Human Papillomavirus Vaccination and Social Media: Results in a Trial With Mothers of Daughters Aged 14–17

Introduction: Parents acquire information about human papillomavirus (HPV) vaccines online and encounter vaccine-critical content, especially on social media, which may depress vaccine uptake. Secondary analysis in a randomized trial of a Facebook-delivered adolescent health campaign targeting mothers with posts on HPV vaccination was undertaken with the aims of (a) determining whether the pre–post-change occurred in self-reports of the mothers on HPV vaccination of their adolescent daughters; (b) describing the comments and reactions to vaccine posts; (c) exploring the relationship of campaign engagement of the mothers assessed by their comments and reactions to posts to change in the self-reports of the mothers of HPV vaccination.

Materials and Methods: Mothers of daughters aged 14–17 were recruited from 34 states of the US (n = 869). A social media campaign was delivered in two Facebook private groups that differed in that 16% of posts in one were focused on indoor tanning (IT) and 16% in the other, on prescription drug misuse, assigned by randomization. In both groups, posts promoted HPV vaccination (n = 38 posts; no randomization) and vaccination for other disease (e.g., influenza, n = 49). HPV and other vaccination posts covered the need for a vaccine, the number of adolescents vaccinated, how vaccines are decreasing the infection rates, and stories of positive benefits of being vaccinated or harms from not vaccinating. Guided by social cognitive theory and diffusion of innovations theory, posts were intended to increase knowledge, perceived risk, response efficacy (i.e., a relative advantage over not vaccinated daughters), and norms for vaccination. Some vaccination posts linked to stories to capitalize on identification effects in narratives, as explained in transportation theory. All mothers received the posts on vaccination (i.e., there was no randomization). Mothers completed surveys at baseline and 12- and 18-month follow-up to assess HPV vaccine uptake by self-report measures. Reactions (such as sad, angry) and comments to each HPV-related post were counted and coded.

Results: Initiation of HPV vaccination (1 dose) was reported by 63.4% of mothers at baseline, 71.3% at 12-month posttest (pre/post p < 0.001), and 73.3% at 18-month posttest (pre/post p < 0.001). Completion of HPV vaccination (two or three doses) was conveyed by 50.2% of mothers at baseline, 62.5% at 12-month posttest (pre/post p < 0.001), and 65.9% at 18-month posttest (pre/post p < 0.001). For posts on HPV vaccines, 8.1% of mothers reacted (n = 162 total), and 68.4% of posts received a reaction (63.2% like; 13.2% love, 7.9% sad). In addition, 7.6% of mothers commented (n = 122; 51 unfavorable, 68 favorable, 1 neutral), and 50.0% of these posts received a comment. There were no differences in pre–post change in vaccine status by the count of reactions or comments to HPV vaccine posts (Ps > 0.05). Baseline vaccination was associated with the valence of comments to HPV vaccine posts (7.2% of mothers whose daughters had completed the HPV series at baseline made a favorable comment but 7.6% of mothers whose daughters were unvaccinated made an unfavorable comment).

Conclusion: Effective strategies are needed in social media to promote HPV vaccines and counter misinformation about and resistance to them. Mothers whose daughters complete the HPV vaccine course might be recruited as influencers on HPV vaccines, as they may be predisposed to talk favorably about the vaccine. Comments from mothers who have not been vaccinated should be monitored to ensure that they do not spread vaccine-critical misinformation. Study limitations included lack of randomization and control group, relatively small number of messages on HPV vaccines, long measurement intervals, inability to measure views of vaccination posts, reduced generalizability related to ethnicity and social media use, and use of self-reported vaccine status.

Clinical Trial Registration: www.clinicaltrials.gov, identifier NCT02835807.