|Contribution title||Short-term association between air pollutant levels and lengths of hospitalizations of paediatric respiratory inpatients|
|Form of presentation||Poster|
Many studies have shown that ambient air pollution affects daily hospital admissions. However, there is a lack of studies on respective effects on the length of hospitalisation.
To examine the association between outdoor air pollution concentrations and length of hospitalisation in children (0-17 years old) from Hanoi with acute respiratory diseases.
We collected all records of children hospitalized for respiratory diseases in the Vietnam National Children’s Hospital from January 2007 to December 2014. Pooled logistic regression was used to assess the association between exposure to air pollution on the days prior to admission and length of hospitalisation (LOH) via the daily probability of being discharged. Children were hospitalized for various respiratory causes, including pneumonia, bronchitis, asthma, upper respiratory disease and other respiratory diseases. Seasonal variation in the length of hospitalisation was modelled using a bspline function. Models were adjusted for meteorological factors, children specific characteristics, holiday, season and number of children hospitalized in the Respiratory Department on the respective day, one . Odds ratios (OR) of the discharge probability were estimated for two exposure windows, namely during the week prior to the admission (lag 0-8) and the week before (lag 9-14).
Daily means of ambient concentrations in Hanoi were 93µg/m3 for PM10 and 56.1µg/m3 for PM2.5. LOH among all respiratory cases showed negative associations with average ambient pollutant levels during lag 0-8. An increase by 66.5μg/m3 in the respective level of PM10 and by 39.4μg/m3 in the one of PM2.5 was associated with an odds ratio of 1.04 (95%CI: 1.01-1.08) and 1.03 (95%CI: 1.00-1.07), respectively, of being discharged according to single-pollutant models. In contrast, a positive association between LOH and average ambient pollutant levels on days 14 to 8 prior to admission was observed. The OR was 0.97 (95%CI: 0.95-0.99) for a 66.5μg/m3 increase in lag 8-14 PM10 and 0.98 (95%CI: 0.95-1.00) for a 39.4μg/m3 increase in PM2.5. The associations remained consistent across all pollutants, except for Ozone.
The finding demonstrated that the length of hospitalizations due to respiratory diseases showed a negative association with average levels of PM10 and PM2.5 in the week before admission but a positive association with respective average levels between 14 and 8 days prior to admission