Geographical heterogeneity and influenza infection within households

BACKGROUND Although it has been suggested that schoolchildren vaccination reduces influenza morbidity and mortality in the community, it is unknown whether geographical heterogeneity would affect vaccine effectiveness. METHODS A 3-year prospective, non-randomized sero-epidemiological study was conducted during 2008-2011 by recruiting schoolchildren from both urban and rural areas. Respective totals of 124, 206, and 176 households were recruited and their household contacts were followed. Serum samples were collected pre-vaccination, one-month post-vaccination and post-season from children and household contacts for hemagglutination inhibition (HI) assay. A multivariate logistic model implemented with generalized estimation equations (GEE) was fitted with morbidity or a four-fold increase in HI titer of the household contacts for two consecutive sera as the dependent variable; with geographical location, vaccination status of each household and previous vaccination history as predictor variables. RESULTS Although our results show no significant reduction in the proportion of infection or clinical morbidity among household contacts, a higher risk of infection, indicated by odds ratio>1, was consistently observed among household children contacts from the un-vaccinated households after adjusting for confounding variables. Interestingly, a statistically significant lower risk of infection was observed among household adult contacts from rural area when compared to those from urban area (OR=0.89; 95% CI: 0.82-0.97 for Year 2 and OR=0.85; 95% CI: 0.75-0.96 for Year 3). CONCLUSIONS A significant difference in the risk of influenza infection among household adults due to geographical heterogeneity, independent of schoolchildren vaccination status, was revealed in this study. Its impact on vaccine effectiveness requires further study.