[teejers1]

Greater numbers of comorbidities? Ok, screen for comorbidities on some agreed upon scale. If you have two individuals in the same category, they rate the same. Disregard race. You do not immunize a young African-American because he is African-American, and ignore an older Asian or Caucasian because of race. That just won't do.

I ran my data through the QCOVID risk prediction algorithm, and got the following:
COVID associated death
0.0027%
1 in 37037
COVID associated hospital admission
0.0433%
1 in 2309

I ran the same exact risk profile, except that I made myself Caribbean (this was derived in the UK, so race/ethnicity categories are somewhat different there), and I get the following:

COVID associated death
0.0055%
1 in 18182
COVID associated hospital admission
0.099%
1 in 1010

In other words, just about twice the risk of death and hospitalization, even though everything else (age, sex, height, weight, risk factors) were exactly identical.  Again, this is UK data and UK prediction algorithm, but the exact same trend would be found if you derived one for the US.  The bottom line is that there is likely an unmeasured factor that is captured by race that is not captured by our conventional risk factors.  Whether that is enough to survive a legal challenge, I don't know, but as of right now, Black and Hispanic patients with COVID-19 have a higher risk of dying than White patients, even with the same risk factors.

BC

That is very interesting.  But are you sure the conclusion you draw is the only one to draw from the data?  It seems to me it's possible that a higher % of the Latino (or AA) community either works in indispensable, in-person workforces or lives in close quarters, or both.  If that's true, then is it possible that race is simply a byproduct of non-race-based risk factors.  BTW, if your conclusion is the correct one, then I think Newsom's distribution plan might pass constitutional muster - it strikes me as a compelling rationale to support a government policy based on race.

[teejers1]

Greater numbers of comorbidities? Ok, screen for comorbidities on some agreed upon scale. If you have two individuals in the same category, they rate the same. Disregard race. You do not immunize a young African-American because he is African-American, and ignore an older Asian or Caucasian because of race. That just won't do.

I ran my data through the QCOVID risk prediction algorithm, and got the following:
COVID associated death
0.0027%
1 in 37037
COVID associated hospital admission
0.0433%
1 in 2309

I ran the same exact risk profile, except that I made myself Caribbean (this was derived in the UK, so race/ethnicity categories are somewhat different there), and I get the following:

COVID associated death
0.0055%
1 in 18182
COVID associated hospital admission
0.099%
1 in 1010

In other words, just about twice the risk of death and hospitalization, even though everything else (age, sex, height, weight, risk factors) were exactly identical.  Again, this is UK data and UK prediction algorithm, but the exact same trend would be found if you derived one for the US.  The bottom line is that there is likely an unmeasured factor that is captured by race that is not captured by our conventional risk factors.  Whether that is enough to survive a legal challenge, I don't know, but as of right now, Black and Hispanic patients with COVID-19 have a higher risk of dying than White patients, even with the same risk factors.

BC

That is very interesting.  But are you sure the conclusion you draw is the only one to draw from the data?  It seems to me it's possible that a higher % of the Latino (or AA) community either works in indispensable, in-person workforces or lives in close quarters, or both.  If that's true, then is it possible that race is simply a byproduct of non-race-based risk factors.  BTW, if your conclusion is the correct one, then I think Newsom's distribution plan might pass constitutional muster - it strikes me as a compelling rationale to support a government policy based on race.