Tuesday, October 20, 2015: 11:30 AM
Grand Ballroom A (Hyatt Regency St. Louis at the Arch)

Melissa Gilkey, PhD1, William Calo, PhD2, Jennifer Moss, PhD2 and Noel T. Brewer, PhD3, (1)Harvard Medical School, Boston, MA, (2)University of North Carolina, Chapel Hill, NC, (3)Department of Health Behavior, Gillings School of Global Public Health, The University of North Carolina at Chapel Hill, Chapel Hill, NC

Purpose: Only 38% of adolescent girls and 14% of adolescent boys complete the three-dose human papillomavirus (HPV) vaccine series, despite its excellent safety profile, mounting evidence of effectiveness, and national guidelines for routine administration. Research indicates that improving the implementation of HPV vaccination programs will require healthcare providers to recommend the vaccine more often and more strongly, but little is known empirically about which types of recommendation are most influential. Thus, we sought to investigate the relationship between provider recommendation quality and HPV vaccine decision making by parents of adolescents.

Method: In 2014 to 2015, we conducted a national, online survey of 1,495 U.S. parents who reported on the immunization history of an 11- to 17-year-old adolescent in their households. For parents who had received provider recommendations for HPV vaccination, our survey assessed recommendation quality on three indicators: 1) strength of endorsement (whether the provider said HPV vaccine was very important); 2) urgency (whether the provider recommended vaccinating “today”); and 3) prevention message (whether the provider said HPV vaccine prevents cancer). Using an index of these indicators, we categorized parents as having received no, low-quality, or high-quality provider recommendations. Separate multivariable logistic regression models assessed associations between recommendation quality and HPV vaccination decisions, including initiation (acceptance of ≥1 dose), follow through (acceptance of 3 doses, among initiators), refusal, and delay.

Result: Almost half (48%) of parents reported no provider recommendation for HPV vaccination, while 16% received low-quality recommendations and 36% received high-quality recommendations. Compared to no recommendation, high-quality recommendations were associated with over nine times the odds of HPV vaccine initiation (OR=9.31, 95% CI, 7.10-12.22) and over three times the odds of follow through (OR=3.82, 95% CI, 2.39-6.11). Low-quality recommendations were more modestly associated with initiation (OR=4.13, 95% CI, 2.99-5.70), but not follow through. Parents who received high- versus low-quality recommendations less often reported HPV vaccine refusal or delay.

Conclusion: Our findings provide early evidence to suggest that high-quality recommendations are more effective than low-quality recommendations for encouraging HPV vaccine acceptance and discouraging refusal and delay. Given that only about one-third of parents received high-quality recommendations, interventions are needed to improve provider communication about HPV vaccine so as to support parents' decision making and address the persistent underuse of a powerful tool for cancer prevention.