DETROIT -- Data showing that airbags did not deploy in thousands of fatal crashes is often inaccurate and does not account for numerous scenarios in which airbags would be programmed not to inflate, according to a 2009 study.
The study by researchers at the Insurance Institute for Highway Safety and University of Maryland found frequent deficiencies in the airbag deployment information collected by the U.S. Fatality Analysis Reporting System. That database was used Thursday as the primary basis for a widely publicized claim by the Center for Auto Safety that 303 deaths occurred in newly recalled Chevrolet Cobalts and Saturn Ions when airbags failed to deploy.
The study found that front airbags were recorded as not deploying in 18 percent of all frontal crashes in which the driver or front-seat passengers died but that further review put the actual number closer to 8 percent. No more than 2 percent of the deaths in those crashes represented “potential system failures,” the researchers said.
The 12 front-seat deaths that General Motors has linked to its recall of 1.6 million vehicles would represent 4 percent of the 303 deaths cited by the Center for Auto Safety.
In 25 percent of frontal crashes examined by the 2009 study, FARS and a separate government database provided conflicting information as to whether the airbags had deployed. They then looked at 43 deaths in which the airbags were verified as not deploying and found that 25 of them -- 58 percent -- were incidents “in which deployment typically would not have been expected based on crash severity and other characteristics.” Six others were classified as borderline calls on whether the airbag should have activated.
The researchers said the airbags improperly failed to deploy in only 11 of the 43 cases, and that only three of the people killed in those instances would have benefited from the airbags.
The Detroit News uncovered the 2009 study on Friday, after The New York Times published an article on the claim of 303 deaths in the recalled GM cars.
GM also disputed the Center for Auto Safety’s claim.
“As knowledgeable observers know, FARS tracks raw data,” the automaker said in an e-mailed statement. “Without rigorous analysis, it is pure speculation to attempt to draw any meaningful conclusions.
“In contrast, research is underway at GM and the investigation of the ignition switch recall and the impact of the defective switch is ongoing. While this is happening, we are doing all we can now to ensure our customers' safety and peace of mind. We want our customers to know that today's GM is committed to fixing this problem in a manner that earns their trust.”
Clarence Ditlow, the head of the Center of Auto Safety, an independent watchdog group in Washington, criticized the National Highway Traffic Safety Administration for failing to investigate the reports of fatalities in crashes without airbag deployment. He called them “evidence of a widespread problem,” while acknowledging that some of the crashes might not have involved a defect and defended publicizing the 303 deaths figure during an appearance Friday on CNBC.
“No one asked the question: ‘Why are these airbags not deploying?’” Ditlow said. “If they would have done that, we would have had a recall years ago. NHTSA closed their eyes to this defect.”
Independent analysis of the FARS data by Automotive News found that vehicles similar to the Cobalt and Ion are logged as having similar rates of fatalities with no airbag deployment. For 2012, the most recent data available, FARS shows 36 such deaths involving a Cobalt, 33 involving a Toyota Corolla, 28 involving a Honda Civic and 28 involving a Ford Focus.