Weekly Link Love – Edition 83

Research of the Week

Diabetic patients taking metformin had lower coronavirus mortality than diabetic patients not taking it.

Increased cardiovascular disease in African-Americans with Covid-19.”

Unsaturated fat intake and deficiencies in calcium and albumin linked to higher coronavirus mortality.

Financial relationships between the leaders of influential US professional medical associations and industry are extensive…”

New Primal Blueprint Podcasts

Episode 425: Robert Glazer: Host Elle Russ chats with Robert Glazer, founder and CEO of the global performance marketing agency Acceleration Partners.

Primal Health Coach Radio, Episode 62: Laura and Erin talk with Dr. Jade Teta about metabolism and hormones.

Media, Schmedia

Women are not just men with “boobs and tubes.”

Who’s still getting sick in New York?

Interesting Blog Posts

The case for red meat.

Social Notes

Astronaut breakfast.

How to slow down time (seriously).

Everything Else

The human voice contains information about dynamic bodily states.”

ApoE4 gene predicts severity of coronavirus infection (in the UK).

A beautiful video about traditional wool clothing. Best part is how they “shear” the sheep without cutting the wool. Looks so satisfying.

Making CBD from orange peels.

Things I’m Up to and Interested In

Fascinating article: How loners help species survive.

Interesting result with big implications for hunger: High-carb meal results in initial increase in available energy followed by a drastic crash.

Great turn of phrase: Fructoholism.

Fact I was unaware of: Mice and many other animal models used in studies do not depend on vitamin D for their innate immune circuitry.

My favorite paradox: High-fat dairy linked to better health, again.

Question I’m Asking

How much nature are you getting these days? (get some)

Recipe Corner

Time Capsule

One year ago (May 24 – May 30)

Comment of the Week

“A note about sunglasses: Some folks should be wearing them. For instance, a lot of the fisherman in native tribes would go blind at a fairly early age from staring into the water for long periods of time. The Eskimos had really cool bone sunglasses, with little slits they would use, so they would not go snowblind. After my own daily sun exposure, I’m not averse to putting on my shades.”

– Nicely said, Nocona.

About the Author

Mark Sisson is the founder of Mark’s Daily Apple, godfather to the Primal food and lifestyle movement, and the New York Times bestselling author of The Keto Reset Diet. His latest book is Keto for Life, where he discusses how he combines the keto diet with a Primal lifestyle for optimal health and longevity. Mark is the author of numerous other books as well, including The Primal Blueprint, which was credited with turbocharging the growth of the primal/paleo movement back in 2009. After spending three decades researching and educating folks on why food is the key component to achieving and maintaining optimal wellness, Mark launched Primal Kitchen, a real-food company that creates Primal/paleo, keto, and Whole30-friendly kitchen staples.

If you'd like to add an avatar to all of your comments click here!

33 thoughts on “Weekly Link Love – Edition 83”

Leave a Reply

Your email address will not be published. Required fields are marked *

  1. Honestly, I don’t buy into the APOE4 and all these other studies that are trying to correlate genetic and dietary habits. Some are good studies, others reference dietary patterns of countries or regions. APOE4’s generally have higher total cholesterol, which is aossicated with better COVID outcomes. But now APOE4 and dementia predicts worse outcomes. So which is it? Better or worse. These are still just theories based on varied data.

  2. Just a note, “Eskimo” is considered a slur. “Inuit” is a better term. Thanks!

    1. Incorrect. I lived and worked in Alaska for 16 years (among the Yupik) and most Alaskans continue to accept the name Eskimo particularly because Inuit refers only to Inupiat of northern Alaska, the Inuit of Canada, and the Kalaallit of Greenland, and it is not a word in the Yupik languages of Alaska and Siberia.

      1. Correct. I spent a few weeks in a Native household in western Alaska and they always referred to themselves as Eskimos, never said Inuit.

  3. I have been sneaking into the forests, although they are somewhat open in WA. Good part about being in a more rural area is you can get away with it. Turns out I was not the only one. Fresh air and a little sun is probably the best remedy anyway.

  4. Slightly less clinical, but I believe the saying you are looking for is “no smoke without fire”

  5. It is always easier in science to disprove a hypothesis than to prove one. As you said – correlation might suggest a causal link, but not prove one. On the other hand, extremely weak or no correlation almost certainly disproves causation. There are now at least five major meta-analyses looking at data from studies for hundreds of thousands of subjects, over six decades, at at costs ranging up to hundreds of millions of dollars, that have failed to find a convincing link between saturated fat and heart disease. Talk about spending up large to try to support a theory, no other nutritional hypothesis has come close in terms of resources committed to support it. And here we have to conclude – after all of that – absence of evidence is by now super strong evidence of absence.

    1. True proof is a concept in mathematical logic but not science per se. The closest science comes is conditional “proof” — supportive findings really. But disproof (of a hypothesis or theory) is achieved by contradictory data.

  6. Presumably you could have causation without correlation in a case with more than one causal factor, for example where necessary and sufficient conditions needed to be met. The necessary factor would be causative but would only correlate when the sufficient condition was also met.

  7. Lots of discussion of correlation and causation in ‘The book of why’, including examples of seemingly uncorrelated variables where there is actually causation. There is a great formal framework around this.

  8. “No correlation means no causation” is wrong for two reasons. One, it is impossible to prove zero correlation and extremely difficult to prove “low” correlation. Studies will often fail to find a significant relationship, but that just means 0 is within the 95% confidence interval. It doesn’t mean the entire confidence interval consists of values very close to zero. Absence of evidence is not evidence of absence. It is possible to show a null effect, but it is rarely done well in practice and almost never in non RCT studies.

    The other point this argument can lead to is that if the correlation is statistically significant one way then surely it can’t be causal the other way (since that would be a more extreme case than zero correlation and causation). But this happens all the time and you mention many of them on this site. Any time you have people making choices, you will have omitted variables which can, and often do, flip the correlation the opposite way of causation.

  9. I have really high LDL and total…like 335 or so. My HDL is high, thank goodness. Dr wanted me on statins (she’s Indian and vegetarian…she suggested don’t eat eggs or beef and exercise more!) I said, how about a CT scan, she agreed and my results were ‘0’! She then said, “but your cholesterol is high so you need to take the statins.” I haven’t done so and haven’t been back to her.

    1. Present institutionalized medicine sucks. Many doctors are doing damage as the medical system pushes the prescribed dogma. Among the many credible informed sites against statins, recommend MIT Stephanie Seneff as extremely credible on this an other matters such as eating clean.

    2. Hallie, it was an intelligence test and you passed with flying colors!

    3. One of the negatives of this continued ‘war on high cholesterol’ is life insurance. My health is in the best state it has been in 30 years thanks to a clean Paleo lifestyle, but my life insurance provider refuses to give me the best rate because my cholesterol is high. I wonder how solid their numbers are to justify their negative view on high cholesterol…

  10. Unrelated question, what is the source of the tapioca fiber in your products, (bars) hoping from the root and not from corn syrup?

  11. I admire and appreciate your knowledge and how you share it, Mark. Many thanks.

    This is not about this post…but I don’t know how else to connect with you Mark….

    I have a question I’d like to put to you before I ask anyone else:

    Is it possible [are you aware of studies or have some personal knowledge or opinion] that I have various health/well-being changes caused by changes in barometric pressure? Hormonal, blood pressure, pulse rate; head pressure, cognitive fog???

    [Is there, and if so] How could a scientific argument support a claim that I do better – from the perspective of well-being – when I am in a warm high-pressure climate or weather environment?

  12. Love the loner article, Mark! For most of my adult life, I have chosen to live and spend, oh, 90 percent of my time in solitude and “doing my own thing.” I crave it with my entire being – even as I’m also continually connecting through remote work and community, which buoys my heart + soul too.

    One of my favourite things to do alone is take long solo walks in nature, ideally in a forest or by the sea. Never feel alone in such moments, but rather, more connected than ever.

  13. Regarding correlation and causation … the XOR function (exclusive or) is one in which there are two independent variables that in combination determine the value of a third, dependent variable. Neither of the independent variables are correlated with each other or with the dependent variable. Yet, the variables in combination absolutely determine the value of the dependent variable. This is a theoretical model, but you can see how it could be applicable in cases where “there may be other factors that determine the causal relationship.” For example, factor A causes effect C, but only in the absence of factor B, and factor B causes effect C, but only in the absence of factor A. I can’t give you a specific example in the medical realm, but it theoretically could exist and would demonstrate causality without bivariate correlation.

  14. Careful. In common use “correlation” such as calculated correlation coefficients is specific to a linear relation between data: visualize drawing a straight line through pairs of points that minimize their total distance from that line. However many nonlinear relations can be constructed where there is no linear correlation. Note a nonlinear cause to effect relation can be far more significant than a linear one. Generally reality is only approximately linear, but we simplify.

    1. But this illustrates hypotheses that are false on their own, in other words, incomplete hypotheses.

  15. This isn’t exactly what you’re looking for but does relate.
    There was research on something, I cant recall at the moment. Results were variable, leading to inability to draw any conclusions. Then someone looked at the time of day the various experiments were done and realized results were consistent if time was taken into account.
    Sorry, but that is all I can remember. My biology classes took place over 45 years ago.

  16. Mark,
    Long time reader. I am a retired engineering problem solver. My career path was quite similar to Ivor Cummins.
    The phase of no correlation means no causation is only true for first oder effects. For interactions it is often possible to observe no correlation when the factor is part of an interaction, ie confounded. This not theory I have seen this many times in my engineering problem solving career.
    A simple example. I was investigating a quality problem in a polymer extrusion process. I varied a process parameter and measured linear decrease in the
    quality problem. Very high correlation in an observational sample. I reported the results to th process owner who was skepical. I conviced I was correct. To prove my hypothesis I designed a randomized control trial which failed, no correlation, altough as it turned causation. The root cause was a 2 factor interaction. At the high level of the second factor the the quality level was linearly dependent on the first factor. At the low of the second factor the process parameter had on effect on the quality level. This can be very frustrating as you can get a great variation in the quality level depending on the levels of the two factors.
    I have two a very useful Excel models with graphs and the statistics and p values that give one better intuition for these kinds of problems.

  17. Connecting dots is something you’re very good at, if there are dots to connect! Thank you for pondering this and sharing with us.

  18. Mandatory face mask protocols and preventing the spread of covid19

    Quarantining healthy people and preventing the spread of any disease

    A naturally occurring coronavirus found in bats that cannot infect human beings and a totally unique corona virus that comes from bats possessing synthetic man-made genetic sequences that allow it to easily infect and spread between human beings

    Is that what you mean?

  19. Disagree. American women are men with vaginas. Femininity has completely disappeared. No accomplished woman is called feminine any more. She’s strong and tough. Just ask her. Want a woman with a vagina? Go to Asia or Europe. Women there are not trying to be men.

  20. Have you noticed that they no longer ask you how much you smoke, but if you have ever smoked? It’s because they figured out that a single exposure can increase health risk. In this case, correlation with disease may seem weak (exposure might have been 30 years ago), but a single exposure can indeed cause a large (potentially fatal) effect.

  21. I think others have pointed this out, omitted variables can make it look like there is a lack of
    correlation/causation when there is actually a causal relationship. I think a great example might be eggs – I would argue all things equal they might cause one to be healthier but it’s you’ve pointed out previously egg eating and healthy people aren’t usually correlated probably because most people eat eggs with other unhealthy stuff and because people were told that eating eggs was unhealthy so people with otherwise healthy habits avoided them (this might be sample selection bias not omitted variable bias for statisticians out there).

    Furthermore, and far more technical, even if you do controlled experiments some percentage of the time you will get things wrong. This can be either missing relationships that are real or falsely confirming relationships that aren’t and most statisticians would err on the side of the former suggesting that actually we say things are too weakly correlated all the time.

    Causality is a hard problem.

  22. “I’m trying to imagine a time where there was no correlation in the data but the one variable still caused the other. Can you think of one?”

    I can think of a whole area of mathematics where this is the case – chaos theory.

    Iterating a chaotic system is entirely deterministic, yet gives results that appear random.