3 ms·

Gosh, for all the bluster in this comment section of supposed technical folks nobody seems to be highlighting what appears to be a very compelling methodologica

Gosh, for all the bluster in this comment section of supposed technical folks nobody seems to be highlighting what appears to be a very compelling methodological advance in the Burden of Proof methodology [0] used to produce this result. > "Here, we propose a complementary approach, in which we quantify the mean relationship (the risk function) between risk exposure and a disease or injury outcome, after adjusting for known biases in the existing studies. Unlike existing approaches, our approach does not force log-linearity in risk functions or make additional approximations, such as midpoint approximations for ranges or shared reference group" [0]: https://www.nature.com/articles/s41591-022-01973-2 https://www.nature.com/articles/s41591-022-01973-2

DennisP 4y ago
Since they don't mention kidneys, I'd like to see someone apply their approach to this study, claiming that red meat increases the risk of end-stage renal disease: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5198288/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5198288/
- tony_cannistra 4y ago
  That's a good idea. As I understand it the Burden of Proof method is a meta-analytical method, though, which means it gets applied to a corpus of studies. I'm sure there are more than 1 study that tries to figure out that relationship, though.
boxed 4y ago
Am I getting this right? 1. MANY studies over many years show red meat consumption is bad 2. This study shows that it's bad, but not a lot 3. This is ONE study we're talking about 4. This study invents a new statistical methodology
- tony_cannistra 4y ago
  Unfortunately, you don't have it quite right. This "one study" is a meta-analysis, which considers many studies. Furthermore, meta-analysis methodology is very important. Doing it wrong is bad. Working to make it better is good. They did that.
  - boxed 4y ago
    This isn't the first meta analysis though...

DennisP 4y ago

Since they don't mention kidneys, I'd like to see someone apply their approach to this study, claiming that red meat increases the risk of end-stage renal disease: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5198288/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5198288/

tony_cannistra 4y ago
That's a good idea. As I understand it the Burden of Proof method is a meta-analytical method, though, which means it gets applied to a corpus of studies. I'm sure there are more than 1 study that tries to figure out that relationship, though.

boxed 4y ago

Am I getting this right? 1. MANY studies over many years show red meat consumption is bad 2. This study shows that it's bad, but not a lot 3. This is ONE study we're talking about 4. This study invents a new statistical methodology

tony_cannistra 4y ago
Unfortunately, you don't have it quite right. This "one study" is a meta-analysis, which considers many studies. Furthermore, meta-analysis methodology is very important. Doing it wrong is bad. Working to make it better is good. They did that.
- boxed 4y ago
  This isn't the first meta analysis though...