OBJECTIVES: Synthesize all available clinical data on the herpes zoster (HZ) vaccine, and determine the best model from which to predict the vaccine effectiveness and waning of effectiveness for the HZ vaccine based on age and time since vaccination.
METHODS: A targeted search of PubMed and Google was performed. The search was limited to literature between January 2006 and December 2016. Literature eligible for inclusion were peer-reviewed papers in scientific journals, conference abstracts and posters.
RESULTS: Six models for the HZ vaccine were produced to estimate the waning of effectiveness. All studies used HZ incidence among vaccinated vs. unvaccinated populations as measures of effectiveness. Four are based on the models presented in Li et al. (2015) clinical study data. One is based on the Long-Term Persistence Study (LTPS) clinical data in Morrison et al. (2015) as modeled by Le and Rothberg (2015). The sixth model is based on the findings of Baxter et al. (2015), real world data from Kaiser Permanente Northern California. These data represent up to 8 years of patient follow up and includes over 400,000 vaccinated patients. One model from Li et al. predicted a drop in effectiveness in year 2 but with slower waning in subsequent years. A second model from Li et al. predicted that vaccine effectiveness over time is similar in all 60+ age groups. Both of these models are consistent with the model based on data from Baxter et al.
CONCLUSIONS: Cost-effectiveness results for herpes zoster prevention is highly dependent on age at vaccination and the waning of effectiveness for the HZ vaccine. Based on the evidence for real world effectiveness against zoster for the HZ vaccine, we recommend utilizing the data from Baxter et al. to predict effectiveness and waning for individuals being vaccinated with the HZ vaccine.
