You are viewing a single comment's thread.

view the rest of the comments →

0
1

[–] doompie 0 points 1 point (+1|-0) ago 

showing just how bad of a predictor race actually is.

of genetic effects. An epidemiologist would have to ignore their many years of training to believe that different prevalences among different races are driven by pure genetics -- especially as heritability doesn't also change between the populations.

But because race is tied to cultural, environmental, and social factors; it's great at summarizing these at a group level. Myocardial infarction is twice as frequent among black men than among white women (source).

I'm all for eliminating race from genetic interpretation, where it is misused in place of ancestry. Epidemiologically, however, the social construct of race tags many non-genetic factors relevant to disease. There's no way around it.