Gruebner Oliver, Lowe Sarah R, Tracy Melissa, Cerdá Magdalena, Joshi Spruha, Norris Fran H, Galea Sandro
1Department of Epidemiology,Mailman School of Public Health,Columbia University,New York,New York.
2Department of Epidemiology and Biostatistics,School of Public Health,University at Albany,State University of New York,Albany,New York.
Disaster Med Public Health Prep. 2016 Apr;10(2):261-73. doi: 10.1017/dmp.2015.172. Epub 2016 Jan 28.
To demonstrate a spatial epidemiologic approach that could be used in the aftermath of disasters to (1) detect spatial clusters and (2) explore geographic heterogeneity in predictors for mental health and general wellness.
We used a cohort study of Hurricane Ike survivors (n=508) to assess the spatial distribution of postdisaster mental health wellness (most likely resilience trajectory for posttraumatic stress symptoms [PTSS] and depression) and general wellness (most likely resilience trajectory for PTSS, depression, functional impairment, and days of poor health) in Galveston, Texas. We applied the spatial scan statistic (SaTScan) and geographically weighted regression.
We found spatial clusters of high likelihood wellness in areas north of Texas City and spatial concentrations of low likelihood wellness in Galveston Island. Geographic variation was found in predictors of wellness, showing increasing associations with both forms of wellness the closer respondents were located to Galveston City in Galveston Island.
Predictors for postdisaster wellness may manifest differently across geographic space with concentrations of lower likelihood wellness and increased associations with predictors in areas of higher exposure. Our approach could be used to inform geographically targeted interventions to promote mental health and general wellness in disaster-affected communities.
展示一种空间流行病学方法，该方法可在灾难发生后用于：（1）检测空间聚集情况；（2）探究心理健康和总体健康状况预测因素的地理异质性。
我们对艾克飓风幸存者（n = 508）进行了一项队列研究，以评估德克萨斯州加尔维斯顿灾后心理健康状况（创伤后应激症状[PTSS]和抑郁最可能的恢复轨迹）和总体健康状况（PTSS、抑郁、功能损害和健康不佳天数最可能的恢复轨迹）的空间分布。我们应用了空间扫描统计量（SaTScan）和地理加权回归。
我们在得克萨斯城以北地区发现了高可能性健康状况的空间聚集，在加尔维斯顿岛发现了低可能性健康状况的空间集中。在健康状况的预测因素中发现了地理差异，在加尔维斯顿岛，受访者离加尔维斯顿市越近，与两种健康状况的关联就越强。
灾后健康状况的预测因素在地理空间上可能表现不同，低可能性健康状况集中，且在高暴露地区与预测因素的关联增加。我们的方法可用于为地理定位干预措施提供信息，以促进受灾社区的心理健康和总体健康。
