1/ Watching this episode of @PeterAttiaMD's podcast with @garytaubes...
On "Pathological Science"...
Taubes: “One of the defining characteristics of pathological science is — people commit themselves publicly to a result based on premature evidence…”
On "Pathological Science"...
Taubes: “One of the defining characteristics of pathological science is — people commit themselves publicly to a result based on premature evidence…”
2/ “…So it’s not ironclad, they haven’t locked it down, there’s still a chance they could be wrong and they don’t understand the likelihood of that chance, no one ever does until you get very good at this when you realize that chance is enormous.”
3/ “And so once you go public, science is supposed to be hypothesis and test, right? You’re supposed to rigorously test your hypotheses and ideally you’re trying to prove that you’re wrong..."
4/ "Richard Feynman – first principle of science – is that you must not fool yourself, and you're the easiest person to fool."
(So incredibly true!)
(So incredibly true!)
5/ "When we talk about pathological science, people tend to think of fraud, right? When you're committing fraud in science, you're knowingly trying to fool other people by manipulating the evidence...
But there are two ways you can manipulate the evidence...
But there are two ways you can manipulate the evidence...
6/ "You can create a signal, or you can do an inadequate job of studying the background.
Remember it's a signal-to-noise problem.
So you can fudge a signal or create a signal, or move a signal...
...or you could ignore the background.
Remember it's a signal-to-noise problem.
So you can fudge a signal or create a signal, or move a signal...
...or you could ignore the background.
7/ "... and do a poor job in the background you'll still end up with a big signal-to-noise ratio. You haven't technically committed fraud.
You're not actively trying to fool anyone else, you're fooling yourself."
You're not actively trying to fool anyone else, you're fooling yourself."
8/ Naturally, this speaks to me very personally.
As mentioned above, the first major endeavor is to seek proving yourself wrong -- which is exactly the value of the coming #LMHRstudy for us.
As mentioned above, the first major endeavor is to seek proving yourself wrong -- which is exactly the value of the coming #LMHRstudy for us.
9/ I continue to hypothesize lipid profiles are often in large part a consequence, not a cause of metabolic health and disease.
FWIW, I've been doing so for years, but I think this could be the "background" that keeps getting overlooked (or maybe it's not relevant after all)
FWIW, I've been doing so for years, but I think this could be the "background" that keeps getting overlooked (or maybe it's not relevant after all)
10/ I likewise believe populations of genetically normal, metabolically healthy populations and their LDL assoc with CVD are surprisingly understudied. Healthy LMHRs will provide a new window of study given very high LDL effect size against resulting CVD outcome. Hence the study.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
