SafeStat Under Scrutiny

Concerns prompt congressional request for investigation.

OLIVER B. PATTON
WASHINGTON EDITOR

      Every truck driver knows that his is a public occupation -- every move he makes is seen by other motorists. But not all drivers know that every mistake they make is also open to electronic inspection by anyone with an online computer.
      Roadside inspections, moving violations, out-of-service orders, compliance reviews -- all are entered into a collection of data bases that are open to public view via the Internet.
      The data is used by the Federal Motor Carrier Safety Administration to rank trucking companies nationally on the basis of safety -- a high rank will trigger a visit from federal investigators. It also is used by insurance companies to adjust their rate quotes (or decline coverage), by shippers to decide which company gets their business, and by personal-injury lawyers to sway juries.
      Obviously, there is a lot riding on the data. If it is correct, it can throw the spotlight on an unsafe operator, maybe even force him out of business if he is unwilling to make the changes he should make.
      If it is incorrect, it can cause great damage to a good company. That is the concern that prompted Rep. Thomas Petri, (R-Wis.), to ask the Inspector General at the Department of Transportation to investigate the system.
      In an Aug. 12 letter to Inspector General Kenneth Mead, Petri said he is worried about possible flaws in the agency's data collection and reporting procedures, as well as its computer model. He asked Mead for a "comprehensive independent investigation."
      Mead's office is not yet sure how it will proceed, said spokesman David Barnes. "We will work with them on this issue," he said.
      Petri's request reflects long-time concerns by trucking companies that there are flaws in the "SafeStat" system, short for Safety Status Measurement System.
      SafeStat gives each company a national safety ranking based on its accident history, roadside inspections, out-of-service orders, driver performance and safety management.
      Chris Licht, senior vice president for safety at Covenant Transport, Chattanooga, Tenn., raises issues that echo throughout the industry.
      For one thing, Licht said, the SafeStat methodology errs because it does not use mileage to calculate a company's score. Typically, accident statistics are used in conjunction with mileage to measure exposure, but SafeStat uses the number of power units.
      Since Covenant uses team drivers, it has relatively fewer tractors, which means that its tractor-to-incident ratio is higher than other carriers, Licht said.
      He also said the SafeStat computation is overly complex. "I have had to become a data analyst instead of a safety manager."
      Further, the process of getting mistakes purged from the record is time-consuming and cumbersome. Licht said SafeStat has included as many as 45 accidents involving Covenant trucks that did not fit the DOT definition of an accident. This problem stems in part from states using their own definition, rather than the federal one.
      According to Licht, Covenant underwent a compliance review this year because of questionable SafeStat grades. He said that in April FMCSA sent the company notice that their accident and safety management performance was unacceptable. "In June, we were found satisfactory -- we went full circle in 90 days."
      He is not opposed to SafeStat but he's not convinced it's working the way it should. "It's supposed to identify carriers with problems but I don't know that it's doing that."
      But there is more to the SafeStat story.
      It is well understood by state and federal officials that mistakes get into the data, and that it is sometimes difficult to get mistakes corrected. Julie Cirillo, the retiring chief safety officer at FMCSA, acknowledged as much in a recent interview. She said the mistakes have to be fixed by the states, and the company must be prepared to substantiate its claims.
      One safety executive who has a lot of experience with this process says that, while it can be laborious, it is not as difficult as some say.
      "SafeStat works as well as can be expected for what it is set up to be," said Jeff Davis, vice president for safety and human resources at Jet Express, Dayton, Ohio.
      While there are flaws -- he estimates that the error rate is between 5% and 10% — they are not difficult to manage.
      He monitors the data as it is posted — particularly the Carrier Safety Profile that is released each month — and immediately corrects mistakes. "It's like weeding the garden," he said. "Getting corrections made is a problem, but it's manageable if you keep up with it."
      He said that if he follows the data month by month and makes sure corrections are made, the mistakes do not show up on the annual SafeStat calculation. "If you don't keep up, it's a major problem."
      The advice is echoed by Timothy Lynch, president of the Motor Freight Carriers Assn., which represents the largest unionized trucking companies. "We routinely tell safety directors that they must be vigilant," Lynch said.
      Davis has a list of SafeStat flaws that need fixing. Like Licht of Covenant, he believes that FMCSA should use mileage instead of power units. That information should be available to the agency through the MCS-150 report companies must file every two years, according to one observer.
      Davis wants a standard way of correcting the data -- the points of contact are different from state to state. One source suggested that FMCSA designate a person in each of its field offices to be the liaison between the state and the company.
      Davis also wants the data to show whether or not the accident was preventable.
      And he wants FMCSA to purge old data from the system more quickly. Right now, an incident stays in the system for 30 months. One year would be a better period, he said.
      For Davis, a big part of the issue comes down to drivers. They must understand the high cost of non-compliance, he said. A single out-of-service order, for example, stays in the system and is visible to all for two and a half years.
      It used to be that the public's perception of truck drivers was shaped solely by what happens on the highway. Now there is what Davis calls the "E-perception" of safety: what the world sees on the Internet.

Sidebar
Keep Tabs On Your Safety Profile