Child restraint system defects.

Bad designs and inadequate testing lead to ill-fitting safety seats and injured children.

The mandatory use of government-approved child restraint systems, starting in Tennessee in 1978, has led to the proliferation of child-seat designs today. Many of these systems are inadequate, leading to unnecessary injuries to the most vulnerable--innocent children.

Consumers are using child restraint systems that can be categorized into five basic designs: rear-facing infant seats, convertible seats, low-shield booster seats, belt-positioning booster seats, and integrated seats. The restraints are generally designed for a specific range of sizes, often designated by age, to provide the best possible fit and crash load distribution, based on the physical characteristics of a child. But many of these systems are unsuitable for some or all of their intended occupants. Others provide superior protection in various types of crashes.

Rear-facing infant seats. For several reasons, the youngest children are best protected when they face rearward in the vehicle. First, more than 50 percent of crashes are frontal impacts. In these crashes, occupants move forward as the vehicle decelerates. Second, given infants' weak skeletal structure, facing rearward in most accidents spreads the crash impact across their backs, the maximum surface area of their bodies, rather than concentrating forces into a smaller area by the belts. These seats also keep the child's head, neck, and chest in alignment, thereby reducing the chance of spinal cord damage. The rear-facing seat provides for early engagement of the restraint and maximum spreading of crash forces.

But the problem with this and other child seat designs is that parents may have difficulty securing the seat to the vehicle using the vehicle's seat belt. Most injuries sustained in infant car seats are a result of not properly securing the seat. The better designs have a separate base module that can be installed in the center rear seat position, the safest position because of its distance from the front and sides of the vehicle. The base can be installed securely. The child carrier unit can then be snapped in and out, reducing installation errors and the hazards associated with them.

But be aware that some seats with separate bases have been recalled due to base-to-carrier attachment defects that resulted in separation during crashes.

Rear-facing infant seats are generally designed for children who weigh up to 20 pounds and are up to one year old. Although it is generally agreed that when children weigh more and are older they can be allowed to face forward, the longer the child is in a seat facing backward, the better. The rear-facing position affords the best protection in higher speed crashes.(1)

A serious hazard exists when a rearfacing seat is used in a front passenger seat position in a car equipped with an air bag. A deployed bag will slam against the seat, causing serious harm or death to the child. A rash of these deaths led the government to mandate that removable stickers warning of this hazard be placed on dashboards of new cars and permanent stickers on their sun visors. Most manufacturers mailed the same stickers to owners driving cars with passenger-side air bags in early 1997. Some automakers will provide, or will retrofit, cars with air bag shutoff switches that will disable the bag.

There are other technologies to detect the presence of a child seat in the front passenger position and prevent air bag deployment. For example, Mercedes sells a rearfacing infant seat with a small device that alerts the car to the seat's presence and automatically disables the passenger-side bag. GM engineers have described an accelerometer that can be installed under the vehicle seat. It detects the presence of a child seat and suppresses the air bag in a collision.(2) However, it has not been used to my knowledge.

Convertible seats. These can be used as rear-facing seats for infants weighing up to 30 pounds and forward-facing seats for toddlers weighing up to 40 pounds. They do not come with separate bases. They have a variety of restraint systems, including three-, four-, and five-point belts. Some have retractors in their belt systems. Others have an additional tray or padded arm that comes in front of the belted child, which may improve or degrade performance depending on its design. These seats generally have high sides that reduce sideways movement in angular and side-impact crashes.

Convertible seats are best attached to the car via a lap belt or lap-shoulder belt that has a retractor that can be locked in place. Otherwise, a locking clip must be used with the three-point shoulder belt. These clips are difficult to install properly and many are installed incorrectly.

Australia and Canada require upper tether anchorages to keep convertible seats from tipping forward during a crash. Tipping can lead to head injuries, neck distractions, and upper cervical injuries.

The U.S. Department of Transportation has until recently not required tethers because they further complicate installation. Recent rulemaking on this issue is discussed below.

Low-shield booster seats. For years, these seats were used for children who outgrew convertible car seats or who could extricate themselves from them. Some manufacturers promoted these seats for smaller children weighing less than 40 pounds, even though they are biomechanically inappropriate for them.

Compared to many other child restraint systems, the low-shield booster is easier to install because it uses a lap-only belt. However, the seat lacks an upper torso restraint and creates a risk of head impact and distraction (axial tension) forces to the upper cervical spine. Also, some designs do not provide sufficient restraint to prevent ejection from the seat.

Although low-shield boosters are tested in accordance with convertible seat regulatory standards, they cannot be installed in cars in the same way they are installed for testing. When a child sits in the seat, it is not possible to tighten the belt as much as it is tightened in the test. Low-shield booster seats are not as safe as forward-facing convertibles and should be banned from the market.(3)

Belt-positioning booster seats. Although relatively new to the market, these seats have a history going back to the early 1980s. They elevate the child so that the car's three-point belts can be properly positioned over the child's body. They are designed for children weighing more than 30 pounds.

Most three-point belts are designed to fit adults, from small women to large men. They are not required to fit children. Belt-positioning boosters compensate for this deficiency and provide protection superior to other restraint systems except for the integrated system described below. Some belt-positioning boosters also have optional harnesses to secure younger children. These harnesses also allow the booster to be used in seat positions that have lap-only belts instead of lap-shoulder belts.

If the optional harness is used, this type of seat offers children who have outgrown rear-facing infant seats better protection than a low-shield booster. New designs about to enter the market should do a better job of protecting occupants of different sizes. Until then, integrated seats provide the best protection for children.

Integrated seats. An integrated child restraint system is built into a vehicle. Typically, it consists of a platform that folds out from the seat back and has a four- or five-point belt to secure the child. Some designs also use the platform as a belt-positioning booster in combination with the vehicle's regular seat belts.

Integrated seats are generally considered the most effective child restraints for several reasons. First, this system eliminates most installation errors. Second, as a component part of the vehicle, it permits maximum ride-down time in a crash. (Ride-down means the child is exposed to lower crash forces over a longer period of time, reducing the risk of injury.) Third, it is tested for the crash pulse of the car. The crash-pulse test closely mimics a real accident, as opposed to a generic-pulse test, which uses a hypothetical crash. The integrated seat's performance characteristics are tuned for the specific vehicle, providing maximum protection. Fourth, the system is placed in the rear seat, the safest location for children because it avoids the hazards of air bags.

Automakers selling vehicles designed for family use should install integrated child seats. Widespread installation and use would reduce injuries to children in vehicle crashes.

Seat safety and performance

Infant and child safety seats were invented in the 1960s. Their purpose was to make the vehicle's belt restraints better fit a child. These restraints should be designed to apply crash forces in the safest manner possible. For small children, the restraints spread the forces more effectively than lap-only belts and in some cases lap-shoulder belts.

Laws requiring children of specific ages to use child safety seats that complied with Federal Motor Vehicle Safety Standard (FMVSS) 213 were passed by all states between 1978 and 1985. By 1988, 84 percent of all children up to age four were using child safety seats.(4)

The original standard went into effect in 1971 with some basic requirements.(5) The 1981 version adopted a 30-mph dynamic test procedure that required the restraint to limit the amount of the child's excursion from the seat and the amount of force applied to the child's head and chest.(6)

The government performs compliance testing of selected seats each year. The work is usually turned over to contractors. Each year, a significant number of seat designs are not in compliance, necessitating a recall.(7)

Manufacturers often attribute injuries caused by child restraints to "misuse" by consumers. That term is an inaccurate description of the problem. The difficulties in installing and using most child seats are due to poor design and instructions.

Surveys have shown that between 61 percent and 85 percent of the seats are not used by consumers in the way the manufacturers intended.(8) Those percentages, which are well known to car seat makers, indicate that manufacturers should apply human factors analysis to make the seats easier to use and therefore more effective in preventing injuries.

In an effort to improve the performance of child restraints and reduce the problem of incorrect installation, NHTSA last year issued new rules requiring significant changes in the design of child seats and passenger vehicles.(9) On February 27,1999, the president announced that seats will be required to have three standard attachments--one on top and two on the bottom--that will fit into new anchors in cars. The new requirements will be phased in and will be fully effective September 1, 2002.

Rather than use the passenger seat belt to hold the child seat in place, easy-to-use rigid bar anchors will hold the lower part of the seat to the vehicle seat bottom and back. Tether anchors at the top of the seat, as used in Canada and Australia, will prevent it from tipping forward. Based on, but not identical to, an International Standards Organization proposal called ISOFIX, this system is expected to reduce incorrect installation, removing the major problem in child restraint performance.

Defect theories

There are several basic defect theories applicable to child car seats.

Inappropriate designation of occupant size. Generally, rear-facing seats are designed for children up to 20 pounds. But just because a child weighs more than 20 pounds does not mean that his or her neck has developed sufficiently to handle frontal impact forces.

Seats intended for children weighing more than 10 kilograms (22 pounds) are required to be tested with a forward-facing 15 kg (33-pound) dummy the size of a three-year-old.(10) But nothing prevents manufacturers from testing seats in a rearfacing position with heavier dummies. Tests conducted by Swedish researchers have demonstrated the superior performance of rear-facing seats for older children. Such seats could and should be sold in the United States.

Low-shield booster seats have also had size designation problems. For years, seat makers instructed consumers to use these seats with children who weigh as little as 30 pounds. However, many designs do not fully restrain small children, permitting partial or complete ejection in frontal crashes.(11)

Structural failure. Testing by NHTSA subcontractors and other independent organizations, such as Consumers Union (publishers of Consumer Reports), has shown that many seats are not strong enough to sustain foreseeable crash loads. Breakage of plastic parts, belts pulling through guide holes, and partial or complete separation of infant seats from their bases have occurred in tests.(12) When government testing discovered the defect, recalls frequently followed.

But for years there was no way for owners to receive notification of recalls. Without a notification system, consumers had to repeatedly check NHTSA publications containing recall lists to ensure a seat did not pose a serious hazard. In recent years, manufacturers have been required to set up registration procedures, but the manufacturers still depend on buyers to send in a card that places them on a mailing list. Many buyers fail to register and miss out on recall notification.

Recently, NHTSA set up a Web site (www.nhtsa.dot.gov) listing recall information. But the owner must check the site for any change in seat status.

Inadequate restraint performance. Seats are typically tested only for compliance with performance criteria set out in FMVSS 213. The basic test requirement sets limits on head injury criteria, forces applied to the chest, and forward movement of the dummy body.

For example, when subjected to a 30-mph frontal sled test, the head can move forward no more than 32 inches from the "Z-plane."

The Z-plane is a point located approximately 4 1/2 inches behind where the vehicle seat bottom cushion meets the seat back.

There are several problems with this. First, the 32-inch limit does not reflect many real-world car and seat adjustment conditions. Smaller cars do not have 32 inches of space between the front and rear seats, especially when the front seats are pushed back on their tracks. Many children may be at risk for head impact, even though the child restraint meets the minimum standard requirements. Head impact is particularly hazardous to young children.

A second problem is that in the test, the lap belt that holds the child seat in place is tightened more than it typically can be in real-world conditions. Most users route the belt around or through the seat without tightening the belt any more than is needed to eliminate slack. Real-world belt tensions are, therefore, far lower than those in the test. A looser belt allows more forward movement of the seat, reduces time and distance for ride-down, and results in higher forces being applied to the child. This is sometimes referred to as "dynamic overshoot." The risk of serious injury is much higher than the test indicates.

Inadequate warnings or instructions. For many years, instructions and warnings provided with child seats inadequately explained installation procedures and child size limits. In the 1990s, manufacturers corrected many deficiencies, but some remain. For example, many of these materials do not comply with the generally accepted methods of providing warnings as outlined by the American National Standards Institute.

Many instruction books are still difficult to understand. A better method, one adopted by some automobile manufacturers, uses an easy-to-read card that on one side illustrates proper installation and on the other provides warnings. Separate cards translated into major foreign languages spoken in the market (for example, Spanish and French for the United States and Canada) would make the information accessible to more buyers.

Litigation and reports by consumer groups of the seats' poor test performance have increased the public's knowledge of child restraint system defects. Until automakers become actively involved in designing and marketing restraints that accommodate the different sizes and vulnerabilities of children, child car seats will continue to be manufactured by playpen and high chair makers with insufficient knowledge of crash dynamics. Although some juvenile furniture makers have increased their research and development efforts, most still use compliance with the minimum federal standard as their design goal.

New regulations with slightly more rigorous performance standards raise the bar a little. New anchorage and tether requirements will allow a phase-in of easier-to-use seats that minimize injuries in real-world crashes. But the new rules do not necessarily protect children as much as we can or should.

Notes

(1.) B. KAMREN ET AL., THE PROTECTIVE EFFECTS OF REARWARD FACING CHILD RESTRAINT SYSTEMS: AN OVERVIEW OF POSSIBILITIES AND PROBLEMS ASSOCIATED WITH CHILD RESTRAINTS FOR CHILDREN AGED 0-3 YEARS (Soc'y of Automotive Engineers Paper No. 933093, 1993).

(2.) ALAN BROWNE ET AL., USING SEAT MOUNTED ACCELEROMETERS TO DIFFERENTIATE BETWEEN NORMALLY SEATED PASSENGERS AND INFANTS IN INFANT SEATS (Soc'y of Automotive Engineers Paper No. 933092, 1993).

(3.) Videotape: Don't Risk Your Child's Life IV: Automotive Safety for Children in the 90's (Shelness Productions 1993).

(4.) NATIONAL HIGHWAY TRAFFIC SAFETY ADMIN., COMPARISON OF RESTRAINT USE RULES, FATAL ACCIDENT REPORTING SYSTEM (1983-1992) VS. RESTRAINT USE IN 19 U.S. CITIES SURVEYS (1983-1991) (1992).

(5.) 49 C.F.R. [sections] 571.213 (1971).

(6.) 49 C.F.R. [sections] 571.213 (1981).

(7.) Recall Analysis Div., National Highway Traffic Safety Admin., Child Seat Safety Recall Campaign Listing (1990-1999) (visited Dec. 8, 1999), http:// www.nhtsa.dot.gov/people/injury/childps/recall/ canister.htm.

(8.) See, e.g., NATIONAL SAFE KIDS CAMPAIGN, CHILD PASSENGERS AT RISK IN AMERICA: A NATIONAL STUDY OF CAR SEAT MISUSE (1999).

(9.) National Highway Traffic Safety Admin., Federal Motor Vehicle Safety Standards; Child Restraint Systems; Child Restraint Anchorage Systems, 64 Fed. Reg. 10,786 (Mar. 5,1999).

(10.) 49 C.F.R. [sections] 571.213.s7.1 (1998).

(11.) Videotape: Don't Risk Your Child's Life IV, supra note 3.

(12.) Child Car Seats, Test: How Safe Is This Child? CONSUMER REP, Jan. 1999, at 24.

Donald H. Slavik practices with Habush, Habush, Davis & Rottier in Milwaukee, Wisconsin.