Assessing the heat and cold effect of temperature on hospitaiization and outpatient by allergic rhinitis in Seoul, Korea

摘要

Background Many studies have investigated the association between the air pollutants and allergic disease with adjusting the temperature effect as confounding factor. The researches assessing the temperature effect on allergic disease are relatively small. The prevalence of allergic rhinitis (AR) in Korea is 14.5%, and it has been increasing steadily. AIMS In this study, we assess the lag effect of temperature on hospitalization or outpatient by AR at different sex or age group in Seoul, Korea. Also, we estimate both the heat and cold effect on AR. METHODS The distributed lag non-linear model was used to examine the lag effect of temperature on daily hospitalization or outpatient by AR. And then generalized additive model was used to estimate the cumulative relative risk (RR) for heat and cold effect of temperature, adjusting air pollutants, humidity, day of week, influenza epidemic, and long-term trend. RESULTS The lag effects of cold and heat temperature in total subject were estimated up to 21 days and on the day of hospitalization, and up to 12 days and on the day of visit in outpatient. The estimated RR for cold and heat effect in total subjects are 1.07(95%CI=1.01,1.33) and 1.33(95%CI=1.17,1.50) for hospitalization, and 0.88(95%CI=0.84,0.93) and 1.02(95%CI=0.93,1.11) for outpatient. The estimated lag effects of temperature by sex or age group were differed from estimated total subject. In addition, the results from estimated RRs for each age groups showed that the adolescent group (13≤age≤18) was most greatly influenced by heat (1.38(95%CI=1.13,1.68)) and cold (2.44(95%CI=1.72,3.46)) temperature in hospitalization by AR. Adjustment of the air pollutants by each or altogether, the estimated RRs do not show significant difference. CONCLUSIONS We found that there were different lag structures by sex or age group on AR. After considering the lag structure for each sex or age group, the result showed that the adolescents were most influenced by temperature on hospitalization by AR. Further investigation is needed in order to confirm these results.