Open Data Challenge to Examine the Impact of Social Determinants of Health on Stroke.
Open Data Challenge to Examine the Impact of Social Determinants of Health on Stroke.
复制标题
开放数据挑战赛旨在检验健康的社会决定因素对中风的影响。
DOI：
10.1161/strokeaha.123.042645
复制
发表时间：
2023
期刊：
影响因子：
8.3
通讯作者：
Roth,GregoryA
中科院分区：
文献类型：
--
作者：
Hall,JenniferL;Roth,GregoryA
美国心脏协会（AHA）向研究人员和临床医生开放了卒中数据挑战，以促进对健康的社会和结构决定因素如何影响卒中表现，护理质量和临床结局的理解。为了解决这些问题，来自华盛顿大学健康评估与评价研究所的县级死亡率和健康社会决定因素的估计值已与AHA获得国家卒中登记数据指南合并。国家卒中登记处包括来自美国2000家医院急性卒中入院的患者水平数据。来自健康评估研究所的数据包括每家医院的当地县中风死亡率以及社会经济特征。当可用时，这些数据也包括每个患者的居住县，当它不同于医院的县位置。由健康评估研究所和AHA汇集的这个独特的数据集将为研究人员和临床医生提供关于现有最大的患者级中风登记处之一的丰富的当地社区数据基础。中风是美国的第五大死因，但却是女性的第三大死因。1到2030年，预计将有340万美国成年人患有中风。1与2012年相比，患病率增加了20.5%。AHA Get With The Guidelines-Stroke是一个通过促进患者始终坚持最新的科学治疗指南来改善卒中护理的院内计划。自2003年该项目启动以来，已有2000多家医院通过该质量改进项目帮助改善了患者治疗效果。由于我们面临着患病率的急剧增加，我们要求研究和临床社区参与这项中风数据挑战，以帮助更好地了解健康的社会和结构决定因素如何影响中风表现，护理质量和临床结果。登记处记录的中风事件与健康评估研究所的美国健康差异数据相匹配项目该项目使用来自国家卫生统计中心、美国人口普查和一系列其他来源的数据，对特定原因死亡率、健康风险和社会经济状况进行可比和一致的估计。中风死亡率估计为县整体以及县内的亚组。跨年龄、空间和时间借用强度的统计模型用于减少小样本量的噪音、处理缺失或解释源数据中未充分指定的信息。2当数据集合并时，每个患者文件都包含了他们的年龄，性别，种族和种族组的特定死亡率，以及中风事件发生年份的中风亚型。还增加了关于社会决定因素的县级数据，包括人均收入、贫困、失业、住房所有权和教育的估计数。
The American Heart Association (AHA) has opened a Stroke Data Challenge to researchers and clinicians to advance the understanding of how social and structural determinants of health influence stroke presentation, care quality, and clinical outcomes. To address these questions, county-level estimates of mortality and social determinants of health from the Institute for Health Metrics and Evaluation at the University of Washington have been merged with the AHA Get With The Guidelines National Stroke registry data. The National Stroke Registry includes deidentified patient-level data from acute stroke admissions across 2000 hospitals in the United States. The data from the Institute for Health Metrics and Evaluation include each hospital’s local county stroke mortality rates as well as socioeconomic characteristics. When available, these data are also included for each patient’s county of residence when it differs from the hospital’s county location. This unique dataset brought together by the Institute for Health Metrics and Evaluation and the AHA will provide researchers and clinicians with a rich foundation of local community data on one of the largest patient-level stroke registries in existence. Stroke is the fifth leading cause of death in the United States, but the third leading cause of death in women. 1 By 2030, it is projected that an additional 3.4 million US adults will have had a stroke. 1 This is a 20.5% increase in prevalence from 2012. The AHA Get With The Guidelines–Stroke is an in-hospital program for improving stroke care by promoting consistent adherence to the latest scientific treatment guidelines for patients. Since the program’s start in2003, over 2000 hospitals have helped to improve patient outcomes through this quality improvement program. As we face a steep increase in prevalence, we are asking the research and clinical community to engage in this Stroke Data Challenge to help better understand how social and structural determinants of health influence stroke presentation, care quality, and clinical outcomes.For this data challenge, stroke events recorded in the registry are being paired with data from the Institute for Health Metrics and Evaluation’s United States Health Disparities Project. This project uses data from the National Center for Health Statistics, the United States Census, and a range of other sources to provide comparable and consistent estimates of cause-specific mortality, health risks, and socioeconomic conditions over time. Stroke mortality rates are estimated for the county overall as well as for subgroups within the county. Statistical models that borrow strength across age, space, and time are used to reduce noise from small sample sizes, handle missingness, or account for under-specified information in the source data. 2 When the datasets are merged, each patient file is enriched with the mortality rates specific for their age, sex, race and ethnicity group, and stroke subtype for the year in which their stroke event occurred. County-level data on social determinants are also added, including estimates of per capita income, poverty, unemployment, home ownership, and education.