Hospitalizations for lower respiratory tract infections in children in relation to the sequential use of three pneumococcal vaccines in Quebec.

Abstract

In Quebec, three pneumococcal conjugate vaccines (PCV) were used sequentially starting in December 2004. The objective of the study was to investigate the association between exposure to different PCV regimens and hospitalizations for lower respiratory tract infection (LRTI). Records with a main diagnosis of LRTI in children born in 2000-2012 and observed up to their second birthday were extracted from the provincial hospital administrative database. Main vaccine regimen in different birth cohorts was derived from the Quebec City Immunization Registry. Hospital admission risk was analyzed by Poisson regression models adjusting for age, season of birth, ambient air temperature, circulation of respiratory viruses, and the weekly hospital admission rate for all other causes excluding LRTI to control for temporal changes in hospital admission practices. In univariate analyses, hospitalizations for LRTI, pneumonia, and bronchiolitis were less frequent in cohorts exposed to PCVs than in unvaccinated cohorts with no difference between PCV regimens. For pneumonia, the difference in cumulative incidence was 16% (13%; 18%). In multivariate analyses, exposure to any PCV schedule was associated with a lower although statistically non-significant hospitalization risk for pneumonia ascompared with unvaccinated cohorts. Again, differences between PCV regimens were minimal. Interpretation of results of this ecological study should be made with care as many factors could influence hospitalizations for respiratory infection in young children. Results are compatible with a modest effect of PCVs in reducing hospitalizations for pneumonia in children. No substantial differences between various PCV schedules were observed.