March 2, 2008
VANDALIA, OHIO – One year after Consumer’s Union called for its removal in a controversial article and the National Highway Traffic Safety Administration defended its safety, the Discovery Infant Car Seat has been voluntarily recalled in advance of a possible defect investigation.
Evenflo announced in early February that it was recalling models 390, 391, 534, 552 – a total of 1 million car seats – based on “recent laboratory tests conducted by Evenflo and the National Highway Traffic Safety Administration, which show that this car seat could potentially become separated from its base in high impact side collisions similar to those in the tests.”
The recall is a curious postscript to a January 2007 uproar after Consumer’s Union released the February issue of its popular Consumer Reports magazine featuring a sensational article rating rear-facing infant car seats in front and side-impact sled tests. The Discovery performed particularly poorly, and the advocacy group urged it to be recalled. CU, however, soon found itself the target of attacks, when it was discovered that the side impact tests had been performed at 70 mph – twice the speed intended.
While questions about the safety of the Discovery were all but drowned out by the uproar over CU’s testing methods, the agency quietly took its own look. From February to October 2007, the Discovery failed when tested in NHTSA’s Side Impact New Car Assessment Program (SINCAP) in four separate 2008 vehicle models: the Chrysler Sebring, the Lexus RX350, the Nissan Versa and the Dodge Grand Caravan. In each test, the seat, which contained an instrumented 12-month-old CRABI dummy, separated from the base. In addition, the agency subjected 12 other infant safety seats in different vehicles to side-impact crash tests at the 38.5 mph speed of the NCAP tests. NHTSA spokesman Rae Tyson said the tests were part of an ongoing research project, which started several years ago, into a possible rulemaking for a side-impact standard for child safety seats.
“When we see failure modes like this, we are going to take steps to protect the consumer,” Tyson says. “A defect investigation is one of the options, if the manufacturer had not stepped forward with a recall.”
NHTSA informed Evenflo of the test results after NCAP testing concluded in October. Evenflo then performed its own tests, and decided to launch a voluntary recall, Tyson said.
In a letter to NHTSA, Evenflo’s Lindsay Harris, vice president of corporate engineering/quality, said that the recall was in response to “anticipated customers concerns” about “recent laboratory side impact crash tests conducted at energy levels well above the regulatory requirements for automobile side impact performance. “While the company believes that the condition observed in the laboratory high impact tests is not a ‘defect’ as this term is understood under federal safety law Evenflo recognizes that this matter will likely raise concerns for its customers.”
As part of the recent announcement, Evenflo said that it would provide current owners of these models with a free supplemental dual-hook fastener that was tested to ensure that the seat remains attached to the base in the event of such collisions.
Evenflo spokeswoman Karen Davis said that the child seat manufacturer was only aware that NHTSA would make the results public, and “even though Evenflo had received no reports of this actually happening in real world crashes, they decided to issue a voluntary safety recall.”
In the summer of 2004, the agency opened and then rapidly closed an investigation into nearly 2.6 million Evenflo Discovery models manufactured from 1999 to 2003. The Preliminary Evaluation was based on 56 crashes in which the Discovery separated from its base, resulting in 23 injuries and 8 fatalities – half of the fatalities occurred in side-impact crashes. In its Closing Resume, NHTSA failed to clearly explain its reasoning, simply noting that the Discovery passed the FMVSS 213 compliance tests and the crashes were high-speed and severe.
More recently, consumer complaints to NHTSA raise concerns about loose-fitting harnesses that allow the infant to slide in the seat or harnesses that fail to latch properly, handles that do not lock in place and break and sun canopies that do not operate properly. Several consumers complained about the stability of the Discovery – typically, the seat tipping over or coming off the base during the course of normal driving, for example:
“While driving one day, my daughter’s car seat actually just tipped over while making a left hand turn. The base was in the correct place and we were not speeding so that couldn’t have made a difference. Luckily my daughter was not injured and we did not get into an accident while I was holding her up while driving to the side of the road.”
“While driving home today and as consumer rounded a corner, the child safety seat came out of the base and turned over. Consumer’s child was in the seat.”
The CU tests showed that only two of the 12 seats performed well in tests and most failed: “The car seats twisted violently or flew off their bases, in one case hurling a test dummy 30 feet across the lab,” the article said.
“In our 35-mph front-impact test, seven car seats failed. They separated from their bases, rotated too far, or would have inflicted grave injuries, as measured by our test dummy, whose sensors record the severity of impact. We retested these to see whether they passed the 30-mph federal standard. All passed except the Evenflo Discovery.” And as part of the story, CU urged the recall of the Evenflo Discovery.
The article – rather than the test results – swiftly became the story. The public and many child passenger safety technicians were shocked by the results. Manufacturers insisted that their products were safe and CU was forced to withdraw its report after NHTSA quickly reviewed the results and found that CU’s testing contractor, Calspan, had conducted its side impact sled tests under conditions that represented being struck at more than 70 mph. When NHTSA tested the same seats installed in sleds at 38.5 mph, the seats stayed in their bases.
The agency, in particular, defended the Evenflo Discovery.
“Our own tests don’t quite agree,” Tyson told one newspaper in January 2007, and stressed that the Evenflo Discovery was compliant with Federal Motor Vehicle Safety Standard 213. “We tested it just this last year, in 2006. All the seats tested do meet federal safety standards.”
CU issued a retraction and appointed a review committee. CU retained Brian O’Neill, retired president of the Insurance Institute on Highway Safety and Kennerly H. Digges, the director of NHTSA’s Vehicle Safety and Biomechanics National Crash Analysis Center to retest the car seats and promised to publish the new results. In the June issue of Consumer Reports, Consumers Union quietly released the results of its re-test. This time, the consumer group abandoned the side-impact tests altogether and simply tested the infant seats according to the FMVSS 213 compliance test. All of the seats passed.
Following the Evenflo recall announcement, CU released a brief statement:
“Consumer Reports has long had concerns about the safety of children in side-impact collisions,” says David Champion, senior director of automotive testing for Consumer Reports. “We’re pleased to see NHTSA addressing the issue.”
CU spokesman Douglas Love declined to comment further.
NHTSA denied that the CU testing influenced this phase of its child safety seats and side-impact research project, but the sequence of events suggest otherwise. The seats installed in its NCAP tests last year were nearly identical to the seats in the Consumer Reports article, and the testing began a month after the CU controversy. (NHTSA tested the Safety First Starter, rather than Designer model, added the Combi Tyro and Connection models and did not include the Eddie Bauer Comfort in its test.)
“Despite the flaws in its initial side-impact test, Consumer’s Union tried to raise the bar of child seat safety and ultimately, the effort proved to be a catalyst for the agency to take another look at the excellent questions CU raised,” says Sean Kane, president of Safety Research & Strategies. “The result has been that one million infant safety seats have been removed from the market.”
It is unclear if these unofficial side-impact research efforts will have a long-term effect in the form of better child safety seat regulations, Tyson said
“We’re continuing to do it. It’s an excellent opportunity,” he said. “We haven’t written off the possibility (of a regulation), but at this point our view is that the side impact tests seem to tell you more about the vehicle than the child safety seat.”
Deborah Stewart, child seat safety advocate and editor of the Safe Ride News believes that in using the NCAP results to encourage a recall, NHTSA has taken a step toward a stricter child safety seat rules.
“This seems to indicate that side-impact performance has become a de facto requirement for child restraints,” she says. “Manufacturers need to be aware of this but are in a difficult position due to the fact that no specific testing procedure has been established that enables them to run reproducible, practicable tests. NHTSA needs to move swiftly to establish such a standard.”
In the meantime, the substantive research and testing on child safety seat performance is being done over the border. Since 2003, Transport Canada has conducted hundreds of tests involving some 77 passenger cars, cross-over vehicles, minivans and SUVs from the 2002-2007 model years, along with low-back and high-back boosters, high-back boosters with a harness latch and tether and a lap and shoulder belt.
Copyright © Safety Research & Strategies, 2008