Editorial: Improving SafeStat, Again

This editorial appears in the July 2 print edition of Transport Topics. Click here to subscribe today.

The votes are in, and it’s clear that while federal referees believe that the SafeStat carrier-safety scoring program has been improved in recent times, they agree that it’s not yet ready for prime time.

The latest word, from the Department of Transportation’s own inspector general, came to us last week: “We found that, although improvements have been made, problems still exist with the reporting of crash data.”

The IG told the Federal Motor Carrier Safety Administration — which runs SafeStat — that until the problems are ironed out, the crash data should not be distributed publicly.
SafeStat analyzes data collected from inspections, compliance reviews and state accident reports to measure the relative safety fitness of motor carriers. But the completeness of crash data still varies among the states.



Trucking has long urged FMCSA to fix the program before releasing crash data. We believe the system still unfairly casts undeserving carriers as safety risks, on one hand, and on the other, serves to shield from greater scrutiny some fleets that need to be watched more carefully.

The IG review becomes public just weeks after the Government Accountability Office released its report on SafeStat, which found similar improvement but also contained recommendations on how to improve its accuracy (6-18, p. 5).

SafeStat scores were removed from FMCSA’s Web site in 2004 after an earlier IG report detailed a series of data inaccuracies.

FMCSA Administrator John Hill would like to return the SafeStat scores to public view and has been working to improve their reliability.

He said the agency has been working with the University of Michigan Transportation Research Institute to improve accident data from the states and to include nonfatal crashes in the calculations. He also said the agency would study the recommendations in both recent reports as it continues its drive to improve the program.

The work with the university is due to be completed in the fall, and Administrator Hill said he expects to make a final decision on SafeStat data some time next year.

We trust that he will continue his diligent efforts to improve SafeStat and won’t rush to repost the scores before he’s convinced that the program’s data accurately reflect the accident history of the nation’s truck fleets.