What do observational studies tell us?
| Happy Sunday, folks. Today I'm talking about another observational study, this time one that maligns animal protein but in a really weird way. Let's get the usual problems with observational studies out of the way. The main author tweeted it out herself in her thread describing the results: "There is no scientific consensus on which macronutrient to restrict (or boost): huge, long-run observational studies find different effects than small, short-run experimental studies that put people on extreme (undoable?) diets." That places observational studies and experimental studies on equal footing. This is not the case. Observational studies cannot prove causation and pale in comparison to experimental studies, which actually have the power to support (or disprove) causation. Has she ever wondered why that might be? Why observational studies conflict with experimental studies? I have to wonder: what is the point of all these observational studies? Are we hoping that the "next one" is going to reveal some hidden secret variable that we hadn't considered and that revolutionizes our knowledge base and finally shows us the perfect diet? They all end up saying the same thing, anyway. They get mired in the healthy user bias and the unhealthy user bias. They fail to account for the buns on burgers and the seed oils everything is cooked in. It makes for good material for me, but isn't it enough? Heck, at least all the other observational studies we've known and loved to hate actually dealt with links to the actual health conditions. This one was a "genomic analysis of diet composition." Rather than link protein intake directly to poor health, it linked protein intake to the genetic markers that other studies have shown to be linked to poor health. In fact, the only actual health condition they studied was BMI, finding that protein intake was linked to higher BMIs. But that wasn't new research. For the BMI/protein link, they just used existing data from other studies. Everything else was a link to a gene which was linked to something else. Those who ate the most protein were more likely to carry an allele that's been linked to heart disease. Those who ate the most protein were more likely to carry an allele that's been linked to high blood sugar. Those who ate the most protein were more likely to carry an allele that's been linked to large waist circumferences. And so on. There were no true end points like death or cancer or diabetes, in other words. The genes were the "end points." Does that seem odd to anyone else? What is this really telling us? Are people with a genetic proclivity to heart disease hungrier for protein? What if you don't have the genes for heart disease and a big waist and poor glucose tolerance but you do eat tons of protein? Or the reverse? What happens if you break the mold? Anyway, I'm struggling with this one. Just trying to understand what the takeaway is here, or whether they just had some cool genetic correlational data laying around that they felt like blending up with Food Frequency Questionnaire results. What about you? What are you struggling with? Any small complaints or concerns? Any existential complaints or concerns? Let's hash it out in the comment section of Weekly Link Love. |
|
|
| | | No longer want to receive these emails? Unsubscribe.
Mark's Daily Apple 1641 S. Rose Ave. Oxnard, CA 93033 |
|
|
|
|
