2477. Impact of Varicella Vaccination in the United States (US): a Dynamic Model-Based Analysis
Session: Poster Abstract Session: Vaccines for Herpes Zoster Virus
Saturday, October 6, 2018
Room: S Poster Hall
  • wolfsonl_201305-0001-IDWeek_Poster_2477_v1.04FINAL.pdf (448.1 kB)
  • Background: Routine childhood immunization with varicella vaccine was first recommended in the US in 1995 as a 1-dose regimen for children aged 12–18 months, with updated recommendations in 2006 for a 2-dose regimen (1st dose at 12–15 months, 2nd dose at 4–6 years). Our objective was to estimate the impact of the US varicella vaccination program.

    Methods: We developed a dynamic transmission model to predict the impact on varicella vaccination on health outcomes in the US.  Vaccine coverage rates were extracted from the US National Immunization Survey (NIS); 1st dose varicella vaccine coverage went from 12% in 1996 to 91% by 2016 for children 18 months old, and 2nd dose coverage starting in 2006 at 5% increasing by 2016 to 94% for children 5 years old;  we assumed that 50% of children   with no history of vaccination or infection by age 13 would become vaccinated .  Interactions between age groups were empirically characterized, and the model was calibrated using age-specific pre-vaccination varicella incidence data.. Vaccine effectiveness was represented via vaccine take and waning immunity estimated from a 10-year trial.

    Results: The model projected reductions of varicella incidence in all ages (and ages <15 years) of 46% (46%) in 2001, 76% (76%) in 2006, 78% (81%) in 2011, and 89% (93%) in 2016 (Figure 1).  The projected reductions in varicella cases and varicella-related hospitalizations and deaths for all ages were 74%, 70%, and 66% by 2006 (one-dose era), respectively, increasing to 89%, 70%, and 69% by 2016 (two dose era), respectively (Figure 2). We estimate that between 1996-2016, 71,885,382 cases of varicella were prevented in the US, together with 178,248 varicella-related hospitalizations and 1,496 deaths.

    Conclusion: Our estimates are slightly lower than previously reported US surveillance data which identified a 97.4% (92.9%-97.9%) reduction between 1993-4 and 2013-14 in IL, MI, TX, and WV (WER 2016). Likely, this difference is related to under ascertainment of milder cases.   This model can be used to estimate the public health benefits of varicella vaccination.   The use of a dynamic transmission model does, however, have limitations, including assumptions about age-specific risk and severity of breakthrough disease and the use of a static population.


    Lara Wolfson, PhD1, Jeffrey Kyle, MS2, Barbara Kuter, PhD, MPH3, Myron Levin, MD, FIDSA4 and Vincent Daniels, PhD1, (1)Center for Observational and Real-World Evidence (CORE), Merck & Co., Inc., Kenilworth, NJ, (2)Merck & Co., Inc., Kenilworth, NJ, (3)Global Vaccines Medical Affairs, Merck & Co., Inc., Kenilworth, NJ, (4)Section of Pediatric Infectious Diseases, University of Colorado School of Medicine, Aurora, CO


    L. Wolfson, Merck & Co., Inc.: Employee and Shareholder , Salary .

    J. Kyle, Merck & Co., Inc.: Independent Contractor , Salary .

    B. Kuter, Merck: Employee and Shareholder , Salary .

    M. Levin, Merck Sharp & Dohme Corp.: Scientific Advisor , Consulting fee , Licensing agreement or royalty and Research grant .

    V. Daniels, Merck & Co., Inc.: Employee and Shareholder , Salary .

    Findings in the abstracts are embargoed until 12:01 a.m. PDT, Wednesday Oct. 3rd with the exception of research findings presented at the IDWeek press conferences.