Skov T, Kristensen T S
National Institute of Occupational Health, København, Denmark.
Am J Ind Med. 1996 Apr;29(4):378-81. doi: 10.1002/(SICI)1097-0274(199604)29:4<378::AID-AJIM19>3.0.CO;2-Z.
In this article a basic distinction is made between etiologic and prevention effectiveness intervention studies. Etiologic intervention studies focus on elucidating causes of disease, while the purpose of prevention effectiveness intervention studies is to study methods of prevention. The design requirements for each of these studies are very different: etiologic intervention studies usually need large study populations, large exposure contrasts, ascertainment of exposure, as well as health outcome. Ideally, randomization and blinding should also be applied. Effective preventive strategies may, on the other hand, be identified in small study populations with exposure as the only outcome measure, and randomization and blinding may be superfluous. At present, intervention studies are in great demand, and often there is a wish that etiologic questions as well as prevention effectiveness be addressed in the same study. We argue that this should not be done without careful consideration of possible conflicting design aspects.
在本文中，对病因学和预防效果干预研究进行了基本区分。病因学干预研究侧重于阐明疾病的病因，而预防效果干预研究的目的是研究预防方法。这两类研究的设计要求非常不同：病因学干预研究通常需要大量的研究人群、较大的暴露对比、暴露的确证以及健康结局。理想情况下，还应采用随机化和盲法。另一方面，有效的预防策略可能在以暴露作为唯一结局指标的小研究人群中得以确定，随机化和盲法可能是多余的。目前，干预研究的需求很大，人们常常希望在同一研究中既解决病因学问题又解决预防效果问题。我们认为，在没有仔细考虑可能存在冲突的设计方面之前，不应这样做。
