Federal database may undercount fatal underride crashes.

Underride crashes, in which passenger vehicles slide beneath large trucks, are much more common than federal statistics indicate, according to researchers at the Insurance Institute for Highway Safety, Arlington, Virginia. Institute researchers compared two databases containing information about fatal car/truck crashes: the Fatal Accident Reporting System (FARS) and the National Accident Sampling System (NASS).

EARS is the standard source of truck crash information the National Highway Traffic Safety Administration (NHTSA) uses when evaluating vehicle standards and regulations. This source provides overall incidence counts based on police reports, vehicle registrations, and driver records.

NASS is the database the Institute uses. It is a sampling system from which estimates are made based on police reports, crash investigations, medical reports, vehicle records, and survivor interviews.

Underride crashes are singled out in both databases because they often lead to fatal injuries, including decapitation. Restraint systems do not protect passenger car occupants from truck or trailer parts that crush or thrust into the upper part of a car.

Both databases agree that passenger vehicle occupants died in about 3,033 car/ truck crashes each year between 1988 and 1993.

However, FARS coded only 136 of these annually as underride crashes while NASS coded 611. Of 275 fatal car/truck crashes included in both databases, FARS coded just 7 percent as underrides while NASS coded 27 percent.

According to Institute researchers, the discrepancy is even wider than it appears. NASS did not code underrides resulting from impacts to the sides of trucks, but FARS did. When side impact data are added to the NASS figures, the percentage of fatal car/truck accidents involving underrides approaches 50 percent.

Elisa Braver of the Insurance Institute said there are two major reasons for undercounts in EARS. "The police reports for about half of the underride crashes didn't contain enough information for FARS analysts to recognize underride had occurred. Second, underrides weren't necessarily coded even when information in the police report indicated underrides."

Braver noted that since 1994 FARS analysts have been better trained to recognize and code underrides, but she believes that "these crashes still are being substantially undercounted."

In 1996, NHTSA issued a truck underride protection standard (49 C.F.R. pt. 571). It requires that newly manufactured heavy trucks and trailers be fitted in 1998 with rear guards that extend downward to within 22 inches of the ground. The standard does not address underrides beneath the side or front of vehicles, and it specifically exempts single-body trucks and trucks equipped with lift gates or other special equipment at the rear.

The recent Insurance Institute study revealed that single-unit trucks are involved in a substantial number of fatal underride crashes--261 per year, including 69 rear impacts.

For a copy of "Incidence of Large Truck/ Passenger Vehicle Underride Crashes in the Fatal Accident Reporting System and the National Accident Sampling System," write to Insurance Institute for Highway Safety, Publications, 1005 N. Glebe Rd., Ste. 800, Arlington, VA 22201.