Backmann Marius
Konstanz University, Constance, Germany.
Med Health Care Philos. 2017 Dec;20(4):513-523. doi: 10.1007/s11019-017-9773-2.
The standardised randomised clinical trial (RCT) has been exceedingly popular in medical research, economics, and practical policy making. Recently, RCTs have faced criticism. First, it has been argued by John Worrall that we cannot be certain that our sample is not atypical with regard to possible confounding factors. I will argue that at least in the case of medical research, we know enough about the relevant causal mechanisms to be justified to ignore a number of factors we have good reason not to expect to be disruptive. I will also argue against an argument provided by Nancy Cartwright and Eileen Munro that RCTs should not be taken to deductively infer probabilistic causal claims, but ampliatively. The paper will end on a discussion of evidence hierarchies and a defence of the stance of evidence-based medicine that RCTs are the best available method to assess a treatment's efficacy.
标准化随机临床试验（RCT）在医学研究、经济学和实际政策制定中极为流行。最近，RCT面临批评。首先，约翰·沃拉尔认为，我们无法确定我们的样本在可能的混杂因素方面并非非典型。我将论证，至少在医学研究的情况下，我们对相关因果机制了解得足够多，有理由忽略一些我们有充分理由认为不会造成干扰的因素。我还将反驳南希·卡特赖特和艾琳·芒罗提出的一个观点，即不应将RCT视为演绎推断概率因果主张，而应视为扩充性推断。本文最后将讨论证据等级制度，并为循证医学的立场辩护，即RCT是评估治疗效果的最佳可用方法。
