We cannot eliminate the box on the form until we first eliminate the biased care.
That is where the article ends: “the need to monitor” etc. But there is no logic to that. The biased care is eliminated by having the care providers not care about race, not by measuring it precisely. You can still record % of people who feel discriminated without doing the discrimination yourself.
The truth is that's very hard. And if we can't measure it, it's impossible to tell which bias-reducing interventions work.
I get why white people, of which I am one, want to just ignore the problem and hope it goes away. But if there's anything that history demonstrates, it's that sustained power differentials don't just go away on their own. In the US, that's what both our Revolutionary War and our Civil War were about, as well as our civil rights movement. And things like the First Nadir and the Second Nadir show that even hard-won victories like those don't last without upkeep.