The News Media’s Dangerous Obsession With “Fake Facts”

What do most people not understand about the news media? I would say two things. First: The most important bias in news media is not left or right. It’s a bias toward negativity and catastrophe. Second: While it would be convenient to blame the news media exclusively for this bad-news bias, the truth is that the audience is just about equally to blame. The news has never had better tools for understanding exactly what gets people to click on stories. That means what people see in the news is more responsive than ever to aggregate audience behavior. If you hate the news, what you are hating is in part a collective reflection in the mirror. If you put these two facts together, you get something like this: The most important bias in the news media is the bias that news makers and news audiences share toward negativity and catastrophe. Jerusalem Demsas, a staff writer at The Atlantic and the host of the podcast Good on Paper, joins to discuss a prominent fake fact in the news—and the psychological and media forces that promote fake facts and catastrophic negativity in the press.

If you have questions, observations, or ideas for future episodes, email us at [email protected].

In the following excerpt, Derek and Jerusalem Demsas explore why the recorded deaths associated with maternal mortality have risen so drastically in the past few decades and why that number may not be what it seems.

Derek Thompson: I want to talk about the phenomenon of fake facts: facts that shape the discourse that in the passage of time and with increasing research we realize aren’t actually true. And I want to start with reporting that you did on something that I absolutely considered a fact for many, many years, which is the idea that maternal mortality in the United States is much higher than in similarly rich countries and that it had meaningfully increased in the last few decades.

I found this factoid everywhere that I looked: Commonwealth Fund, which is a health care–focused think tank that I’ve relied on for many articles, claimed many times that maternal mortality in the U.S. had risen. In 2023, The Wall Street Journal said maternal mortality was the highest it had been since 1965. Politicians talk about this a lot. Stacey Abrams made it a core plank of her policy platform. U.S. Representative Emanuel Cleaver, a Democrat from Missouri, recently hosted a roundtable of experts to discuss the ostensible rise in maternal mortality. Before we investigate the actual factualness of this fact, was this a factoid that you were vaguely aware of as well?

Jerusalem Demsas: Oh, yeah. I feel like as someone, particularly a young woman, who’s entering when I’m starting to think about having kids of my own and my friends are all starting to think about that too. I mean, this has just been a background fact I feel like of my life for the last decade, is knowing that here in the U.S., maternal mortality is rising. It’s rising across age groups. It’s rising in comparison to our peer countries. This kind of general sense of doom around it is like, “We’re getting worse. We don’t really know why. We know the U.S. health care system has tons of problems.” But it really has been just a background fact of my entire thinking about the problem.

Thompson: So is it true? Is it true that maternal mortality in the U.S. is rising and that it has for decades been significantly higher than, as you said, our peer countries?

Demsas: So it is true that there is a line on a graph that goes up when you look from 1999 to 2019. You will see that recorded deaths as associated with maternal mortality have risen, and they’ve doubled actually. There was a big study that came out about data from 1999 to 2019 that showed that maternal mortality had more than doubled over that 20 years. That’s true. That did happen. But what’s happening recently is that there are a bunch of researchers, and I think the one that I really want to call out here as having really put together a bunch of this is Saloni Dattani at Our World in Data. And what she really puts together is this picture that shows that a lot of the rise is due to a change in measurement, a change in how we measure maternal mortality. Not that there are actually more women dying in 2019 versus 1999 from maternal mortality deaths but that we’re just measuring it differently.

Thompson: As specifically as you can, what is happening here? How is the different way that states are counting death certificates creating what might be a false impression that maternal mortality is going up?

Demsas: So measuring maternal mortality is hard, right? Because this is just not ... I mean, there are some clear-cut cases, right? If a woman dies in childbirth, it can be pretty clear to say, “Look, this is why someone died.” But if you’re thinking about someone who may have a ton of underlying health issues—maybe they have HIV, maybe they have high blood pressure, maybe they have diabetes—there are a bunch of different things that can happen that pregnancy can exacerbate. And thus doctors are often making difficult calls around what is and isn’t maternal mortality: what is related to the death, what is exacerbated. And at the end of the day, you have to pick one. There’s not a world in which you can have multiple causes of death.

And so what ends up happening is that there’s a big push because there’s a recognition that around the world, including in the United States, there’s actually an underreporting going on. A bunch of women who clearly had been pregnant within the last year, their health had deteriorated as a result. And so we needed to figure out a way to count those women’s deaths as associated with their pregnancy.

And so there was a change made to institute what’s called a pregnancy checkbox. And this checkbox would be on death certificates in the United States and would be required for the doctor/medical examiner filling out that document to check that box if the woman had been pregnant. And these were just automatically coded as maternal deaths in many cases. And so what’s really interesting, what happens is this doesn’t happen all at once, right? Because if you had seen one massive jump in a year, I think everyone would’ve realized what had happened. But states implemented this in a staggered fashion. So four states here, a couple states here over the course of two decades. You see a bunch of states implementing this change to how their death certificates look.

And so you see this really alarming but steady rise in maternal mortality in the United States as a result. And there are researchers that have looked into this, and they’re finding that the pregnancy checkboxes are definitely overcounting maternal mortality deaths. So there was a quality assurance study done of four states: Ohio, Michigan, Georgia, and Louisiana. And they find that 21 percent of women who had been counted as pregnant were not pregnant when they died or had not been pregnant in the year previously; 7 percent could not even be confirmed.

And so nearly 30 percent, potentially, of women who had been counted as having been pregnant within a year of their death maybe were not at all. And so that’s not even entering into the kind of murky phase of whether or not someone died related to pregnancy. That’s just on the simple question of: “Were you or were you not pregnant?” And so this kind of measurement error, which really was born out of a correct concern that we were undercounting maternal mortality, pushed us in overcounting, in many cases, the maternal situation.

And what’s really important to note is that there are many countries outside of the United States who did not institute this pregnancy checkbox, who have not been trying to address this undercounting in the way that we have. And so their numbers in many cases remained steady, remained low, probably because they’re still undercounting the number of women who are dying as a result of pregnancy.

This excerpt was edited for clarity. Listen to the rest of the episode here and follow the Plain English feed on Spotify.

Host: Derek Thompson
Guest: Jerusalem Demsas
Producer: Devon Baroldi

Subscribe: Spotify