Air bag systems were installed in the United States in thousands of automobiles during the model years 1974 through 1976. These air bags were designed and installed in large size vehicles by both General Motors Corporation and Ford Motor Company. At the time, the intent was to provide frontal crash protection for the general population of motorists who, on the whole, were not wearing seat belts. Consequently, the air bags installed in the 1970s were quite large in circumference -- occupying fully the entire front of the vehicle from door to door. The very few accident studies of early vintage air bag crashes concluded that these systems were working to minimize injury potential. Those studies were consistent with laboratory testing which demonstrated that these passive restraint systems provided protection in barrier impacts through 30 to 40 mph crash speed range.
The production vehicles of the 1970's which included air bag systems used compressed gas (sodium azide) in a chemical can which rapidly inflated the airbag by ignition and conversion to harmless nitrogen gas when sensors detected a pre-planned "severe" frontal crash. Publications in recent years have analyzed the field performance of these older systems. Looking just at driver's side airbag system performance, these authors determined that the system was found to be 21% effective in preventing AIS 3+ injuries, but was -34% ineffective for AIS 2+ passenger injury. Many of the airbag inflation injuries found in the research of the systems performance -- as well as the much more recent data obtained since 1989 -- occurred in moderate severity crashes without distortion of the occupant compartment. Thus, the deployment of an airbag under some circumstances has the potential to seriously injure or kill, and it does not always depend upon the severity of the crash. As David Viano of the GM Biomedical Science Department stated a few years ago,
"The high energy release of an airbag may injure an occupant against the system at the instant of deployment. Blocking the path of deployment increases pressures in the cushion during gas generation and develops high forces on the occupant. Since the force occurs with high velocity, there is a risk of injury by a Viscous mechanism."
As a result of the federal government's delay in the starting date of mandatory passive restraint criteria, [FMVSS 208] automobile manufacturers chose not to pursue the mass production of motor vehicles with air bags until the late 1980s. In the calendar years 1980 and 1981, Daimler Benz introduced in Europe a test fleet with full frontal protection by air bag restraint. Then in the 1982 model year, Daimler Benz introduced a driver's side air bag system (along with a manual lap and shoulder belt) in Europe, along with a three point passenger side manual seat belt with a pretensioner retractor. Daimler Benz then introduced air bag systems to the U.S. market in 1984, offering it first as an optional piece of equipment and then it became standard equipment. Chrysler Corporation became the first American manufacturer in the 1980's to introduce air bags into its vehicles. And, with time, the remaining manufacturers have installed air bags for both driver and passenger positions -- as a direct result of the mandatory passive restraint performance criteria adopted by the NHTSA. It is estimated that by the year 2000 there will be over 50 million vehicles on the roads with air bags, and each year there will be literally hundreds of thousands of air bag deployments.
Over the past five years, researchers have made a concerted effort to collect and synthesize accidents in air bag equipped vehicles. The intent of these papers was to generally detail the injury patterns of air bag protected occupants, so that vehicle designers and medical practitioners can consider the nature of the harm and benefit of this safety device.
The air bag system, like any other safety device, needs to be viewed in connection with the benefits and risks it poses; that analysis is important if efforts are to be made to minimize the risks associated with air bag equipped vehicles. Some researchers have extrapolated current accident data to estimate that by the year 2000 there will be over 200,000 injuries per year in the United States alone, which are induced by air bag deployment. What kinds of injury patterns are evolving and are there vehicle design features implicated by these patterns of injury?
Before analyzing this issue, it is critical to stress that air bags and seat belts save lives. Every vehicle which includes an air bag system provides for incremental safety that was unavailable to Americans during the 1970s and 1980s. While inflatable restraints are not a panacea, they are a significant advancement in the quest to reduce trauma on our highways. How well these devices function in specific accident scenarios does, however, vary because of design features, variability in accident modality and occupant kinematics.
A review of over a dozen recently published papers has allowed for a study of the details of one hundred and twenty-two frontal accidents involving air bag deployment. While there are differences of opinion about the causative relationship between air bags and injury, it is necessary to account for the trends in the accident data. It is also important to study the causative relationship between some of the events precipitating fatalities and the deployment characteristics of the air bags installed in vehicles. This analysis can only lend credence to the adage that everything we do can be done better.
Injuries versus Fatalities
As one would expect, most of the reported injuries and deaths due to air bag deployment involved drivers rather than front seat passengers. In these data there were 16 driver fatalities and 4 passenger fatalities -- exclusive of the more than 20 fatalities to children positioned in the front passenger position. While the proportionate number of fatalities to injury is not necessarily out of the ordinary statistical range, what is unusual are other patterns emerging from these fatalities.
In the twenty fatalities reviewed here, it was observed that a majority of these motorists were female and under 50 years of age and unbelted. However, there was no injury pattern between the belted and unbelted fatalities. Between the two groups, there was a proportionately equal number of deaths due to brain injury and spinal cord trauma. What was somewhat unexpected is that the majority of these deaths -- whether the motorist was belted or unbelted -- occurred in accident sequences which can only be characterized as low velocity collisions.
The risk of injury from an inflating air bag has been a consideration for all occupants for many years. Laboratory studies confirm what we are now seeing in field accidents: "high" loads can be developed on occupants who are within the path of an inflating bag. While many writers have called this situation the "out of position condition," it is more accurately described as the "near position condition." This is so because circumstances causing impact to the occupant happen when he or she is properly using the vehicle and "happens" to be positioned too close to the bag. As in other dynamic circumstances, the risk of injury depends on the system tested, the environment and the alignment and position of the occupant at the time of bag inflation. Clearly the human body parts at greatest risk are the head, neck and torso because of their (foreseeable) close proximity to the system. Real world "near position" injuries are associated with a wide range of crash severities, but most troubling are those circumstances when the Delta V is below 25 mph, because these are accident circumstances which should, ordinarily cause little or no serious injury. This exposure to injury occurs under a variety of scenarios, including the driver or passenger who has slumped forward from illness or drowsiness; or moved forward because of multiple or singular minor impacts, displacing the occupant forward prior to a more severe impact causing inflation. Likewise, during pre-impact braking both the restrained and unrestrained occupant will displace forward; most seat belt systems do not provide in their design for retractor lock-up until the vehicle has experienced a .6 to .7 g, which would allow the belt to spool out somewhat before impact. Additional problems allowing for the "near position" are: occupant size, seat adjustment, steering column or wheel tilt adjustment, seated posture, and inflation timing.
Laboratory studies have demonstrated that "near positioning" can result in substantially higher head, neck and chest accelerations then one would ordinarily expect for the seat belt restrained occupant -- because of air bag inflation rates. Melvin, Horsch and others at GM have reported that testing reveals a real potential for increases in head acceleration, neck forces and moments, and a high risk of serious chest injury.
One detailed example of fatal injury from air bag inflation at low velocity (reported in the literature) is presented here.
Illustrative Case Example
The accident in question involved a 1990 Dodge Shadow that was involved in a center frontal impact sequence with utility pole, which deployed its supplemental driver air bag system. The vehicle was driven by a 36 year old female, who weighed about 115 pounds and stood 5'1." At the time of the crash the driver was alone in the car and she was wearing her seat belt system. For some unknown reason, the driver of the Shadow veered out of her path of travel, crossed over one lane of travel and left the roadway. The left front corner of the car impacted a fence post and sightly redirected the car into a 10" diameter utility pole, located a few feet away from the fence. The PDOF was 12 o'clock and there was a velocity change of 14.4 mph (calculated based upon 14.5" of bumper crush).
The driver of the Shadow had her seat adjusted to a forward position and the tilt steering column set to the center adjustment point. The following injuries were found:
|Rupture of abdominal aorta||AIS 5||Air bag/steer wheel|
|Multiple rib fractures||AIS 4||Air bag/steering wheel|
|Ruptured spleen||AIS 3||Air bag/steering wheel|
|Contusions of eyelids||AIS 1||Air bag|
|Contusions on left shoulder||AIS 1||Shoulder belt|
|Abrasions with contusion chin||AIS 1||Air bag|
|Abrasions of neck||AIS 1||Air bag/shoulder belt|
|Contusions over breasts||AIS 1||Air bag|
|Lacerations/contusions to knees||AIS 1||Knee bolster|
The kinematics ascertained by the Calspan investigation team were that the driver's close pre-impact proximity with the steering wheel caused very little time for the seat belt to lock-up and restrain the driver. The driver did load the shoulder belt webbing enough to cause contusions and her knees did load the knee bolster. However, because she was within the deployment range of the bag, she suffered substantial chest loading, which caused her fatal injuries. The vehicle was subsequently repaired and is out on the road again.
Injuries attributable to accidents with airbag deployment have also been categorized in this review. In the 122 cases studied, 102 injury cases were reviewed. The majority of the injured motorists in this study were seat belted.
Injuries in Airbag Deployed Accidents
From the standpoint of a societal analysis, it was somewhat disconcerting that six motorists in reportedly low speed crashes suffered devastating spinal cord trauma. In all but one of these cases the injury was to the cervical spine, with several at the C1-C2 level to youngsters under the age of 10.
Air bag induced injuries are, by the very terms used, injuries caused when a motorist is impacted by an inflating air bag. NHTSA has informed Congress that air bags can have adverse effects for small-statured people, older people and out of position occupants. This review of the cases reported in the literature confirm those conclusions and identify additional risk factors. The questions raised by our analysis of these accident cases are the following:
1. What can be done to minimize the risk of injury or death when an occupant is unrestrained?
2. What can be done to improve the performance of seat belt restraint systems to maximize the distance available to the air bag to inflate -- without impacting the occupant?
3. What can be done to lessen the impact forces of an inflating air bag?
The accident data suggests that air bag induced injuries to unrestrained occupants are a result of the occupant moving "out of position" and being struck by the inflating air bag during low speed collisions. The obvious solutions to this dilemma are to compel seat belt usage, change the deployment timing, and alter the deployment rate. To minimize the risk of deployment induced injuries, air bags should be designed to deploy faster when the seat belt is not worn. This allows for complete inflation before the occupant moves substantially forward in the seat.
A high number of head and neck injuries have been reported in air bag deployments even when the motorist is restrained. The mechanics of this type injury suggest that the seat belted occupant is moving forward in a relatively upright position and coming into contact with the inflating air bag so that the neck is either placed in compression or extension. To minimize seat belt excursion to the "out of position" stage the seat belt must be designed to lock-up when deceleration of the vehicle first begins -- through braking -- or as a result of occupant movement within the seat belt -- by using web sensitive retractors which are tuned to lock up at .35g. Additionally, the environment of the driver's position needs to be studied to deal with people of small stature. Some possible incremental solutions would include adjustable pedals, a pretensioning retractor and changing the characteristics of deployment -- including a reduction of the deployment rate, tethering the bag, and widening its coverage which would also reduce "bag slap."
Lessen Impacting Forces
The current state of the art would allow air bags to be designed with improved folding patterns, special fabric coatings, and modifying the deployment rates. These changes will "soften" the impacting forces to the air bag. Undoubtedly, these changes will provide incremental improvements and changes in injury potential.
Educating the public to the need to remain as far away from the air bag container is critical to a reduction in deployment injuries and deaths. Both motor vehicle manufacturers and the government have a responsibility to alert consumers to these inherent risks in the current air bag modules. Until other design changes can be filtered into the restraint systems in vehicles, the only answer to this risk-benefit is one of instruction and warnings. Motorists need to be informed of the grave risk that they can be exposed to by a safety device that must still be viewed as a life saver.