— Part 2 of a review of the Plant Positive video, How Much LDL?
Honestly, trying to review this excellent second video in Plant Positive’s Primitive Response series, has smeared egg on my title and caused me to re-evaluate what I know about the subject. Seriously, in watching the 42-slide video as it laid down brick after brick of the Lipid Hypothesis (LH) argument, Plant Positive’s very apropos image of Sisyphus again came to mind as I contemplated the enormity of the task I was undertaking . Really:
What I’m trying to say in my roundabout way is that my first take on the video was something like: “WOW! There’s really more here than I had any idea existed. There’s no point even discussing this.” But as soon as I finished the video and got into verifying references and reading actual journal articles, doubts began to re-assert themselves. Not that I want to defend the CD case. I am not a cholesterol denier and not anti-vegan; I’m more of a fence-sitter.
But before I get into my doubts, I want to make it clear: I can’t say enough how thrilled I am to have found these videos from Plant Positive. When I first started becoming confused by all the conflicting assertions about heart disease and cholesterol, I went to Kahn Academy and was disappointed in my search there for an in-depth, online course. These videos from Plant Positive are exactly what I was looking for. I expected to find them as a course offering — but at least they’re here — thank you Plant Positive! What an incredible service you have provided! 🙂
My Problem With The Lipid Hypothesis
My first problem is philosophical and if philosophy irritates you might want to skip to the next section. So, a huge issue in philosophy is the discussion about what can be known or proven. A “hypothesis” is anything but that; it can only be falsified; never proven. From Scientific Hypothesis, Theory, Law Definitions
A hypothesis is an educated guess, based on observation. Usually, a hypothesis can be supported or refuted through experimentation or more observation. A hypothesis can be disproven, but not proven to be true.
So doesn’t the very name of it mean that the scientific / medical community still considers the evidence to be inconclusive? Yet, frequently in the literature, the LH is said to be proven. Moreover, from the LH statement presented by Plant Positive in the video I last reviewed:
the claim is on the table that high serum cholesterol is NOT an innocent bystander, but rather a significant cause of heart disease.
I don’t believe I’m nitpicking here when I object to the premature assignation of causation. When we convict the wrong party, the guilty gets away with murder. As I go through the evidence presented in the PP video, I’ll be looking for studies that (instead of just establishing that cholesterol was there at the scene of the crime) attempt to distinguish bystander from perpetrator. Not only that, if it can’t be proven, it might not just be some philosophical BS; it might be because it’s just plain wrong. That happened to me back in the day when I first came to really understand the difference between appearing to be exceedingly unlikely and actually being impossible.
The Lesson of Humility and the Law of Large Numbers
Early in my career as a software engineer consultant, I was part of a four person team developing custom solutions for big business. We were close to the deadline for going online with it, but there was a nasty little randomly intermittent “bug” in the system we’d created that was eluding us and each of us were claiming our code was innocent. I decided I would write up a rigorous, formal mathematical proof that would at least exonerate MY work. A few hours later, I was stunned to find I could not do that. There were 4 or 5 places in my code where, although the chances the code was flawed seemed inconceivably remote, correctness could not be proven!
It was easy to plug those few holes in the code and, thus, make my proof unassailable. I did that, anticipating being able to sit back and laugh at the other guys as they scrambled to fix THEIR problem. But that’s not what happened. Instead, when I uploaded the changes, lo(!) and behold(!), the problem disappeared. It was my problem all along and I had too much ego investment to see it. In everyday life, we are accustomed to automatically and unconsciously ruling out the unlikely and coming up with really good solutions that will generally be correct. But in the world of computer code where instructions are executed <or interrupted!> at the rate of billions per second, anything that can go wrong, will go wrong; count on it! The only question is, how often.
I believe the same adage is true in biologic systems. With 7 billion people on the planet and enzyme facilitated reactions that are consummated <or interrupted!> at the rate of umty-illion per second, we have to use a different kind of non-intuitive thinking: if you can’t prove it doesn’t happen, you are wrong! it.will.happen. The only question is, how often.
The Law of Large Numbers contains a seeming paradox: A predictable result follows from a large number of experiments, even though the outcome of a single experiment is random. So even though the outcome odds ratio is amazingly stable, ABSOLUTELY nothing is said about the result in any individual experiment and certainly nothing can be inferred about causation. Causation occurs at the level of individual experiments. Collectively, in complex systems, cause cannot be attributed. To be clear, the individual experiments here are you or I showing up with heart disease in, say the next year. The large number of experiments is epidemiology. Or even a RCT in which we modify one variable in a population and measure the collective result.
Alternatives to the LH have been proposed. My favorite was put forth by Bill Lands, so I’ll call it the Bill Lands’ Hypothesis (BLH). I have not really subjected his hypothesis to intense scrutiny. My current belief is that the BLH does better at explaining the facts than the simpler LH; perhaps that belief is unjustified. In subsequent posts, I’ll be discussing the BLH along with the LH, plus trying to look for evidence already in place that would falsify it.
Greed and The Lesson of the Golden Boy
When I was in graduate school, there was a golden boy in our department who could do no wrong. He was the professor’s pet and he was brilliant. Years later, when his results could not be duplicated, it was determined that the unspeakable had occurred and he had falsified his experimental results in order to get his degree.
My take is that he was between a rock and a hard place. Perhaps he didn’t feel confident enough in his own work to challenge the prevailing wisdom. Perhaps, his character was such that the easy way out seemed to be preferable and he rationalized to himself that fraud was the only alternative. Perhaps he had such confirmation bias that he unconsciously corrected his results in the expected direction. Perhaps none or all of the above.
I hate to bring this up, but it’s the elephant in the room. I’m referring to the Big Pharma money that has to have contaminated medical research to some extent. The cholesterol deniers invariably mention that committees that control NIH funding and set guidelines are dominated by doctors with clear conflict of interest issues.
Not only that, as in the Golden Boy case, it invites unconscious and well as intentional deception to attribute more certainty to results than actually exist. We must endeavor to always be clear about the limits to what we know we know, what we know we don’t know, so that we can even hope to find what we don’t know, we don’t know.