5 ms·

I'm curious what people would propose we do about this? What if it turns out that doctors of the same ethnic group treat their co-ethnics better, even if it isn

I'm curious what people would propose we do about this? What if it turns out that doctors of the same ethnic group treat their co-ethnics better, even if it isn't conscious? I fear the conclusion people will draw is just a new form of segregation in every area of life.

whateveracct 8y ago
Train them better. There’s nothing inherent about these studies’ results. Doctors are well-educated human beings and are capable of correcting for their subconscious biases.
geofft 8y ago
If that is true (which I think there isn't a real reason to believe a priori, but sure, it's a possible hypothesis), I feel like I'd prefer a world where more people live to a world with less segregation, if I had to choose.
skybrian 8y ago
Or maybe it's an argument in favor of diversity: "The male doctors in their study were better at treating women with heart attacks when they had more experience treating such patients—and especially when they worked in hospitals with more female doctors. This suggests that whatever female doctors are doing that’s better is also transferable." https://www.theatlantic.com/science/archive/2018/08/women-more-likely-to-survive-heart-attacks-if-treated-by-female-doctors/566837/ https://www.theatlantic.com/science/archive/2018/08/women-mo...
- Rapzid 8y ago
  "These differences persisted even after the team accounted for factors like the doctors’ years of experience.." So, I'm having a little bit of trouble reconciling these two statements. Who are "such patients" and why wouldn't doctors with more years of experience have more experience treating them?
  - skybrian 8y ago
    "Such patients" means women who had heart attacks. Apparently, "years of experience" doesn't necessarily mean the same thing? Maybe some kinds of experience are better than others? But if you really want to dig into this you'll probably have to read the paper.
h4b4n3r0 8y ago
Try to replicate in a more rigorous statistical setting. Do nothing until the result is replicated. I’d be willing to bet a significant amount of money that this result is due to a missed latent variable (or a few) that explains everything without making it seem like gender is causal to the outcome.

whateveracct 8y ago

Train them better. There’s nothing inherent about these studies’ results. Doctors are well-educated human beings and are capable of correcting for their subconscious biases.

geofft 8y ago

If that is true (which I think there isn't a real reason to believe a priori, but sure, it's a possible hypothesis), I feel like I'd prefer a world where more people live to a world with less segregation, if I had to choose.

skybrian 8y ago

Or maybe it's an argument in favor of diversity: "The male doctors in their study were better at treating women with heart attacks when they had more experience treating such patients—and especially when they worked in hospitals with more female doctors. This suggests that whatever female doctors are doing that’s better is also transferable." https://www.theatlantic.com/science/archive/2018/08/women-more-likely-to-survive-heart-attacks-if-treated-by-female-doctors/566837/ https://www.theatlantic.com/science/archive/2018/08/women-mo...

Rapzid 8y ago
"These differences persisted even after the team accounted for factors like the doctors’ years of experience.." So, I'm having a little bit of trouble reconciling these two statements. Who are "such patients" and why wouldn't doctors with more years of experience have more experience treating them?
- skybrian 8y ago
  "Such patients" means women who had heart attacks. Apparently, "years of experience" doesn't necessarily mean the same thing? Maybe some kinds of experience are better than others? But if you really want to dig into this you'll probably have to read the paper.

h4b4n3r0 8y ago

Try to replicate in a more rigorous statistical setting. Do nothing until the result is replicated. I’d be willing to bet a significant amount of money that this result is due to a missed latent variable (or a few) that explains everything without making it seem like gender is causal to the outcome.