The population impact of a large school-based influenza vaccination campaign.

Abstract

The optimal vaccination strategy to mitigate the impact of influenza epidemics is unclear. In 2005, a countywide school-based influenza vaccination campaign was launched in Knox County, Tennessee (population 385,899). Approximately 41% and 48% of eligible county children aged 5-17 years were immunized with live attenuated influenza vaccine before the 2005-2006 and 2006-2007 influenza seasons, respectively. We sought to determine the population impact of this campaign.Laboratory-confirmed influenza data defined influenza seasons. We calculated the incidence of medically attended acute respiratory illness attributable to influenza in Knox and Knox-surrounding counties (concurrent controls) during consecutive seasons (5 precampaign and 2 campaign seasons) using negative binomial regression and rate difference methods. Age-stratified analyses compared the incidence of emergency department (ED) visits and hospitalizations attributable to influenza.During precampaign seasons, estimated ED visit rates attributable to influenza were 12.39 (95% CI: 10.34-14.44) per 1000 Knox children aged 5-17 years and similar in Knox-surrounding counties. During the campaign seasons, annual Knox influenza-associated ED visit rates declined relative to rates in Knox-surrounding counties: rate ratios 0.55 (95% CI: 0.27-0.83) and 0.70 (95% CI: 0.56-0.84) for the first and second campaign seasons, respectively. Overall, there were about 35% or 4.86 per 1000 fewer influenza-associated ED visits among Knox County children aged 5-17 years attributable to the campaign. No significant declines in Knox compared to surrounding counties were detected for influenza associated ED visits in children aged <5 years, all adults combined or selected adult age subgroups, although power for these analyses was limited. Alternate rate-difference analyses yielded consistent results.Vaccination of approximately 45% of Knox school-aged children with influenza vaccine was associated with a 35% annual reduction (4.86 per 1000) in ED visit rates attributable to influenza. Higher vaccination coverage and/or larger studies would be needed to determine whether similar interventions have indirect benefits in other age groups.