/sci/ - Science & Math

[Anonymous No.8311568]

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3988204/
>Among subjects with no diabetes at baseline those in the high protein group had a 73-fold increase in risk (HR: 73.52; 95% CI: 4.47–1209.7)

Any epidemiologyfags around? Is this legit or meme statistics?

[Anonymous No.8311650]

>only reported hazard ratios for human data in the paper itself
>left actual rates in the supplemental
also you should be leery any time an epidemiologist talks about conclusions drawn from interaction effects because 99% of epidemiologists can't actually accurately define what an interaction effect means

i dont really care to spend the time to dig through and try to find the actual rates that those HR values are drawn from but my first question would be whether the increases are biologically meaningful