Previous studies have demonstrated the importance of weather variables in influencing the incidence of influenza. However, the role of air pollution is often ignored in identifying the environmental drivers of influenza. This research aims to examine the impacts of air pollutants and temperature on the incidence of pediatric influenza in Brisbane, Australia. Lab-confirmed daily data on influenza counts among children aged 0-14years in Brisbane from 2001 January 1st to 2008 December 31st were retrieved from Queensland Health. Daily data on maximum and minimum temperatures for the same period were supplied by the Australian Bureau of Meteorology. Winter was chosen as the main study season due to it having the highest pediatric influenza incidence. Four Poisson log-linear regression models, with daily pediatric seasonal influenza counts as the outcome, were used to examine the impacts of air pollutants (i.e., ozone (O3), particulate matter≤10μm (PM10) and nitrogen dioxide (NO2)) and temperature (using a moving average of ten days for these variables) on pediatric influenza. The results show that mean temperature (Relative risk (RR): 0.86; 95% Confidence Interval (CI): 0.82-0.89) was negatively associated with pediatric seasonal influenza in Brisbane, and high concentrations of O3 (RR: 1.28; 95% CI: 1.25-1.31) and PM10 (RR: 1.11; 95% CI: 1.10-1.13) were associated with more pediatric influenza cases. There was a significant interaction effect (RR: 0.94; 95% CI: 0.93-0.95) between PM10 and mean temperature on pediatric influenza. Adding the interaction term between mean temperature and PM10 substantially improved the model fit. This study provides evidence that PM10 needs to be taken into account when evaluating the temperature-influenza relationship. O3 was also an important predictor, independent of temperature.
Keywords: Air pollution; BIC; Bayesian information criterion; CI; Influenza; Pediatric; RR; Temperature; confidence interval; relative risk.
© 2013 Elsevier Ltd. All rights reserved.