Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Mortality rate for Black babies plummets when they’re delivered by Black doctors

washingtonpost.com

Post reply on HN
Print view View original

61–70 of 100 posts

Re: Mortality rate for Black babies plummets when they’re delivered by Black doctors

#61

Post by sz4kerto »

croissants wrote: post #57
AzzieElbab wrote: post #51

Earlier quoted context omitted.

Maybe they picked up outstanding doctors who deliver outstanding results regardless of patients race? I mean the dataset is so small, one off can derail any conclusion

> I mean the dataset is so small, one off can derail any conclusion What do you mean? Per the Supplement, their data comes from a pool of ~4k white doctors and ~500 black doctors, with hundreds of thousands of observations/patients. This dataset doesn't seem "so small"? [1] https://www.pnas.org/content/pnas/suppl/2020/08/12/191340511...

It really all depends on the methodology. I recommend some reading about 'p-hacking'. It is not that hard to find to find really shocking findings in large datasets -- the question is how hard you look. In this particular case, the question is how many hypotheses have been evaluated, i.e. has the author 'mined the data' for correlations. If yes, then the acceptable p value is much smaller than normal.

Re: Mortality rate for Black babies plummets when they’re delivered by Black doctors

#63

Post by dang »

sockjuice wrote: post #22
x86_64Ubuntu wrote: post #12

Earlier quoted context omitted.

Black America is well aware of the fact that being black means that we will run into sub-part outcomes in US society. This isn't news to our community. Just because it causes feelings of racial anxiety in other groups doesn't mean it shouldn't be mentioned and be studied. Before such a study, when black Americans would have noted that something seemed off, we would have been met with "You need data to back that up, X…

Exactly this, but you will of course be downvoted here for pointing out this truth. For a purportedly open-minded set of hackers who claim to follow their intellectual curiosity, too many people here are very closed-minded about racism.

That comment is heavily upvoted. Your comment here, however, has rightly been downvoted and flagged, not just for being false, but for being flamebait and a supercilious dismissal. Please read https://news.ycombinator.com/newsguidelines.html and stick to the rules when posting here.

Re: Mortality rate for Black babies plummets when they’re delivered by Black doctors

#64

Post by jart »

klyrs wrote: post #38
sz4kerto wrote: post #4

It would be really-really important to use pre-registration especially in socially sensitive studies like this one. It is possible that these numbers are artifacts even if the stated p value is low. I'm not saying that the study is wrong, just that shocking results like this one need extreme methodological rigor behind them before we start telling everyone that white doctors are killing black babies etc. https://en.m…

> ...before we start telling everyone that white doctors are killing black babies etc. We aren't yelling that, though, and you're the first one to say it here. This fear of being accused of racism is intense; as if acknowledging that black people are actually treated worse by our society is actually more harmful than the loss of life that black people experience. This is a preliminary study, and yes, it deserves a fo…

> If you're saying that preliminary studies should always be kept under wraps and not discussed, that's weird and anti-science.

See https://www.medrxiv.org/

> Preprints are preliminary reports of work that have not been certified by peer review. They should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.

It's annoying when news websites report on research that isn't reviewed. It's been happening at an increasing rate over the last few years because these news websites are so hungry for advertising revenue. It undermines the public's trust in science. It threatens our ability to do research openly and inclusively.

Re: Mortality rate for Black babies plummets when they’re delivered by Black doctors

#66

Post by lern_too_spel »

CyberRabbi wrote: post #6

First: correlation is not causation. Second: what percentage of obgyns are black? Could Simpson’s paradox be at play?

The percentage of obgyns that are black is not a factor for Simpson's paradox. The only thing that matters is if the black patient population for black doctors is different from that of white doctors (like if the black patient population for black doctors has a smaller percentage of diabetics).

Re: Mortality rate for Black babies plummets when they’re delivered by Black doctors

#67

Post by watwut »

polishdude20 wrote: post #13

Just speculating here but could it be that, in neighborhoods where a black person is more likely to become a doctor, those neighborhoods have a better healthcare system in place? The cities/neighborhoods with higher black poverty probably have less black doctors in general.

There was no doctor race correlation if baby was white. Which means that it is not just general healthcare quality issue.

Re: Mortality rate for Black babies plummets when they’re delivered by Black doctors

#68

Post by watwut »

lawnchair_larry wrote: post #52
DoreenMichele wrote: post #50

Earlier quoted context omitted.

And how is it 'more blacks are poor' persistently across the nation for hundreds of years having nothing to do with the color of their skin? Blacks are last hired, first fired. There is a history of laws being written to explicitly and intentionally exclude them and not all those laws have been expunged from the books. Redlining and similar practices are well documented. I don't know how anyone can suggest that pover…

> Blacks are last hired, first fired. Sounds like you don’t have much exposure to large companies these days. If anything it’s the opposite. You’re incorrect that there are any racist laws remaining that might influence hiring, unless you are referring to govern contracts, in which minority owned businesses are given preference. It is silly to introduce historical artifacts as though they are the situation today.

Common, even employment statistics for last year show that white males gained while black women lost the most.

It is not just racism/sexism but, seriously, white people are not oppressed.

Re: Mortality rate for Black babies plummets when they’re delivered by Black doctors

#69

Post by watwut »

croissants wrote: post #46
uniqueid wrote: post #31

I wonder if anyone has conducted a study on how many internet comments on articles alleging bias are variations of 'now hang on a second, there may be an innocent explanation here'.

Eh. It's not like the internet has a shortage of people happy to read a study's editorialized conclusion on Twitter, take it as truth, and go on with their day. A little critical thinking --- and this means going beyond "I don't like this, so I'll say correlation does not imply causation and move on to something else" --- seems fine if you're trying to do it honestly and carefully?

But like, whenever there is positive study about computer games, HN is not critical at all. Instead, it is completely non critical, extrapolates findings, even if the n is like 5 and study weak.

My point is, some studies are knee jerkly believed and others are knee jerkly doubted.

Re: Mortality rate for Black babies plummets when they’re delivered by Black doctors

#70

Post by busythrowaway22 »

croissants wrote: post #19

If you're interested in the raw data behind these conclusions, it looks like the paper supplement [1] has some. One potentially meaningful point from Table S1A of the supplement: it looks like there are 360k white doctor-black patient observations and 97k black doctor-black patient observations, and the mean length of stay for white doctor-black patient observations is 4.73 days vs. 3.37 days for black doctor-black p…

If you have the time, take a look at the supporting information for Materials and Methods. This is interesting to me as well, but my understanding of the English language is not as good.

Some key takeaways:

-They use delivery data from different categories of births, including 'complications occurring during the course of labor and delivery'. Do they account for the conditions in which the babies were born, and then attribute that to the types of doctors handling those deliveries? In other words, do they make a distinction saying emergency white doctors perform worse than emergency black doctors? Because I cannot find that information in the research.

-Since doctor race data isn't codified in Florida's database (where they use the data from), they used web pictures and determine if the doctor is white or black. They narrowed this task down to 4 people who were good at it.

-Black female doctors performed the best, they mention some control of the gender but I do not fully understand how they did it.

-They used a Latino doctor group as well to account for differences, but found that while there is a distinction, it is much smaller than Black-White.

I also don't understand if/how they control for the quality of the hospital. Is there one bad low income hospital with white doctors in a predominantly black area?

Post reply on HN