There is a problem in the medical field with people considering randomized controlled trials the *only* source of evidence. It is the *gold standard* of evidence, but that does not mean other inferior sources of evidence have no value. It is an excessively binary way of thinking.
Randomized controlled trials should, as a generalization, trump other inferior forms of evidence (although RCTs are not 100% infallible). But in the absence of RCT data, *the best alternative sources of evidence should be used*, and it should not be assumed to be useless.
An analogy: In a murder trial, the gold standard of evidence might be video footage of the murder occurring. However, we don't say that because we often lack the very best form of evidence, that it is impossible to infer by any other means/evidence what is likely to have happened
Video evidence, in this analogy, should trump inferior sources of evidence, *if available*. If not, inferior sources of evidence should be used, with an appropriate awareness of their limitations and thus the increased probability they are wrong.
How does this manifest? Consider the initial WHO/CDC opposition to the use of masks early in the pandemic. We didn't have gold standard scientific evidence on their efficacy. However, we *did* have less rigorous evidence that their use seemed to correlate with slower spread.
This insistence on only using gold standard evidence and disregarding all other forms of evidence, even when gold standard evidence is lacking, can lead to some decisions that lack basic common sense, such as those with respect to masks.
Furthermore, there is seemingly no understanding of the idea of "risk vs. return". Masks are harmless. The cost of being wrong was *extremely low*, whereas the *potential benefits were huge*. Insisting on gold standard evidence before encouraging their use lacked common sense.
Quality decisions are based on the "totality of the evidence", and recognize that there is an "evidence hierarchy" where some evidence trumps other evidence due to its superior rigor, and that there are varying "degrees of uncertainty" associated with all forms of evidence.
Evidence is not binary - it is a spectrum. We can know more or less, with more or less confidence. The degree of evidence and our confidence in it should be balanced against the risk/reward of being right vs. being wrong. That is the best way to make intelligent decisions IMO.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
