Are black babies in the US really more likely to die under the care of white doctors?

This BBC podcast is supported by ads outside the UK.

How do you make an Airbnb a Verbo?

Picture a vacation rental with a host who's showing you every room like you've never seen a house before.

Now get rid of them.

There you go.

No host ever.

Now it's a Verbo.

Make it a Verbo.

What's going on, California?

It's Bluff here.

You know what's better than a relaxing day at the beach?

A relaxing day at the beach with SpinQuest.com, America's number one social casino with over a thousand slots and table games available to the comfort of your own phone, with instant cash prize redemptions, and new users get a $30 coin package for only 10 bucks.

That's S-P-I-N-Q-U-E-S-T.com.

I'll see you there.

SpinQuest is a free-to-play social casino.

Voidwear prohibited.

Visit spinquest.com for more details.

Hello, and thank you for downloading the More or Less podcast.

We're the program that looks at the numbers in the news and the world around us.

And I'm Lizzie McNeil.

Data can tell us a lot of things.

Useful things like how many vaccines need to be administered and to who, what series you should binge-watch next, or how your ranking stands up in Diabolo.

But sometimes the people interpreting data get the wrong end of the statistical stick.

Correlations get assumed to be causations and fingers get pointed in directions they probably shouldn't.

One such case was this stark headline.

The chance of survival is tripled if a black baby is looked after by a white doctor.

This came from a peer-reviewed paper published in the Proceedings of the National Academy of Sciences in 2020.

The researchers were looking into factors that could reduce the mortality risk for black babies.

And they found that if a black baby is looked after by a white doctor, they're twice as likely to die.

This headline didn't quite get that bit right.

This finding obviously, and quite rightly, caused quite the media commotion.

As it was during the time of George Floyd's death death in the Black Lives Matter movement, it exploded.

It even ended up being misquoted in the Supreme Court.

But is it true?

A black baby's cared for by white doctors really twice as likely to die?

What happens when you look at the data in a slightly different way?

For the last few years, I've been interested in the fragility of the stuff we do as researchers.

Some researchers look at a data set and they get result A.

Other researchers look at the same data set, they get result B.

And A and B completely contradict each other, right?

That's Harvard professor George Borjas.

He spends a lot of his time looking into how immigration data can be, and often is, misinterpreted.

When he heard about claims that white doctors were directly responsible for the high death rate of black babies, he needed to know more.

So he started to sift through the data.

Now, the specific data in this instance is very good data indeed.

It's from the state of Florida, where they've recorded hospital events such as births, operations, and deaths over the past 30 years.

It is very rare to have this kind of data, where every single hospital event has been recorded for decades.

Although Florida is not representative of the whole of the US, the picture of mortality for black babies born to American mothers across the US is depressingly similar.

Non-Hispanic Black or African American babies do have the highest mortality rate of any ethnic group in America, with a mortality rate rate of over 10 per 1,000 live births, compared to just over four deaths per 1,000 live births for white babies.

The original paper was extremely important.

It's actually a very important issue.

The mortality of black numerals in the US is not...

the rate is not small.

So if this paper were true, that would be an incredibly important finder that could save tons of lives, right?

To save lives, you first need to work out what the problem is.

Is it, like the first paper suggests, due to the care the babies receive from their attending doctor?

George started out by trying to replicate the result found by the original paper.

Controlling for the 65 most common events that could go wrong during the birth, they found that the black babies attended by black doctors had a much higher rate of survival than the black babies attended by white doctors.

The events listed in the original paper include conditions such as napi rash, compression of the umbilical cord, septicemia, jaundice, preterm births, and whether hallucinogens were found in the placenta or mother's breast milk.

When you control for the same things the paper controlled for, you get the same result.

In fact, they're correct.

When you do what they did, you find this connection between the race of the baby and the race of the doctor.

But then we started looking in deeper.

George discovered that something had been forgotten, something important.

We noticed that that in the literature on newborn mortality, there's one factor that really, really is very important, and that's the birth weight of the baby.

Babies born under 1500 grams have a very high chance of something bad happening, you know, and not surviving the process.

Remarkably, the original paper did not control for babies who were born at very low birth weights.

In the US, that's defined as being under 1,500 grams.

Birth weight is an extremely important factor in an infant's chance of survival.

In the US, babies born weighing 2,500 grams or less account for two-thirds of neonatal deaths.

Those who have a very low birth weight, that's under 1,500 grams, account for half of neonatal deaths.

So why weren't they included in the first place?

Once you see the answer, it's sort of obvious.

But believe me, when we started out, it wasn't obvious to us.

I mean, we didn't say, oh my god, it has to to be low birth weights.

It turned out that very low birth weights didn't have its own category.

So George and his co-author had to really sift through the data.

But once they controlled for low birth weight, the paper's original finding became clearer.

Does the baby weigh less than 1500 grams?

That's it.

The whole effect disappears.

The majority of babies dying in Florida were not regular healthy babies that had been attended at birth by a doctor of a different race.

They were instead low weight, high-risk babies with a slew of other associated health issues.

It turns out that the most vulnerable black babies, meaning those weighing less than 1500 grams, tend to be attended by white doctors.

One possible story is that the most vulnerable babies, whether black or white, tend to be assigned to the most experienced doctors.

Given the way the U.S.

medical schools and admissions has worked over 30, 40, 50, 60 years, most of the older doctors, more experienced doctors, are probably white.

So that meant that the most vulnerable black babies were being attended by white doctors.

So it's not that there was a connection between the race of the doctor and whether the baby survived or not.

It's just that the most vulnerable babies happened to be treated by the more experienced white doctors.

There's possibly a different story of discrimination going on here, of unequal opportunities within the workplace, both historical and present.

But that isn't what the original paper tackled.

And as such, we've been looking to answer the wrong question.

Why are non-Hispanic black and African American babies at a higher risk of low birth weight?

It might have to all do with maternal care.

That might be where a connection between the race of the doctor and the race of the baby shows up.

This is a very important topic.

We know the reasons aren't likely to be just genetic, as low birth weights are not particularly prevalent in middle-class Africa or even Africans giving birth in the US.

In fact, a recent study by Princeton University found that 7.8% of babies born to foreign-born black women are low birth weight, which is under 2,500 grams, compared to 11.8 American babies born to US-born black women.

For whatever reason, it's black American babies that are the most at risk.

The reason why this group has lower weight babies is currently unknown.

The correlation between low birth weight and a less privileged socioeconomic background is strong, but low birth weight does also affect African Americans and non-Hispanic black people from across American society.

Unfortunately, the Florida data does not cover care during the pregnancy stage, and this is really the missing link to telling us what the actual issue is and how we can solve it.

So, there is a problem.

but not the one the original paper suggested.

And that in itself is a problem.

If you take the original study at face value you might put in a lot of resources into something that will not solve the problem one iota.

You could actually make it worse.

If what you really care about is the the actual mortality rate of the black and white newborns you really want to get the right answer to this.

We spoke to the authors of the original paper and they told us that I'm glad that there is a continued and robust dialogue in the scholarly community about the means by which the racial mortality gap in birthing outcomes can be ameliorated.

It is important work that needs to be continued.

I think we can all agree on that.

Thanks to George Borjas.

And that's all we have time for this week.

As usual, please do keep your questions and comments coming in to more or less at bbc.co.uk.

Until next week, goodbye.

Sucks!

The new musical has made Tony award-winning history on Broadway.

We the demand to be home.

Winner, best score.

We demand to be seen.

Winner, best book.

We demand to be quality.

It's a theatrical masterpiece that's thrilling, inspiring, dazzlingly entertaining, and unquestionably the most emotionally stirring musical this season.

Suffs.

Playing the Orpheum Theater, October 22nd through November 9th.

Tickets at BroadwaySF.com.

Mike Kempel here, Patriot Software's founder.

My first business startup was in the basement of a factory.

We had no heat, no air conditioning.

Only office visitors were rats, birds, flies, and snakes.

We used blankets as our office walls.

When the floors would flood, we'd have to write software while sitting on top of our desks.

It took us four years to earn a normal paycheck after quitting our corporate job.

We know how hard it is to start a business.

That's why Patriot Software charges so little for our accounting and payroll software.