Impact of influenza vaccine on childhood otitis media in Taiwan: A population-based study

PURPOSE: Acute otitis media (AOM) is a common infectious disease in children and usually accompanied by a preceding viral respiratory tract infection, especially in the preschool-age population. The study aimed to evaluate impact of influenza vaccine on childhood otitis media. METHODS: This retrospective cohort study included data for 803,592 children (<10 years old) recorded in Taiwan’s National Health Insurance Research Database. AOM incidence and tympanostomy tube insertion incidence in each influenza season before and after the introduction of traditional injectable trivalent influenza vaccine (TIV) were compared using the Poisson regression analysis to estimate the incidence rate ratios (IRRs) and 95% confidence intervals (CIs). RESULTS: In children < 2 years old, the age group eligible for free influenza vaccination, there was a significant reduction in seasonal AOM incidence after TIV introduction in 2004 (from 98.4 episodes/1000 person-seasons [95% CI: 96.4–100.5] to 66.1 episodes/1000 person-seasons [95% CI: 64–68.1]). In addition, with the increased vaccine coverage rate, the outpatient visits for AOM in the influenza season of 2005 and 2006 were significantly lower than that in 2004 (IRR = 0.85 and 0.80, respectively, p < 0.0001). CONCLUSIONS: A significant reduction in primary care consultations for children <2 years old was observed after the introduction of the TIV in Taiwan in 2004. With the increased vaccine coverage, there was an additional decline in 2005 and 2006. In addition of the direct protection provided by the vaccination, we believe that TIV may have induced some herd immunity that further contributed to the reduction in influenza attack rates and the rates of associated AOM in that age group. These reductions were observed only in vaccine-eligible children, while older children, who were not enrolled in the influenza vaccination program during the study period, have experienced increases in the AOM incidence during the 2004–2006 period compared to the 2000–2003 period.