Bailey K R
Section of Biostatistics, Mayo Clinic, Rochester, Minnesota 55905.
Control Clin Trials. 1991 Dec;12(6):741-52. doi: 10.1016/0197-2456(91)90037-m.
In 1980-1981 the National Heart, Lung, and Blood Institute (NHLBI) sponsored a multicenter study of animal models for protection of ischemic myocardium (AMPIM). After data collection for the study was complete, problems were identified with data from one of the labs. This article deals with the issue of statistical evidence versus hard evidence as well as the process of crossing the boundary from suspicion to certainty regarding the nature and extent of the problem.
1980年至1981年，美国国立心肺血液研究所（NHLBI）发起了一项关于缺血性心肌保护动物模型的多中心研究（AMPIM）。在该研究的数据收集完成后，发现其中一个实验室的数据存在问题。本文探讨了统计证据与确凿证据的问题，以及在问题的性质和程度方面从怀疑到确定的跨越过程。
