Exposure to ambient fine particulate matter (PM_2.5) is the largest environmental health risk in India. All previous studies in India focused on PM₁₀ exposure for acute impact assessment; even the single study examining the PM_2.5 acute exposure impact used data from a single site. Here we estimated the impact of acute PM_2.5 exposure on non-trauma all-cause mortality in the megacity Delhi using a quasi-Poisson regression model and satellite-derived high-resolution (1-km) exposure data. Our satellite-PM_2.5 dataset was calibrated and validated against coincident measurements from ground-based reference-grade monitors. This allowed us to minimize exposure misclassification, which otherwise would have happened due to reliance on limited ground-based monitoring sites with a large spatial gap. The annual average (median) PM_2.5 exposure (with interquartile range) during this period was 108.5 (63.5–133.6) μg/m³. We found a 0.52% (95% confidence interval, CI: 0.42–0.62%) increase in non-trauma all-cause mortality for every 10 μg/m³ increase in 6-days cumulative PM_2.5 exposure with a relatively higher impact on men (0.57%, 95% CI: 0.46–0.69%) than on women (0.52%, 0.38–0.65%). The impact of PM_2.5 was almost two-fold higher in winter (0.55%, 0.45–0.66%) per 10 μg/m³ increase in PM_2.5 than in summer (0.29%, 0.14–0.45%). The district-level aggregated 6-days cumulative PM_2.5 exposure had the largest impact on mortality (1.10%, 0.84–1.35%). The impact of acute exposure to PM_2.5 on all-cause mortality was larger than that of PM₁₀ exposure (from previous studies). Our study adds to the global evidence pool, fills the critical gap of evidence of acute PM_2.5 exposure impact on mortality in India, and most importantly, demonstrates the significance of disaggregated exposure assessment using satellite data to reduce exposure misclassification in health impact studies, particularly in data-poor regions.

Main finding

We estimate a 0.52% increase in all-cause mortality for every 10 μg/m³ rise in short-term PM_2.5 exposure with a two-fold higher impact in the winter than in the summer and a higher impact on male (0.57%) than on female (0.52%) in the megacity Delhi.

中文翻译：

---

急性暴露于环境 PM _2.5对大城市德里的非创伤全因死亡率的影响

暴露于环境细颗粒物 (PM _2.5 ) 是印度最大的环境健康风险。之前在印度的所有研究都集中在 PM ₁₀暴露以进行急性影响评估；甚至一项研究 PM _2.5急性暴露影响的研究也使用了来自单一地点的数据。在这里，我们使用准泊松回归模型和卫星衍生的高分辨率（1 公里）暴露数据估计了急性 PM _2.5暴露对大城市德里的非创伤全因死亡率的影响。我们的卫星-PM _2.5数据集根据来自地面参考级监视器的重合测量值进行校准和验证。这使我们能够最大程度地减少暴露错误分类，否则由于依赖具有较大空间差距的有限地面监测站点而可能会发生这种情况。在此期间，年平均（中位数）PM _2.5暴露（四分位距）为 108.5 (63.5–133.6) μg/m³。我们发现，在 6 天累积 PM _2.5暴露中每增加 10 μg/m³，非创伤全因死亡率增加 0.52%（95% 置信区间，CI：0.42–0.62%），对男性的影响相对较大（ 0.57%, 95% CI: 0.46–0.69%) 比女性 (0.52%, 0.38–0.65%)。PM _2.5的影响PM _2.5每增加 10 μg/m³，冬季 (0.55%, 0.45–0.66%) 几乎是夏季 (0.29%, 0.14–0.45%) 的两倍。地区级累计 6 天累积 PM _2.5暴露对死亡率的影响最大（1.10%、0.84-1.35%）。急性暴露于 PM _2.5对全因死亡率的影响大于暴露于PM _{10 的}影响（来自之前的研究）。我们的研究增加了全球证据库，填补了急性 PM _2.5暴露对印度死亡率影响的关键证据空白，最重要的是，证明了使用卫星数据进行分类暴露评估的重要性，以减少健康影响研究中的暴露错误分类，特别是在数据匮乏的地区。

主要发现

我们估计，短期 PM _2.5暴露每增加 10 μg/m³，全因死亡率就会增加 0.52% ，冬季的影响是夏季的两倍，对男性的影响（0.57%）比夏季高大城市德里的女性 (0.52%)。