The 'smart' air bag case: air bags are intended to save lives, but they may also cause injury. As air bag technology has become more complex, so has air bag litigation.

Today, we live in the so-called smart air bag era. Overall, air bags fire with significantly less force than they did just a few years ago, and air bag systems are designed to detect children in the front seat and drivers sitting too close to the air bag. In such cases, air bags may either not deploy or deploy with minimal force.

Depending on the severity of a crash, air bags may deploy at different levels of force: High-speed crashes require air bags to deploy more aggressively than low-speed crashes do. All these changes were implemented after an outcry over air-bag-related deaths, mostly of children and women, in the 1990s.

In theory, air bag litigation should be assigned to the ranks of cases like those involving breast implants and intrauterine devices--cases that faded away as the products got safer or were removed from the marketplace. However, that has not proved to be the case. The sad fact is that air bags remain dangerous and continue to cause injuries and deaths. While regulations and air bags have changed, auto manufacturers continue to cut corners to save costs. Side air bags have also come into play, and they are completely unregulated by the government.

Standard 208

Federal Motor Vehicle Safety Standard 208 is the controlling federal regulation with which auto manufacturers must comply. (1) In 1984, it specified that by 1987 to 1990, manufacturers had to phase in passive restraints, which include air bags. It was later amended to allow manufacturers to delay including passenger air bags in cars until the 1994 model year, if the driver's side of those cars did have an air bag. Additional amendments required vans and light trucks to include air bags. By 1994, most passenger cars had both a driver and passenger side air bag, and by 1998 the same was true for vans and light trucks. (2)

FMVSS 208 has undergone several changes, and different versions apply to any given case, depending on the vehicle's model year. The basic changes lie in the type of testing and the warnings to be posted in the vehicle.

Warnings were not mandatory in vehicles until 1994. (3) Then, warnings on sun visors were required to warn occupants to "not install rearward-facing child seats in any front passenger seat position," "not sit or lean unnecessarily close to the air bag," and "not place any objects over the air bag or between the air bag and yourself." Occupants were also told, "For maximum protection in all types of crashes, you must always wear your seat belt." In addition, manufacturers had to include information and warnings about the air bag in the owner's manual. (4)

Certain changes were made for vehicles manufactured after February 25, 1997; probably the most basic one was that occupants were no longer warned to "not sit or lean unnecessarily close to the air bag." Instead, they were told to "sit as far back as possible from the air bag." (5) Also, temporary labels had to be affixed to the steering wheel or dashboard on the passenger side with warnings similar to those on the visor. (6)

Originally, manufacturers had to perform frontal barrier crash tests-crashing a vehicle into a fixed barrier at speeds up to and including 30 mph. The barrier had to be perpendicular to the line of travel or at any angle up to 30 degrees left or right. The tests were to be run with seat-belted and unbelted 50th-percentile male dummies.

Generally, manufacturers complied with the regulations by running six tests: a first set with belted dummies in vehicles traveling at 90 degrees, 30 degrees right, and 30 degrees left. The second set used unbelted dummies at the same angles. In each case, certain injury criteria, as measured by sensors in the dummies, had to be met. Accelerations to the head and forces to the chest and the femur could not exceed specified limits. (7)

In March 1997, the National Highway Traffic Safety Administration (NHTSA) adopted a rule to address the urgent problem of air bags causing injuries in low-speed crashes, expediting the development of depowered air bags. (8) Under the rule, which was intended to be temporary, the unbelted crash-test requirements were amended so that manufacturers could conduct sled tests instead of barrier crash tests, which are more expensive. (9) In addition, neck-injury criteria were added to the test, including measurements of flexion and extension bending, as well as axial tension and axial compression.

Then, in May 2000, NHTSA adopted what is commonly referred to as the smart air bag regulation. The rule purports to "improve protection and minimize risk by requiring new tests and injury criteria and specifying the use of an entire family of test dummies." (10) It was to be phased in during two stages: the first to cover the period from September 1,2003, to August 31, 2006, and the second to cover the period from September 1, 2007, to August 31, 2010.

The rule is complex, and there are numerous different options for testing. For this discussion, the rule may be summarized best by dividing it into frontal crash tests and out-of-position tests (in which air bag systems are to either suppress the air bag completely or deploy it with low risk).

Frontal crash tests include:

* 30-mph belted tests into a rigid barrier with both female and male dummies in the perpendicular crash

* 25-mph unbelted tests into a rigid barrier with both female and male dummies (11)

* the addition of a 25-mph offset deformable barrier using a fifth-percentile dummy (meant to represent a woman). (12)

Out-of-position tests include both drivers and passengers. On the driver's side, a female test dummy is to be used. On the passenger side, dummies representing one-, three-, and six-year-old children are to be used. For each test, the air bag must either be suppressed or meet injury criteria applicable to each dummy.

All these frontal crash tests and out-of-position tests were to be completed during the first phase-in period. The only change to be made during the second phase-in period is that the rigid-barrier belted test is to be done at 35 mph instead of 30 mph, and it eliminates the need to use a fifth percentile dummy, using only the male dummy.

FMVSS 208 shapes and informs air bag litigation. It raises its head in countless ways. Given that the standard is a federal regulation, an immediate concern is whether it preempts any litigation at all. For the most part it does not. (13) The types of air bag cases described below are all actionable. To date, the only type of air bag case preempted by FMVSS 208 is the claim that a car is defective if it is not equipped with a frontal air bag. (14)

The Sixth Circuit has held that auto manufacturers need not add any warnings to the inside of the vehicle, except those that FMVSS 208 requires on the visors. (15)

In dealing with FMVSS 208, it is important to note that the very meaning of a "motor vehicle safety standard" is "minimum standard for motor vehicle or motor vehicle performance." (16) Moreover, NHTSA itself has stressed that FMSS 208 is only a performance standard, not a design standard. (17) That is, NHTSA does not require any particular design; rather, it mandates only a minimum performance standard with which to comply, regardless of the particular design chosen.

Air bag anatomy

An air bag might cause injury because it inflates too quickly, because it is too large, because it fails to use tethers (internal straps that prevent the air bag from pushing too far back in the passenger compartment), or because its folding pattern is dangerous.

NHTSA recognized that these characteristics, among others, directly affect air bag safety, and in 1997, the agency queried auto manufacturers about the engineering specifications of their air bags. It published its findings in 1999. (18) Since this study was published, any air bag's characteristics can be compared to those of others--providing a very useful method of determining whether any given air bag is an outlier.

Sensors are the brains of the air bag system. They detect the collision and tell the air bag whether, when, and how forcefully to fire. But their placement and calibration are problematic, sometimes causing air bags to fail to deploy when they should, to deploy late (allowing occupants to get too close to the air bag), and to deploy in minor collisions when they should not deploy at all. Consequently, claims of sensor system defects are common in air bag cases.

Auto manufacturers use many types of systems. They vary in the number and placement of sensors they employ and in the amount and type of testing they undergo.

Related to the sensor system are deployment thresholds. If the manufacturer sets the thresholds too low, air bags deploy in relatively minor collisions. Every air bag system has "must-fire," "may-fire," and "no-fire" thresholds. For example, both Ford and Chrysler in the 1990s designed their-vehicles so that air bags must fire in any collision with a barrier equivalent velocity (BEV) of 14 mph. (19) They specified that the air bag must not fire in any collision with a BEV of 8 mph or less and that it may fire in any collision with a BEV greater than 8 mph and less than 14 mph. (20) The may-fire zone is also known as the gray zone.

A common allegation in minor-collision cases is that the deployment threshold was set too low. This claim is based on dictates set by the Society of Automotive Engineers: "A restraint system or component which is stored remotely and moved rapidly into place immediately before or during an impact should to the extent possible not pose a greater hazard to the vehicle occupants or to service personnel than if the system were not present...." (21)

Auto manufacturers themselves started raising the thresholds in 2000. For example, the 2000 Ford Taurus's deployment threshold was raised to 18 mph and the no-fire threshold to 14 mph if the occupant is seat-belted. For unbelted occupants, the no-fire threshold was raised to 11 mph. Today, several vehicles have higher thresholds.

An inflator is the component that inflates the air bag. Some early types used pyrotechnic squibs to inflate sodium azide, and others used stored gases. One technology to mitigate the force of inflation is a dual-level inflator, which regulates the force of the air bag deployment, depending on numerous factors. Typically, the inflator is tied to a collision's severity. Air bags need to inflate most quickly in severe collisions, as the occupants move forward more quickly than they do in less forceful collisions.

With a dual-level inflator, if the sensor system detects a severe collision, the air bag deploys aggressively; if the collision is mild, the air bag deploys less forcefully. For example, the 2001 Mercury Marquis air bag inflates at reduced power when the crash has the force of a BEV of 17, and it inflates more aggressively when the collision has a BEV of 31 mph. (22)

Under the latest version of FMVSS 208, dual-level inflators are one option for dealing with the presence of children or drivers sitting too close to the steering wheel. In these situations, the manufacturer may either use a dual-level inflator to inflate the air bag less aggressively or suppress deployment altogether.

Defective-design claims

Air bag cases generally fit into two categories: deployment and nondeployment cases. Both an air bag that deploys too forcefully and an air bag's failure to deploy may cause injury. Beyond these broad generalizations, however, case specifics vary greatly.

Swanson v. Nissan is one example of a deployment-injury case. (23) Norma Swanson was a passenger in a 1994 Nissan Altima when it ran over a curb in a minor collision, causing the air bags to deploy. The passenger air bag smashed directly into her eyes, blinding her immediately. (24) A simple examination of the vehicle clearly showed her eye makeup on the air bag.

Further investigation showed why Swanson's injury occurred: The air bag was simply too big. When it deployed, it struck Swanson while still inflating. Also, its folding pattern directed the air bag right into her eyes.

Swanson claimed that Nissan engineers knew about this defect. At trial, the plaintiff presented photos showing crash tests in which the air bag hit crash dummies in the eyes. The case was settled before it went to the jury.

The basic premise underlying Swanson--and all deployment-injury cases--is the well-recognized axiom that an air bag should never strike an occupant while still inflating. Instead, the occupant should fall into a fully inflated air bag. (25)

Crash sensor claims

In Foster v. General Motors, Robert Foster was driving his 2002 Buick Century when he rear-ended a vehicle that had stopped to make a left turn. (26) Although the crash was severe, his air bag failed to deploy. He suffered devastating head and spinal injuries and died three weeks later.

Foster's estate sued General Motors, claiming that the sensor system was defective. Crash data retrieved from the car's black box proved invaluable in supporting the case. It not only confirmed that Foster was wearing his seat belt at the time of the crash, but it also demonstrated that the crash was significantly more severe than GM's air bag deployment threshold. The damage to the vehicle was also significantly greater than the damage to other vehicles in cases where GM had argued that air bag deployment was proper and warranted.

Plaintiff counsel's investigation revealed that Foster's Buick Century lacked a crash sensor in the engine compartment, although older models had them. Further investigation found that GM included a crash sensor at the front of the car, near the radiator, in Chevrolet versions of this vehicle platform--but not in Buick versions. Although GM had included predrilled screw holes for attaching such a sensor near the radiator, and although it had molded a plastic cover to accommodate the device, it failed to include the sensor in the Buick models. The parties settled the case early in discovery.

Gaytan v. DaimlerChrysler provides another example of the issues that can arise with a sensor system. (27) The plaintiff, Maria Gaytan, was driving a 2002 Dodge Neon and rear-ended a vehicle at a low speed. Her air bag deployed.

Although Gaytan was belted and her seat was midway back in the seat track, the air bag struck her in the eye. The deployment was so forceful that her right eye was crushed, and it had to be surgically removed and replaced with a prosthetic.

The eyeglasses Gaytan was wearing were broken into pieces. The low-speed collision would not have been forceful enough to crush her glasses or her eye; only the force of an air bag could have caused the injury.

However, there is a crucial difference between this case and Swanson. While the Swanson crash test film showed the air bag hitting the crash dummy before it was filly inflated, the crash tests for Gaytan's Dodge Neon did not. In fact, the air bag appeared to do exactly what it was supposed to do--inflate fully before the crash dummy fell into it. The question became how Gaytan was hurt by the air bag when crash tests showed it fully inflating as it should.

Sensors normally give the signal to fire the air bag within about 20 to 30 milliseconds of when the collision begins--before crash forces can move the occupant forward significantly. But if the air bag fires late--perhaps 50 to 80 milliseconds after the crash commences--the vehicle's occupant moves directly into the path of the inflating air bag.

Further investigation of the sensor system in Gaytan showed that it used a single-point sensor system--that is, a single sensor to detect and interpret crashes, located in the passenger compartment. Single-point sensor systems are notorious for their unreliability. (28) In low-speed crashes, as in Gaytan, they do not pick up the crash signal quickly enough to fire the air bags in time. They may also fail to interpret crash signals properly and order an air bag to fire when the collision is below the no-fire threshold.

According to Gaytan's experts, the single-point sensor system in the Neon was at the heart of the problem. It ordered the air bag to fire late, when Gaytan had already moved into the bag's path. Also, it ordered the air bag to fire even though the "delta v" (change in velocity at impact) was below the air bag's no-fire threshold. The experts argued that a good sensor system has sensors at the front of the vehicle, where they can receive a crash signal early and properly interpret it.

The case settled once plaintiff counsel located an offset deformable barrier test, which is meant to replicate a softer crash, such as a car-to-car crash. (29) The test demonstrated that while the air bag should have fired within 31 milliseconds, it actually fired at 68 milliseconds--37 milliseconds too late. In that time, the dummy had moved far too close to the air bag. When the air bag did fire, the dummy's head and neck were pushed far back and to the left. Sensors in the neck showed that the criteria for a safe deployment were far exceeded.

In Bock v. General Motors, 76-year-old Marjorie Bock was attempting a left turn and failed to see another car, which struck her 1997 Chevrolet Malibu on the right passenger side. Bock's air bag went off, striking her in the face. The impact caused a subdural bleed in her brain, resulting in paralysis on her left side and speech deficits. (30)

Air bags are not supposed to deploy in side-impact collisions, yet they do to a surprising extent. Even NHTSA has identified this problem and published an article about it for the Society of Automotive Engineers. (31)

In Bock, it was clear that the air bag should not have deployed as a result of the side impact. Download of the car's black-box data showed a maximum delta v of 5.3 mph. It also showed that the air bag had fired within 25 milliseconds of impact, at a velocity change of less than 2 mph.

Investigation showed that the vehicle had only a single-point sensor rather than remote sensors on the sides of the vehicle. Bock sued General Motors, claiming that the sensor system was defective.

Documents produced in discovery designated the vehicle's single-point sensor system as the "reduced cost" sensor. The plaintiff learned that, for cost reasons, the manufacturer removed the front-discriminating sensor originally designed for the vehicle. This essentially left the sensor system with a "hair trigger," capable of firing the air bag even in side collisions. (32)

The plaintiff also alleged that GM had misled NHTSA when the agency launched an investigation into inadvertent deployments in Malibus and similar vehicles. The case was settled before trial.

Side air bags

Air bag litigation now increasingly involves claims related to side air bags--typically, either the manufacturer's failure to install a side air bag system or the failure of the side air bags to deploy properly.

In Estate of Dereck Lopez v. General Motors, a drunk driver ran a red light and struck the driver's side of a 1998 Chevrolet Cavalier, driven by Dereck Lopez. (33) The Cavalier was not equipped with side air bags for head protection. Lopez suffered severe head injuries, which were her only injuries, and she died six days later without ever regaining consciousness. (34)

The plaintiff sued GM, claiming that the vehicle should have been equipped with a side air bag. GM asserted that the accident was caused solely by the drunk driver's negligence, that the vehicle met the relevant government standard for side-impact protection--FMVSS 214--without a side air bag, that few vehicles in 1998 used side air bags for head protection, and that the forces from the truck impact to the driver's side were too severe for a side air bag to avoid Dereck's brain injuries and death. The case was settled on the eve of trial.

Discovery in Lopez and another case, Burry v. General Motors, (35) revealed that GM engineers had recommended that side air bags be introduced as early as 1994 and that a head-thorax side air bag system be adopted as a standard feature in 1996.

The plaintiff in Burry contended that he was injured in his 2001 Suburban for two reasons: first, that a side air bag, sometimes called a torso bag, failed to deploy when it should have; and second, that the vehicle failed to have an air bag or curtain to protect the head.

At trial, the plaintiff used exhibits to show that GM's management did not want to include that head protection on all vehicles and instead chose to use the torso bag as a marketing device on vehicles that the company perceived to be lagging behind the competition. The plaintiff also argued that GM cut corners in developing its single-remote-sensor side air bag system, and that the company should have implemented a multiple-remote-sensor system at a cost of $6 per vehicle. The case ended with a verdict for the plaintiff.

In Kaplan v. DaimlerChrysler, 74-year-old Elwood Kaplan was driving his brand-new 1999 Mercedes Benz $400 through an intersection when the car was struck on the driver's door by a Dodge Durango. (36) Although Kaplan's car was equipped with a side air bag mounted inside the door, the bag failed to deploy. The force of the impact shattered the window in the driver's door, and Kaplan's left arm went outside the window and was ripped off at the shoulder by the Durango.

DaimlerChrysler claimed that the collision was not severe enough to warrant air bag deployment, and that even if it had deployed, the bag was designed to protect only the driver's torso, not his arm.

The plaintiff argued that the wire that was supposed to carry the electric impulse to trigger the bag was severed in the collision because it was placed, unprotected, behind the sheet metal of the door's exterior. A diagram of a fully deployed Durango side air bag clearly showed the bag pressed up against the driver's left arm, keeping it inside the vehicle and protected from harm in a collision. Also, the Kaplan vehicle was much more severely damaged than DaimlerChrysler test vehicles were in less severe events in which the company's internal videos showed the air bag deploying. The jury found in the plaintiff's favor.

Automatic suppression system

The latest version of FMVSS 208 requires that, under certain conditions, the air bag must be depowered or suppressed. Generally, the system must suppress or depower the air bag for adult drivers too close to the air bag or for children in the passenger seat. (37) Unfortunately, the complexity of this system can cause problems of its own.

In Banden v. General Motors, a passenger was injured when the air bag failed to deploy in a moderately severe collision. (38) Information from the black box showed that the problem was the automatic suppression system: It suppressed the air bag, but it should not have, as the passenger was a 28-year-old adult and the seat was all the way back in its track. Had the air bags fired, Banden would never have suffered a closed head injury that resulted in a coma.

Air bag cases are expensive and difficult to prosecute. But when the facts support a claim, careful consideration of air bag issues may lead to a recovery that can make all the difference in an injured client's quality of life.

Despite the advent of so-called smart air bags, air bag litigation is expanding rather than shrinking. Motor vehicle practitioners would be well advised to consider air bag issues in any auto crash case.

Notes

(1.) 49 C.F.R. [section] 571.208 (2005).

(2.) See generally 58 Fed. Reg. 46551 (Sept. 2, 1993).

(3.) 49 C.F.R. [section] 571.208 S4.5.1 (b) (1) (1994).

(4.) Id. at [section] 571.208 S4.5.1 (e).

(5). Id. at [section] 571.208 S4.5.1 (b) (2).

(6.) Id. at [section] 571.208 S4.5.1 (e); see also 61 Fed. Reg. 60206 (Nov. 26, 1996).

(7.) A so-called head injury criteria (HIC) score could not exceed 1,000, thoracic accelerations could not exceed 60 g's, compression deflection of the dummy's chest could not exceed 3 inches, and the forces applied through the femur could not exceed 2,250 pounds. See generally 49 C.F.R. [section] 571.208 S5, S6 (1994).

(8.) 62 Fed. Reg. 12960-01 (Mar. 19, 1997).

(9.) In a sled test, a vehicle is placed on a sled, the dummies are placed in position in the seats, and the sled is pulled (jerked) backward according to a certain pulse. The test is said to simulate the forces placed on a dummy in a barrier test.

(10.) 65 Fed. Reg. 30680, 30685 (May 12, 2000).

(11.) The requirement that the unbelted test be done at 25 mph is said to be an interim rule. Consideration will be given to changing the speed to 30 mph as data is collected. (65 Fed. Reg. 30680, 30687 (May 12, 2000)). Note that this test replaces the sled test for unbelted dummies.

(12.) A deformable barrier is a barrier that is soft and deforms relatively slowly as it absorbs the energy of the crash. An offset test is one in which the vehicle is driven into the barrier so that only a portion of the front of the vehicle--one side of the vehicle--strikes the barrier. The current rule requires the driver's side to strike the barrier. Videos of offset tests can be seen on the Web site for the Insurance Institute for Highway Safety at www.iihs.org/ratings/frontal_test_info.html (last accessed Dec. 22, 2006).

(13.) See Perry v. Mercedes Benz, 957 F.2d 1257, 1266 (5th Cir. 1992).

(14.) See Geier v. American Honda Motor Co., 529 U.S. 861 (2000). The case does spell out an exception, so a closer review is necessary if considering a similar case.

(15.) See Fisher v. Ford Motor Co., 224 F. 3d 570, 574 (6th Cir. 2000). For an analysis of this case, see Lawrence Baron & Robert C. Sanders, No Dead End for Air Bag Cases, TRIAL 52 (Feb. 2001).

(16.) 49 U.S.C. [section] 30102(a) (9) (2000) (emphasis added).

(17.) 62 Fed. Reg. 2996-03, 2998 (Jan. 21, 1997).

(18.) See Air Bag Technology in Light Passenger Vehicles, NHTSA Office of Research and Development (2001).

(19.) A 14 mph BEV collision has the same force as a collision into a solid wall at 14 mph.

(20.) Determining any particular air bag deployment threshold is hit or miss. Because of the many cases against Ford and Chrysler, their 1990s deployment thresholds are well known. Deployment thresholds can be found in the owner's manual of some vehicles, but most vehicles' thresholds are not in any vehicle literature and are often regarded as proprietary information. In some cases, deployment thresholds can be found in a manufacturer's correspondence with NHTSA, through communication with attorneys who may have had cases involving the vehicle at issue, or through discovery.

(21.) Socy. Automotive Engrs. Information Report J128, Occupant Restraint System Evaluation--Passenger Cars and Light-Duty Trucks [section] 3.1.1.5 (Nov. 1994).

(22.) These must-fire thresholds apply only to a seat-belted occupant.

(23.) No. CV97-315-ST (D. Or. filed Feb. 2, 1997).

(24.) Later, Swanson recovered vision in one eye.

(25.) See Atsumi Kikuchi, Injury to Eye and Facial Skin (Rabbit) on Impact with Inflating Air Bag, 2d Intl. Conf. of Biomechanics of Serious Trauma (1975).

(26.) No. 8:04-CV-2499-T-26TBM (M.D. Fla. filed Nov. 17, 2004).

(27.) No. 6:05-CV-1763-AA (D. Or. filed Nov. 23, 2005).

(28.) David Breed et al, A Critique of Single Point Sensor Sensing, Socy. Automotive Engrs. Paper No. 920124 (1992).

(29.) Many experts believe that an offset deformable barrier test does not replicate a car-to-car crash well enough. Ford actually performs car-to-car crashes as part of its sensor development.

(30.) No. 01CV680420 (Mo., Jasper Co. Cir. filed Apr. 3, 2001).

(31.) Han Sun Chan et al., Frontal Airbag Deployment in Side Crashes, Socy. Automotive Engrs. Tech. Paper No. 980910 (1998).

(32.) One problem with single-point sensors is that without sensors at the front of the car, their sensitivity must be enhanced to be able to detect low-impact collisions.

(33.) No. 17-197104-03 (Tex., Tarrant Co. Dist. filed Jan. 21, 2003)).

(34.) Danny Hakim & Norm Alster, Lawsuits: This Year's Model, Crashes with Cars Join Rollovers in Legal Fight over Light Trucks, N.Y. Times (May 30, 2004).

(35.) 203 S.W.3d 514 (Tex. App. 2006).

(36.) 77 Fed. Appx. 507 (table), 2003 WL 22023325 (11th Cir. 2003) (unpublished).

(37.) 65 Fed. Reg. 30680, 30755-68 (May 12, 2000).

(38.) No. 06-CA-1364-10G (Fla., Seminole Co. Cir. filed July 21, 2006). At press time, this case was pending.

LAWRENCE BARON practices law in Portland, Oregon.