NEIGHBORHOOD EFFECTS IN COMMUNITY TRIALS FOR CVD

目的是评估社区心血管疾病(CVD)预防试验中的“邻里效应”。在20世纪80年代，NHLBI/NIH资助了三项社区试验，以测试旨在降低心血管疾病风险的计划。虽然每项研究都报告了有利的危险因素变化，但治疗地点和对照地点之间的许多总体净差异不大和/或在统计上不显著。新的研究表明，这些分析可能掩盖了重要的站点内邻里终点差异。这可能是(1)“地点”，高于和超过个人水平的衡量标准，影响心血管疾病相关的行为和(2)整体项目结果可能会被社区水平的影响所混淆。这项研究使用了来自(1)Pawtucket心脏健康计划(PHHP)和(2)1980年和1990年十年一次的人口普查的丰富和现有的数据。使用多水平统计方法和地理信息系统(GIS)进行分析。简而言之，具体目标是：确定个人水平的心血管疾病相关变量因社区而异的程度；在对个人水平的测量进行调整后，确定社区水平的心血管风险预测因素；检查社区的终点趋势；根据社区分析剂量效应(即干预过程)；评估不同社区定义对结果的影响；当人口普查数据每十年收集一次时，找到处理人口趋势的最佳方法；推进心血管流行病学的多层次和地理信息系统方法；促进流行病学理论的“地点”。这项为期三年的具有成本效益的研究具有重大意义、创新性和及时性。它将促进21世纪公共卫生实践所需的多学科实质性理解和方法工具。

The objective is to evaluate "neighborhood effects" in community trials for cardiovascular disease (CVD) prevention. During the 1980s, the NHLBI/NIH funded three community trials to test programs aimed at reducing CVD risks. While each study reported favorable risk factor changes, many overall net differences between treatment and control sites were modest and/or statistically insignificant. New research suggests that these analyses may mask important within-site neighborhood-wise endpoint differences. It may be that (1) "place," above and beyond individual level measures, influences CVD-related behaviors and (2) overall program outcomes may be confounded by neighborhood level influences. This study employs rich and existing data from the (1) Pawtucket Heart Health Program (PHHP) and (2) the 1980 and 1990 decennial censuses. Multilevel statistical methods and a geographic information system (GIS) are used for analyses. Briefly, the specific aims are to: Determine the extent to which individual-level CVD related variables vary by neighborhood; Identify neighborhood-level predictors of CVD risk after adjusting for individual-level measures; Examine neighborhood-wise endpoint trends; Analyze dose effects (i.e., intervention process) in terms of neighborhoods; Evaluate the effect of different neighborhood definitions on findings; Find best methods for handling demographic trends when census data are gathered decennially; Advance multilevel and GIS methodologies in cardiovascular epidemiology; Contribute to an epidemiologic theory of "place." This cost-effective three year study is significant, innovative and timely. It will advance the multidisciplinary substantive understanding and methodological tools necessary for public health practice in the twenty-first century.