Home » Status Report » 2000 » Article
Status Report, Vol. 35, No. 4 | April 15, 2000 Subscribe

Child restraints take their punches in repeated high-speed crash tests

It's intuitively clear that a child restraint visibly damaged in a serious car crash should be replaced. But the need for replacement in less obvious cases has long been debated. A California law requires insurers to cover the cost of replacing a restraint used by a child in any crash, regardless of crash severity or whether the damage is visible. Illinois is considering a similar law. But recent Institute crash tests indicate there's no objective reason to toss out all crash-involved restraints indiscriminately. Most child restraints are just as safe after a typical fender-bender as they are out of the box.

"There's no reason to assume child restraints become less effective just because they've been in a crash, especially a minor one," says Institute president Brian O'Neill. "On the contrary, child restraints are remarkably durable. Even when we have subjected them to successive crash tests at high speeds, most of the restraints kept their structural integrity despite minor damage."

The Institute conducted the vehicle crash tests with child restraints to determine the extent of the damage, if any, in high-speed impacts. Twelve different child restraint models, both toddler seats and rear-facing infant restraints, were tested in car-to-car frontal offset crashes, both vehicles traveling 30 mph. A dummy representing an infant or child was in each restraint. Some of the restraints came through one round of tests without damage, though most sustained minor damage. Four of the damaged restraints were then tested again in crashes at the same speed.

In the initial crashes, most of the child restraint damage was readily apparent from a visual inspection, but the damage was structurally insignificant — minor plastic deformation, slight fraying of the harness webbing, bending of harness buckle latch plates, and some minor cracking. These results add to existing evidence that blanket replacement laws overreach their goal of improving safety.


Damage to child restraints after 30 mph car-to-car crash tests:

Century Assura booster

Century Assura: Plastic deformation occurred after the first crash test and worsened after the second.

Century Room-to-Grow booster

Century Room-to-Grow: After one crash test small cracks were observed at the base of structural stiffeners.

Kolcraft Secura booster

Kolcraft Secura: Structural stiffeners were observed pushing through restraint walls after one crash test.

Cosco Booster

Cosco booster: After one crash test, there was damage from the restraint harness straps. After a second test, additional deformation of the plastic was observed in places not previously damaged.


Much of the damage "wasn't different from what's typically seen with normal wear and tear," O'Neill notes. "This suggests that most effects of crash involvement aren't substantially different from the effects of extended use over a period of years." The four damaged restraints subjected to a second round of crash tests showed further damage similar to what was caused in the first round. But none of the twice-tested restraints failed catastrophically, and all succeeded in restraining the test dummies through the second impact.

Last year, the Insurance Corporation of British Columbia (ICBC) conducted similar investigations of restraint durability. Testing various child restraint types in offset barrier crashes, ICBC found most of them unlikely to be damaged in impacts at 30 mph or even subsequent crashes at 40 mph.

Even more dramatic are results of ICBC's low-speed testing. After 50 consecutive sled tests at about 9 mph, a selection of restraints representing major manufacturers and all seat types showed no visible damage at all. X-rays revealed no hidden damage lurking beyond the scope of visual inspection. And after 50 crashes, 3 of the restraints still passed a barrier test at 30 mph required by the Canadian government. These results are particularly impressive given that some of the child restraints were up to 5 years old.

Low-speed crashes like those conducted by ICBC occur far more often than the high-speed crashes simulated in the Institute's testing. So while minor damage did occur at high speeds, no damage at all occurred in the repeated tests representing the much more common kind of crash.

One argument used to justify replacing restraints after even minor crashes is that a greater rate of replacement would help get old or recalled child restraints off the road. Old seats may be more prone to crack due to exposure to extreme temperature changes or sunlight or simply due to wear and tear. But as the ICBC tests show, restraints continue to perform well even after significant years of use.

The bottom line is that child restraints shouldn't have to be replaced after most crashes. A restraint always should be inspected carefully after a crash and replaced if there's visible damage or, to be cautious, even in the absence of visible damage if the crash was severe.

Biggest threat to children in cars still is riding with no restraint at all

How child restraints perform doesn't matter if children aren't properly buckled up to begin with. Nonuse of restraints still is the biggest threat.

The National Center for Injury Prevention and Control reports that restraint use, seating position, and death rates among children ages 4-8 changed little during 1994-98. Fewer than 40 percent of the children this age who died from crashes were restrained. More than half who survived fatal crashes were buckled up.

According to earlier studies, restraint use is about 85 percent among infants younger than 1 but only about 60 percent among toddlers 1-4. Among kids who have outgrown child seats, the problem becomes nonuse of adult belts and booster seats. And despite lots of public attention to the risks, too many children still ride in front seats where airbags might harm them (see "Car seat anxiety," Jan. 16, 1999). In a random survey by the Harvard School of Public Health, 1 of every 4 cars with children younger than 13 included a child in the front seat.

While crash tests by the Institute and Insurance Corporation of British Columbia should put to rest the idea that mandatory child seat replacement is a pressing need, the real need is clear — getting kids into restraints and out of the front seat.

©1996-2016, Insurance Institute for Highway Safety, Highway Loss Data Institute | www.iihs.org