Background. The ongoing development of a vaccine against human papillomavirus (HPV) raises important questions about the impact of various vaccination strategies.
Methods. Two mathematical models are developed to explore the population-level impact of an HPV vaccine. The first model focuses on the infection process and the second on the disease process (specifically, cervical carcinoma in situ and cancer).
Results. Both population characteristics (ie, sexual mixing and rates of sex partner change) and vaccine characteristics affect the steady state prevalence of HPV that would be expected if a vaccine program is implemented. Under a particular set of assumptions, we find that vaccinating both men and women against a specific HPV type would result in a 44% decrease in prevalence of that type whereas vaccinating only women would result in a 30% reduction. We also find that if a vaccine gives protection against some, but not all, high risk types of HPV, the reduction in disease may be less than the reduction in HPV because the remaining high risk HPV types may replace the disease caused by the eliminated types.
Conclusions. A multivalent vaccine containing the majority of disease-causing HPV types would greatly reduce the need for colposcopy, biopsy and treatment. However, it is unlikely that Pap-screening programs would become redundant unless the vaccine is highly effective and coverage is widespread. In contrast to less common infections that are primarily restricted to core groups, targeting the vaccine towards the most sexually active individuals is less effective for a common sexually transmitted infection such as HPV.