Current data makes it clear: child safety seats and booster seats save lives, prevent injury

First things first, and let there be no mistake about it: in a motor vehicle crash, a seat belt alone does not deliver the same level of protection as a child safety seat or booster seat.

In a recently released, peer-reviewed article for the journal Pediatrics, authors Kristy Arbogast and Dennis Durbin conclude that:

"Children aged 4 to 8 and using Belt Positioning Booster seats were 45% less likely to sustain injuries than similarly aged children who were using the vehicle seat belt."

I bring this up today because of a disturbing segment on Tuesday's Good Morning America. The audience for this piece heard authors Steven Levitt and Stephen Dubner--who were promoting their latest book--suggest that child safety seats and booster seats are no more effective than adult seat belts at protecting children during motor vehicle crashes.

Now, if you want to slice up the data to be provocative, have at it. As a grandfather and as Secretary of an agency whose number one mission is safety, I don't have that luxury.

I prefer to help parents make the kind of choices that can keep their kids not just alive, but safe.

It won't raise any eyebrows, but if advocating child safety seats and booster seats can reduce the number of deaths and injuries among children from motor vehicle crashes, I'm okay with that.

At the websites of Children's Hospital of Philadelphia, the Insurance Institute for Highway Safety, and DOT's National Highway Traffic & Safety Administration, you'll find plenty of analysis supporting the common-sense decision to use child safety seats and booster seats.

But, here are a few quick points to make that decision even easier:

• Those arguing against the effectiveness of child restraint systems base their work on only one national set of crash outcome data that is already biased toward less effectiveness;
• Arbogast and Durbin's analysis, based on 2 sets of data and limited to crashes where children were restrained with more recently developed seats, indicates that children in car seats have 28% less risk of dying in a motor vehicle crash than children using adult seat belts;
• Even after accounting for serious car seat misuse, children in car seats have 21% less risk of dying; and
• Every car seat on the market must meet federal safety standards. Child safety seats can be found in a wide range of prices; many effective models are affordable, and many programs distribute them for free or at reduced cost.

While I admire  Mr. Levitt and Mr. Dubner's call for better child restraint systems, there are better ways to urge progress on that front. And I ask Mr. Levitt and Mr. Dubner this question: were it not a state law, would you really choose not to buckle your own children into age-appropriate restraints?

I'm pretty sure I know the answer.

So please, parents, choose wisely; buckle those kids securely into age-appropriate restraints.

Comments

[Michael E. Bailey]

Child safety seats should be required for children under 8 years old. and the parents of newborns should get special car safety seats given to them by the hospital when the mother checks out. It would be great too if fire departments held child safety seat programs once a year or so to distribute free seats to poor families that need them just likhe many fire departments do now with smoke detectors. Best wishes, Michael E. Bailey.

[Robin]

With all due respect Mr. Secretary, I urge you to read Steven Levitt's response and answer his questions. I fully agree with you that this is a serious issue, and as such doesn't it require serious analysis?

http://freakonomics.blogs.nytimes.com/2009/10/27/what-the-secretary-of-transportation-has-to-say-about-my-car-seat-research/

[Charles]

I'm deeply disappointed in your brush off of Levitt's research. You should note his response and reach out. What is the risk in actually meeting with him? It's pretty clear you didn't read the argument set forth in the book or the papers. Insecurity is a terrible thing in a person who's in a power position.

[Kyle Maxwell]

Why not have someone in your agency look at the new analysis and understand what led to a different conclusion? "One test result is worth one thousand expert opinions." Don't misunderstand, I'm keeping my kids in safety seats, because I won't throw out accepted wisdom based on one study. But that doesn't mean the new study shouldn't get some attention.

[Paul Heaton]

The Secretary's statement that "Those arguing against the effectiveness of child restraint systems base their work on only one national set of crash outcome data that is already biased toward less effectiveness" is incorrect. The research cited in Mr. Levitt's book uses data from the Fatality Analysis Reporting System (FARS), data from individual state crash data systems (such as Wisconsin's CODES data), and the General Estimates System (GES), which is designed to provide a nationally-representative portrait of fatal and non-fatal crashes in the U.S. As a researcher and a parent of two young children I was highly skeptical of Mr. Levitt's claims regarding the effectiveness of safety seats for older children when I first saw one of his working papers several years ago. I set out to debunk his findings using data on millions of accidents in New Jersey that I had obtained for another project. But when I conducted the analysis for myself, I ended up confirming rather than refuting his results.

[dullgeek]

If it weren't a law, and the literature showed that there was no difference in outcomes, then of course I'd buckle my kids into an adult seatbelt. Why wouldn't I?

The only question remaining is whether or not to trust Dubner & Levitt's research or the research cited above.

[twitter.com/jeanniez]

I disagree that children up to 8 need to be in booster seats. It would be better if car manufacturers were required to put seat belts in vehicles that can optimally restrain older children (5-8 year olds) without extra booster seats. This would be safer for smaller adults too.

I always followed seat belt laws when my children were younger, but once they reached school age very often they are driven by other people and I drive other children and it is difficult to always have car seats for everyone even though the law requires it. Levitt's research says to me that we are not being reckless by simply buckling in our 7 year olds. This question IS worthy of your time and attention.

[Ken]

Levitt has responded to your blog post on his blog (found at http://freakonomics.blogs.nytimes.com/). He makes some very good points and I, as a tax-payer and concerned citizen, would be interested to see your response. "As Secretary of an agency whose number one mission is safety," it seems that it would be the responsibility of your agency to take advantage of all available data and analysis within your subject area. Additionally, you have a nationally recognized economist offering his direct help to your agency. As a tax-payer, I know I would be disappointed if the people who use my money to run the country ignore this opportunity. Sincerely- Ken

[Steven Maese]

Mr. LaHood,

You're title is misleading; current data, analyzed properly, is inconclusive regarding child-seat efficacy. Instead of dismissing Steven Levitt and Stephen Dubner's research, why don't you address it thoroughly and intellectually?

[Sam]

I can't imagine that both you and the SuperFreakonomics folk are correct (data doesn't work that way). I think in the spirit of sound research both you and they should accept the possibility that the other is more accurate and then either do new research or analyze previous research together (or using an impartial third party).

I have read both this post and the Freakonomics blog's reply, and it seems very clear to me that there needs to be more discourse on this topic.

[Chris Auld]

It isn't "Levitt and Dubner's research." Levitt (researcher) and Dubner (journalist) wrote a pop book, but the material in question in that book is is based on actual research by Levitt.

Levitt's paper was published in the highly-ranked peer-reviewed journal the Review of Economics and Statistics, which has much, much higher methodological standards than Pediatrics, a journal largely written by and for physicians. The fact that the authors of the Pediatrics article refuse to release their data makes their results irreproducable and seriously damages their credibility.

Levitt, S. (2008) "Evidence that Seat Belts Are as Effective as Child Safety Seats in Preventing Death for Children Aged Two and Up," Review of Economics and Statistics 90(1):158-63.

[Abhi Chaudhuri]

I find it disconcerting that you, as a policy-maker, approach this issue without reviewing the statistical analysis that led to Levitt's conclusion - and instead rely on how you "feel" about it.

To be perfectly safe, everyone should only buy the best in-car protection for their child even if it cost $10,000. But the government is not mandating that. Clearly, there isn't a "zero-risk-tolerance" policy, so why not evaluate how much risk is really being mitigated by car seats vs. better optimized seat belts?

I don't see anything wrong with spending $200 less if it doesnt increase the risk significantly.

I also find it curious that you are discounting the FARS data that the NHTSA compiles, and instead rely on a "recently released article" that does not provide any transparency to the data its based on.

[William Loeffler]

You give no account for the time and effort that goes into using such seats, especially for large families on very short trips. They also take up a large amount of space in a vehicle and make it all but useless for older riders in the back seat.

I can tell you for a fact that many short trips are foregone in large families because of the annoying hassle of trying to get four children strapped in and out of child safety seats.

You also make no reference to the environmental hazards caused by the much larger vehicles which are required to seat four or more children in restraints. More resources go into vehicle production, fuel and carbon waste. Of course, that probably doesn't fit under your job description so you ignore it.

I do use them as it is the law, but in answer to your question, yes, there are times when I would not use them if my state did not mandate it, on very short trips and when space in the vehicle is at a huge premium.

I am a Democrat but this is one of the examples where Democrats assume that safety (or assumed safety) comes cheap because they refuse to incorporate all the costs besides the pure monetary ones.

[Michael Kelly]

This is a very disappointing response, Mr. Secretary. Good policy is not created by dismissing contrary analysis, especially when the initial analysis is based upon a dataset that is unavailable to the public.

Perhaps you are satisfied just telling people what to do because it feels "right". We are not.

[John Guenther]

Mr. Secretary
How many more years will it take for NHTSA give us a side impact standard. Given the fact that side impacts are the most harmful ... why is there not one now?

[Brian]

I agree with all the other comments, and have nothing insightful to add except my voice of encouraging looking into this issue.
When school, church, and shopping are all just down the road in a 25 MPH zone, I'd love to not have to use the carseats for more carpooling ease. Can you tell me if my kids are really safer in carseats in these short drives? (For the 5 years and older kids).

[create a free blog]

The research cited in Mr. Levitt's book uses data from the Fatality Analysis Reporting System (FARS), data from individual state crash data systems (such as Wisconsin's CODES data), and the General Estimates System (GES), which is designed to provide a nationally-representative portrait of fatal and non-fatal crashes in the U.S. As a researcher and a parent of two young children I was highly skeptical of Mr. Levitt's claims regarding the effectiveness of safety seats for older children when I first saw one of his working papers several years ago.

[car dvd gps]

I always followed seat belt laws when my children were younger, but once they reached school age very often they are driven by other people and I drive other children and it is difficult to always have car seats for everyone even though the law requires it.