by Jason S. Brinkley, PhD, MA, MS
On the Brink addresses topics related to data, analytics, and visualizations on personal health and public health research. This column explores current practices in the health arena and how both the data and mathematical sciences have an impact. (The opinions and views represented here are the author’s own and do not reflect any group for which the author has an association.)
This month marks the two-year anniversary of On the Brink, and I want to mark the occasion with a return to roots, of sorts. My first blog “What Kills Us” is still one of my favorites. There we saw that the two highest sources of US mortality were related to cancer and cardiovascular disease, contributing to almost half of all yearly American deaths. Here we will revisit the concept of US mortality but focus on a different (and most vulnerable) cohort. Today we ask, What kills kids?
Child deaths are overall quite rare, and their measurement has to be distinguished between infants (less than 1 year old) and those aged 1-18. I spent some time pulling mortality data from the CDC WONDER Database. WONDER is a comprehensive resource that tracks a number of public health metrics; among its many features is a comprehensive listing of mortality across the US from 1999-2016. For my first blog I used multiple sources and discussed everything in terms of death rates, but here I want to look at this single source and use it to follow a cohort of children. One interesting feature of WONDER is that it will show anyone the top 15 causes of death for any subgroup of Americans. So let’s follow children born in 1999 across the first 17 years of life and determine what kills these kids over that time frame.
By the Numbers
Before diving into specifics, let’s give some big-picture numbers and discuss some limitations and assumptions. In 1999, the United States had approximately 3.8 million children aged 0 to 12 months. The overwhelming majority of these children were live births with a handful being immigrant children brought to the US before their first birthday. We’ll end 2016 with about 4.2 million seventeen-year-olds; tallying up all the deaths across the 17 years of WONDER data, we find 43,092 deaths. And while we need to attribute the difference between the 4.2 and 3.8 million figures to immigration, there is no way to fully know how many of the 43,092 death were US born and non-US born. But in order to err on the side of conservative estimates, I choose to assume that all the deaths would have come from the 3.8 million originally born in 1999. Doing this means that the first thing we can know is that at most 1.1% of the 3.8 million children born in 1999 (let’s call them 99ers) died before reaching adulthood.
I don’t care what it is as long as it is healthy. That was the sentiment from my father when I told him the ultrasound revealed that my first-born child was a boy. This is the most common fear of new parents, and their fears are reasonable; the first year of life is the one that sees the highest number of deaths. From our 3.8 million newborn 99ers, 27,937 of them died before their first birthday. That is 65% of all the deaths of the aforementioned 43 thousand deaths of kids between 0-17 born in 1999. Many of these babies died due to birth-associated issues: birth defects (5,473 deaths), early term (4,392), maternal complications (1,399), respiratory distress (1,110), and so on. At least 10 of the top 15 causes of death were related to birthing, and the data seem to show that it is getting worse. A recent Time article suggests that the United States continues to fall behind other developed nations on early childhood mortality.
But keep in mind, that this too is a numbers game. 3.8 million live-births and 28 thousand deaths show that over 99% of newborns made it to age 1.
In spite of this victory, we unfortunately lost 2,648 children to SIDS, a disorder that is still not well understood. Beyond that, 845 babies died in accidents and another 331 were victim of homicides; these combined to yield more deaths than average-per-year deaths of the 99ers from age 3 to age 15 as we’ll see below.
Big Picture Ages 1 to 17
The risks shifted greatly after children reached age 1, so much so that the CDC catalogs infant mortality with an almost completely different set of codes. But the ~35% of remaining childhood deaths didn’t occur evenly across the years; in fact, total deaths bottomed out in preteen years only to rise again among teenagers. The visual below shows the total number of deaths among our 99ers across all causes; we see that we lost 1,946 one-year-olds in the year 2000 and 2,055 seventeen-year-olds in 2016. But these years were only a fraction of the number of deaths we saw getting to age one.
Ages 1 to 5
The top five causes of death among 99ers were the same every year from age 1 to 5 in both cause and rank. Accidents, birth defects, homicides, cancer, and diseases of the heart claimed 1,165 one-year-old 99ers in 2000 (60% of all deaths) to 469 deaths in five-year-old children in 2004 (71% of all deaths). Accidents were far and away the leader here, but the numbers go downhill each year as children approached elementary school. They reached the lowest numbers of all-cause mortality.
Ages 6 to 11
2005 to 2009 seemed to be the “safest” period for our 99er cohort. The likelihood of dying from genetic and other diseases was lower, accident rates were among the lowest, and the children were the least likely to be victim of a homicide. While many of the main early causes seemed to “tail” out, this era opened the door to the emerging threats of the teenage years while maintaining a few constants. For example, deaths from influenza and pneumonia stayed steady as the seventh to ninth ranked cause of death but were responsible for fewer than 15 deaths per year across the United States. Likewise, meningitis killed roughly ten 99ers per year and was ranked off and on in the bottom part of the top 15 causes of death up until 2009 when the 99ers were aged 11. After age 11, meningitis fell off the list of top 15 causes of death and didn’t reappear in any of the 2010-2016 rankings at all. Note that the US guidelines for meningitis vaccine is to take the first dose at 11 years old.
In 2008, suicide first entered our list of top 15 causes of death at number 12 with a tragic 4 suicides among 9-year-old 99ers. Suicide went on to become the second leading cause of death in 17-year-old 99ers with over 400 deaths in 2016.
Ages 12 to 15
Cancer kills. Cancer kills children as well as it kills adults and was the second highest cause of mortality in young children starting at age 3. Cancer killed roughly one hundred 99ers per year and would continue to reign as the number 2 risk factor among teens except for the rise of the teen risk factors. Specifically, cancer got knocked off its place at number 2 with increases in suicides, homicides, and a dramatic increase in accidental deaths which combined to lead to 633 deaths in 2014 (62% of all deaths).
Ages 16 and 17
The 2015 and 2016 years saw the final transition of causes of mortality, with accidents, suicides, and homicides claiming almost 2,600 of the roughly 3,500 deaths (74%) while the numbers of other “adult” causes of death (namely cardiovascular disease and cancer) began to trend upward as well. And while the numbers continued to be tragic, there were many silver linings. 98% of children born in 1999 lived to age 17, and government data from other sources suggest that about 80% of children graduated high school and 70% of those that graduated started some form of post high school education.
Putting It All Together
So what kills kids? Early on, birth complications and defects tend to be the biggest culprit. Those issues tend to work themselves out as children approach elementary school but later on give rise to other culprits in later years stemming from suicide and homicides. Accidents are an ever-present threat, but it’s hard to break those down into accidents due to negligence versus those that are unforeseeable.
You Have Died of Dysentery is written on my son’s favorite t-shirt with a scene from the classic video game Oregon Trail. His grandfather got his wish and the child was born happy and healthy. His shirt is supposed to evoke some humor but is also reminiscent of a time when child deaths were much more common. The current data are pretty clear that while we were over 98% effective at getting children born in 1999 to adulthood, there is room for improvement. As a statistician I often hear the question what is the chance that’ll happen? This 1.1% is just for one birth class across one set of years. Online sources suggest that the average Facebook user has over 300 friends. Imagine that each of those people has 10 children in his/her life (children, grandchildren, nieces, nephews, cousins, neighbors, and so on). The numbers suggest that at least 33 of those children may not make it to adulthood. So, as our social network grows and we become ever more connected, we have to keep in mind that rare is not impossible.
I’ve compiled the data I used here from WONDER and have made it available as a spreadsheet which you can download here.
Jason S. Brinkley, PhD, MS, MA is a Senior Researcher and Biostatistician at Abt Associates Inc. where he works on a wide variety of data for health services, policy, and disparities research. He maintains a research affiliation with the North Carolina Agromedicine Institute and serves on the executive committee for the NC Chapter of the American Statistical Association and the Southeast SAS Users Group. Follow him on Twitter. [Full Bio]
Previous posts in this series:
- The Golden Age of Health Research Funding
- Does Living on a Prayer Work?
- The Opioid Data Crisis
- Income Lost from Snow Days*
- What the #$@&*! Is Blockchain?
- Opportunistic Research Opportunities
- Text Mining UFO Data: Little Green Aliens or Santa’s Elves?
- Should You Know Your Doctor’s Home Address?
- The Population Bullet
- The Unknown Unknowns of Missing Data
- Communicating Science–More Than Just Good Words?
- Counting Alabamas
- The Third World in Your Own Backyard
- The Unrealistic Gold Standard
- Does MACRA Signal the Beginning of the End for Medicare Claims Data?
- Think You Aren’t Extraordinary? Odds Are You’re Wrong
- Mapping by Words
- Are We Asking Too Much From Surveys?
- Making Better Comparisons
- What Kills Us?