Two separate studies have put Federal Motor Carrier Safety Administration’s (FMCSA) database of commercial vehicle crashes, known as Motor Carrier Management Information System (MCMIS), under intense scrutiny.

On one hand a November report from the U.S. Government Accountability Office (GAO), the Congressional watchdog arm, found that the quality of overall crash data have improved, thanks to states improving the timeliness of submitting crash data and the number of reportable crashes to the agency. Although the report praised the agency’s progress in maintaining reliable data, non-fatal accident reporting is said to remain inconsistent.

On the other hand, a report from Oak Ridge National Laboratory (ORNL), contracted by the DOT Office of the Inspector General to audit SafeStat, an MCMIS-based rating system that ranks unsafe carriers, found that untimely and incomplete data continue to keep SafeStat unreliable.

MCMIS is a major aspect of SafeStat results, but other factors include roadside inspections, traffic violations and compliance reviews. The report was completed in October 2004, but only this week was made publicly available.

“Incomplete [MCMIS] data resulted in some carriers being ranked “at risk” when they would not have been with complete data,” stated the ORNL report. “Both the timeliness and completeness of MCMIS data are improving.”

Underreporting to MCMIS was estimated at about 33%, with 80% crashes with injuries reported and 50% tow-away crashes, according to the ORNL report. According to GAO, states input crash data to MCMIS anywhere between 13 days to 339 after the time of an accident. These variances are enough to break the SafeStat system, ORNL said.

“Selecting carriers with the highest (SafeStat) scores, without addressing the accuracy of the scores, results in the selection of many carriers due to random variations and not true change in carrier risk,” ORNL said.

“[The ORNL report] calls into question the underlying algorithm of SafeStat,” David Osiecki, vp- safety for the American Trucking Assns. (ATA) told FleetOwner “It calls into question its validity and whether it should be continued to be used. And if it is continued, how it should be changed and modified to increase its objectivity and accuracy.”

FMCSA had provided states with nearly $21 million in discretionary grants aimed to improve reporting. GAO has recommended that the agency set up guidelines for assessing and awarding state funding requests to improve data.

“The agency has criteria and guidelines they provide to states on what could be a reportable truck crash, but it’s complex,” Sharon Silas, GAO project director told FleetOwner. “It’s hard to make sure make sure everyone is reporting it the same way. FMCSA can do the training and provide funds, but they can’t tell the states what to do, or force the states to report a certain way. But they can provide the guidance to make sure they’re doing it more consistently.”

Bob Inderbitzen, director of safety & compliance for the National Private Truck Council, thinks the government is not doing enough to improve reporting.

“It’s beyond me why this problem is being allowed to exist,” Inderbitzen told FleetOwner. “[The government is] not trying to use a heavy hand with the states and is instead giving them money and time to fix the problem.”