of 3.2 drinking days in January 2021. The “near daily/early” group (7.6%) began with an average of 5.6 drinking days per week in March, which continued increasing. Males, whites, middle-aged and older adults, college graduates, those consistently working, and those above the poverty limit were overrepresented in the increasing (vs. minimal/stable) frequency trajectories.

For drinking intensity, 85% of participants showed a “minimal/stable” trajectory, with a very low probability of binge drinking throughout. One group (7.4%) showed a “low-to-moderate/fluctuating” pattern, with varying binge-drinking probabilities across timepoints. The remaining 7% of participants increased their probability of binge drinking during the pandemic. Just over 4% showed a “moderate/mid increasing” pattern, with an average 39% probability of binge drinking in April rising to 65% during August−September; a further 2.7% of participants had a “high/early increasing” trajectory, with an 84% binge-drinking probability rising to 96% by June, and remaining high. Males, whites, non-married, those without a college degree, 18−39-year-olds, and the middle-aged were overrepresented in the increasing (vs. minimal/stable) intensity trajectories.

In summary, the data suggest that several sociodemographically distinct US adult subpopulations acquired new drinking patterns during the pandemic’s first 42 weeks. Previous research has revealed stress, lack of non-alcohol alternatives, reduced healthcare access, economic distress, experiencing discrimination, and easy access to alcohol delivery as possible causes of increased pandemic drinking. Conversely, closure of bars and fewer social gatherings might have reduced drinking for others. The current findings could inform precision prevention strategies for reducing alcohol-related health disparities and harm in different demographic communities. 

Alcohol use trajectories among U.S. adults during the first 42 weeks of the COVID-19 pandemic. A. M. Leventhal, J. Cho, L. A. Ray, R. Liccardo Pacula, B. P. Lee, N. Terrault, E. Pedersen, J. Olivia Lee, J. P. Davis, H. Jin, J. Huh, J. P. Wilson, R. C. Whaley (pages xxx)

ACER-21-5042.R1