MPH 510 Module 8 Evaluating a Vaccination Program Example

From Vaccine to Population Impact: Evaluating HPV Vaccination With Epidemiological Evidence

Student Name

Master of Public Health Program, Aspen University

MPH 510: Epidemiology in Public Health

Instructor Name

Month Day, Year

From Vaccine to Population Impact: Evaluating HPV Vaccination With Epidemiological Evidence

Vaccines are tested for efficacy in trials, but a program's real value is measured in populations over years: whether infections fall, whether disease falls and whether unvaccinated people are protected too. Human papillomavirus vaccination offers an instructive example, because its main goal, preventing cervical cancer, takes decades to measure. This paper evaluates the evidence on HPV vaccination's population impact and proposes a surveillance plan for a composite state.

The Challenge of Measuring Impact

Trials showed that HPV vaccines prevent infection and precancerous lesions, but they could not follow participants long enough to measure invasive cancer. Programs therefore relied on intermediate outcomes, infection and precancer, while waiting for cancer data. Epidemiologists used registers, surveillance and ecological comparisons to fill the gap.

The Swedish Register Study

Using national registers, researchers followed 1,672,983 Swedish girls and women aged 10 to 30 from 2006 through 2017. Invasive cervical cancer was diagnosed in 19 women who had received the quadrivalent vaccine and in 538 who had not. After adjustment for age, calendar year, county and parental characteristics, the incidence rate ratio was 0.37 for all vaccinated women, 0.12 for those vaccinated before age 17 and 0.47 for those vaccinated at 17 to 30 (Lei et al., 2020).

Population-Level Effects

A systematic review and meta-analysis of 65 studies in 14 high-income countries compared periods before and after vaccination programs began (Drolet et al., 2019). The table summarizes selected results.

Herd Effects

The decline in anogenital warts among boys and young men in countries that vaccinated only girls indicates herd effects: fewer infected women meant fewer infections passed to men (Drolet et al., 2019). Herd effects amplify a program's value and strengthen the causal case, since no alternative explanation predicts declines among unvaccinated groups timed to vaccination programs.

Applying Hill's Viewpoints

The evidence meets Hill's viewpoints for causation (Hill, 1965). The association is strong. It is consistent across countries and study designs. Vaccination precedes the outcome. A gradient appears, with larger benefits when vaccination occurs younger, before exposure. The finding is biologically plausible, since HPV causes nearly all cervical cancers. And it is coherent with trial evidence on precancer.

Confounding and Bias

Observational studies of vaccination can be confounded. Women who are vaccinated may also be more likely to attend cervical screening, which removes precancerous lesions and prevents cancer. Conversely, higher screening among vaccinated women might detect more cancers. The Swedish study adjusted for parental education, income and other factors, but residual confounding cannot be fully excluded. The consistency of findings across designs reduces concern.

Ecological Comparisons

Before-and-after comparisons at the population level can be affected by changes in screening practices, sexual behavior or testing methods over time. The meta-analysis addressed this by focusing on studies with consistent methods across periods and by comparing age groups with different vaccination coverage. Declines concentrated in vaccinated age groups argue against secular trends as the explanation.

Surveillance Plan for a Composite State

A state with adolescent HPV vaccination coverage of 62% plans to evaluate its program over ten years. Indicators include vaccination coverage by county, sex and insurance from the immunization registry; HPV type prevalence in residual specimens from routine screening among women aged 20 to 24; rates of high-grade cervical lesions from pathology reports; and cervical cancer incidence from the cancer registry, analyzed by birth cohort.

Linking Data Sources

With appropriate privacy safeguards, linking the immunization registry to the cancer and pathology registries would allow cohort analyses similar to the Swedish study. Where linkage is not possible, comparisons across birth cohorts with different coverage provide ecological evidence. Reporting by county and race would reveal whether the program narrows or widens disparities.

Using Results

Evidence of impact supports efforts to raise coverage, such as school-based vaccination, reminder systems and clinician recommendations. Disparities in coverage identified by surveillance would direct outreach to communities with lower uptake. Results also inform screening policy, since vaccinated cohorts may need less frequent screening in the future.

Vaccination Coverage and Timing

Impact depends on coverage and on vaccinating before exposure. The register study's gradient, much lower cancer rates when vaccination occurred before 17, reflects that girls vaccinated young are less likely to have been infected already. Programs that vaccinate at 11 or 12, as US recommendations advise, therefore gain the most benefit per dose.

Changing Vaccine Products

The studies reviewed mostly involved bivalent or quadrivalent vaccines. The nine-valent vaccine now used in the United States covers additional high-risk types, so future impact may be larger. Surveillance of type-specific infection will show whether types covered only by the newer vaccine also decline, adding another test of the program's effect.

Communicating Impact

Clear evidence of falling precancer and cancer rates can counter hesitancy. Messages that describe how many cancers were prevented in a real population may persuade parents more than trial efficacy figures. State reports should translate relative reductions into numbers of cases prevented in local terms.

Global Relevance

The heaviest toll from cervical cancer falls on poorer nations where screening is scarce. Evidence from high-income countries supports the World Health Organization's elimination goal for cervical cancer, which relies on high vaccination coverage, screening and treatment. Single-dose schedules, now supported by newer evidence, may make high coverage more achievable where resources are scarce.

Limitations of the Evidence

Most population data come from high-income countries with organized screening, which may limit generalization. Follow-up for invasive cancer remains relatively short for cohorts vaccinated as adolescents. Continued surveillance will show whether the early reductions persist and grow as vaccinated cohorts reach the ages when cervical cancer is most common.

Conclusion

HPV vaccination programs have produced large, consistent declines in infection, warts and precancer, herd effects among unvaccinated groups and, now, lower invasive cervical cancer rates, especially when vaccination occurs early. Epidemiological methods, from register-based cohorts to meta-analysis and causal reasoning, made these conclusions possible and provide a model for evaluating other long-horizon prevention programs.

References

Drolet, M., Bénard, É., Pérez, N., Brisson, M., Ali, H., Boily, M.-C., Baldo, V., Brassard, P., Brotherton, J. M. L., Callander, D., Checchi, M., Chow, E. P. F., Cocchio, S., Dalianis, T., Deeks, S. L., Dehlendorff, C., Donovan, B., Fairley, C. K., Flagg, E. W., ... Yu, B. N. (2019). Population-level impact and herd effects following the introduction of human papillomavirus vaccination programmes: Updated systematic review and meta-analysis. The Lancet, 394(10197), 497-509. https://doi.org/10.1016/S0140-6736(19)30298-3

Hill, A. B. (1965). The environment and disease: Association or causation? Proceedings of the Royal Society of Medicine, 58(5), 295-300. https://doi.org/10.1177/003591576505800503

Lei, J., Ploner, A., Elfström, K. M., Wang, J., Roth, A., Fang, F., Sundström, K., Dillner, J., & Sparén, P. (2020). HPV vaccination and the risk of invasive cervical cancer. New England Journal of Medicine, 383(14), 1340-1348. https://doi.org/10.1056/NEJMoa1917338