Cost-effectiveness of infant pneumococcal vaccination in the Netherlands

Objectives: The Dutch National Immunization Program offers the 10-valent pneumococcal conjugate vaccine (PCV10). Also licensed for use in the infant population is the 13-valent PCV (PCV13). To update cost-effectiveness (CE) estimates of PCV13 over PCV10, using current epidemiological and economic data. Methods: We modeled vaccinating a birth cohort with either PCV10 or PCV13 (3+1 dose schedule), and calculated costs and effects linked to resulting disease. We modeled invasive pneumococcal disease (IPD), non-invasive pneumonia and acute otitis media, and considered death and lifetime impairments after IPD. We calculated direct effects in the vaccinated cohort and indirect effects -herd immunity for the vaccine-type (VT) serotypes and replacement for the non-VT serotypes- in the rest of the population. Since no price is available, we use a price difference of € 11 per dose and vary this price difference in sensitivity analyses. Epidemiological and economic data are taken as current as possible. A set of scenarios explore different assumptions, including different sets of epidemiological data, assumptions on vaccine efficacy and indirect effects. Results: Taking only direct effects into account PCV13 cannot be considered cost-effective, unless the price difference is much lower than € 11 per dose. In three scenarios, PCV10 dominates PCV13; in the other scenarios the ICER is between € 89000 and € 153000 per QALY gained. If indirect effects are also taken into account, the ICER of PCV13 compared to PCV10 is below € 20,000 per QALY for all scenarios. Scenarios do not have a large impact on the policy decision, unless we assume extra efficacy of PCV10 against non-typeable Haemophilus influenzae. Conclusions: Replacing PCV10 with PCV13 is not likely to be cost-effective in preventing invasive pneumococcal disease in young children. Taking potential benefits in elderly into account, PCV13 is likely cost-effective. The CE of PCV13 was highly sensitive for indirect effects our analysis.