Time series analysis of meteorological factors and air pollutants and their association with hospital admissions for acute myocardial infarction in Korea. (1st January 2021)

• Record Type: Journal Article
• Authors: Kim, Arim; Jung, Jaehun; Hong, Jinwook; Yoon, Seok-Jun
• Is Part Of: International journal of cardiology. Volume 322(2021)
• Page Start: 220
• Page End: 226
• Publication Date: 2021-01-01
• Subjects: Weather -- Air pollution -- Myocardial infarction -- Interrupted time series analysis; Cardiology -- Periodicals; Electronic journals; 616.12
• DOI: 10.1016/j.ijcard.2020.08.060
• Languages: English
• ISSNs: 0167-5273

Abstract

Abstract: Background: We assessed the association between multiple meteorological factors and air pollutants and the number of acute myocardial infarction (AMI) cases using a multi-step process. Methods: Daily AMI hospitalizations matched with 16 meteorological factors and air pollutants in 7 metropolitan provinces of the Republic of Korea from 2002 to 2017 were analyzed. We chose the best fit model after conducting the Granger causality (GC) test and examined the daily lag time effect on the orthogonalized impulse response functions. To define dose-response relationships, we performed a time series analysis using multiple generalized additive lag models based on seasons. Results: A total of 196, 762 cases of AMI in patients older than 20 years admitted for hospitalization were identified. The distribution of meteorological factors and air pollutants showed characteristics of a temperate climate. The GC test revealed a complex interaction between meteorological factors, including air pollutants, and AMI. The final selected factors were NO2 and temperature; these increased the incidence of AMI on lag day 4 during summer (NO2 : population-attributable fraction [PAF], 3.9%; 95% confidence interval [CI], 3.6–4.0; mean temperature: PAF, 3.3%; 95% CI, 2.7–3.9). Conclusions: This multi-step time series analysis found that average temperature and NO2 are the most important factors impacting AMI hospitalizations, specifically during summer. Based on the model, we were able to visualize the effect-time association of meteorological factors and air pollutants and AMI. Highlights: Meteorological factors and air pollutants affecting acute myocardial infarction show complex interplay. Simple linear regression dose not accurately explain for the effect of these risk factors. This multi-step time series analysis found that average temperature and NO2 are the most important factors, impacting AMI.