General Motors says it cannot rule out that more than 12 fatalities will ultimately be linked to the problem in cars it recalled last month, but it also says that a death toll of 303 being floated by the Center for Auto Safety (CAS) is “pure speculation” based on unreliable data.
“If, in the course of our internal review, we identify any others, we will, of course, promptly bring that to the attention of the regulators,” GM spokesman Greg Martin said.
Warnings against taking the 303 count seriously also come from auto-safety specialists at the Insurance Institute for Highway Safety (IIHS) and, indirectly, from the National Highway Traffic Safety Administration (NHTSA).
The problem is that the 303 number in a CAS-commissioned study relies upon, and may strain the limitations of, data in the federal Fatality Analysis Reporting System (FARS). It’s a federal database, managed by NHTSA, that collects information supplied by local and state police agencies who responded to crashes.
Reports on air-bag performance, the key safety issue in the GM ignition switch recall, seem particularly problematic. An extensive 2009 IIHS study looking at air bag deployment found that front air bag malfunctions “appear to be relatively uncommon and far less frequent that suggested by FARS data.”
The death toll in the CAS study was culled from FARS data on 2003-2007 Saturn Ion and 2005-2007 Chevrolet Cobalt compact sedans that make up most of the 1.62 million cars and six models that GM recalled last month. The center hired Friedman Research to pick out fatalities in which front air bags did not deploy.
In the recalled cars, faulty ignition switches can unexpectedly move from their normal “run” position into “accessory,” causing the engine to stall and cutting off power to a variety of systems, including air bags. The Friedman study didn’t determine how many of the 303 total deaths involved a link between faulty ignition switches and failed bags.
Clarence Ditlow, head of the CAS, agrees that FARS lacks “that level of detail.” The point of using the number, he says, is to make the point that “NHTSA has to get the actual police reports and send out investigators.”
The Friedman study excluded only rear-end crashes. That means it almost certainly counted crashes where the front bags were not supposed to deploy, such as side crashes, rollovers and cases where vehicles drove under higher vehicles such as semi-tractor trailers.
Front bags are designed to inflate only when sensors detect an abrupt, dramatic front impact. The force of the bags inflating can injure people, making it important to keep the bags from inflating unnecessarily.
“Most people don’t know when the front air bag is supposed to deploy. Their expectations are incorrect,” says Sean Kane, safety researcher who works mainly for lawyers.
GM acknowledges it knew of a problem with the ignition switches as early as 2001. NHTSA is demanding the automaker explain why it only now conducted the recall.
GM’s recall is for ignition switches that can slip out of normal “run” position, shutting off the engine and cutting power to air bags and other safety systems. GM says it knows of 31 frontal crashes in which 12 people died as an apparent result of the flaw.
NHTSA says it “uses a variety of tools to evaluate the more than 40,000 complaints it receives each year, including special crash investigations, searches for similar complaints and comparisons to other vehicles. In this case, the data available to NHTSA at the time did not contain sufficient evidence of a possible safety defect trend that would warrant the agency opening a formal investigation.”