BACKGROUND Though observational studies have widely linked air pollution exposure to various chronic diseases, evidence comparing different exposures in the same people is limited. This study examined associations between changes in air pollution exposure due to relocation and the incidence and mortality of 14 major diseases. METHODS We included 50,522 participants enrolled in the UK Biobank from 2006 to 2010. Exposures to particulate matter with a diameter ≤2.5μm (PM2.5), particulate matter with a diameter ≤10μm (PM10), nitrogen oxides (NOx), nitrogen dioxide (NO2), and sulfur dioxide (SO2) were estimated for each participant based on their residential address and relocation experience during the follow-up. Nine exposure groups were classified based on changes in long-term exposures due to residential mobility. Incidence and mortality of 14 major diseases were identified through linkages to hospital inpatient records and death registries. Cox proportional hazard models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for incidence and mortality of the 14 diseases of interest. RESULTS During a median follow-up of 12.6 years, 29,869 participants were diagnosed with any disease of interest, and 3,144 died. Significantly increased risk of disease and all-cause mortality was observed among individuals who moved from a lower to higher air polluted area. Compared with constantly low exposure, moving from low to moderate PM2.5 exposure was associated with increased risk of all 14 diseases but not for all-cause mortality, with adjusted HRs (95% CIs) ranging from 1.18 (1.05, 1.33) to 1.48 (1.30, 1.69); moving from low to high PM2.5 areas increased risk of all 14 diseases: infections [1.37 (1.19, 1.58)], blood diseases [1.57 (1.34, 1.84)], endocrine diseases [1.77 (1.50, 2.09)], mental and behavioral disorders [1.93 (1.68, 2.21)], nervous system diseases [1.51 (1.32, 1.74)], ocular diseases [1.76 (1.56, 1.98)], ear disorders [1.58 (1.35, 1.86)], circulatory diseases [1.59 (1.42, 1.78)], respiratory diseases [1.51 (1.33, 1.72)], digestive diseases [1.74 (1.58, 1.92)], skin diseases [1.39 (1.22, 1.58)], musculoskeletal diseases [1.62 (1.45, 1.81)], genitourinary diseases [1.54 (1.36, 1.74)] and cancer [1.42 (1.24, 1.63)]. We observed similar associations for PM10 and SO2 with 14 diseases (but not with all-cause mortality); increases in NO2 and NOx were positively associated with 14 diseases and all-cause mortality. CONCLUSIONS This study supports potential associations between ambient air pollution exposure and morbidity as well as mortality. Findings also emphasize the importance of maintaining consistently low levels of air pollution to protect the public's health. https://doi.org/10.1289/EHP14367.

中文翻译：

---

搬迁导致空气污染物暴露变化与 14 种主要疾病类别和全因死亡率发生率之间的关联：一项自然实验研究。


背景 尽管观察性研究已将空气污染暴露与各种慢性疾病广泛联系起来，但比较同一人不同暴露的证据是有限的。本研究考察了搬迁导致空气污染暴露变化与 14 种主要疾病的发病率和死亡率之间的关联。方法 我们纳入了 2006 年至 2010 年在英国生物样本库注册的 50,522 名参与者。根据每个参与者的居住地址和搬迁经历，估计每个参与者暴露于直径≤2.5μm （PM2.5）、直径 ≤10μm 的颗粒物 （PM10）、氮氧化物 （NOx）、二氧化氮 （NO2） 和二氧化硫 （SO2） 在随访期间根据他们的居住地址和搬迁经历。根据住宅流动性引起的长期暴露变化对 9 个暴露组进行分类。通过与医院住院记录和死亡登记的联系确定了 14 种主要疾病的发病率和死亡率。Cox 比例风险模型用于估计 14 种相关疾病的发病率和死亡率的风险比 （HRs） 和 95% 置信区间 （CIs）。结果 在中位 12.6 年的随访期间，29,869 名参与者被诊断出患有任何感兴趣的疾病，3,144 名参与者死亡。在从空气污染较低地区迁移到较高空气污染地区的个体中观察到疾病风险和全因死亡率显着增加。与持续的低暴露相比，从低 PM2.5 暴露转向中度 PM2.5 暴露与所有 14 种疾病的风险增加相关，但与全因死亡率无关，调整后的 HR （95% CI） 范围为 1.18 （1.05， 1.33） 至 1.48 （1.30， 1.69）;从低 PM2.5 区域移动到高 PM2.5 区域会增加所有 14 种疾病的风险：感染 [1.37 （1.19， 1.58）]、血液疾病 [1.57 （1.34， 1.84）]]、内分泌疾病 [1.77 （1.50， 2.09）]、精神和行为障碍 [1.93 （1.68， 2.21）]、神经系统疾病 [1.51 （1.32， 1.74）]、眼部疾病 [1.76 （1.56， 1.98）]、耳部疾病 [1.58 （1.35， 1.86）]、循环系统疾病 [1.59 （1.42， 1.78）]、呼吸系统疾病 [1.51 （1.33， 1.72）]、消化系统疾病 [1.74 （1.58， 1.92）]、皮肤病 [1.39 （1.22， 1.58）]、肌肉骨骼疾病 [1.62 （1.45， 1.81）]、泌尿生殖系统疾病 [1.54 （1.36， 1.74）] 和癌症 [1.42 （1.24， 1.63）]。我们观察到 PM10 和 SO2 与 14 种疾病（但不与全因死亡率）有类似的关联;NO2 和 NOx 的增加与 14 种疾病和全因死亡率呈正相关。结论 本研究支持环境空气污染暴露与发病率和死亡率之间的潜在关联。研究结果还强调了保持始终如一的低水平空气污染以保护公众健康的重要性。https://doi.org/10.1289/EHP14367。