In medical school, we teach our future doctors to put the patient's race right up front.
"X-year-old-Race-Gender person with a chief complaint of bunnyhopping."
Race and ethnicity are based on appearance or self-reported, the latter being the gold standard in the clinic. We have no way to confirm or deny someone's identity in an objective way unless genealogy studies are available*.
Even if you believe that race can be important for a patient, how do you factor it in? Do I assess the blonde-haired blue-eyed girl as the white person she appears to be or as the Cherokee that she legally is?
Society at large, like Native American populations, has mixed over time. Even in my career, race has not been as helpful in diagnosis or prognosis as we were led to believe in medical school. I have seen black children with cystic fibrosis and white children with sickle cell disease.
Race and ethnicity matter for health, of course, since they associate with social, cultural, and economic factors that are far more important in treating and preventing disease than the color of skin, the texture of hair, or the shape of facial features. While there may be some genetic risks that segregate by race, we should identify the genes rather than using a poor proxy like race. We also need to be an inclusive society for people of all identities to minimize some of these social, cultural, and economic factors that segregate with race and ethnicity (hey, I can dream).
I have quit asking and reporting race in my patient notes. No one is grading me at this point in my career.
That blonde Cherokee girl can get a lot of care through her tribal health center at no cost to her family. Now that's a good thing to know for her medical management.
*And you never know how accurate these are since children may be fathered by someone not on the birth certificate