Research On The Effects Of Mereorological Factors And Air Pollutants On Health Based On The GAM Model

Meteorological factors and air pollutants have an important influence on outbreak of infectious disease. Statistical methods about the effects of air pollution and meteorological factors on infectious disease would provide scientific foundations for China’s government sector of the Center for Disease Control and Prevention to make policy and methods for preventing and controlling the infectious diseases.The generalized additive models（GAM） were used to investigate the effects of meteorological factors on daily hospital admissions for tuberculosis disease from January 1, 2010 to November 30, 2014 in Wenzhou. Long term time trends, the day of the week and public holidays were controlled in the model. After controlling for the three confounding factors above mentioned, as well as extremely hot weather, cold weather and extremely relative humidity were controlled. And the same methods above were used to study the effects of air pollutants on daily hospital admissions for tuberculosis disease in Shaoxing from January 2013 to November 2014. The main conclusions of this thesis were as follows.The meteorological factors have some effect on the daily hospital admission of tuberculosis disease. An interquartile range increase for mean temperature, lowest temperature and lowest surface temperature on the current day were will lead to 16.95%, 19.52% and 10.38% increasement in hospital admissions for tuberculosis diseases, respectively. Every interquartile range increase in highest temperature and mean surface temperature was associated with 8.38% and 7.13% increase in tuberculosis diseases, respectively. The highest surface temperature significant has effects on tuberculosis with a strongest effect on lag 6 day. The effect of the estimate of air pressure estimates was relatively small. An interquartile range increase of mean pressure and lowest pressure on lag 4 day were significantly associated with 5.78% and 5.01% increase in tuberculosis diseases, respectively. And a 5.01% increase in tuberculosis with an interquartile range increased in highest pressure at lag 7 day. Moreover, the average wind speed on the current day and the maximum wind speed on the lag 1 day have the strongest negative estimated effect on it, respectively. Adjusting for other meteorological factors had little influence on their effect estimates. The results of stratified analysis by the gender showed that the effects of meteorological factors on men was more significant than women.Air pollutants have the significant lagging effects on daily hospital admissions for tuberculosis disease. A 6.66% increase in hospital admissions for tuberculosis diseases was associated with a 40-μg/m3 increase in the 7-day moving average concentration of PM2.5. An interquartile range increase in 3-day moving average concentration of PM10 and SO₂ were significantly contributed to the percentage increase of 6.82%, 14.41% for all hospital admissions for tuberculosis diseases, respectively. A 23-μg/m3 increase in the 4-day moving average concentration of O3 was lead to a 15.55% increase in hospital admissions for tuberculosis diseases, and this effect was attenuated after adjusting for PM25, PM10 and SO₂. Adjusting for gaseous pollutants also had significant influence on particulate pollutants effect estimates. But adjusting for other pollutants had little influence on the effect estimates of SO₂. The altering models assumptions also had little influence on pollutants effect estimates. The results of stratified analysis by the gender and occupation showed that the effects of pollutants on migrant worker was more significant than others.