Infant Seats and Crash Testing: What Do We Know?


are nothing new.  This issue has come up a number of times in the 10 years I’ve been involved in child passenger safety.  Testing seats with different methods is not a new thing, either.  This new research by the NHTSA put child seats in actual vehicles (rather than a test sled), during a frontal crash test.  Keep in mind that the vehicle crash test itself isn’t new, it basically the same as those that have been done for about 30 years in the NCAP program.  The result?  Some infant seats failed in a major way.  Consumer Reports made a big blunder over a year ago with an infant seat test.  Their report had a number of flaws, but the largest one was a lack of oversight that led to a test that simulated a much, much higher speed than intended.  Ironically, part of their mistake was because they didn’t consider that putting a child seat in a real vehicle allows the vehicle to absorb some of the crash energy.  This must be considered when you simulate a real crash on a sled.  The result?  Some infant seats failed in a major way.

Anyway, there’s still a lot we don’t know about these tests and the results.  It’s both premature and a little unfair to speculate on why some infant seats failed, on whether or not the new NHTSA research test was fair or even on the actual risks posed to infants in these particular seats.

There are a couple things we do know for certain, though.  First, some infant seats didn’t detach from their bases.  That’s right.  In both tests, there were models that remained intact, despite the alleged severity of the test.  Second, at least one company, Combi, has already done something about it.

Those two things make me a little suspect of the JPMA response that Heather mentioned in her blog.  In reading it, I have to think they’ve overlooked the fact that at least one manufacturer believes it is a real concern, despite the cost and weight issues that might be a consequence.  The JPMA has also overlooked that some manufacturers have designs that don’t separate in these tests.  Given the number of failures reported, I think they’ve overlooked that there may be a real issue involving the safety of infant seats in actual vehicles, perhaps even at speeds lower than NCAP speeds.  Speed is not the only issue involved here.

Like many parents, I have to wonder why there is such a furor over testing in real vehicles and/or at higher speeds, in the same exact test that vehicles have undergone for decades to get a “star” safety rating.  Sure, maybe it’s not a reasonable test for an updated minimum pass/fail standard on child seats.  On the other hand, given the results from the research, I think it appears to be a very reasonable basis to show parents which models may be safer than others in a supplemental test, like NCAP does for vehicles.

This is the key difference.  NCAP testing is not used to determine if a car or SUV is safe enough to be on the market.  There are other federal standards for that.  NCAP is a supplemental test that shows consumers that differences in safety may exist among vehicles and helps them decide what to buy.  So, sure, like all the vehicles being sold today, maybe all the infant seats being sold are safe enough to protect babies in most frontal crashes.  In that, the JPMA and all these other agencies are probably right.  Even so, is it such a bad thing to allow parents to know which models might be safer than the minimum requirements allow?

I sure don’t think it’s a bad thing.  I think it’s a great thing.  I think the JPMA is very disingenuous when they said, “As noted by these and other experts, there is no evidence that infant child restraints would protect children better in real world crashes if they were designed to meet a crash test conducted at a higher speed.”  Well, that’s because we don’t have a system that tests child restraints to a higher standard and real life data to determine if the test helped!  It’s unscientific to assume the outcome of such a test when you haven’t even done the test yet.  Duh.  The same chicken-and-egg reasoning could have been used by the NHTSA and IIHS against the implementation of their crash testing programs, yet it turns out these programs have correlated well with real world reduction in fatal injury.  In fact, this very NHTSA research test clearly shows that some important differences exist among infant seats in a scenario consistent with the NCAP program.

So, I think the JPMA is very short-sighted with their response.  We’re not just talking about infant seats rotating a little more than the standard allows or exceeding dummy criteria by some percentage.  No, we’re talking about the possibility of carriers detaching completely from the base and being violently thrown through the passenger compartment.  While the JPMA and others seem concerned about what this might do to the cost and weight of infant seats, I’m concerned about what this might do to the baby and the other passengers in the vehicle.

There also seems to be an implication that it would be a hardship for manufacturers (and thus consumers) to comply with the requirements to pass a test like this.  As it turns out, there is no compliance requiring vehicles to do well in NCAP tests.  If you are willing to do just the minimum required to get a vehicle on the market, then you may be content with a 1-star, 2-star or 3-star rating.  This could be the same for child restraints if we had a similar program for supplemental testing.  I’m pretty sure it wouldn’t be too much of an expense or hardship to get 4-star or 5-star ratings, as the JPMA might lead us to believe.  After all, some seats didn’t detach in these tests and at least one manufacturer is already doing something in response.

Just because, “No product is more effective at reducing fatalities and injuries to our precious children,” as the JPMA states, that doesn’t mean we shouldn’t be given the means to decide which of those products are safer than others.  The JPMA goes on to say that, “It is irresponsible to suggest that infant child restraints may not perform well in crashes.”  I happen to think it is irresponsible to deny the fact that some models clearly perform better than others in crashes and that one manufacturer has already shown that it is possible to improve designs in order to prevent carriers from flying off bases in situations other than a sled test with a very narrow set of parameters.

Maybe we don’t need a new standard for child seats, at least maybe not any major changes beyond those already being implemented.  On the other hand, these tests have shown us that there clearly are differences among child seats.  It sure doesn’t seem like a bad idea to know which ones perform better than others.  That would also encourage improvement, as it has in vehicle crash-worthiness for many years.  Again, not a bad thing in my opinion, when the life of your child may depend on it.

That’s my soap box for today, and I’m sticking to it.


As much as I disliked the response from the JPMA, I liked the response from Safe Kids USA.  Instead of focusing on what shouldn’t be done, they focused on what should be done to keep your kids safe.  They also agreed that, “More testing for car seats and vehicles can only continue to advance the child passenger safety field and improve the level of protection we can offer children when they ride. The more we know about car seats and how they react in crashes, the better equipped we will be to push for new technology and improvements that will keep children safer.”  As for keeping your kids safe, like I’ve said over and over, there is one thing we know for sure.  Driving undistracted/unimpaired and keeping your child properly restrained in a rear seating position with an appropriate car seat will offer them the best possible chance in a crash.  Do that much, and the choice of car seat is a relatively smaller issue in terms of risk.


  1. Wineaux March 8, 2009
  2. CPSDarren March 7, 2009
  3. aunt83me02 March 6, 2009
  4. Steve March 6, 2009
  5. Kecia March 6, 2009
  6. CPSDarren March 6, 2009
  7. Steve March 6, 2009
  8. Bookmama March 5, 2009
  9. LISmama810 March 5, 2009
  10. murphydog77 March 5, 2009