Skegg D C, Doll R
J Epidemiol Community Health. 1981 Mar;35(1):25-31. doi: 10.1136/jech.35.1.25.
A study was carried out to assess the feasibility of using record linkage for drug monitoring. For two years, three types of records were collected for a total of 43 117 people: (1) details of basic attributes, such as sex and age; (2) details of prescriptions dispensed; and (3) records of hospital admissions, obstetric deliveries, and deaths. The records about each person were linked together, and analyses were performed to reveal associations between drugs and diagnoses. The study suggested that record linkage would be useful both for generating and for testing hypotheses about the adverse effects of drugs. The method would be especially valuable for detection of delayed effects (such as the induction of cancer), sudden deaths outside hospital, and effects of the fetus-all of which are difficult to study by other means. A full-scale project would need to cover a large population, and some of the practical issues that would arise are discussed.
开展了一项研究以评估使用记录链接进行药物监测的可行性。在两年时间里,收集了三类记录,涉及总计43117人:(1)基本属性细节,如性别和年龄;(2)所配发处方的细节;以及(3)住院、产科分娩和死亡记录。将关于每个人的记录链接在一起,并进行分析以揭示药物与诊断之间的关联。该研究表明,记录链接对于生成和检验有关药物不良反应的假设均有用。该方法对于检测延迟效应(如癌症诱发)、院外猝死以及胎儿效应尤为有价值,而所有这些用其他方法都难以研究。一个全面的项目需要覆盖大量人群,文中还讨论了将会出现的一些实际问题。