When are alcohol-related blackout Tweets written in the United States?

Addict Behav. 2022 Jan:124:107110. doi: 10.1016/j.addbeh.2021.107110. Epub 2021 Sep 4.

Abstract

Background: Alcohol use varies throughout the year and often peaks on weekends or during celebrations (e.g., New Year's). There is not a perfect correlation between alcohol use and negative consequences, and the extent to which one particularly risky consequence-an alcohol-related blackout-is more common during certain times of the year is unknown. Identifying when blackouts occur may help identify which periods are associated with more risk and be critical in designing public health campaigns. Thus, we examined Twitter data to ascertain whether alcohol-related blackouts occur more during certain holidays/celebrations than typical weekends and whether they differed in timing from general alcohol-related Tweets.

Methods: We used a Twitter-sponsored platform to access unique Tweets written in the United States referencing blackouts (e.g., "blackout") and alcohol generally (e.g., "drunk").

Results: The final dataset included 3.5 million blackout Tweets and 591 million alcohol Tweets (written between 2009 and 2020). Both blackout and alcohol Tweets were written in the late evening, on weekends, and during certain holidays (New Years, St. Patrick's). However, relative to typical weekends, only blackout Tweets were more common during Thanksgiving and only general alcohol-related Tweets were more common during Cinco de Mayo.

Conclusion: While blackout and alcohol-related Tweets were similar in time of day (peaking in the evening) and day of week (peaking on weekends), they differed during certain celebrations/holidays, suggesting that while alcohol use may be more common during some celebrations, others are more associated with serious harms.

Keywords: Blackouts; Celebratory drinking; Social networking sites; Twitter.

MeSH terms

  • Alcohol Drinking / epidemiology
  • Alcoholic Intoxication*
  • Alcoholism*
  • Amnesia, Anterograde*
  • Ethanol
  • Humans
  • Social Media*
  • United States / epidemiology

Substances

  • Ethanol