A study developed and validated deep learning models to glean a better understanding of the public’s perception of human papillomavirus (HPV) vaccines per discussions on social media.
HPV, the most common sexually transmitted disease in the United States, causes more than 33,000 cases of cancer each year. The vaccine has been available for nearly 15 years and is recommended in adolescents aged 9 through 26 years (in some cases, patients up to 45 years old may be candidates). Despite this, HPV vaccination rates remain low compared to other adolescent vaccines.
“Improving understanding of the public perceptions of HPV and the HPV vaccine is essential to developing tailored educational efforts and increasing HPV vaccination rates,” the researchers explained. “Social media platforms offer a unique opportunity to examine the unfiltered opinions, comments, and discussions of large populations, while mitigating the limitations of traditional surveys, which include resource costs and the difficulties of tracking changes in real-time.”
The study authors performed a retrospective cohort study from April to August 2019. The study used longitudinal and geographic analyses of public perceptions surrounding HPV vaccines and incorporated sampled conversations related to the HPV vaccine that took place on Twitter from January 2014 to October 2018.
Final analysis included 1,431,463 posts created by 486,116 unique users. When mapping discussions to the constructs of behavior change theories, the deep learning algorithms attained F-1 scores that ranged from 0.6805 (95% confidence interval [CI], 0.6516–0.7094) to 0.9421 (95% CI, 0.9380–0.9462). The prevalence of perceived barriers was at its highest in July 2015 (56.2%) and at its lowest in October 2018 (28.4%), for a 27.8% difference (P<0.001). The HPV vaccine was regarded more positively from 2017 (30.7%) to 41.9% when the study concluded—a 11.2% difference (P<0.001)—and at the same time, decreases were observed in negative attitudes (42.3% to 31.3%; difference, 11.0%; P<0.001). Perceptions of the HPV vaccine varied by state. Significant perceived barriers were observed in Ohio (11,531 of 17,106; 67.4%) and Maine (1,157 of 1,684; 68.7%), as were negative attitudes (9,655 of 17,197; 56.1% and 1,080 of 1,793; 60.2%, respectively).
The study was published in JAMA Network Open.
“[This study] provided a method to automatically understand population-level and individual-level health beliefs and attitudes toward the HPV vaccine. This can then inform rational and directed programmatic efforts to improve actual immunization coverage rates by allowing for real-time monitoring of beliefs and intentions and adjustment of educational and public health campaigns and messaging as warranted,” the researchers wrote in their conclusion. “Such data-enabled real-time information is invaluable to the design of such efforts and can assist in realizing the benefits of increased population vaccine coverage levels.”