PIN70 ASSOCIATION BETWEEN INFLUENZA VACCINE EFFECTIVENESS AND HEALTHCARE UTILIZATIONS AMONG US ELDERLY POPULATION

Abstract

Influenza vaccine effectiveness (VE) can vary from year to year due to vaccine mismatch with the influenza virus. However, it was not known well on how fluctuating VE affected on health. This study aims to investigate how the VE level affected broad health outcomes among elderly population. We examined the effect of the 2014-15 flu season on all-cause hospitalization and emergency department (ED) visits and compared these results to the 2013-14 and 2015-16 seasons, utilizing 2014-2016 National Health Interview Survey (NHIS) data. The target population was 65+ years old. We conducted a pooled cross-sectional study employing an adjusted logistic regression model, where outcome variables include hospitalization and ED visit in the survey year. We applied a difference-in-difference (DID) approach, which indicated the low VE effect on hospitalization and ED visit without any time trend bias. A key covariate was an interaction term between a flu vaccination status and year variable indicating seasonal flu season (Low VE or High VE). We found that vaccinated elderly in the high VE year of 2013 were 25% (Odds Ratio=0.75, p=0.031) less likely to get hospitalization than those in the low VE year of 2014. However, there was no difference of hospitalization between high VE year 2015-vaccinated elderly and low VE year 2014-vaccinated elderly. In addition, there was no difference in ED visits among elderly vaccinated patients over the three-year period. Hospitalization could increase in the low VE year compared to the high VE year, but the influence might vary over years. Thus, it is necessary to further pursue this line of investigation using more complete health records in a long period of time.