Source Specific Air Pollution's Link to Hospitalizations and Emergency Department Visits for Influenza or Bacterial Pneumonia in Adults: The New York State Accountability Study

Fine particulate air pollution (PM2.5) has been linked to respiratory infections including influenza and bacterial pneumonia, but the response of respiratory infections to source specific PM concentrations (e.g. traffic, wood smoke) remains an area of active research. Using source specific PM2.5 concentrations generated from Positive Matrix Factorization analyses at 6 urban sites in New York State (Buffalo, Rochester, Albany, Queens, Bronx, and Manhattan), we examined the association between source specific pollutants and the excess rate (ER) of hospitalizations and emergency department (ED) visits for influenza or bacterial pneumonia. We retrieved all hospital admissions and ED visits with a primary diagnosis of influenza or bacterial pneumonia from 2005 to 2016 from the Statewide Planning and Research Cooperative System (SPARCS) database. Using a case-crossover design and conditional logistic regression, we estimated the rate of influenza or bacterial pneumonia (for both hospitalizations and ED visits) associated with increased concentrations of source specific PM2.5 including ammonium sulfate (AS), ammonium nitrate (AN), biomass burning (BB), pyrolyzed organic carbon (OP), road dust (RD), diesel (DES) and spark-ignition (GAS) vehicles. In preliminary analyses, interquartile range increases (IQR) in GAS concentrations in the previous 3 days were associated with an increased rate of influenza hospitalizations (ER = 13.8%; 95% CI: 5.4, 22.9) and ED visits (ER = 13.4%; 95% CI: 6.4, 20.8) with similar findings for DES and AN in both care settings. The association between bacterial pneumonia and source specific PM was inconsistent, with unexpected findings of reduced bacterial pneumonia hospitalizations associated with increases in both OP and DES, but increased ED visits associated with increases in RD. Future analyses will adjust for PM2.5 mass, but increases in secondary PM2.5 appear to be associated with an increased rate of influenza but not bacterial pneumonia