Modeling the impact of a new vaccine on pneumococcal and nontypable Haemophilus influenzae diseases: a new simulation model.

BACKGROUND

A heptavalent pneumococcal conjugate vaccine (PCV-7) is available to immunize infants against pneumococcal disease. However, a recently developed vaccine, pneumococcal nontypable Haemophilus influenzae protein D-conjugate vaccine (PHiD-CV), has recently been licensed. PHiD-CV contains 3 additional Streptococcus pneumoniae serotypes and may provide protection against nontypable H influenzae (NTHi) infection. New health economic models are required to model the impact of PHiD-CV and compare its effectiveness with PCV-7.

OBJECTIVES

The aim of this article was to design a model capable of projecting the pneumococcal and NTHi disease burden on the entire UK population under different schedules of PCV-7 and PHiD-CV. This model should also be capable of modeling the net indirect effect of vaccination (ie, the sum of serotype replacement and herd protection).

METHODS

A static, deterministic, age-compartmental model was created based on published information and the input of a board of experts in pneumococcal disease. The model presents results from both a payer-based and societal perspective. A 1-way sensitivity analysis was used to demonstrate the robustness of the model. Key parameters included the case fatality ratio for bacteremia, the hospitalization rate for acute otitis media (AOM), and parameters surrounding the extent of the net indirect effect of vaccination.

RESULTS

Excluding net indirect effect, 325 cases of invasive pneumococcal disease (IPD), 619 hospitalizations for pneumonia, and 9016 general practitioner visits for AOM would be prevented annually with the current PCV-7 2 + 1 program. These numbers would increase to 374, 755, and 30,920, respectively, using a PHiD-CV 2 + 1 regimen, or to 503, 994, and 47,180 using a PHiD-CV 3 + 1 regimen. When a net indirect effect of 38% is considered, health benefits could be much larger; 2417, 2451, and 3045 IPD cases would be prevented in the 3 scenarios, respectively.

CONCLUSIONS

It is predicted that any vaccination program in the United Kingdom would have a striking impact on the incidence of all outcomes analyzed. A PHiD-CV 3 + 1 schedule is predicted to have a greater effect than PCV-7 in all scenarios. While the primary purpose of vaccination would be to prevent IPD and pneumonia hospitalizations, an additional benefit would be a noticeable reduction in AOM incidence. While the predictions made by the model were based on informed reasoning, all of its projected estimations remain approximations that are dependent on the inputs used to configure it, a limitation that is common to all simulation models.