The relationship between changes in daily air pollution and hospitalizations in Perth, Australia 1992-1998: a case-crossover study.

A case-crossover study was undertaken to investigate the relationship between daily air pollutant concentrations and daily hospitalizations for selected disease categories in Perth, Western Australia. Daily measurements of particles (measured by nephelometry and PM2.5), photochemical oxidants (measured as ozone), nitrogen dioxide (NO2) and carbon monoxide (CO) concentrations were obtained from 1992 to 1998 via a metropolitan network of monitoring stations. Daily PM2.5 concentrations were estimated using monitored data, modelling and interpolation. Hospital morbidity data for respiratory, cardiovascular (CVD), gastrointestinal (GI) diseases, chronic obstructive pulmonary diseases (COPD) excluding asthma; pneumonia/influenza diseases; and asthma were obtained and categorized into all ages, less than 15 years and greater than 65 years. Gastrointestinal morbidity was used as a control disease. The data were analyzed using conditional logistic regression. The results showed a small number of significant associations for daily changes in particle concentrations, nitrogen dioxide and carbon monoxide for the respiratory diseases, CODP, pneumonia, asthma and CVD hospitalizations. Changes in ozone concentrations were not significantly associated with any disease outcomes. These data provide useful information on the potential health impacts of air pollution in an airshed with very low sulphur dioxide concentrations and lower nitrogen dioxide concentrations commonly found in many other cities.