Misinformation and other elements in HPV vaccine tweets: an experimental comparison

Our study examined how misinformation and other elements of social media messages affect antecedents to human papillomavirus (HPV) vaccination of adolescents. In 2017-2018, we randomly assigned a national sample of 1206 U.S. parents of adolescents to view one tweet using a 2 × 2 × 2 × 2 between-subjects factorial experiment. The 16 experimental tweets varied four messaging elements: misinformation (misinformation or not), source (person or organization), narrative style (storytelling or scientific data), and topic (effectiveness or safety). Parents reported their motivation to vaccinate (primary outcome), trust in social media content, and perceived risk about HPV-related diseases. Tweets without misinformation elicited higher HPV vaccine motivation than tweets with misinformation (25% vs. 5%, OR = 6.60, 95% CI:4.05, 10.75). Motivation was higher for tweets from organizations versus persons (20% vs. 10%, OR = 2.47, 95% CI:1.52, 4.03) and about effectiveness versus safety (20% vs. 10%, OR = 2.03, 95% CI:1.24, 3.30). Tweets with misinformation produced lower trust and higher perceived risk (both p < .01), with impact varying depending on source and topic. In conclusion, misinformation was the most potent social media messaging element. It may undermine progress in HPV vaccination.