Early Warning Data Becomes Recall Tracking System, Agency Release of Data Lags
Reprinted from The Safety Record, Volume 6, Issue 4, August / September 2000
CROWNSVILLE, MD – A new analysis of Early Warning Data by statisticians Randy and Alice Whitfield, of Quality Control Systems Corporation, appears to show that the system is functioning more like an identifier of failed recalls than a spotter of developing trends.
Using 2003 to 2008 EWR data, QCS identified their top 10 vehicles with troubling trends for injuries and deaths. The 1999 Ford Explorer remains, hands down, the most worrisome vehicle causing fatalities – despite a 2000 recall of the vehicle’s OEM Firestone Wilderness tires, and despite its role as the mother of the Transportation Recall Enhancement, Accountability and Documentation (TREAD) Act. Passed by Congress in October 2000, TREAD was supposed to usher in a new era of prevention. NHTSA would collect manufacturer information related to death, injury, property damage, warranty claims, field reports and production numbers on a quarterly basis, and sift the numbers for emerging problems. Much of the EWR data was to be public, affording other individuals and organizations an opportunity to do the same.
“Judging from consumer complaints that may be related to death and injury claims continuing through 2008 on our top ten issues list, many of the ‘early warning’ signals we see are anything but early,” Randy Whitfield said. “Finding a root cause – if there is one – requires good detective work by competent engineers. Our rankings help to show where it makes the most sense to look. Because the data for deaths and injuries since the end of 2008 are still secret, it’s harder to get a good jump on new, potential problems.”
Using five and a half year’s worth of death and injury claims, the Whitfields were able to show which vehicles are in volved in unusual patterns of deaths and injuries. They then correlated that data with consumer complaints in the NHTSA database. Finally, the Whitfields associated injuries/deaths and consumer complaints with recalls that were launched to address related problems. For the first time since the Whitfields began mining the EWR deaths and injuries information, each of the vehicle and component combinations on their top 10 list appeared to be clearly linked to specific consumer complaints. These problems, in turn, appeared to be well-recognized issues supposedly addressed by earlier recalls.
However, the links to specific vehicle defects, Randy Whitfield said, is an open question. EWR codes the injuries and deaths by broad component categories, such as “suspension.” The VOQs are similarly coded, but they also contain the actual text of consumer complaints, making it easier to see specifically what about a problematic suspension may be to blame. From there, complaints, injuries and deaths can be correlated to recalls.
“We don’t offer this as statistical proof,” “he said. We don’t think these are the top 10 most defective vehicles. What we are saying is that this is a useful tool for finding red flags.”
The results demonstrate, at a minimum, that EWR appears to be documenting the continuation of old problems. For example, the 2004 Toyota Tundra was eight in the EWR injuries/death list due to a suspension issue, and was also the target of many VOQs, in which consumers reported a separation in the ball-joint leading to a loss-of-control crash. This continues to show up in Toyota’s EWR and NHTSA’s VOQ data, despite a recall in 2005 initiated to repair the problem.
In March 2007, the Whitfields filed a Freedom of Information Act lawsuit against the agency for the release of Early Warning Reports related to Ford Explorer fatalities. The Whitfields had been requesting the numbers to better analyze a rise in tire-related Explorer fatalities. In September 2008, after a protracted legal battle over EWR accessibility among the agency, the Rubber Manufacturers Association and Public Citizen ended with a ruling that Congress had not mandated that EWR data be confidential, NHTSA released some of the information to the Whitfields. They finally settled the lawsuit in November, for $4,000 meant to cover reasonable attorneys’ fees and expenses.
In the meantime, the Whitfields, who have submitted eight FOIA requests for EWR data, are still waiting for the agency to provide the first two quarters of 2009 reports, so that they can continue their analyses.