The incidence of symptomatic infection with influenza virus in the Netherlands 2011/2012 through 2016/2017, estimated using Bayesian evidence synthesis

Abstract

Due to differences in the circulation of influenza viruses, distribution and antigenic drift of A subtypes and B lineages, and susceptibility to infection in the population, the incidence of symptomatic influenza infection can vary widely between seasons and age-groups. Our goal was to estimate the symptomatic infection incidence in the Netherlands for the six seasons 2011/2012 through 2016/2017, using Bayesian evidence synthesis methodology to combine season-specific sentinel surveillance data on influenza-like illness (ILI), virus detections in sampled ILI cases and data on healthcare-seeking behaviour. Estimated age-aggregated incidence was 6.5 per 1000 persons (95% uncertainty interval (UI): 4.7-9.0) for season 2011/2012, 36.7 (95% UI: 31.2-42.8) for 2012/2013, 9.1 (95% UI: 6.3-12.9) for 2013/2014, 41.1 (95% UI: 35.0-47.7) for 2014/2015, 39.4 (95% UI: 33.4-46.1) for 2015/2016 and 27.8 (95% UI: 22.7-33.7) for season 2016/2017. Incidence varied substantially between age-groups (highest for the age-group <5 years: 23 to 47/1000, but relatively low for 65+ years: 2 to 34/1000 over the six seasons). Integration of all relevant data sources within an evidence synthesis framework has allowed the estimation - with appropriately quantified uncertainty - of the incidence of symptomatic influenza virus infection. These estimates provide valuable insight into the variation in influenza epidemics across seasons, by virus subtype and lineage, and between age-groups.