II. Overview of Problem and the Agency's Remedial Actions

A. Introduction.

While air bags are providing significant overall safety benefits, NHTSA is concerned that current air bags have adverse effects on certain groups of people in limited situations. Of particular concern, NHTSA has identified 87 primarily low speed crashes in which the deployment of an air bag resulted in fatal injuries to an occupant, as of November 1, 1997.(4) NHTSA believes that none of these occupants would have died if they had not been seated in front of an air bag.

The primary factor linking these deaths is the proximity to air bags at the time of their deployment. All of these deaths occurred under circumstances in which the occupant's upper body was very near the air bag when it deployed.

There were two other factors common to many of the deaths. First, apart from 12 infants fatally injured while riding in rear-facing infant seats, most of the fatally injured people were not using any type of child seat or seat belt. This allowed the people to move forward more readily than properly restrained occupants in a frontal crash. Further, the air bags involved in those deaths were, like almost all current air bags, so-called "one-size-fits-all" air bags that have a single inflation level.(5) These air bags deploy with the same force in very low speed crashes as they do in higher speed crashes.

The most direct behavioral solution to the problem of child fatalities from air bags is for children to be properly belted and placed in the back seat whenever possible, while the most direct behavioral solution for the adult fatalities is to use seat belts and move the driver seat back as far as practicable. Implementing these solutions necessitates increasing the percentage of children who are seated in the back and properly restrained in child safety seats. It also necessitates improving the current 68 percent rate of seat belt usage by a combination of methods, including the enactment of State primary seat belt use laws.(6)

The most direct technical solution to the problem of fatalities from air bags is to require that motor vehicle manufacturers install advanced air bags that protect occupants from the adverse effects that can occur from being too close to a deploying air bag.

All of these solutions are being pursued by the agency. However, until advanced air bags can be developed and incorporated into production vehicles, behavioral changes based on improved information and communication about potential hazards and simple, manually operated technology are the best means of addressing fatalities from air bags, especially those involving children.

To partially implement these solutions, and preserve the benefits of air bags, while reducing the risk of injury to certain people, NHTSA issued two other final rules in the past year. One rule requires new passenger cars and light trucks whose passenger air bags are not advanced to bear new, enhanced warning labels. (61 FR 60206; November 27, 1996) The other final rule provides vehicle manufacturers with the temporary option of ensuring compliance by conducting a sled test using an unbelted dummy instead of conducting a vehicle-to-barrier crash test using an unbelted dummy. (62 FR 12960; March 19, 1997) The purpose of the option is primarily to enable vehicle manufacturers to expedite their efforts to lessen the force of air bags as they deploy.

On the behavioral side, the agency has initiated a national campaign to increase usage of seat belts through the enactment of primary seat belt use laws, more public education, and more effective enforcement of existing belt use and child safety seat use laws.

In conjunction with the National Aeronautical and Space Administration, as well as Transport Canada, and in cooperation with domestic and foreign vehicle manufacturers, restraint system suppliers and others through the Motor Vehicle Safety Research Advisory Committee (MVSRAC), NHTSA is undertaking data analysis and research to address remaining questions concerning the development and introduction of advanced air bags. As noted above, the Federal motor vehicle safety standards have permitted, but not required, the introduction of advanced air bags. NHTSA recognizes that, if it were to require advanced air bags, it would have to take into consideration the differing leadtimes for the various kinds of advanced bags under development, and the fact that the longest leadtimes will be those for the most advanced bags. The agency also recognizes the engineering challenge and potential costs associated with incorporating some of the advanced air bag design features into the entire passenger car and light truck fleet. A proposal to require the installation of advanced air bags is expected this winter.

B. Background.

1. Air Bags: Safety Issues.

a. Lives Saved and Lost.

Air bags have proven to be highly effective in reducing fatalities from frontal crashes, the most prevalent fatality and injury-causing type of crash. Frontal crashes cause 64 percent of all driver and right-front passenger fatalities.

NHTSA estimates that, between 1986 and November 1, 1997, air bags have saved about 2,620 drivers and passengers (2,287 drivers (87 percent) and 332 passengers (23 percent)).(7) Of the 2,620, 1,800 (69 percent) were unbelted and 700 (31 percent) were belted. These agency estimates are based on comparisons of the frequency of front seat occupant deaths in vehicles without air bags and in vehicles with air bags. Approximately half of those lives were saved in the last two years. These savings occurred primarily in moderate and high speed crashes. Pursuant to the mandate in the Intermodal Surface Transportation Efficiency Act of 1991 (ISTEA) for the installation of air bags in all passenger cars and light trucks, the number of air bags in vehicles on the road will increase each year. As a result, the annual number of lives saved by air bags will continue to increase each year. Based on current levels of effectiveness, air bags will save more than 3,000 lives each year in passenger cars and light trucks when all light vehicles on the road are equipped with dual air bags. This estimate is based on current seat belt use rates (about 68 percent, according to State-reported surveys).

While air bags are saving large numbers of people in moderate and high speed crashes, they sometimes cause fatalities, especially to children, in lower speed crashes. As of November 1, 1997, NHTSA's Special Crash Investigation program had confirmed a total of 87 crashes in this country in which the deployment of an air bag resulted in fatal injuries. Forty-nine of those fatalities involved children. Three adult passengers have also been fatally injured. Thirty-five drivers are known to have been fatally injured.

In addition to the 87 confirmed air bag related deaths, there were 18 deaths under investigation, as of November 1, 1997, 1 involving a 1996 crash and 17 involving 1997 crashes. The single 1996 death still under investigation involved a driver. The 17 deaths in 1997 involved 1 infant, 11 children ranging in age from 1 to 11 years, and 5 drivers. Although the agency cannot predict how many of the deaths under investigation that will ultimately be categorized as confirmed air bag related deaths, the agency notes that roughly 80 percent of the deaths investigated to date have ultimately been confirmed.

The trends in the annual numbers of child and adult deaths differ significantly. The annual number of confirmed fatally-injured children increased significantly in 1993 through 1996 (1 in 1993, 5 in 1994, 8 in 1995 and 22 in 1996), while the number of confirmed fatally-injured drivers did not increase appreciably in the same period (4 in 1993, 7 in 1994, 4 in 1995, and 6 in 1996). As of November 1, 12 children and 6 drivers had been confirmed as having been fatally injured by air bags this year. However, as noted above, additional deaths are under investigation. The total number of confirmed deaths for this year will not be known until some time next year.

The number of vehicles with either driver air bags or both driver and passenger air bags increased steadily over the last four years. Since the fall of 1996, the number of vehicles with both driver and passenger air bags has been increasing at the rate of 1 million vehicles per month. The ratio of driver deaths to vehicles with driver air bags decreased significantly between 1993 and 1996. The ratio of child deaths to vehicles with passenger air bags also decreased, but not nearly so much.

b. Causes of Air Bag Fatalities.

The one fact that is common to all who died is not their height, weight, sex, or age. Instead, it is the fact that they were too close to the air bag when it started to deploy. For some, this occurred because they were sitting too close to the air bag. More often this occurred because they were not restrained by seat belts or child safety seats and were thrown forward during pre-crash braking.

Air bags are designed to save lives and prevent injuries by cushioning occupants as they move forward in a front-end crash. They keep the occupants' head, neck, and chest from hitting the steering wheel or dashboard. To accomplish this, an air bag must move into place quickly. The force of a deploying air bag is greatest in the first 2-3 inches after the air bag bursts through its cover and begins to inflate. Those 2-3 inches are the "risk zone." The force decreases as the air bag inflates further.

Occupants who are very close to or in contact with the cover of a stored air bag when the air bag begins to inflate can be hit with enough force to suffer serious injury or death. In contrast, occupants who are properly restrained and who sit 10 inches away from the air bag cover will contact the air bag only after it has completely or almost completely inflated. The air bag then will cushion and protect them from hitting hard surfaces in the vehicle and thus provide a significant safety benefit, particularly in moderate to serious crashes.

The confirmed fatalities involving children have a number of fairly consistent characteristics. First, all 12 infants were in rear-facing infant seats. Second, the vast majority of the older children were not using any type of restraint.(8) Third, almost all of the small number of older children who were using some type of restraint were improperly restrained or were leaning so far forward that benefits of being restrained were largely negated. For example, some were too small to be using just a vehicle lap and shoulder belt. Fourth, as noted above, the crashes occurred at relatively low speeds. If the passenger air bag had not deployed in those crashes, the children would probably not have been killed or seriously injured. Fifth, the infants and older children were very close to the dashboard when the air bag deployed. Properly installed rear-facing infant seats are always very close to the dashboard. For essentially all of the older children, the non-use or improper use of occupant restraints or the failure to use the restraints most appropriate to the child's weight and age, in conjunction with pre-impact braking, resulted in the forward movement of the children. (9) As a result, they were very close to the air bag when it deployed. Because of their proximity, the children sustained fatal head or neck injuries from the deploying passenger air bag.

As in the case of the children fatally injured by air bags, the key factor regarding the confirmed adult deaths has been their proximity to the air bag when it deployed. The most common reason for their proximity was failure to use seat belts. Only 11 of the 35 drivers were known to be properly restrained by lap and shoulder belts at the time of the crash. Moreover, of those eleven, two appeared to be out of position (blacked out, due to medical conditions, and slumped over the steering wheel) at the time of the crash. As in the case of children, the deaths of drivers have occurred primarily in low speed crashes.

The other cause of air bag fatalities is the design of current air bags. Air bag fatalities are not a problem inherent in the concept of air bags or in the agency's occupant restraint standard, Standard No. 208 (49 CFR 571.208). That standard has long permitted, but not required, a variety of design features that would reduce or eliminate the fatalities that have been occurring, e.g., higher deployment thresholds that will prevent deployment in low speed crashes,(10) different folding patterns and aspiration designs, dual stage inflators, (11) new air bag designs like the Autoliv "Gentle Bag" that deploys first radially and then toward the occupant, and advanced air bags that either adjust deployment force or suppress deployment altogether in appropriate circumstances. While some of these features are new or are still under development, others have been around for more than a decade. The agency identified a number of these features in conjunction with its 1984 decision concerning automatic occupant protection and noted that vehicle manufacturers could choose among those features to address the problems reported by those manufacturers concerning out-of-position occupants.

Although Standard No. 208 permits vehicle manufacturers to install air bags incorporating those advanced features, very few current air bags do so. Instead, vehicle manufacturers have thus far used designs that inflate with the same force under all circumstances. Although the vehicle manufacturers are now working to incorporate advanced features in their air bags, the introduction of air bags with those features is only just beginning. Introduction of significant numbers of advanced air bags may not begin for another several model years.

With the help of a recent amendment to Standard No. 208, vehicle manufacturers have been able to expedite the introduction of depowered air bags. While these new air bags will reduce, but not eliminate, the likelihood of air bag-caused deaths, they still deploy with the same force in all crashes, regardless of severity, and regardless of occupant weight or location. Many manufacturers have introduced substantial numbers of these less powerful air bags in the current model year (1998).

2. Air Bag Requirements.

Today's air bag requirements evolved over a 25-year period. NHTSA issued its first public notice concerning air bags in the late 1960's. However, it was not until the fall of 1996 that manufacturers were first required to install air bags in any motor vehicles.(12)

When the requirements for automatic protection (i.e., protection by means that require no action by the occupant) were adopted in 1984 for passenger cars, they were expressed in broad performance terms that provided vehicle manufacturers with choices of a variety of methods of providing automatic protection, including automatic belts and air bags. Further, the requirements allowed broad flexibility in selecting the performance characteristics of air bags. Later, those requirements were extended to light trucks. Ultimately, strong market demand led manufacturers to begin to install air bags in all of their passenger cars and light trucks.

In 1991, Congress included a provision in ISTEA directing NHTSA to amend Standard No. 208 to require that all passenger cars and light trucks provide automatic protection by means of air bags. ISTEA required at least 95 percent of each manufacturer's passenger cars manufactured on or after September 1, 1996, and before September 1, 1997, to be equipped with an air bag and a manual lap/shoulder belt at both the driver and right front passenger seating positions. Every passenger car manufactured on or after September 1, 1997, must be so equipped. The same basic requirements are phased-in for light trucks one year later.(13) The final rule implementing this provision of ISTEA was published in the Federal Register (58 FR 46551) on September 2, 1993.

Standard No. 208's automatic protection requirements, whether for air bags or (until the provisions of ISTEA fully take effect) for automatic belts, are performance requirements. The standard does not specify the design of an air bag. Instead, vehicles must meet specified injury criteria, including criteria for the head and chest, measured on test dummies. Until recently, these criteria had to be met for air bag-equipped vehicles in barrier crashes at speeds up to 30 mph, both with the dummies belted and with them unbelted.

However, on March 19, 1997, the agency published a final rule amending Standard No. 208 to temporarily provide the option of testing air bag performance with an unbelted dummy in a sled test incorporating a 125 millisecond standardized crash pulse instead of in a vehicle-to-barrier crash test. This amendment was made primarily to expedite manufacturer efforts to reduce the force of air bags as they deploy.

Standard No. 208's current automatic protection requirements, like those established 13 years ago in 1984, apply to the performance of the vehicle as a whole, and not to the air bag as a separate item of motor vehicle equipment. The broad vehicle performance requirements permit vehicle manufacturers to "tune" the performance of the air bag to the specific attributes of each of their vehicles.

The Standard's requirements also permit manufacturers to design seat belts and air bags to work together. Before air bags, seat belts had to do all the work of restraining an occupant and reducing the likelihood that the occupant will strike the interior of the vehicle in a frontal crash. Another consequence of not having air bags was that vehicle manufacturers had to use relatively rigid and unyielding seat belts that can concentrate a lot of force along a narrow portion of the belted occupant's body in a serious crash. This concentration of force created a risk of bone fractures and injury to underlying organs. The presence of an air bag increases the vehicle manufacturer's ability to protect belted occupants. Through using energy managing devices, such as load limiters, a manufacturer can design seat belts to give or release additional belt webbing before the belts can concentrate too much force on the belted occupant's body. When these new belts give, the deployed air bag is there to prevent the belted occupant from striking the vehicle interior.

Further, Standard No. 208 permits, but does not require, vehicle manufacturers to design their air bags to minimize the risk of serious injury to unbelted, out-of-position occupants, including children and small drivers. The standard gives the manufacturers significant freedom to select specific attributes to protect all occupants, including attributes such as the crash speeds at which the air bags deploy, the force with which they deploy, air bag tethering and venting to reduce inflation force when a deploying air bag encounters an occupant close to steering wheel or dashboard, the use of sensors to detect the presence of rear-facing child restraints or the presence of small children and prevent air bag inflation, the use of sensors to detect occupant position and prevent air bag inflation if appropriate, and the use of dual stage versus single stage inflators. Dual stage inflators enable air bags to deploy with lower force in low speed crashes, the type of crashes in which children and drivers have been fatally-injured, and with more force in higher speed crashes.

C. Comprehensive Agency Plan to Address Air Bag Fatalities.

In late November 1996, NHTSA announced that it would be implementing a comprehensive plan of rulemaking and other actions (e.g., consumer education and encouragement of State seat belt use laws providing for primary enforcement of their requirements) addressing the adverse effects of air bags.(14) While there is a general consensus that the best approach to preserving the benefits of air bags while preventing air bag fatalities will ultimately be the introduction of advanced air bags, those air bags will not be widely available in the next several years. Accordingly, the agency has focused on rulemaking and other actions that will help reduce the adverse effects of air bags in existing vehicles as well as in vehicles produced during the next several model years. The actions which have been taken, or are being taken, include the following:

1. Interim Rulemaking Solutions.

a. Existing and Future Vehicles- in-Use.

This final rule exempts, under certain conditions, motor vehicle dealers and repair businesses from the "make inoperative" prohibition in 49 U.S.C. 30122 by allowing them, beginning January 19, 1998, to install retrofit manual on-off switches for air bags in vehicles owned by people whose request for a switch is approved by NHTSA. The purpose of the exemption is to preserve the benefits of air bags while reducing the risk that some people have of being seriously or fatally injured by current air bags. The exemption also allows consumers to have new vehicles retrofitted with on-off switches after the purchase of those vehicles. It does not, however, allow consumers to purchase new vehicles already equipped with on-off switches.

b. New Vehicles.

On March 19, 1997, NHTSA published in the Federal Register (62 FR 12960) a final rule temporarily amending Standard No. 208 to facilitate efforts of vehicle manufacturers to depower their air bags quickly so that they inflate less aggressively. This change, coupled with the broad flexibility already provided by the standard's existing performance requirements, provided the vehicle manufacturers maximum flexibility to quickly reduce the adverse effects of current air bags.

On November 27, 1996, the agency published in the Federal Register (61 FR 60206) a final rule amending Standards No. 208 and No. 213 to require improved labeling on new vehicles and child restraints to better ensure that drivers and other occupants are aware of the dangers posed by passenger air bags to children, particularly to children in rear-facing infant restraints in vehicles with operational passenger air bags. The improved labels were required on new vehicles beginning February 25, 1997, and were required on child restraints beginning May 27, 1997.

On January 6, 1997, the agency published in the Federal Register (62 FR 798) a final rule extending until September 1, 2000, an existing provision in Standard No. 208 permitting vehicle manufacturers to offer manual on-off switches for the passenger air bag for new vehicles without rear seats or with rear seats that are too small to accommodate rear-facing infant restraints.

2. Longer-Term Rulemaking Solution.

The longer term solution is advanced air bags. The agency has established a working group under the Crashworthiness Subcommittee of MVSRAC to work cooperatively with the vehicle manufacturers, restraint system suppliers and other organizations regarding advanced air bags. Activities include sharing data and information from research, development and testing of advanced air bags and providing test procedures that could be used in evaluating the advanced air bag technologies. While some of these technologies are complex, others are relatively simple and inexpensive. NHTSA plans to issue an NPRM to require a phasing-in of advanced air bags and to establish performance requirements for those air bags. While Standard No. 208 has provided vehicle manufacturers with the flexibility necessary to introduce advanced air bags, the Standard has not required them to take advantage of that flexibility. Among other things, the agency anticipates proposing tests using a 5th percentile female dummy (15) and advanced child dummies and specify appropriate injury criteria for those dummies, including neck injury criteria, as part of its rulemaking regarding advanced air bags.

3. Educational Efforts; Child Restraint and Seat Belt Use Laws.

In addition to taking these actions, and conducting extensive public education efforts, the Department of Transportation announced this past spring a national strategy to increase seat belt and child seat use. Higher use rates would decrease air bag fatalities and the chance of adverse safety tradeoffs occurring as a result of turning off air bags. The plan to increase seat belt and child seat use has four elements: stronger public-private partnerships; stronger State seat belt and child seat use laws (e.g., laws providing for primary enforcement of seat belt use requirements); active, high-visibility enforcement of these laws; and effective public education. Substantial benefits could be obtained from achieving higher seat belt use rates. For example, if observed belt use increased from 68 percent to 90 percent, an estimated additional 5,536 lives would be saved annually over the estimated 9,529 lives currently being saved by seat belts. In addition, an estimated 132,670 injuries would be prevented annually. The economic savings from these incremental reductions in both fatalities and injuries would be $8.8 billion annually.








4. The vast majority of the deaths appear to have occurred in crashes in which the vehicle was traveling at less than 15 miles per hour when the air bag deployed. Almost all occurred at vehicle speeds under 20 miles per hour. NHTSA notes that Federal safety standards do not specify a vehicle crash speed at which air bags must deploy.

5. The Federal safety standards do not require a "one-size-fits-all" approach to designing air bags. They permit a wide variety of technologies that would enable air bags to deploy with less force in lower speed crashes or when occupants are out-of-position or suppress deployment altogether in appropriate circumstances.

6. In States with "secondary" seat belt use laws, a motorist may be ticketed for failure to wear a seat belt only if there is a separate basis for stopping the motorist, such as the violation of a separate traffic law. This hampers enforcement of the law. In States with primary laws, a citation can be issued solely because of failure to wear seat belts.

7. Studies published in the November 5, 1997 issue of the Journal of the American Medical Association by IIHS and by the Center for Risk Analysis at the Harvard School of Public Health confirm the overall value of passenger air bags, while urging action be taken quickly to address the loss of children's lives due to those air bags. IIHS found that passenger air bags were associated with a substantial reduction in crash deaths. The Center evaluated the cost-effectiveness of passenger air bags and concluded that they produce savings at costs comparable to many well-accepted medical and public health practices.

8. 29 (or 78%) of the 37 forward-facing children who were fatally injured by air bags were not using any type of belt or other restraint. This included 4 children who were sitting on the laps of other occupants. The remaining 8 children included some who were riding with their shoulder belts behind them and some who were wearing lap and shoulder belts but who also should have been in booster seats because of their small size and weight. Booster seat use could have improved shoulder belt fit and performance. These various factors and pre-crash braking allowed the children to get too close to the air bag when it began to inflate.

9. For information on the restraint most appropriate for a particular child, see the table at the end of the information brochure in Appendix A in the regulatory text.

10. Mercedes Benz offers passenger air bags whose deployment threshold is 12 mph if the passenger is unbelted and 18 mph if the passenger is belted.

11. The air bags installed in approximately 10,000 GM cars in the 1970's were equipped with dual stage inflators. Today, Autoliv, a Swedish manufacturer of air bags, has a "gas generator that inflates in two steps, giving the bag time to unfold and the vent holes to be freed before the second inflation starts. Should the bag then encounter an occupant, any excessive gas -- and indeed bag pressure -- will exit through the vent holes."

12. Air bag firsts--In view of the confusion evident in some public comments on this rulemaking and even now in some media accounts about when air bags were first required, and by whom, the agency has set forth a brief chronology below:

13. At least 80 percent of each manufacturer's light trucks manufactured on or after September 1, 1997 and before September 1, 1998 must be equipped with an air bag and a manual lap/shoulder belt. Every light truck manufactured on or after September 1, 1998 must be so equipped.

14. For a discussion of the actions taken by NHTSA before November 1996 to address the adverse effects of air bags, see pp. 40787-88 of the agency's NPRM published August 6, 1996 (61 FR 40784).

15. A 5th percentile female dummy has a standing height of 5 feet and a weight of 110 pounds.


[Back to section 1]   [Table of Contents]   [Next Section]