Abstract
Recent findings suggest that the rate of certain cancers can be reduced by increasing human papillomavirus (HPV) vaccination coverage, resulting in considerable research interest on the antecedents of HPV vaccine receipt to identify avenues to promote vaccination. The current article continues this stream of research by (1) studying the three HPV vaccination outcomes of willingness, receipt, and word-of-mouth, (2) investigating the antecedent effects of sociodemographic characteristics, health insurance status, provider conversation, and political orientation, and (3) testing the mediating role of vaccine hesitancy dimensions using a recently developed conceptualization. We achieve these goals by conducting a cross-sectional study with 404 participants (Agemean = 37.64; AgeSD = 14.91; 57% female; 72% white; 100% located in U.S.). Our results show that provider conversations and political orientation had the most consistent and strong effects of the personal characteristics, whereas perceptions that vaccines pose health risks and perceptions that vaccines are not needed for healthy individuals produced the most consistent and strong effects of the vaccine hesitancy dimensions. Other personal characteristics and vaccine hesitancy dimensions also produced intermittent significant effects, including age, education, and race. Together, these results support that personal characteristics and vaccine hesitancy dimensions relate to a broader range of outcomes associated with HPV vaccination than previously known, and these results also support that the recently developed conceptualization of vaccine hesitancy is apt for understanding HPV vaccine perceptions. Our discussion concludes with highlighting avenues for future research and practice that can leverage our discovered relations to improve HPV vaccination outcomes.
Internal review board (IRB) statement
All procedures for the current study were approved by the IRB of the primary author's institution (#1663301).
Disclosure statement
The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
Notes
1 It could be argued that the MVHS (and any modifications) should instead be modeled as nine independent latent factors. When modeled as nine covaried latent factors, the model fit slightly improves (CFI = .96, IFI = .96, RMSEA = .05, SRMR = .06, χ2/df = 2.06) to meet standard cutoffs. Therefore, both approaches to modeling the modified MVHS produce appropriate model fit, providing support for the continued use of the measure.