Medical flip-flopping and Evidence-based concept

Adequate (size and design) scientific study is often prohibitively complicated and expensive. This forces us to rely on less definitive work or even lab-based studies.  This may lead to ever-changing conclusions.

The human body is so complicated and different from one person to another. What works on one person or a certain group may not work for another. Also our understanding of human body is so superficial, that what we believe should work often does not.

The disturbing thing is that it is the treatments that make the most theoretical sense that are often shown later on to be ineffective. One thing that they have in common is that they are all reasonable, logical and scientifically appealing. Therefore, we have no way of knowing that which of the reasonable, logical and scientifically appealing treatments will not work.

The flip-flopping, unsurprisingly, is not uncommon in dental literature. There are series of studies starting in 1995 that looked into the impact of restoration quality and root filling quality on periapical health. They employed similar study design. The first study found that quality of restoration was more influential on periapical health. The second, both were equally important. The third, root filling quality was more important and so on and so forth. It’s a pure flip-flop. It is frustrating because clinicians use this kind of information to make clinical decisions. So, which is it? Which of these studies should be used as a basis for clinical decision?

How should we deal with this?

Evidence-based concept may alleviate confusion. All the studies on the graph spanned over three decades. They looked into the same issue i.e. healing rates of initial endodontic treatment, but produced very different results (less than 50% to almost 100%).  The studies may appear similar, but they employed different study design, had different ability to establish causative relationship and had different level of the risks of bias.  Therefore, they are not the same and we can rank them up by their credibility (level of evidence).  Once ranked, we look at the top level.  If there are more than one, we could apply statistical method (meta-analysis) to combine all different results into one, on which we based our decisions.

Are all these different results (sometimes in the opposite directions) an incremental progress of science or the result of flawed truth seeking methods? This is important to ponder because the first requires acceptance, but the latter requires correction/improvement.

Sources

https://www.nytimes.com/2015/11/03/science/book-review-ending-medical-reversal-laments-flip-flopping.html