Can social vulnerability indices predict county trauma fatality rates?
Can social vulnerability indices predict county trauma fatality rates?
复制标题
DOI：
10.1097/ta.0000000000003228
复制
发表时间：
2021-08-01
期刊：
影响因子：
--
通讯作者：
Brown JB
中科院分区：
文献类型：
--
作者：
Phelos HM;Deeb AP;Brown JB
创建社会脆弱性指数是为了根据不同地理区域的社会经济和人口特征来衡量对环境灾害的恢复能力。它们由多个经过验证的结构组成，这些结构也可能在伤害发生后识别地理上的脆弱人群。我们的目的是确定这些指数是否与美国的伤害死亡率相关。
我们评估了三个社会脆弱性指数：灾害与脆弱性研究所的社会脆弱性指数（SoVI）、美国疾病控制与预防中心的社会脆弱性指数（SVI）以及经济创新集团的困境社区指数（DCI）。我们分析了SVI的子指数和常见的个体人口普查变量作为社会经济地位的指标。结果包括每10万人口中按年龄调整的县级总体、枪支和机动车碰撞（MVC）死亡人数。线性回归确定了伤害死亡率与SoVI、SVI和DCI的关联。双变量等值区域图确定了总体死亡率、SoVI和DCI的地理差异和空间自相关。
研究涵盖了美国3137个县。对于所有三个指数，只有24.6%的县处于相同的脆弱性四分位数。尽管如此，所有指数都与总体、枪支和MVC死亡率的上升相关。在总体伤害死亡率方面，DCI在模型拟合、方差解释和诊断性能上表现最佳。美国各县的SoVI、DCI和伤害死亡率存在显著的地理差异，SoVI具有中等程度的空间自相关（莫兰指数为0.35，p<0.01），伤害死亡率和DCI具有高度自相关（莫兰指数分别为0.77，p<0.01和0.53，p<0.01）。
虽然这些指数提供了独特的信息，但在所有指数中，较高的社会脆弱性都与较高的伤害死亡率相关。这些指数可能在伤害相关死亡率的流行病学和地理评估中有用。有必要进一步研究以确定这些指数是否优于创伤研究中使用的传统社会经济地位衡量指标及相关结构。
Social vulnerability indices were created to measure resiliency to environmental disasters based on socioeconomic and population characteristics of discrete geographic regions. They are comprised of multiple validated constructs that can also potentially identify geographically vulnerable populations after injury. Our objective was to determine if these indices correlate with injury fatality rates in the US. We evaluated three social vulnerability indices: The Hazards & Vulnerability Research Institute’s Social Vulnerability Index (SoVI), the CDC Social Vulnerability Index (SVI) and the Economic Innovation Group’s Distressed Community Index (DCI). We analyzed SVI sub-indices and common individual census variables as indicators of socioeconomic status. Outcomes included age-adjusted county-level overall, firearm, and motor vehicle collision (MVC) deaths per 100,000 population. Linear regression determined the association of injury fatality rates with the SoVI, SVI, and DCI. Bivariate choropleth mapping identified geographic variation and spatial autocorrelation of overall fatality, SoVI, and DCI. 3,137 US counties were included. Only 24.6% of counties fell into the same vulnerability quartile for all three indices. Despite this, all indices were associated with increasing fatality rates for overall, firearm, and MVC fatality. The DCI performed best by model fit, explanation of variance, and diagnostic performance on overall injury fatality. There is significant geographic variation in SoVI, DCI, and injury fatality rates at the county-level across the US, with moderate spatial autocorrelation of SoVI (Moran’s I 0.35, p<0.01) and high autocorrelation of injury fatality rates (Moran’s I 0.77, p<0.01) and DCI (Moran’s I 0.53, p<0.01). While the indices contribute unique information, higher social vulnerability is associated with higher injury fatality across all indices. These indices may be useful in the epidemiologic and geographic assessment of injury-related fatality rates. Further study is warranted to determine if these indices outperform traditional measures of socioeconomic status and related constructs used in trauma research.