Duke 2008 | ASHE | Equity and Efficiency in Health and Healthcare

Human papillomavirus (HPV) is the most common sexually transmitted disease. Nearly 80% of all people get HPV during their lifetime. While most cases are harmless and never detected, about 10% of people contract genital warts and 1% of women develop cervical cancer, which is caused almost exclusively by HPV.

In 2006, a HPV vaccine (Gardasil) was approved by the U.S. Food and Drug Administration (FDA) for females ages 9-26. The vaccine protects against viruses responsible for the majority of cervical cancers and genital warts. Studies suggest that the vaccine may also be cost-effective. Consumer-direct advertising and an Advisory Committee on Immunization Practice (ACIP) recommendation of routine vaccination for girls age 11-12 has raised substantial public awareness in the U.S. However, perceptions are mixed and inoculation remains low, especially among younger adolescents. A competing vaccine is now under FDA review and will target different strains of HPV.

In the U.S., the HPV vaccine remains controversial for several reasons, including: (1) the link between HPV and sexual activity; (2) the young age of routine recommendation; (3) limited vaccine safety data; (4) substantial expense ($360 for 3 doses); and (5) the low incidence and fatality rate of cervical cancer (12th most prevalent, 14th most deadly among women). In contrast, cervical cancer is the leading cause of cancer death in several developing countries. The policy environment is muddled by a lack of rigorous economic studies about vaccine demand.

We developed a conjoint analysis (CA) survey of adolescent girls (age 13-17) and their mothers to understand their preferences and willingness to pay for HPV vaccination. CA methods are particularly valuable here because most of the population has not had the vaccine, and because a competing vaccine with different features is under FDA review and additional competitors are in clinical trial stages. The market and the public health setting are expected to change rapidly in the future, and our results may be used to predict the acceptability of other vaccines and the impact on uptake.

Following a literature review, discussions with policymakers, and extensive pretesting, we developed an optimal experimental design and fielded a conjoint analysis survey to 300 mother-daughter pairs in the Knowledge Networks online panel. Mothers were asked about preferences for vaccines their daughters; daughters were asked about vaccines for themselves. Key attributes were price, duration, and effectiveness against cervical cancer and genital warts. Each dyad completed individual surveys and 8 choice tasks in a three-part format: public health recommendation, personal vaccine decision, and prediction of the other respondent's preference.

The data were analyzed using both conditional and mixed logit models. We also estimated dyad discordance as a function of vaccine attributes and sociodemographics. Mothers and daughters reported significant, positive willingness to pay (WTP) values for HPV vaccination, but this was driven by different features among each: mothers had a higher WTP and placed greater value on genital wart protection than daughters. Both valued cancer protection much more overall. Age was significantly and positively associated with likelihood of vaccination in the next year for both.

Differences between parent and child preferences indicate that information campaigns should be tailored by age group and emphasize different features to maximize coverage of HPV vaccination. We predict that a bivalent vaccine with lesser protection against genital warts will be less valued. High WTP for preventing cervical cancer makes HPV vaccination suggests that full cost-benefit calculations based on these results may be favorable.