Background: Exposure to wildfire smoke has been associated with increases in measures of respiratory health. Forecasts of smoke exposure can provide prospective information to support public health responses. The BlueSky Wildfire Smoke Forecasting System (BlueSky) predicts hourly PM2.5 (particulate matter ＜ 2.5 micron in aerodynamic diameter) concentrations from wildfires for western Canada. Previous evaluation suggests reasonable agreement of BlueSky predictions with PM2.5 measurements and remote sensing data. Assessing the association between BlueSky predictions and health indicators is another important step for evaluating the role of BlueSky in public health decision making. This association has not been investigated in any previous evaluation of forecasting models. Aims: To assess the associations between BlueSky predictions and population-scale indicators of respiratory health during the 2010 fire season in British Columbia, Canada. Methods: The daily counts of prescription asthma reliever dispensations and asthma outpatient physician visits were aggregated for each geographic health unit. Smoke exposures predictions from BlueSky and daily PM2.5 measurements from monitoring stations were assigned to each unit. Poisson regression was used to estimate the association between these exposures and health indicators. Analyses were also stratified by areas with different extents of smoke impact. Results: For an increase of 30μg/m3 in BlueSky forecasted PM2.5, the relative rate (RR) for asthma reliever dispensations was 1.04 (95% CI 1.02 -1.05), compared with 1.12 (95% CI 1.08 -1.15) for PM2.5 measurements. Significant effects were observed in areas constantly covered by smoke plumes, but not in less smoke-impacted areas. Smaller but significant effects were estimated for asthma physician visits. Conclusions: BlueSky forecasts were significantly associated with respiratory health indicators. These results support its utilityfor informing public health decision making.
展开▼
