The question of methodology and that allegedly higher risk of stroke for vegans

Gus Duffy is making light of a very scary topic:

I guess I should comment on the aforementioned study. Or maybe just let the article I read on the matter do it for me:

“There is some evidence which suggests that very low cholesterol levels might be associated with a slightly higher risk of hemorrhagic stroke,” she said. Similarly, other research points to deficiencies of some nutrients, like vitamin B12, may be linked to a higher risk of stroke, said [Tammy] Tong.[1]

The problem with hemorrhagic strokes is this:

Hemorrhagic strokes make up about 13 percent of stroke cases. It’s caused by a weakened vessel that ruptures and bleeds into the surrounding brain. The blood accumulates and compresses the surrounding brain tissue.[2]

It’s the kind of stroke you don’t want to treat with anticoagulants because anticoagulants can actually be a causal factor.[3] And in an emergency situation, it can be hard to tell which kind of stroke a patient is suffering.[4]

Still, some researchers were skeptical of the stroke finding.

The research shows that people who cut out meat from their diet are significantly healthier than meat eaters, Dr. Malcolm Finlay, consultant cardiologist at Barts Heart Centre, Queen Mary University of London, told the Science Media Center.

But he said the study put “too much weight on a complex statistical method to try and correct for the fact that the vegetarians were very much healthier than meat eaters.”

“While this method can say the risk of stroke isn’t as low as one might expect it to be in vegetarians considering how much healthier they are in general compared to meat-eaters, their overall risk of a major life-changing cardiovascular event happening still appears much lower,” said Finlay, who was not involved in the study.[5]

Which is, to say the least, a bit of a different story from the headlines that flashed all over the fucking place.

I’m remembering the dilemma I found myself with as I was still at California State University, East Bay, where I finished my Bachelor’s and my Master’s degrees. I took two statistics classes because at that point I was still enamored with a quantitative approach. And I was taking seriously (and still do) the bit about how when reading a scholarly article, one should examine the methodology, which is very often statistical.

I was discovering rather rapidly that the very last thing I wanted to do was become a statistician. Mathematics is not my thing—I hit a brick wall with trigonometry—and I know statisticians say that what they do is not math but it’s close enough to math that I can’t tell the difference.

I also realized that one would have to be a statistician to evaluate the appropriateness of these often exotic statistical methods, let alone whether or not they were done correctly. Which would entail taking on a second major in the topic, which I most certainly didn’t want to do.

By the time I was working on my Master’s, I understood very well that I couldn’t tell if researchers weren’t using statistical methods the way some car aficionados worked on their cars, substituting a carburetor here, a different exhaust manifold there, trying to optimize performance. Which would be cherry picking a method to obtain a particular result.

And of course, that would be a no-no. A researcher is supposed to choose a methodology that is appropriate to the research question and actually asks the question, not in any way predetermining the result.

The reality of that is that there’s some grey area: When I was recruiting a dissertation committee, I had to acknowledge right up front that conservatives were unlikely to look good under the lens of a critical theory method, which is what I employed, and I acknowledged doubt that they (conservatives) would accept my results.

This is inherent to the method: Critical theory seeks to expose power relationships. It inclines toward social justice. Conservatives, in their various tendencies, generally tend to prioritize other concerns.

That wasn’t a reason not to conduct the study: Such an argument would be to delegitimize critical theory itself because the powerful will nearly always act to protect their power; they are thus nearly all, to some degree, functionalist conservatives. To attack my work on these grounds would be to argue that no legitimate research can ever challenge the power relationships of the status quo.

Such an attack would take out a much broader swathe of research than that using critical theory methodology. But the entire idea of a positivist, quantitative approach is to reify an objectivity that doesn’t really exist—none of us have a “God’s eye” view. Critical theory doesn’t pretend to be objective; we are instead to state our social location relative to the subject at hand. Statistical methods do pretend that objectivity, so the problem of bias is actually more insidious in quantitative methods than it is in qualitative methods.

Which is how I come back to that “complex statistical method.” Just like those kids and their carburetors.

Certainly there is an interest in showing the health and environmental benefits of a plant-based lifestyle. There is also, correspondingly, an interest in showing the inverse, attempting to minimize those benefits. And the question of an “objective” method, when objectivity doesn’t actually exist, means the choice between arcane statistical methods isn’t nearly so black and white as one might assume.

And that, in essence, is the challenge that Malcolm Finlay was raising. So what we can really take from this is that vegans, including myself, should make sure we’re getting our vitamins. To any extent there is any “slightly higher risk of hemorrhagic stroke,” it may be occurring in people with “very low cholesterol levels” (not me). Finally, statistical methods show correlations, not causation, so we don’t actually know what is elevating that risk, even if that risk is only relative to expectations for “much healthier” people.[6]

  1. [1]Nina Avramova, “Vegetarians might have higher risk of stroke than meat eaters, study says,” CNN, September 4, 2019, https://www.cnn.com/2019/09/04/health/vegetarian-vegan-diet-stroke-heart-disease-risk-intl/index.html
  2. [2]American Stroke Association, “Hemorrhagic Stroke (Bleeds),” n.d., https://www.stroke.org/en/about-stroke/types-of-stroke/hemorrhagic-strokes-bleeds
  3. [3]Mary Ellen Ellis, “Hemorrhagic Stroke,” HealthLine, May 23, 2018, https://www.healthline.com/health/hemorrhagic-stroke; MedLine Plus, “Ischemic Stroke,” n.d., https://medlineplus.gov/ischemicstroke.html
  4. [4]Kenneth S. Yew and Eric Cheng, “Acute Stroke Diagnosis,” American Family Physician 80, no. 1 (2009): 33-40.
  5. [5]Nina Avramova, “Vegetarians might have higher risk of stroke than meat eaters, study says,” CNN, September 4, 2019, https://www.cnn.com/2019/09/04/health/vegetarian-vegan-diet-stroke-heart-disease-risk-intl/index.html
  6. [6]Nina Avramova, “Vegetarians might have higher risk of stroke than meat eaters, study says,” CNN, September 4, 2019, https://www.cnn.com/2019/09/04/health/vegetarian-vegan-diet-stroke-heart-disease-risk-intl/index.html

One thought on “The question of methodology and that allegedly higher risk of stroke for vegans

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.