Institute for Urban and Regional Development, University of California Berkeley, Berkeley, CA, USA.
Urban Health Collaborative, Drexel Dornsife School of Public Health, Philadelphia, PA, USA.
Health Place. 2021 Nov;72:102703. doi: 10.1016/j.healthplace.2021.102703. Epub 2021 Nov 6.
Recent studies highlight the equigenic potential of greenspaces by showing narrower socioeconomic health inequalities in greener areas. However, results to date have been inconsistent and derived from high-income countries. We examined whether urban greenness modifies the associations between area-level education, as a proxy for socioeconomic status, and life expectancy and cause-specific mortality in Latin American cities.
We included 28 large cities, >137 million inhabitants, in nine Latin American countries, comprising 671 sub-city units, for 2012-2016. Socioeconomic status was assessed through a composite index of sub-city level education, and greenness was calculated using the normalized difference vegetation index. We fitted multilevel models with sub-city units nested in cities, with life expectancy or log(mortality) as the outcome.
We observed a social gradient, with higher levels of education associated with higher life expectancy and lower cause-specific mortality. There was weak evidence supporting the equigenesis hypothesis as greenness differentially modified the association between education and mortality outcomes. We observed an equigenic effect, with doubling magnitudes in the violence-related mortality reduction by education in areas with low greenness compared to medium-high greenness areas among men (16% [95% CI 12%-20%] vs 8% [95% CI 4%-11%] per 1 SD increase in area-level education). However, in contradiction to the equigenesis hypothesis, the magnitude in cardiovascular diseases (CVD) mortality reduction by education was stronger in areas with medium-high greenness compared to areas with low greenness (6% [95% CI 4%-7%] vs 1% [95% CI -1%-3%] and 5% [95% CI 3%-7%] vs 1% [95% CI -1%-3%] per 1 SD increase in area-level education, in women and men, respectively). Similarly, each 1-SD increase in greenness widened the educational inequality in life expectancy by 0.15 years and 0.20 years, in women and men, respectively. The equigenic effect was not observed in violence-related mortality among women and in mortality due to communicable diseases, maternal, neonatal and nutritional conditions (CMNN).
Our results confirm socioeconomic health inequalities in Latin American cities and show that the equigenic properties of greenspace vary by health outcome. Although mixed, our findings suggest that future greening policies should account for local social and economic conditions to ensure that greenspaces provide health benefits for all, and do not further exacerbate existing health inequalities in the region.
Wellcome Trust (Grant, 205177/Z/16/Z).
最近的研究强调了绿地的等基因潜力，表明在绿化程度较高的地区，社会经济健康不平等程度较小。然而，迄今为止的结果并不一致，而且来自高收入国家。我们研究了城市绿化是否改变了以地区教育水平（代表社会经济地位）与拉丁美洲城市的预期寿命和特定原因死亡率之间的关联。
我们纳入了 9 个拉丁美洲国家的 28 个大城市，超过 1.37 亿居民，包含 671 个次城市单元，研究时间为 2012 年至 2016 年。社会经济地位通过次城市级教育的综合指数来评估，绿化程度使用归一化植被指数计算。我们采用多水平模型，将次城市单元嵌套在城市中，以预期寿命或对数死亡率作为结果。
我们观察到一种社会梯度，较高的教育水平与较高的预期寿命和较低的特定原因死亡率相关。有微弱的证据支持等基因假说，因为绿化程度不同程度地改变了教育与死亡率结果之间的关联。我们观察到一种等基因效应，在绿化程度较低的地区，与教育相关的暴力相关死亡率降低的幅度是绿化程度中等偏高地区的两倍（男性中每增加 1 个标准差的地区教育水平，死亡率降低 16%[95%CI 12%-20%]；而在中等偏高地区，死亡率降低 8%[95%CI 4%-11%]）。然而，与等基因假说相反，在心血管疾病（CVD）死亡率方面，与教育相关的死亡率降低幅度在绿化程度中等偏高的地区比绿化程度较低的地区更强（女性中每增加 1 个标准差的地区教育水平，死亡率降低 6%[95%CI 4%-7%]；而在绿化程度较低的地区，死亡率降低 1%[95%CI -1%-3%]；男性中，每增加 1 个标准差的地区教育水平，死亡率降低 5%[95%CI 3%-7%]；而在绿化程度较低的地区，死亡率降低 1%[95%CI -1%-3%]）。同样，绿化程度每增加 1 个标准差，女性的预期寿命和男性的预期寿命的教育不平等程度分别增加 0.15 年和 0.20 年。女性的暴力相关死亡率和传染性疾病、孕产妇、新生儿和营养状况（CMNN）所致死亡率均未观察到等基因效应。
我们的结果证实了拉丁美洲城市的社会经济健康不平等现象，并表明绿地的等基因特性因健康结果而异。尽管结果不一致，但我们的研究结果表明，未来的绿化政策应考虑当地的社会和经济条件，以确保绿地为所有人带来健康益处，并且不会进一步加剧该地区现有的健康不平等现象。
惠康信托基金（资助，205177/Z/16/Z）。
