3 ms·

I think this is the most important figure in the actual study, and it absolutely doesn't justify the bold claims of bigthink. The upper right quadrant is the da

I think this is the most important figure in the actual study, and it absolutely doesn't justify the bold claims of bigthink. The upper right quadrant is the danger zone. https://www.nature.com/articles/s41591-022-01968-z/figures/4 https://www.nature.com/articles/s41591-022-01968-z/figures/4 The study tries to account for things like publication bias (though seems to not find any evidence for it or do any accounting for it in 5/6 of the comparisons under study). I don't think it's wise to conclude from the fringe cutting through the x-axis that only shoddy research would claim unprocessed red meat has a health risk (which is to say nothing about processed red meat, the subject of most of the headline-grabbing studies about health risks).

tony_cannistra 4y ago
> I don't think it's wise to conclude from the fringe cutting through the x-axis that only shoddy research would claim unprocessed red meat has a health risk You're right, but they don't claim that. The y-axis here is "relative risk." There's no such thing as "no risk" in these studies. Practically speaking, of course, there is -- most of us live our lives with "low risk" = "no risk," necessarily. The takeaway from this figure is exactly what's written in the article: "As shown in Fig. 4, the relationship between unprocessed red meat intake and combined-cause incidence and mortality was increasing across the entire exposure domain."
- dundarious 4y ago
  > The takeaway from this figure is exactly what's written in the article: "As shown in Fig. 4 [...]" That's from the study that I didn't criticize, not the bigthink article that I do criticize. The claims about "shoddy research" and "not a health risk" are from the bigthink article, not the study. If bigthink rewrote the article to focus on the quotation you provided, it would be far more informative. Edit: And to somewhat address your point about relative risk (RR). If someone told me that an activity had almost no upside health risk, and a wide range of downside health risk, I would not think "no health risk" is an appropriate summary statement. Again, this is a criticism of the bigthink (or really, RealClearScience) article.
  - pessimizer 4y ago
    The upside of consuming calorie and nutrient-rich foods is not hidden. They allow us to sustain life.
    - dundarious 4y ago
      True, but vapid.

tony_cannistra 4y ago

> I don't think it's wise to conclude from the fringe cutting through the x-axis that only shoddy research would claim unprocessed red meat has a health risk You're right, but they don't claim that. The y-axis here is "relative risk." There's no such thing as "no risk" in these studies. Practically speaking, of course, there is -- most of us live our lives with "low risk" = "no risk," necessarily. The takeaway from this figure is exactly what's written in the article: "As shown in Fig. 4, the relationship between unprocessed red meat intake and combined-cause incidence and mortality was increasing across the entire exposure domain."

dundarious 4y ago
> The takeaway from this figure is exactly what's written in the article: "As shown in Fig. 4 [...]" That's from the study that I didn't criticize, not the bigthink article that I do criticize. The claims about "shoddy research" and "not a health risk" are from the bigthink article, not the study. If bigthink rewrote the article to focus on the quotation you provided, it would be far more informative. Edit: And to somewhat address your point about relative risk (RR). If someone told me that an activity had almost no upside health risk, and a wide range of downside health risk, I would not think "no health risk" is an appropriate summary statement. Again, this is a criticism of the bigthink (or really, RealClearScience) article.
- pessimizer 4y ago
  The upside of consuming calorie and nutrient-rich foods is not hidden. They allow us to sustain life.
  - dundarious 4y ago
    True, but vapid.