Fife D
New Jersey Department of Health, Trenton 08625-0360.
Accid Anal Prev. 1989 Feb;21(1):79-83. doi: 10.1016/0001-4575(89)90050-x.
Motor vehicle traffic fatalities in the United States are described by two major data sources, the Fatal Accident Reporting System (FARS) and the National Center for Health Statistics Multiple Cause of Death data (NCHS). Certain data, such as the age and sex of the fatality, are reported by both sources. However, each source contains data absent from the other. For example, only the FARS describes the precise circumstances of injury, and only the NCHS identifies the anatomic injuries listed on the death certificate. Thus, it would be useful to have a single file that offers for each case all of the data provided in each of the separate files. Creation of such a file is impeded by the fact that neither file contains personal identifiers for the cases listed. The present paper describes a method of matching cases from the two files based on simultaneous agreement of several variables common to both files (age, sex, date of death, role in the crash, and state in which the injury occurred). Using this method, 85% of the FARS cases can be uniquely matched with a case in the NCHS data.
美国机动车交通死亡事故由两个主要数据源进行描述,即致命事故报告系统(FARS)和国家卫生统计中心的多死因数据(NCHS)。某些数据,如死亡者的年龄和性别,这两个数据源都会报告。然而,每个数据源都包含对方所没有的数据。例如,只有FARS描述了受伤的确切情况,只有NCHS识别出死亡证明上列出的解剖学损伤。因此,拥有一个为每个案例提供每个单独文件中所有数据的单一文件会很有用。由于这两个文件都不包含所列出案例的个人标识符,所以创建这样一个文件受到了阻碍。本文描述了一种基于两个文件共有的几个变量(年龄、性别、死亡日期、在撞车事故中的角色以及受伤发生的州)同时一致来匹配两个文件中案例的方法。使用这种方法,85%的FARS案例可以与NCHS数据中的一个案例唯一匹配。