Objectives: To assess self-reported population oral health conditions amid the COVID-19 pandemic using user reports on Twitter.
Method and materials: Oral health-related tweets during the COVID-19 pandemic were collected from 9,104 Twitter users across 26 states (with sufficient samples) in the United States between 12 November 2020 and 14 June 2021. User demographics were inferred by leveraging the visual information from the user profile images. Other characteristics including income, population density, poverty rate, health insurance coverage rate, community water fluoridation rate, and relative change in the number of daily confirmed COVID-19 cases were acquired or inferred based on retrieved information from user profiles. Logistic regression was performed to examine whether discussions vary across user characteristics.
Results: Overall, 26.70% of the Twitter users discussed “Wisdom tooth pain/jaw hurt,” 23.86% tweeted about “Dental service/cavity,” 18.97% discussed “Chipped tooth/tooth break,” 16.23% talked about “Dental pain,” and the rest tweeted about “Tooth decay/gum bleeding.” Women and younger adults (19 to 29 years) were more likely to talk about oral health problems. Health insurance coverage rate was the most significant predictor in logistic regression for topic prediction.
Conclusion: Tweets inform social disparities in oral health during the pandemic. For instance, people from counties at a higher risk of COVID-19 talked more about “Tooth decay/gum bleeding” and “Chipped tooth/tooth break.” Older adults, who are vulnerable to COVID-19, were more likely to discuss “Dental pain.” Topics of interest varied across user characteristics. Through the lens of social media, these findings may provide insights for oral health practitioners and policy makers.
Schlagwörter: attitude, big data, data mining, dental anxiety, logistic models, social media