RFA-CE-22-008, ASTRAL (Aquatic Safety Through Research, Association, and Linkage of Data)

摘要 该项目的长期目标是了解影响发病率和死亡率的条件， 意外溺水，特别是在溺水率较高的亚群中。的 具体目标是：1）通过链接EMS、医院和死亡率的数据，开发分析数据库 通过症状监测数据验证记录，并丰富社会经济背景 有关溺水风险因素的信息; 2a）审查性别、种族和民族差异 溺水率，并确定导致这些差异的风险因素; 2b）进行空间- 对溺水事件进行时间分析，以确定这些事件发生的地点和时间段。 3）计算与致命和致命疾病相关的总医疗和其他费用， 按年龄、少数民族身份和水生物分类的非致命性溺水。 这是一项横断面研究的所有受害者或病人的意外溺水发生在 德克萨斯州哈里斯县，2016-2023年。我们将把死亡报告、EMS和 住院数据，并从其他数据源中添加有关溺水的上下文信息， 分析数据库。溺水者的个人和环境信息将得到匹配 在邻里层面。我们将通过以下方式匹配人口统计学、暴露和结果变量： 空间重叠，并将确定与以下方面有关的人口统计、医疗和溺水特征： 致命和非致命溺水。空间回归建模将产生一个模型来预测致命的， 按性别、种族、西班牙裔和其他族裔进一步细分的非致命性溺水 差异 我们将研究溺水与邻里特征、游泳距离、 场地和游泳课的参与。将检查溺水事件的空间集中度 使用最近邻分层聚类算法。为了将溺水与预测因素联系起来， 马尔可夫链蒙特卡罗泊松-对数正态-条件自回归空间回归 模型将在人口普查区块组一级进行测试。 将分析医院账单记录，以计算直接医疗成本， （终身）成本和与溺水相关的残疾调整生命年。该项目将整合 从多个数据源中获取信息，以确定导致溺水增加的关键因素 rates.研究结果将为伤害预防工作提供信息，并有助于优先考虑保护 人们溺水。

Abstract The long-term goal of the project is to understand conditions affecting morbidity and mortality from unintentional drowning, especially in subpopulations that experience higher rates of drowning. The specific aims are: 1) Develop an analysis database by linking data from EMS, hospital and fatality recordsverified by syndromic surveillance data and enriched with socio-economic contextual information pertaining to risk factors of drowning; 2a) Examine gender, racial and ethnic differences in drowning rates and identify risk factors that account for these differences; 2b) Perform spatial- temporal analysis of drownings in order to identify locations and time periods when these events are most likely to occur and 3) Compute the aggregate medical and other costs associated with fatal and non-fatal drowning by age, minority status and aquatic body. This is a cross-sectional study of all victims or patients of unintentional drowning that occurred in HarrisCounty, Texas between 2016-2023. We will link data from fatality reports, EMS and hospitalization dataand add contextual information on drowning from other data sources to prepare an analysis database. Individual and environmental information of drowning victims will be matched at the neighborhood level. We will match demographic, exposure and outcome variables through spatial overlap and will identify demographic, medical and drowning characteristics associated with fatal and nonfatal drowning. Spatial regression modeling will produce a model to predict fatal and nonfatal drowning, further broken down by gender, race, Hispanic ethnicity and other ethnic differences. We will examine associations of drowning with neighborhood characteristics, proximity to swimming venues and swim lesson participation. Drowning incidents will be examined for spatial concentrations using the Nearest Neighbor Hierarchical Cluster algorithm. To relate drownings to predictive factors, a Markov Chain Monte Carlo Poisson-Lognormal-Conditional Autoregressive spatial regression model willbe tested at the census block group level. Hospital billing records will be analyzed to compute the direct medical costs, comprehensive (lifetime) costs, and disability adjusted life years associated with drowning. This project will integrate informationfrom multiple data sources to identify key factors that contribute to increased drowning rates. The results will inform injury prevention efforts and help prioritize strategies to safeguard people from drowning.