Assessment of virus interference in a test-negative study of influenza vaccine effectiveness

Abstract

BACKGROUND: The test-negative study design (TND) is an observational study design increasingly used to estimate influenza vaccine effectiveness (VE). In this variant of the case control study, one important assumption is that receipt of influenza vaccination does not affect the risk of being infected with a virus other than influenza virus. We used data from the Influenza Incidence Surveillance Project in the United States to evaluate the association between receipt of influenza vaccination and the risk of infection with non-influenza viruses. METHOD: Patients of age ≥ 6 months presenting for ambulatory care with acute respiratory infections were tested for influenza and 5 other common respiratory viruses. We used conditional logistic regression to obtain the odds ratio of influenza vaccination in cases versus controls, adjusting for age group and sex, and estimated VE as one minus the adjusted odds ratio. We evaluated the sensitivity of VE estimates by choosing three control groups: all patients that tested negative for influenza virus (VE(ANY-)), patients that tested negative for influenza but positive for another respiratory virus (VE(ORV+)) and patients that tested negative for influenza and other respiratory viruses (VE(PAN-)). We further examined the association between influenza vaccination and detection of other respiratory viruses among patients test negative for influenza viruses. RESULTS: During the 2010-11, 2011-12 and 2012-13 influenza seasons, influenza was detected in 3,743 of 10,650 patients (35.1%). The overall VE was found to be modest across three years: VE(ANY-) was 47% (95% CI: 42%, 52%), VE(ORV+) was 51% (44%, 57%), and VE(PAN-) was 44% (38%, 50%). VE estimates with each control group were consistent overall or when stratified by age groups, influenza season, early/middle/late phase within each season and influenza type/subtype. We found no statistically significant association between influenza vaccination and detection of RSV, rhinovirus, PIV 1-3, MPV and adenovirus for each age group compared with pan-negative controls. CONCLUSION: In this 3-year test-negative study in the United States, we did not find any evidence that receipt of influenza vaccination affected the risk of infection with another respiratory virus. Among patients testing negative for an influenza virus, we found no significant associations between detection of other respiratory viruses and receipt of influenza vaccination.