Li Deshun, Dong Jiyuan, Liu Xiaoju, Shu Juan, Zhu Lisha, Bao Hairong
The 1st School of Clinical Medicine, Lanzhou University, Lanzhou, 730000, China.
School of Public Health, Lanzhou University, Lanzhou, 730000, China.
Sci Rep. 2025 Aug 2;15(1):28252. doi: 10.1038/s41598-025-10116-9.
Numerous studies have identified air pollutant as a primary risk factor for chronic obstructive pulmonary disease (COPD), particularly for fine particulate matter (PM). However, the daily average levels of PM may not accurately reflect the actual exposure level due to the fluctuating air pollutant levels throughout the day and different human activity patterns. Few studies have comparatively analyzed the association between different exposure metrics of PM and COPD hospitalization. We aimed to explore the association between PM and COPD admission in Tianshui city using five different exposure metrics (a) Daily mean concentration (DailyPMmean), (b) Daily hourly peak concentration (DailyPMmax), (c) Daily morning concentration (DailyPMmor), (d) Daily evening concentration (DailyPMeve) and (e) Daily excessive concentration hours (PMDECH). The PMDECH was defined as daily total concentration hours > 25 µg/m. We collected daily records of COPD admission and hourly data on air pollutants and meteorological factors from Tianshui, China, 2018-2019. DailyPMmean, DailyPMmax, DailyPMmor, DailyPMeve and PMDECH were used as the exposure variables. A generalized additive model (GAM) based Poisson-distribution combined with a distributed lag non-linear model (DLNM) was applied to quantify the association between five different exposure metrics of PM and COPD admission. Stratified analyses were conducted to examine the effect modifications of gender, age, and season. The analysis revealed significant associations between five different exposure metrics of PM and COPD admission. PM-related risks of COPD admission differed by five exposure metrics. An interquartile range (IQR) increment in DailyPMmean at lag06, DailyPMmax at lag06, DailyPMmor at lag04, DailyPMeve at lag06 and PMDECH at lag07 was associated with a 21.64%(95%CI: 7.46%, 36.68%), 12.84%(95%CI: 1.47%, 24.58%), 16.47%(95%CI: 0.96%, 33.01%), 31.54%(95%CI: 16.49%, 47.58%), and 30.37%(95%CI: 13.91%, 46.88%) increase in Tianshui's COPD admissions, respectively. Associations of COPD admission with five different exposure metrics of PM appeared to be stronger in the cold season and among the females, while we found significantly higher effects in the population < 65 years old. The five metrics of PM exposure and COPD hospitalization show an exposure-response relationship that resembles a function curve that increases monotonically. This study added evidence for increased risk of admissions associated with exposure to the five metrics of ambient PM in Tianshui City, China, and DailyPMeve or PMDECH may be a significant risk factor for the increased risk of COPD admission, indicating that we should focus more on the relationship between air pollution and related diseases during the evening peak hours.
众多研究已将空气污染物确定为慢性阻塞性肺疾病（COPD）的主要风险因素，尤其是细颗粒物（PM）。然而，由于全天空气污染物水平波动以及不同的人类活动模式，PM的日均水平可能无法准确反映实际暴露水平。很少有研究对PM的不同暴露指标与COPD住院之间的关联进行比较分析。我们旨在使用五种不同的暴露指标（a）日均浓度（DailyPMmean）、（b）日每小时峰值浓度（DailyPMmax）、（c）日早晨浓度（DailyPMmor）、（d）日傍晚浓度（DailyPMeve）和（e）日超标浓度小时数（PMDECH），探讨中国天水市PM与COPD入院之间的关联。PMDECH定义为日总浓度小时数>25μg/m³。我们收集了2018 - 2019年中国天水市COPD入院的每日记录以及空气污染物和气象因素的每小时数据。将DailyPMmean、DailyPMmax、DailyPMmor、DailyPMeve和PMDECH用作暴露变量。应用基于泊松分布的广义相加模型（GAM）并结合分布滞后非线性模型（DLNM）来量化PM的五种不同暴露指标与COPD入院之间的关联。进行分层分析以检验性别、年龄和季节的效应修正。分析揭示了PM的五种不同暴露指标与COPD入院之间存在显著关联。与PM相关的COPD入院风险因五种暴露指标而异。滞后06时的DailyPMmean、滞后06时的DailyPMmax、滞后04时的DailyPMmor、滞后06时的DailyPMeve和滞后07时的PMDECH每增加一个四分位数间距（IQR），分别与天水市COPD入院增加21.64%（95%置信区间：7.46%，36.68%）、12.84%（95%置信区间：1.47%，24.58%）、16.47%（95%置信区间：0.96%，33.01%）、31.54%（95%置信区间：16.49%，47.58%）和30.37%（95%置信区间：13.91%，46.88%）相关。COPD入院与PM的五种不同暴露指标之间的关联在寒冷季节和女性中似乎更强，而我们发现在<65岁人群中效应显著更高。PM暴露的五个指标与COPD住院呈现出类似于单调递增函数曲线的暴露 - 反应关系。本研究补充了证据，表明在中国天水市，暴露于环境PM的五个指标与入院风险增加相关，并且DailyPMeve或PMDECH可能是COPD入院风险增加的重要风险因素，这表明我们应更多地关注傍晚高峰时段空气污染与相关疾病之间的关系。
