The failure of news coverage supportive of human papillomavirus vaccination: The investigation of the effects of online comments on female college students' vaccination intention.

BACKGROUND

News coverage supportive of human papillomavirus (HPV) vaccination is widespread on the Internet. Meanwhile, large numbers of online comments questioning HPV vaccine safety, attacking vaccine manufacturers or providing personal experience of side effects may reduce the persuasion effect. This study explored how and under what conditions such comments influence the behavioral intention of subsequent youth female viewers or readers.

METHODS

We recruited a random sample of 239 Chinese undergraduate female students from different, non-medicine-related disciplines and conducted a 2 (valence of online comments: positive vs. negative) × 2 (presentation of comments: narrative vs. descriptive) between-subjects factorial design study. After carrying two ANOVAs to examine the effects of experimental conditions, we established a simple mediation and a moderated mediation analysis to test hypothesis.

RESULTS

Perceived risk of HPV emerged as a significant mediator in the moderated mediation model. Particularly under the negative condition, the narrative (vs. descriptive) comments increased behavioral intention to receive the vaccine (B = -0.20, SE = 0.10, 95% CI = -0.40 to -0.01) while they increased participants' perceived risk of HPV (B = -0.14, SE = 0.05, 95% CI = -0.27 to -0.06). However, under the positive condition, there was no significant difference in either risk perception or behavioral intention between the effects of narrative and descriptive comments on participants.

CONCLUSIONS

Online comments do influence participants' perceived risk of HPV and vaccination intention. Perceived risk of HPV fully mediates the effect of comments on vaccination intention. More importantly, the indirect effect of narrative (vs. descriptive) comments on behavioral intention through risk perception is moderated by the valence of comments, and significant only under the negative comments condition.