Is my baby going to die in the middle of the night?

A common nightly scene in households with young children: exhausted parent shuffles kid through dinner, bathtime, storytime. Then, after an hour or so of trying to get the child to sleep, exhausted parent collapses on the couch in relief. But five minutes later, exhausted parent scans a poor-quality video feed from the kid's room to see if her belly is moving up and down. 

New parents seem to spend a lot of time worrying that their kids are going to spontaneously die. And it's easy to understand why. Parenting resources share terrifying details about the rare conditions that might lurk within a seemingly healthy baby. The 2014 CDC report on infant mortality, for example, opens with the terrifying figure of 23,000 infant deaths. Just reading it probably makes you want to risk waking up your sleeping munchkins in order to check for breathing. But what does that number really mean?

To figure that out, we have to travel through some scary and sad data. In 2013, there were 23,446 deaths per 3,932,181 births. Put another way, 6 in 1,000 babies will die within their first year. This figure ranks the US considerably lower than other wealthy countries

But part of the reason the US's infant mortality rate is so much higher than other countries is that not all countries count pre-term births. In their investigation of infant mortality, Alice Chen, Emily Oster, and Heidi Williams eliminated infants born before 22 weeks' gestation, a figure more in keeping with how other countries measure live births. That calculation brought the US deaths per 1,000 live births to 4.2.

Chen, Oster, and Williams also studied how the infant mortality rate varies across different populations within the US. Some areas of the country have rates much higher than the national average, and infant mortality appears linked with socioeconomic status. While in other nations infant mortality is more evenly spread across demographic groups, in the US, where you live and how much education you have can dramatically affect the infant mortality rate. 

The work of economists like Chen, Oster, and Williams to more accurately determine the infant mortality rate, and medical and public health professionals to lower that rate, are critical work. But it is possible for individual parents to place too much emphasis on these numbers. In focusing on our country's relatively high infant mortality rate when compared to other wealthy nations, we forget the importance of the world "relative." Whether the number is 6 or 4.2 in 1,000 live births, the United States' infant mortality rate is extremely high when compared with other similar nations. But the incidence of infant mortality itself is extremely low. 

If you haven't read much about probability and statistics since picking M&Ms out of a bag in elementary school, now's the time. We tend to think of numbers as concrete expressions of reality, but as Edward MacNeal has argued in the wonderful primer Mathsemantics, the same number can often sound different when expressed in different ways. 

For example, hearing a percentage expressed negatively versus positively impacts how people interpret risk, and therefore the choices that they make. In "Rhetorical Numbers," [free to read with a MYJSTOR account, also free] Joanna Wolfe describes a friend concerned about a pregnancy book that gave her a 1 in 50 chance of a particular pregnancy complication. That friend was reassured by her doctor, who told her that she had a 97% chance of a healthy child. The second scenario is actually worse than the first (97% vs. 98% chance of no complication), but people perceive it to be better because it is expressed in abstract terms.

When a number is expressed as 1 in x, we seem to view ourselves as the 1 because we can visualize it. As Wolfe's friend put it, "I know fifty people."  She could imagine herself being that one of fifty, even though the probability was quite low. It's the same kind of logic, Wolfe notes, that encourages people to buy lottery tickets: a 1 in 20 chance of winning sounds much more likely to happen than a 5% chance of winning, even though those statements are identical. 

The 6-in-1000 ratio used to describe infant mortality is a helpful way of comparing infant mortality in the US to infant mortality in other countries (and also from state to state, as Chen, Oster, and Williams have done). But that 6-in-1000 chance is also what makes young parents so preoccupied about infant deaths: we can picture ourselves as the parents of those 6 in 1000 kids. A stronger grasp of statistics can help us separate risks into significant, small, and near-imaginary categories. It might help us stop needlessly checking to see if our kids are breathing at night, or racing through showers because we're convinced they're gravely wounded on the other side of the door.

US infant mortality rate represented in 1000 dots. 994 black dots represent US infants who survive their first year. 6 red dots represent US infants who die before their first birthday.

US infant mortality rate represented in 1000 dots. 994 black dots represent US infants who survive their first year. 6 red dots represent US infants who die before their first birthday.

So let's look at that infant mortality rate a bit differently. 6 in every 1,000 infants will die in their first year. When represented in this way, we we have a tendency to imagine our babies as part of the 6 instead of part of the 1,0000. But what if we represented this number graphically? The black dots in this chart represent babies who survive their first year. The red dots represent US infants who die before their first birthday.

Let's look at the infant mortality rate one more way. 6 in every 1,000 infants will die in their first year. That means 994 of them--99.4%--will not. When you can't sleep in the middle of the night because you're worrying about whether or not your child is breathing, think of 99.4% and perhaps you'll rest a little bit easier. Or if you're past the first 28 days, think 99.8%, which is the postneonatal survival rate

My point here is not to be cavalier in my child care, but to recognize that the vast majority of babies in this country survive their first year. Of course, it's important for various health organizations to research ways to reduce this number further. Of course, it's important for parents to be vigilant about real dangers. But our collective panic about things that might kill the baby are almost certainly overblown, in part because so many of us have poor mathematical literacy. So this week, each time you feel the urge to check to see if the baby is still alive, reach for a book about math instead. Or, perhaps more realistically, every time you check on the baby, go read a chapter of MacNeal's Mathsemantics or John Allen Paulos' Innumeracy to help put that check in mathematical perspective. 

This article originally appeared on