US child death rates fell 89% since 1935 but have been higher than peer nations' since 1975, study finds

Americans under 20 die at a far lower rate than they did in 1935. The death rate for that age group has fallen by 89%. Yet children in other wealthy countries have gained even more. A study tracing 89 years of records finds that the United States once had the lower death rate, lost that lead in 1975, and has trailed its peers ever since.

The study, published on October 5 in Pediatrics, counted deaths from all causes together. That makes it a reliable guide to when the gap opened and which ages it affects. It cannot say by itself what lies behind the gap. For that, the picture has to be pieced together from earlier research.

Why the timing of the US child mortality gap matters

That American children die at higher rates than those in comparable countries was already established. The new paper takes it as the starting point: children in the United States experience higher all-cause mortality than high-income peer nations. All-cause mortality means deaths from every cause counted together.

The same research group, led from Children’s Hospital of Philadelphia, has documented this twice before. A 2018 paper compared the US with 19 other countries from 1961 to 2010. It framed the puzzle in terms of money: the United States has poorer child health outcomes than other wealthy nations despite greater per capita spending on health care for children. Its conclusion was that mortality in the US has been higher than in peer nations since the 1980s.

A second paper, in 2025, looked at 2007 to 2023 and went well beyond deaths. It concluded that the health of US children has worsened across a wide range of health indicator domains over the past 17 years. Neal Halfon of the University of California, Los Angeles, was the senior author of that paper and is the last-named author of the new one. Halfon commented on the 2025 results when they appeared. “This study confirms what many pediatricians, educators, and parents have been sensing for years; that our children are facing a growing health crisis,” Halfon said.

What neither paper could show was the long view. In the new paper’s words, the timing, durability, and age and sex contributions to this gap remain unclear. When a gap began says something about its causes. A disadvantage that had always existed would point to something fixed. One that appeared at a particular moment points to something that changed.

How US child mortality was compared with peer nations

Christopher Forrest is the first author. The team used vital statistics, meaning the official records of births and deaths that governments keep. They took them from the Human Mortality Database, a research archive that describes itself as providing detailed high-quality harmonized mortality and population estimates. Harmonized means processed in the same way for every country so that the figures can be set side by side.

The comparison covered everyone from birth to age 19, for each year from 1935 to 2023. On one side was the United States. On the other were 13 high-income countries, treated as a single group. All belong to the Organization for Economic Cooperation and Development (OECD), a club of mostly wealthy democracies. The paper calls the group OECD13.

For each year the researchers calculated the death rate per 100,000 children and adolescents. They compared the two rates directly, measured how fast each was falling, and estimated excess deaths. Excess deaths are the number of US deaths over and above what would have occurred at the peer nations’ rate.

Because the records cover whole national populations and not a sample, there is little statistical uncertainty in the basic rates.

What the study found about child mortality since 1935

Both lines fall steeply. In the United States, deaths before age 20 dropped from 477.6 per 100,000 in 1935 to 52.4 per 100,000 in 2023, a decline of 89.0%. In the 13 peer nations they dropped from 785.1 to 26.9 per 100,000, a decline of 96.6%.

The starting points are worth a second look. In 1935 the American rate was about 40% lower than the peers’. The United States was then the safer place to be a child.

The lines met four decades later. US mortality crossed over OECD13 in 1975 and remained higher thereafter.

By 2023 the American rate was 1.95 times the peer rate, a difference of 25.5 deaths per 100,000. Put another way, about 1 in 1,900 Americans under 20 died that year, against about 1 in 3,700 in the peer nations.

The gap has kept widening in recent years. From 2006 to 2023 the US rate fell by 1.2% a year, with a range from 0.6% to 1.7%. The peer rate fell by 2.3% a year, with a range from 2.0% to 2.7%. Both were still improving, but the peers improved about twice as fast.

The authors’ summary is that the US disadvantage dates to the mid-1970s and has persisted for 5 decades.

That date is a little earlier than the one in the group’s 2018 paper, which put the divergence in the 1980s. The two studies used different sets of comparison countries, 19 then and 13 now, and the earlier one began only in 1961.

Which children carry the US mortality gap

The disadvantage is not spread evenly across childhood. By 2023, the US disadvantage was largest among those aged 15 to 19 years, whose death rate was 2.9 times that of their peers abroad.

That age group also accounted for a large share of the excess deaths. In 2023, 40.4% of them were among 15- to 19-year-olds. The authors say the gap is now concentrated among older male adolescents.

The 2018 paper had found two pressure points. Its authors named infants and children ages 15-19 as the two age groups with the greatest disparities. For 2001 to 2010 it put the risk of death at 76% greater for American infants and 57% greater for those aged 1 to 19. Over its 50 years of data, the lagging US performance amounted to over 600,000 excess deaths.

The 2025 paper found infants and older children about equally affected in more recent years. In its mortality comparison, which ran from 2007 to 2022, infants were 1.78 times as likely, and those aged 1 to 19 were 1.80 times as likely, to die in the US as in a group of 18 comparison countries.

What earlier research says about causes of US child mortality

The new study did not separate deaths by cause. The earlier ones did.

For babies, the 2025 paper identified two causes with the largest gap: prematurity, at 2.22 times the comparison rate, and sudden unexpected infant death, at 2.39 times. A study by economists reached a related conclusion by a different route. They began from the observation that the United States has higher infant mortality than peer countries, and checked whether that was an artifact of how very early births are recorded. It was not. The gap arose mainly after the first month of life, and that later disadvantage, they wrote, is driven by poor birth outcomes among lower socioeconomic status individuals. Socioeconomic status means a person’s standing by income, education and occupation.

For older children and teenagers, two causes stood out in the 2025 paper: firearms and motor vehicle crashes. Deaths from motor vehicle crashes were 2.45 times as common in the US. Deaths involving firearms were 15.34 times as common. The 2018 paper had reported something starker for a narrower group, stating that children ages 15-19 were eighty-two times more likely to die from gun homicide in the US.

Independent studies fill in the picture from inside the country. One report gave the scale, detailing the 10 leading causes for the 20,360 deaths of children and adolescents in the United States in 2016. A later analysis of federal injury data found a turning point: in 2019, firearm injuries surpassed motor vehicle collisions to become the leading cause of death for youth aged 0 to 19 years in the United States. Homicide was the most common kind of firearm death, and the authors noted that suicide represents a large proportion of firearm deaths in 10- to 19-year-old youth.

A comparison of the US, Brazil, Mexico and Colombia found the American rate still climbing. Among Americans aged 1 to 19, firearm deaths rose from 3.57 per 100,000 in 2015 to 5.92 per 100,000 in 2022. Over those years, youth firearm mortality rates increased in the US and Mexico and decreased in Colombia and Brazil. The authors did not claim to know why, though they suggested that increased firearm availability may be a factor.

These causes fit the age pattern in the new study, with its concentration among older teenage boys. They do not explain 1975. The new study’s contribution is to show that the disadvantage is half a century old, older than the rise in firearm deaths of the past decade.

What the child mortality comparison cannot show

Causes. The study counted all deaths together. It shows when and at what ages the US fell behind. The reasons come from other studies, covering shorter periods.

Thirteen countries as one. The peer nations were pooled into a single rate. Individual countries within that group will differ, and the comparison depends on which countries were chosen. The group’s earlier papers used 19 and 18.

National averages. A single US rate conceals large differences within the country. The analysis of firearm deaths, for example, found that in 2019 Black youth had a firearm mortality rate 4.3 times higher than that of White youth.

How deaths are recorded. Countries differ somewhat in how they register births and deaths, especially for babies born at the edge of viability. The economists’ study found that the US infant disadvantage persists after adjusting for potential differential reporting of births near the threshold of viability, which suggests recording differences are not the main explanation.

Description, not explanation. A time series shows what happened. It does not test why other countries improved faster.

Rare events. Even at the US rate, about 1 in 1,900 people under 20 died in 2023. The study describes populations and carries no information about any one child.

Whether a health concern in a particular child needs attention is assessed by their doctor.

The United States cut its child death rate by nearly nine tenths in 89 years, and the study’s point is that its peers, starting from further behind, cut theirs by more.