Objective: This study aimed to determine the spatiotemporal distribution and epidemiological characteristics of hospital admissions for carbon monoxide poisoning (COP) in Guangdong, China, from 2013 to 2020.
Methods: Data on age- and sex- specific numbers of hospital admissions due to COP in Guangdong (2013-2020) were collected. Daily temperatures were downloaded through the China Meteorological Data Sharing Service System. We analyzed temporal trends through time series decomposition and used spatial autocorrelation analysis to detect spatial clustering. The distributed lag nonlinear model was used to quantify the effects of temperature.
Results: There were 48,854 COP admissions over the study period. The sex ratio (male to female) was 1:1.74. The concentration ratios (M) ranged from 0.73-0.82. The highest risk occurred in January (season index = 3.59). Most cases were concentrated in the northern mountainous areas of Guangdong with high-high clustering. COP in the study region showed significant spatial autocorrelation, and the global Moran's Ivalue of average annual hospital admission rates for COP was 0.447 ( P < 0.05). Low temperatures were associated with high hospital admission rates for COP, with a lag lasting 7 days. With a lag of 0 days, the effects of low temperatures [5th (12 °C)] on COP were 2.24-3.81, as compared with the reference temperature [median (24 °C)].
Conclusion: COP in Guangdong province showed significant temporal and spatial heterogeneity. Low temperature was associated with a high risk of COP, and the influence had a lag lasting 7 days.
Keywords: Carbon monoxide poisoning; Distributed lag nonlinear model; Hospitalization; Spatial-temporal model.
Copyright © 2022 The Editorial Board of Biomedical and Environmental Sciences. Published by China CDC. All rights reserved.