[scoofy]

If you want to feel some dissonance about this dissonance, you might note that a meta-analysis was done on the evidence that using a parachute is effective at saving your life when jumping from an aircraft. It's a real, but satirical approach to worshiping randomized controlled trials for things that should effectively be obvious. There are obvious limits to empiricism, and we shouldn't proudly flaunt lack of evidence ≠ evidence of absence in cases where it's perfectly reasonable to expect results. I mean, we could be wrong about theses things, but it would be a shift in thinking akin to newtonian -> einsteinian physics.

Paper: Parachute use to prevent death and major trauma when jumping from aircraft: randomized controlled trial

Conclusions: Parachute use did not reduce death or major traumatic injury when jumping from aircraft in the first randomized evaluation of this intervention. However, the trial was only able to enroll participants on small stationary aircraft on the ground, suggesting cautious extrapolation to high altitude jumps. When beliefs regarding the effectiveness of an intervention exist in the community, randomized trials might selectively enroll individuals with a lower perceived likelihood of benefit, thus diminishing the applicability of the results to clinical practice.

>https://www.bmj.com/content/363/bmj.k5094

[Mathnerd314]

As the paper's title says, it's a randomized controlled trial, not a meta-analysis. A meta-analysis would likely exclude the RCT due to poor experimental design.

[scoofy]

You are correct, but a meta-analysis would still suggest that there is no evidence that parachutes protect people jumping out of airplanes, because we don’t have evidence because we can’t have evidence.

[Mathnerd314]

Actually meta-analyses are not limited to RCTs - in my experience they can be quite subjective.

So for example, a meta-analysis might say that there's a fair amount of evidence that people not wearing parachutes die after falling from airplanes (e.g. https://www.washingtonpost.com/world/2021/08/26/story-an-afg...), and plenty of evidence that people can land successfully with a parachute (skydiving videos, too many to count). The conclusion would be a strong correlation between parachute use and not dying. The only thing there isn't is an RCT, to prove that the correlation is causal. The purpose of the parachute RCT paper seems to be to prove that the design of an RCT is just as fallible as any other reasoning that attempts to prove causality, so in fact RCTs are not the "gold standard" and other forms of reasoning may be just as valid. (e.g. basic physics in the case of the parachute)

[scoofy]

>basic physics

I totally agree, the point that organizations like the World Heart Federation are using. We are taking basic physics as a truism, but it's the result of experimentation, which creates a framework. The same is done for the way the body functions. That framework can be used to make sweeping statements about alcohol / ionizing radiation, etc.

I've been totally blown away recently listing to This Week in Virology, as the panel of experts regularly discuss their concerns about how many people are discussing the viral/vaccine mechanics as though the framework we have of disease is perfect. It's always fascinating to hear what absurdly qualified people have concerns about when discussing their area of expertise.

[spider-mario]

To be clear, the problem here is in equating “evidence” with randomized controlled trials. We have plenty of evidence that parachutes work, just not in the form of RCTs.

David Gorski of sciencebasedmedicine.org calls it “methodolatry”:

https://twitter.com/gorskon/status/1484220420295376900

https://twitter.com/gorskon/status/1479900075602194445

https://twitter.com/gorskon/status/1483268592674217986