Background: Asthma attack shows strong seasonality. The purpose of the present study was to quantify the contribution of climate variables and other seasonal factors on the incidence of emergency visits for childhood asthma in Tokyo, Japan.
Methods: The number of children who visited emergency rooms at Jikei university hospitals in Tokyo during 1998-2002 (5559 visits) was retrieved retrospectively from files from the Department of Pediatrics, and compared with 45 climate parameters from the Meteorological Agency using multiple regression models with a stepwise backward elimination approach.
Results: The number of visits (3.7 +/- 3.1) per night increased significantly when climate conditions showed a rapid decrease from higher barometric pressure, from higher air temperature and from higher humidity, as well as lower wind speed. The best-fit model demonstrated that a 22% variation in the number of visits was explained by a linear relationship with 12 climate variables, which increased to 36% after adjusting for calendar month and day of the week. Moreover, when the number of asthma visits was cut off at nine per night, the area under the receiver operator characteristics curve was 0.91 (95% CI: 0.89-0.94) in the multiple logistic regression model using the same variables.
Conclusions: These results suggest that these models might quantify contributions of specific climate conditions and other seasonal factors on the number of emergency visits per night for childhood asthma attack in Tokyo, Japan.