Introduction: Swisher Sweets, a leading brand of little cigars and cigarillos in the United States, switched its Twitter account to protected status, limiting access to its tweets. This study examines how the protected status of Swisher Sweets tweets influences post engagement, aiming to inform regulatory strategies for branded tobacco promotions on social media.
Method: Using natural language processing, we predicted the demographics of individuals replying to Swisher Sweets' public and protected tweets. Engagement with public versus protected tweets was compared using a Mann-Whitney U test, and a mixed-effects logistic regression assessed the likelihood of different demographics replying to each tweet type. Chi-square analyses examined word frequencies related to any flavor, concept flavors, and characterizing flavor in replies to public and protected tweets.
Results: Overall, 16 % of individuals replying to Swisher Sweets' tweets were predicted to be under 21, and 65 % were Black. No significant difference was found in average reply counts to public versus protected tweets (p = .78). Black individuals were 2.61 times more likely than White individuals to engage with protected tweets after the status change (OR = 2.61, 95 % CI [1.36, 5.06], p = .004). Replies to protected tweets contained more words related to any flavor (adjusted p < .001) and concept flavors (adjusted p < .001) compared to public tweet replies.
Conclusions: Our study suggests that the protected status on Twitter was ineffective in preventing underage engagement with Swisher Sweets' branded tweets and may facilitate targeted marketing among Black individuals.
Implications: More stringent age verification procedures and promotional regulatory measures are needed to prevent targeted tobacco brand marketing on social media.
Keywords: Flavored cigars; Machine learning; Social media; Tobacco marketing; Twitter.
Copyright © 2024 The Authors. Published by Elsevier B.V. All rights reserved.