According to the CDC leading causes of death reports from the WISQARS database, the total fatalities by age in the “MV Traffic” category from 2001 to 2010 (the most recent decade available) are as follows:
Age 0-12 months: 1,208
Age 1: 1,147
Age 2: 1,201
Age 3: 1,119
Age 4: 1,085
Age 5: 1,075
Age 6: 1,040
Age 7: 982
Age 8: 991
Age 9: 1022
Age 10: 997
There is no information provided about the whether a child restraint was present or not. If a restraint was present, there is no information about misuse. We don’t know if alcohol or distracted driving contributed to the fatality. We really know nothing else about this data, other than the total number of children killed at each age.
Even given the lack of specifics on this raw data, do these numbers surprise you? Would you have expected toddlers age 1, 2 or 3 to suffer fatal motor vehicle related injuries more frequently than babies under 1 year old, those presumably more likely to have been rear-facing in 2010 and earlier? Would you have expected booster age children age 4 and up to have more fatalities than younger kids that are more likely to be in a 5-point harness? Or, is a pretty even distribution just what you would have expected?
The Freakonomics piece was countered by researchers at CHOP in a LTE to the New York Times. The gist of their criticism is that Levitt and Dubner considered only fatal crashes (FARS database) rather than all crashes. The letter can still be found online at http://www.nytimes.com/2005/07/24/magazine/24LETTERS.html?_r=0 (scroll down to the second-to-last letter).
I would rather see the numbers broken down by state. I live in a state with a relatively low population so when there’s a fatality, especially a child fatality, it makes big news and we’re actually able to do education with it. But in higher population states like CA or NY, I can see a fatality being just another number. I’m also curious what happens at age 9 to bump the numbers up. It’s clear that around age 7, kids start fitting into seatbelts better so the fatality numbers go down, but why do they go back up at 9? I’d like to see the data past age 10 to see if the numbers trend lower or if the 10 yr olds are an anomaly.
Yes, for 2010 or 2011, I think motor vehicle crashes had fallen to #2 behind drowning for some age ranges.
I think it’s odd… since my mind has it broken down from the last CDC chart I looked at where it was ‘only’ the number 2 cause of death for kids over 3. Seeing it broken out in another way is interesting. And worse than I remembered it. :/.
those numbers are just staggering…. and sadly i am not surprised. i would say a good 75% of infant seats i see are improperly installed. people just don’t read the manual and i would bet misuse plays a large part. oh and the not yet 1yr olds who are turned FF early (which is common here!!) because they are such a “big boy/girl”…..
fascinating topic….
AK CPS Dad, I agree. I don’t really know that there is any conclusion that can be drawn here, but it’s an interesting topic for discussion. It may show that there isn’t a sudden spike of fatalities for 1-year olds, once they are turned front-facing? Hard to say for sure, obviously. Freakonomics is a whole other issue. I suspect their study had some merit, as it used the same database of fatalities as everyone else uses. I don’t think their claim was ever really countered directly, but I may be mistaken. Other studies used other data and controlled for variables differently to reach different conclusions that contradicted the Freakonomics results.
Hi Brigala,
You can look at the results for individual years, too. For example, the default at the link above is 2010 only. The totals are staggering, especially when compared to other causes. As you will see in my next blog:-(
I’d be interested in seeing a breakdown by year. Just out of curiosity.
Running these numbers through my calculator, that’s an average of 1,187 fatalities per year in the 10 & under age group. That’s a lot of kids. 🙁
I feel like as much as I’d like to come to some conclusion, I do think there are too many variables involved without a careful dissection of the data to come to a logical conclusion. That’s why I can’t believe those Freakanomics guys came to such a poor conclusion a while back saying that CRs made no difference in crash outcomes – seems illogical to make any conclusion about the effectiveness of the CR when we’ve got misuse rates in the 75%+ range.
I think the best we can do is look at the empirical data from crash tests and use that to promote best practice. I do wish there were more open and free sharing of information about the lab results, but that’s a different subject (for instance, since Diono does NCAP testing at 35mph, how can I compare their seats to most other manufacturers that don’t?)
Pretty even…what I would expect with maybe a slight increase in infants because of gross misuse with unknowing new parents.