Identifying Topics Around Nicotine Gum: A Machine Learning Approach with Twitter Data

Area of Science:

• Public Health
• Social Media Analysis
• Tobacco Product Research

Background:

• Nicotine gum products represent a new category in the tobacco market.
• Research on nicotine gum use correlates is limited.
• Social media data offers a valuable resource for understanding user experiences with novel tobacco products.

Purpose of the Study:

• To identify prevalent topics discussed on Twitter regarding nicotine gum.
• To analyze public perceptions and user experiences with nicotine gum.
• To inform future research, policy, and health communication strategies.

Main Methods:

• Utilized Twitter's Application Programming Interface (API) to collect data.
• Employed machine learning (Top2Vec) for topic modeling on 16,940 tweets from 10,353 unique users.
• Conducted a string search for popular brands and flavors.

Main Results:

• Identified eight distinct topics, with 'smoking and vaping cessation' being the most frequent.
• Other key topics included promotion, pricing, marketing, appeal, product comparisons, perceived benefits, distrust in institutions, health concerns, and COVID-19 misinformation.
• Nicorette was the most mentioned brand, and 10% of tweets referenced a flavor.

Conclusions:

• Understanding public discourse on nicotine gum is crucial for public health.
• Findings can guide the development of targeted research, policy interventions, and health campaigns.
• Social media analytics provide timely insights into emerging consumer behaviors and perceptions related to nicotine products.