Evidence Will Absolve Me — R.G.
The title here is tongue-in-cheek. But I do want to focus on the concept of evidence. Not every piece of information equals to evidence. Philosophers have been hotly debating the grounding of evidence since the last century. Just because informaiton is constructed, presented, and published does not mean it is evidence of anything. Evidence Based Medicine (EBM) properly conceived has to be viewed to hold the highest quality of evidence as the gold standard; as I have argued previously on this blog. And therefore just because a study is published does not magically ordain it as evidence qua evidence. It can just be held as information until further notice. Doctors need to be more discerning. Essentially, they need to be better social scientists to deal with EBM. But that can be a problem.
W.W raises the exceedingly valid point about X-study being published and then being contradicted not long after. This is a common feature of social science research that calls into question the concept of another hotly debated philosophical concept: objectivity in science. I’ve discussed this previously as well elsewhere. EBM may be unsettling because it drags core social science concerns into the medical field which is already fraught with its own internal contradictions.
Unfortunately I do not have much to add to my previous post since W.W does not seem to have contention with my core argument: that levels of evidence is important.
I fully agree that doctors should not bend in the breeze of fragile studies coming at them from every direction. Clinical practice requires a firm grasp of theoretical foundations in which doctors can call upon to make decisions to best benefit their patients. If, however, doctors jettison their theoretical foundations for the new flavor of the week study then they are simply doing a disservice to their patients. It definitely is plausibly that not every clinical decision can be reduced to an empirical decision regurgitation. Unfortunately doctors have to make incurably complex decisions as a daily normality.
W.W makes another important point: doctors (perhaps especially young ones) likely do not have sufficient training in the business of interpreting data. I think this is true. Empirical study interpretation (concerning people and populations) is something that is hammered into your brain when you endure rigorous social science training. But that is not what medical school is for. When EBM is thrust upon young doctors I would expect misuse and abuse. This, however, is not the fault of EBM since that is just the trade-off that must be accepted: keep the data coming.
W.W mentions that not enough “negative studies” are published. I agree with this. But this is not a fatal counterpoint. In social science there is an active research agenda on replication studies. These are studies that attempt to replicate the results of previous studies. If they cannot replicate the results then the original results are likely bogus. In social science, Psychology has a replication crisis, for example, where most of the famous studies cannot be replicated. This is a counterbalance to the frequent publishing of studies.
In branches of economics and political science, researchers now offer their entire raw datasets to public review so as to prove that the data was not manipulated (or p-hacked) to get specific results. These are methods to deal with the negative effects of frequent bad studies without needing to “reform” the core concept of EBM.
W.W. says: “the clinician is expected to utilize his judgement, but this is clearly not possible if ‘the best available evidence’ is about as constant as the seasons.” I agree with the spirit of this point. But the nuance I want to add is that just because a study was published does not mean it is the “best available evidence”. In most cases it is just the currently available information.
