Who Pays the Bill?

The Center for Auto Safety is the nation’s premier independent, member driven, non-profit consumer advocacy organization dedicated to improving vehicle safety, quality, and fuel economy on behalf of all drivers, passengers, and pedestrians.


Motor vehicle injuries are a major public health problem. Vehicle crashes affect both the individual crash victims and society as a whole in numerous ways. The cost of medical care is borne by the individual through payments for uninsured expenses, by society through higher insurance premiums and through the diversion of medical resources away from other needs, such as disease control or medical research. Significant costs also are associated with the productivity that is lost when an individual’s life is claimed at an early age or as a result of an impaired person’s disability. Those dependent on the victim suffer the immediate economic hardship from foregone income, but society also suffers through efforts to support the victim or victim’s dependents and, eventually through foregone contributions to the nation’s productivity. Victims bear pain, suffering and lost quality of life.

A growing number of studies have focused on determining the enormous social and economic consequences for the injured individuals, their families, and society as a whole.1 This report examines the costs to society of deaths and injuries from burns suffered in fire crashes in 1973-87 General Motors pickups with side saddle gas tanks.


Although fires in motor vehicle crashes occur in a small fraction of all crashes, fire can significantly increase the risk of injury beyond that caused by crash forces. Fire is of particular concern where occupants are trapped inside the vehicle, where injuries prevent occupants from escaping themselves or where there is insufficient time for occupants to escape before the vehicle is enveloped in flames. Federal Motor Vehicle Safety Standard (FMVSS) 301, “Fuel System Integrity,” was issued to reduce the risk of crash fires occurring in survivable crashes but has failed.

Motor vehicle fires in all police reported crashes occur in about 3 of every 1000 crashes but occur in 26 of every 1000 fatal crashes, nearly 9 times the rate for all crashes.2 Fire is associated with the more severe impact crashes which also tend to be fatal crashes so the 9-fold increase in frequency is not attributable to the incidence of fire alone. In 1988, there were nearly 1800 deaths in 29,000 fire crashes. In 1975, there were 500 fewer fatalities in fire crashes with only 1300. Vehicle burn victims require more medical attention than other vehicle accident victims with the average hospital stay for vehicle burn injuries being 16.1 days versus 6.7 days for all motor vehicle injuries.3

Tragically, while automotive technology has done little, if anything, to reduce crash fire casualties, medical technology has made dramatic advances in the past 25 years in treatment of patients with severe burn injuries.4 Today, burn victims who would have died, live. Burn victims who live have far better lives due to advances in medical treatment and rehabilitation. While costs of medical treatment for those who survive may go down, productivity and quality of life costs go up with the number of increased survivors. What is needed now is more prevention of crash fire injuries to match the better treatment.


In the course of its investigation of side saddle pickups, the National Highway Traffic Safety Administration (NHTSA) conducted an in-depth analysis of its Fatal Accident Reporting System (FARS) data base to determine the incidence of fire-related deaths in 1973-87 GM pickups.5 To determine whether a burn death was involved in a vehicle crash, NHTSA utilized the “most harmful event (MHE)” variable from the FARS database, which has been available since 1979. The MHE category describes the leading cause of death in each vehicle crash based on the police accident report and other supporting state documents. NHTSA’s analysis enables one to sort out deaths in fire crashes due to trauma from deaths in fire crashes due to fire which was not done in earlier studies.

For the years 1979-90, NHTSA determined there were 349 GM pickup accidents which had at least one death attributable to fire. Including calendar years 1973-78 and 1991 to present, there were 490 GM pickup accidents which had at least one death attributable to fire. While General Motors has criticized NHTSA’s analysis as being too high,6 NHTSA’s report understates the number of fire deaths in at least two aspects. First, some burn victims such as Donna Romine who died more than 30 days after the accident are not counted by FARS even though the death is clearly due to fire. Second, by focusing on the number of pickup accidents with at least one death versus number of fire deaths, the NHTSA analysis fails to account for multiple fire deaths in the same accident such as the August 9, 1989 Vergara accident in Texas which resulted in 5 deaths due to burn injuries.

Although the FARS data base could not give a total for burn injuries in fire crashes because it does not include non-fatal crashes, two independent sources indicate that for every burn death in a GM pickup, there are five burn injuries. First, NHTSA’s 1990 report on “Motor Vehicle Fires in Traffic Crashes” shows this relationship. Second, General Motors engineering personnel have testified to this relationship.7


Given the number of GM pickup fire deaths and injuries are determined, the economic costs to victims could be determined through the computer models and data bases developed by Dr. Ted Miller in association with The National Public Services Research Institute and The Urban Institute.8 Dr. Miller concluded that the aggregate cost of deaths and injuries in fires in these vehicles was $2.0 billion during from 1973 to 1993, or about $93 million per year as shown in Table 1.

Table 1

Annual Costs of GM Truck Fire Crashes 1973-93

(Millions of 1992 Dollars)

Component CostPercentMedical & Transport2.12.3 Productivity Loss17.819.0Pain & Suffering72.477.4Legal & Administrative1.21.3TOTAL93.6100.0

The total direct costs of GM Truck fire deaths and injuries to date are approximately $2 billion (in 1992 dollars). This comprises 490 fatal injuries and nearly 2,500 treated injuries in crashes of 1973-87 GM C/K trucks with side saddle gas tanks from 1973 to present. Property damage is not included. Fatal injuries account for $1.3 billion of this total; non-fatal burn injuries account for $700 million. For each GM pickup burn death, the cost is $2.6 million; for each hospitalized burn injury, the cost is $814,000; and $15,000 for a non-hospitalized burn injury as shown in the following table:

Table 2

Costs per Motor Vehicle Fire Burn Victim 9
This analysis assumes that for burns, police-reported A-injuries (disabling injuries) correspond to hospitalized injuries and B and C injuries (other evident and possible injuries) to nonhospitalized injuries. It uses the ratios of motor vehicle burn injuries by police-reported severity from NHTSA’s report, “Motor Vehicle Fires”, supra n. 2.

(November 1992 Dollars at a 2.5% Discount Rate)

Fire &MedicalEMSPoliceProductivity Admin LegalQ(Life)Total Fatal 11,199453579680,96081622,3201,862,7152,599,042Hosp43,29426850542,7663,04914,376710,003 814,251No Hosp698265052,35551-0-11,44715,082

The total direct costs of GM Truck fire deaths and injuries from now until the last ones are off the road is approximately $0.8 billion (in 1992 dollars). This comprises 200 fatal injuries and nearly 1,000 treated injuries in crashes of 1973-87 GM C/K trucks with side saddle gas tanks unless they are recalled which would reduce this toll by up to half.

The model and data used in this study are the same as used in CPSC’s fire injury study, “Estimating the Costs to Society of Cigarette Fire Injuries.” The CPSC study estimated medical costs, transport costs, productivity losses, lost quality of life (including what is commonly referred to as “pain and suffering”), and legal and health insurance administrative costs for fire related injuries of all types and then did a regression analysis specific to cigarette fires. For this Truck Fire study, Dr. Miller did a similar regression analysis to develop costs of motor vehicle fires which had not previously been done.

One of the principal sources of data relied on vehicle fire medical costs is the California Hospital Discharge Survey for the second half of 1990 when the state mandate that hospitals record the causes of burns and other traumatic injuries first took effect. These data recorded 2129 hospitalized burns cases including 92 moving motor vehicle fires. As shown in Table 3 on the following page, the California Hospital data show moving motor vehicle fire burns to be more serious than any other type of burn with a mean hospital stay of 16.1 days versus 9.7 for all burn injuries.

Vehicle burn injuries result in longer and more expensive hospital stays than other motor vehicle accidents. According to NHTSA, the average length of hospital stay for all motor vehicle accidents is 6.7 days with the hospital costs being $9,790.10 The length of stay for vehicle burn injuries is over twice that at 16.1 days while the hospital cost is over four times greater at $43,294. While it is beyond the scope of this study, individual cases suggest GM Truck fire crashes are more costly.

Table 3

Number of Hospitalized California Burn Survivors & Mean Length of Stay

CauseCases% of Cases Length of StayFlame 61428.811.4Electrical 833.97.3Chemical 1286.05.2Scald 65030.58.1Radiation 160.88.4Moving Motor Vehicle 924.316.1Other Motor Vehicle 341.613.4Other Vehicle/Machine 231.115.0Hot Object/Heat 1999.47.1Not Classified 381.811.5Unknown 25211.811.6 TOTAL 2129100.09.7

Society has traditionally abhorred burn injuries due to the extreme pain and suffering as well as the nature of the injuries themselves. The non-fatal injury quality of life costs are reflective of amounts juries typically award in burn cases. The figures above are conservative since auto burn cases rank at the very top of jury verdicts with three of the five largest product liability awards since 1978 being for auto fuel tank cases. For fatal injuries, the quality of life costs are derived from statistical valuations in 49 studies of what people pay to reduce fatality risks in their daily lives in terms of dollars, time and inconvenience.11


The following discuss some of the many GM Truck fire crashes that resulted in deaths and injuries to show the impact on the victims.

Kelly Tyler

Kelly Tyler is a 28-year old college student from Antioch, California. on March 31, 1982, she was a high school teenager who borrowed her father’s GM pickup to visit a friend after school. on a rain slick highway, her truck was impacted by another vehicle, skidded out of control, flipped on its side, and burst into flames. Ms. Tyler was trapped inside her truck, unconscious, until she was pulled from the flames by a friend driving in a car just ahead.

As a result of her crash, Ms. Tyler received severe second and third degree burns over the upper third of her body. She spent the next six weeks in the Alta Bates hospital in Berkeley, CA receiving skin grafts and recuperating from her burn injuries. Since her accident, Ms. Tyler has undergone approximately 20 operations to help correct the damage inflicted by the burns; each procedure has required a minimum two day stay in the hospital.

Douglas Worden

Douglas Worden is a retired teacher from Puyallup, Washington. A little over a year ago on November 3, 1992, Mr. Worden was driving his 1978 Chevy pickup when it was struck in the driver’s door by another vehicle. There was minimal impact damage to his truck, but the side saddle fuel tank ruptured causing the pickup to erupt into flames.

As the cab of the truck became engulfed in fire, Mr. Worden struggled to exit the inferno. The driver’s side door was jammed, so he scrambled to open the passenger door to escape. once free of his pickup Mr. Worden had to roll on the ground to extinguish the flames burning his clothing.

The only injuries Mr. Worden received were burns over 55% of his body. He was treated for six months at the Harborview Medical Center in Seattle, where he accumulated medical costs in excess of $380,000. Mr. Worden’s mobility has been severely diminished by his injuries, and he still requires additional physical therapy to maintain the remaining flexibility in his limbs.

Robert & Michael Stephens

On May 8, 1992, Robert and Michael Stephens were driving in a 1974 Chevy pickup in Oakdale, California when their truck was forced off the road. In an attempt to avoid a collision with another vehicle, the Stephens’ pickup collided with a utility pole rupturing the side saddle fuel tank. The truck caught on fire causing severe and disabling burn injuries to both men.

Michael Stephens received burns over 80% of his body while Robert Stephens was burned over 50%. Robert Stephens was hospitalized in a burn center for six weeks while Michael Stephens was hospitalized for six months for treatment at UC-Davis’ burn center. Both will require extensive future medical treatment. Their medical bills totaled nearly 3.5 million dollars before the Shriners agreed to provide future medical care at their burn center in Galveston, Texas without charge.

Donna Romine

Ray Romine and his wife Donna were on their way to a 1976 Christmas Eve family dinner in South Dade, Florida when tragedy struck. At a local intersection, the Romine’s gift-loaded 1976 Chevy Silverado with two full tanks of gas was struck in the passenger side by a vehicle running a stop sign. The Romine truck exploded into flames that could be seen as far as a half-mile away.

Mr. Romine was burned over 15% of his body, while his wife received third degree burns over 65% of her body. Mrs. Romine never left the hospital and died 120 days later due to her burn-related injuries. The cost of her medical treatments was estimated to be approximately $120,000 in 1976 dollars. Because her death occurred more than 30 days after the accident, she is not considered to be a fire death by NHTSA’s Fatal Accident Reporting System.


The tragedy of the GM Truck fire deaths and injuries could have been avoided and can still be reduced. Known technology exists in safer tank locations, breakaway fuel lines and safety valves, fuel cells and bladders and better packaging to prevent all fire deaths and injuries in crashes. In 1973, GM Engineers internally recommended to top management a level of performance for fuel systems in crashes such that “fuel leaks, should not occur in collisions which produce occupant impact forces below the threshold of fatality” — i.e., if you survive the crash forces, you should not be burned by fires from fuel leaks.12

This is the very principle which the US Army adopted in 1970 after suffering numerous burn casualties in helicopter accidents in Vietnam. In April 1970, Bell Helicopter changed production to use a crashworthy fuel system (CWFS) consisting of a fuel cells, breakaway fuel lines, and cutoff valves. “In the ensuing 39 months the Army experienced 702 accidents with CWFS-equipped helicopters with one very remarkable result – there wasn’t a single fatality or injury due to thermal trauma.” During the same period, helicopters without CWFS experienced 52 burn fatalities and 31 burn injuries in 895 accidents.

If GM Trucks had been built using design principles and technology known and available in 1973, then no one should have died from burn injuries in GM fire crashes. Due to heavy industry lobbying, NHTSA issued FMVSS 301 at levels far lower than what the auto makers could meet and what their own engineers said should be met. In a 1990 evaluation of FMVSS 301, NHTSA concluded that the standard had not reduced motor vehicle crash fire fatalities and injuries. FMVSS 301 should be amended to prevent fuel leaks and fires at crash speeds at which occupants survive as originally proposed by GM Engineers in 1973. Due to advances in crashworthiness since 1973, FMVSS 301 should be revised to preclude any fuel leakage in vehicle-to-vehicle crashes up to 55 mph front, rear and side impact.

GM Trucks were not even built to the same level of fuel system safety as that of their competitors. Both Ford and Chrysler installed the gas tanks inside the frame rail because of concerns over fuel leakage and fire in side crashes. As a result, the incidence of fatal injury in side impact fire crashes where death is caused by fire is 3.5 times higher for GM than Ford and 4.7 times higher than for Chrysler. Our analysis indicates that $465 to $910 million of the $2 billion cost of GM Truck fires to date is attributable to fuel tank location.

By recalling the 1973-87 GM Trucks with side saddle gas tanks to remedy just the fuel tank location, we could reduce the future crash fire costs of these trucks by up to $370 million. The repair remedy in the recall could be to move the tank inside the frame rail or install bladder lined tanks and protective cages outside the frame rail as GM proposed but never implemented for its 1983 Trucks.


The $2.0 billion in accident costs for GM Truck fire crashes to date is a tragedy that could have been prevented if industry lobbying had not forced the National Highway Traffic Safety Administration (NHTSA) to set an inadequate standard. GM engineers whose views were not known to NHTSA made the right recommendation which should be implemented today — there should be no fuel leakage in any crash where the occupants would survive the trauma from the crash forces. When GM’s engineers made their recommendations in 1972, occupant survivability was less than what it is today with airbags and other improvements in crashworthiness so their recommendation of 45 mph rear moving barrier, 30 mph side moving barrier and 30 mph front fixed barrier.

Since crash protection technology in new vehicles today enables occupants to survive 45 mph frontal fixed barrier crashes, 45 mph side moving barrier and 45 mph fixed rear barrier, FMVSS 301 should be amended to require no fuel leakage at these impact velocities or 45 mph in the near term. As crashworthiness protection improves in the future, FMVSS 301 should be further amended to provide for no fuel leakage in the specified crash modes at 55 mph.

Up to half the cost of GM Truck fire crashes could have been avoided if GM had installed their fuel tanks inboard of the frame rails as did Ford and Chrysler. Unless they are recalled, GM Trucks will be involved in fire crashes with future costs to victims of $800 million. By recalling the trucks, up to $370 million of this cost could still be avoided.

The victims and society have paid the overwhelming proportion of the $2.0 billion cost of GM Truck fire crashes. Government, both State and Federal, pay 29% of the medical cost with insurance picking up 52%. When lost productivity and other economic costs are included, the public bears 48% of the costs through direct federal payments and loss of income taxes, victims pay 31% and employers pay 21%, for sick leave, cost of retraining and hiring replacement workers. GM Truck fire victims and their families receive minimal compensation for their lost quality of life.

[1] This report relies on the methodology used in the Federal Highway Administration’s October 1991 report, “The Costs of Highway Crashes,” which examined the costs for an estimated 14.8 million motor vehicle crashes involving 47,000 deaths and almost 5 million injuries in 1988. More than 4.8 million years of life and functioning were lost. Crash costs totalled $334 billion. They included $71 billion in property damage medical, emergency services, workplace, travel delay, legal, and administrative costs (with half of this being property damage); $46 billion in wages and household production, and $217 billion in pain, suffering, and lost quality of life.This methodology has also been used in “Cost of Injury In the United States: A Report to the Congress” (1989) produced by The Johns Hopkins University Injury Prevention Center and the University of California Institute for Health and Aging for the National Highway Traffic Safety Administration and the Centers for Disease Control. The report relies on data on motor vehicle fire burn and on flame burns generally in Estimating the Costs to Society of Cigarette Fire Injury, Ted Miller et al., transmitted to Congress by the Consumer Product Safety Commission, August 1993.<?xml:namespace prefix = o ns = “urn:schemas-microsoft-com:office:office” />

[2] NHTSA, “Motor Vehicle Fires in Traffic Crashes and the Effects of the Fuel System Integrity Standard,” HS-807-625 (Nov. 1990).

[3] NHTSA, “Source of Payment for the Medical Cost of Motor Vehicle Injuries in the United States, HS-807-800 (January 1992).

[4] Appendix A summarizes the advances in burn treatment.

[5] “Fatalities, Fire-Related Fatalities, and Fatal Burns in Crashes Involving Certain Full-Sized American Pickup Trucks” (April 26, 1993).

[6] GM has contended that some of the fatalities coded with the most harmful event as fire are actually due to trauma based on analysis of autopsy reports. Such error also goes the other way where examination of the autopsy report reveals the death to be due to fire where fire was not listed as the most harmful event.

[7] In Moseley v. GM, former GM engineer Ronald Elwell testified that he had investigated about 1500 post-collision fuel fed fire cases and found that for every death “at the minimum it [number of burns per death] would be four and probably six.” Trial transcript of Jan. 15, 1993.

[8] Dr. Miller used the data and models developed in his work for the Consumer Product Safety Commission on “Estimating the Costs to Society of Cigarette Fire Injuries” and for the Federal Highway Administration in “The Cost of Highway Crashes”. Using regression analysis, Dr. Miller can estimate the cost to society of different sources of burn injuries. To date Dr. Miller has run the model for cigarette fires for the CPSC and for motor vehicle fires for the Center for Auto Safety.

[9] Definitions:

Hosp is a hospitalized injury.

Not Hosp is an emergency-room treated non-hospitalized injury.

Medical include payments for hospital and physician care, as well as prescriptions, allied health services, medical devices, and nursing home care.

EMS includes ambulance and helicopter transport.

Fire and police includes one fire call per crash and one police officer responding per injury. The police costs are net of average police response costs for a crash without injuries.

Productivity losses include wages, fringe benefits, and housework lost by the injured. They exclude productivity lost by people stuck in traffic jams caused by crashes; by co-workers and supervisors investigating crashes, repairing vehicles, recruiting and training replacements for disabled workers; and by family and friends caring for the injured.

Administrative costs cover insurance claims processing.

Legal costs cover lawyers fees, and court costs. The cost estimate omits time that plaintiffs and defendants spend on the case.

Q(life) represents costs of pain, suffering, and lost quality of life. The nonfatal quality of life costs are reflective of typical jury verdicts for these costs for burn injuries.

[10] “Source of Payment for the Medical Cost of Motor Vehicle Injuries,” supra n. 3.

[11] For a detailed discussion, see “The Costs of Highway Crashes,” pp 74-80, , supra n. 1; “Cost of Injury In The United States,” Chapter 4, supra n. 1; NHTSA, “The Economic Cost of Motor Vehicle Crashes,” HS-807-876, App. B (Sept. 1992).

[12] GM Engineers Ronald Elwell, Jim Steger and Paul Judson made the presentation to GM Vice President for Engineering Frank Winchell which is found in Appendix B. This document was authenticated by GM in Moseley v. GM, January 15, 1993 trial transcript at pages 34-42 and in a March 17, 1993 letter to Transportation Secretary Pena from James A. Durkin, GM Legal Staff.

[13] “Summary of U.S. Army Crashworthy Fuel Systems Accident Experience, 1970-73”, reproduced in Appendix C.