Hardy R J, Schroder G D, Cooper S P, Buffler P A, Prichard H M, Crane M
University of Texas Health Science Center, School of Public Health, Houston.
Am J Epidemiol. 1990 Jul;132(1 Suppl):S32-42. doi: 10.1093/oxfordjournals.aje.a115786.
A statistical procedure for monitoring the health status of a community potentially exposed to a hazardous environment is presented. It utilizes two levels of investigation. Level I studies monitor routinely collected vital statistics and routes of community exposure whereas level II studies require additional data collection and are further distinguished by their design and duration. In a level I study, routine vital statistics for specified end points over a specified period of time are monitored, and the observed number of events is compared with the expected number of events for a given population. The statistical model used with this procedure employs a two-step decision rule based on the standardized mortality ratio for the study community. An "alert status" is invoked when the number of events exceeds a prescribed excess. An "action status" is indicated if the excess noted in the initial period persists or if the observed number of events in the initial period greatly exceeds expectation. Should an "action status" be justified, level II studies to determine the likely explanation for the significant excess are initiated. This could include the conduct of a "case-control" study using the exposure data available from monitoring the community.
本文介绍了一种用于监测可能暴露于危险环境的社区健康状况的统计程序。它采用两级调查。一级研究监测常规收集的生命统计数据和社区暴露途径，而二级研究需要额外的数据收集，并根据其设计和持续时间进一步区分。在一级研究中，监测特定时间段内特定终点的常规生命统计数据，并将观察到的事件数量与给定人群的预期事件数量进行比较。此程序使用的统计模型基于研究社区的标准化死亡率采用两步决策规则。当事件数量超过规定的超额数量时，会调用“警报状态”。如果初始阶段记录的超额情况持续存在，或者初始阶段观察到的事件数量大大超过预期，则表明处于“行动状态”。如果“行动状态”合理，则启动二级研究以确定显著超额的可能原因。这可能包括使用从社区监测中获得的暴露数据进行“病例对照”研究。
