Allan Sniderman 加拿大麦吉尔大学医学部

    <International Circulation>: Right now evidence-based medicine is very popular.  What do you think are the basic flaws in the case for evidence-based medicine? 

  《国际循环》：Sniderman教授您好。目前循证医学是非常受欢迎的。请问您认为循证医学最基本的缺点是什么？
 
    Prof. Sniderman:  Evidence-based medicine， like most ideas that come along， get a lot of positive when they become accepted.  Then they can become dominant and the negatives do not get seen and as a result corrupt the purpose they were adopted for in the first place.  Evidence-based medicine originally was supposed to be that the doctor making their rounds would consult the literature to see how it informed the individual care of a particular patient.  That is at the core of the original evidence-based medicine construct as it was originally enunciated.  It was too complicated for doctors to the analysis of the literature and they had no time to do it.  What it has evolved into is a specialized group of people who evaluate the evidence according to certain pre-specified rules.  Rule #1 is that the randomized clinical trial is the highest form of clinical evidence and is considered the highest form of evidence because this method should eliminate most bias.  The biases that you know and the ones you do not know， which is what randomization should do.  What many of us do not appreciate that it is an assumption and may not always happen that way.  If you have an outcome， is that outcome generalizable to the everyday patient?  That is a limitation of a lot of randomized clinical trials.  Randomized clinical trials are designed to test one point， but we use multiple drugs in multiple doses.  You could learn that a drug is good but many of the other decisions a doctor has to make they don’t have information on to look at a patient and offer choices.  If a patient takes the choice it is their choice， if they don’t then they can change their mind.  If a doctor starts a treatment and realize that wasn’t the best choice they can take it away， because for statins， for instance， the risk is so low.  That is my point and in my talk I got a strong rebuttal from the opposing viewpoint.  I got such a strong rebuttal because I did not express my argument in a way that was acceptable to the audience.  

    Sniderman教授: 循证医学，像其它一些出现的观念一样，当积累的阳性证据达到一定程度时，这种观点就会被大家所接受。这时这种观点会占据主导地位，同时负面结果不再被重视，结果就破坏了他们开始想要达到的效果。循证医学最初应该是指医生在临床实践中通过查阅文献来了解并参考文献中报道的如何对特殊病人进行个体化治疗。这是循证医学起初被提出时对其构想的核心组成部分。但是医生很难对全部文献进行分析并且他们根本没有足够的时间来完成这样的工作。完成这些工作需要一个专业团队的参与，这些人能根据预先制定的特定规则对可用的证据进行评价。第一条规则是随机临床试验是最高级别的临床证据，由于这样的试验方法能消除大多数偏倚因此也被认为是最高级别的证据。随机的目的就是要消除那些你已经知道和还不知道的偏倚。我们大多数人不知道的是，这种试验也是一种假设而且可能常常并非为我们想要的结果。假设你得到一个结果，这个结果是否能概括地适用于日常的患者呢？这是大多数随机临床试验的缺陷。设计随机临床试验是为了检验一种观点是否正确，但是在临床上我们使用着多种不同的药物，每种药物都有不同的剂量。你知道一个药物对患者是有效的，但是医生根据一位患者的病情需要做出的很多其他决策并没有信息可供参考，并且医生也没有把这些选择提供给患者。如果患者选择了一种治疗方法，那这种治疗方法就是他们想要的，如果这种治疗方法并不是他们想要的，那么他们还能选择其他的方法。如果医生开始了一种治疗并且意识到这种治疗并不是最佳选择，那么他们就可以停止这种治疗方法而选择其他的方法，因为举例来说，像他汀类药物的副作用就很小。这就是我的观点并且在我的谈话中我提出了从相反角度看待问题的强有力的反驳。我使用了如此强烈的反驳是因为我并没有把我的论据通过听众可以接受的方式传递给大家。