Preliminary Economic Assessment
Advanced Notice of Proposed
To Add A Side Impact Test
to FMVSS No. 213
Office of Regulatory Analysis and Evaluation
Plans and Policy
TABLE OF CONTENTS
I. EXECUTIVE SUMMARY
IV. THE ANPRM
VI. DATA AND TESTING
VIII. INJURY REFERENCE VALUES
IX. PERFORMANCE TEST
XI. COST EFFECTIVENESS
XII. INITIAL REGULATORY FLEXIBILITY ANALYSIS
XIII. SMALL BUSINESS IMPACT
XIV. UNFUNDED MANDATES REFORM ACT
I. EXECUTIVE SUMMARY
At the present time there are no side impact tests required for child restraints in Federal Motor Vehicle Safety Standard No. 213. Child restraints are tested to several frontal crash simulations. The agency is considering two tests. The first is a 90-degree ½ sine lateral sled test at 20 mph. To protect the child, limitations are set on the amount of head excursion and on the amount of force that can be exerted on the head, chest, and neck of a child test dummy during the dynamic test. The second test has a rigid wall at the end of the sled seat and will be run at 15 mph with a different pulse. Booster seats are not being considered, because booster seats have no sides and adding side structures to booster seats will make them too bulky. It is assumed that the structure of the vehicle will provide the same protection to children in booster seats as it does for adults. There are no dynamic side impact protection test required anywhere in the world at this time.
Target Population: Approximately 32 percent of the children who died in a motor vehicle crash were involved in a side impact crash. Approximately 16 percent of the injuries to children were sustained from side impact crashes. For the period 1993 to 2000, there were approximately 3,630 non-fatal injuries per year and in 1999 there were 91 fatalities to children in child restraints in side impact crashes.
Countermeasures: 1) Given that some child restraints could meet the excursion and injury limits under consideration and that child restraint manufacturers have never had to design for a side impact test, it is possible that minor changes in design for forward facing child restraints, could allow some child restraints to pass the test. For forward-facing infant restraints the agency estimated the costs and benefits of adding padding around the head area of the infant. However, the agency has also considered three other countermeasures, which involve:
2) Extending the sides (i.e., larger wings) on a forward-facing child restraint, 3) Adding another tether to child restraints, and 4) using Rigid LATCH connectors. None of these potential countermeasures have even been tested to determine their feasibility and benefit in a side impact test.
Benefits: The countermeasures under consideration could result in a reduction in head excursion and injuries, with potential benefits as shown below:
Estimated Benefits of the Various Countermeasures
Costs: The cost of the various countermeasures (in year 2000 economics) are: 1) No cost to the forward facing child restraint, just redesign and testing costs; the incremental cost of adding one-inch thick padding to the head area of an infant restraint is estimated to be $2.50. 2) Adding larger wings is expected to add $15.00 to the price of an average convertible child restraint system. 3) The cost of a tether is $4.14 per child restraint and the cost of adding two anchorages to each vehicle is $1.40. 4) The incremental cost of rigid LATCH connectors versus flexible connectors is estimated to be $25.15 per child restraint. Testing costs are approximately $468,000.
The cost per equivalent life saved (in millions) at the seven percent discount rate is:
Design Changes = $0.25
Larger Wings = $17 .20
Add Tether = $3.90
Rigid Latch = $10.24
The purpose of this ANPRM is to get comments on the possibility of changing the child restraint standard to increase the likelihood that children in child restraints are better protected in side impact crashes, especially with respect to head and neck injuries. The Fatality Analysis Reporting System (FARS) data files for 1999 show that approximately 32 percent of the children who died in a motor vehicle crash were involved in a side impact crash. From the National Automotive Sampling System (NASS) Crashworthiness Data System (CDS) database, approximately 16 percent of the injuries to children were sustained from side impact crashes. The Transportation Recall Enhancement, Accountability, and Documentation (TREAD) Act, Public Law 106 - 414(114 Stat. 1800), which contains provisions on improving child restraints safety performance standards and testing requirements, was signed into law on November 1, 2000. Section 14(b)(3) directs the agency to consider whether to require improved protection from head injuries in side-impact and rear-impact crashes.
In an effort to learn more about the relative injury risks of side impact crash modes, the agency has studied the types of injuries to children in side impact crashes by analyzing the FARS and NASS data files. The children's head seems to be the area most affected in side impact crashes. One reason for this is the size of the child's head relative to the body. Children's heads and the necks that support the heads are more vulnerable than the heads and necks of adults because children's heads are larger in proportion to the rest of their bodies.
There are Federal motor vehicle safety standards to protect motor vehicle occupants in side impact crashes. These regulations tend to be geared towards adults rather than children, because of the unavailability of instrumented child dummies when these regulations were promulgated.
The National Highway Traffic Safety Administration's (NHTSA's) side impact protection requirements are set forth in Federal Motor Vehicle Safety Standard No. 214, Side Impact Protection. The standard has two sets of requirements for passenger cars: 1) quasi-static door strength requirements and 2) dynamic requirements.
The passenger car quasi-static door requirements, which have applied to passenger cars since 1973, are designed to mitigate occupant injuries in side impact crashes by limiting the amount of side door intrusion into the passenger compartment. The requirements specify that side doors must resist crush forces that are applied against the door's outside surface in a laboratory test. NHTSA extended the requirements to light trucks, buses, and multipurpose passenger vehicles (MPVs) <= 10,000 lbs. gross vehicle weight rating (GVWR) (called LTV's) in a final rule published on June 14, 1991 (see 56 FR 27427) to be effective September 1, 1993 (MY 1994).
In 1990, NHTSA issued a final rule pertaining to dynamic side impact requirements for passenger cars, (see 55 FR 45722, October 30, 1990) which became fully effective on September 1, 1996 after a three-year phase-in. In a final rule published July of 1995, the agency extended the passenger car dynamic side impact protection requirements to trucks, buses, and multipurpose passenger vehicles with a gross vehicle weight rating of 6,000 lbs or less.
These standards have resulted in vehicle improvements, including side impact air bags in some vehicles. Under Standard No. 214's dynamic requirements, a passenger car must provide protection to the thorax and pelvis regions of a specified side impact dummy (SID) (see 49 CFR 572, subpart F) in a full-scale crash test in which the test car (known as the "target" car) is struck on either the left or right hand side by moving a deformable barrier (MDB) simulating the striking vehicle in a typical, vehicle-to-vehicle, side impact crash. The SID represents a fifth-percentile male.
The test procedure includes placing a SID dummy in the outboard front and rear seats of the target car or test vehicle on the side to be struck by the barrier. For the thorax, the performance limit is expressed in terms of an injury criterion known as the Thoracic Trauma Index (dummy) or TTI (d). This injury criterion represents the average of the peak acceleration values measured on the lower spine and the greater of the acceleration values of the upper and lower ribs of the SID dummy. The TTI (d) limit is 85 g's (max.) for 4-door and 90 g's for 2-door passenger cars. For the pelvis, the performance limit is specified in terms of the peak acceleration measured on the pelvis of the SID dummy. The standard specifies 130 g's (max) for the pelvis for passenger cars. Again, the SID represents an adult occupant. There is no SID child dummy available.
This section provides a chronological discussion of past rulemakings on Standard 213 relating to protection in side impact crashes.
In the original NPRM for FMVSS No. 213, published on January 24, 1969 (34 FR 1172) the agency expressed concern regarding the rigidity of surfaces of a child restraint likely to be contacted by a child's head or torso. In this NPRM, the agency proposed
"Any rigid component of a child seating system that, during forward, right-side, left-side, or rearward impact, may be contacted by the torso or head of a child …shall covered with energy-absorbing material having a thickness of at least one-half inch."
The subsequent final rule issued in March 1970 (35 FR 5120) changed the proposed impact requirements. It revised the impact protection requirements to (1) provide a more definitive description of energy-absorbing material to give manufacturers a better basis for material selection and to preclude the use of soft sponge rubber, which offers little impact protection, and (2) exclude rigid sides of child seats which have a surface area of at least 24 square inches (it is believed that large, flat surfaces tend to distribute impact loads over a sufficient area of a child's torso so as to make padding unnecessary).
Petitions for reconsideration of the March 1970 final rule requested, led to a NPRM published on September 23, 1970 (35 FR 14786) that, among other things, proposed revisions to the impact protection requirements that would add more exact performance requirements and test procedures to the standard. Specifically, the NPRM proposed to require that the energy-absorbing material have (1) a closed cell structure, and (2) a 25 percent compression-deflection resistance of not less than 5 and not more than 13 pounds per square inch when tested in accordance with Sections 17, 19, and 20 of ASTM Standard D1056-68, "Testing Sponge and Expanded Cellular Rubber Products," December 1968.
These proposed revisions were largely not adopted, as the agency subsequently announced its intention to institute a dynamic test as the method for testing child seating system performance. A final rule dated March 23, 1973 (38 FR 7562) slightly modified the head impact requirements of the standard by (1) changing "energy-absorbing material" to "force-distributing material," (2) exempting belt adjustment hardware from the padding requirements, and (3) eliminating the exception to the padding requirements for components contactable by the head.
The March 1974 NPRM that initially proposed dynamic testing for the evaluation of child restraint performance concluded that limitation of head motion so as to prevent impact with the vehicle interior, combined with padding requirements, was the most practicable method for providing an appropriate level of safety for children. The proposed rule provided for impact protection when a child occupant impacts the restraint. Impact force in head and facial contact areas would be controlled by impact testing with a test headform, and torso impact areas in child seats would be required to have minimum surface dimensions. The proposed rule stated that each component (except belts, belt buckles, and belt adjustment hardware) of a child restraint system other than a recumbent position restraint that may contact the head of the test dummy when the system was tested dynamically (frontal, lateral, and rear impacts) shall (1) limit the impact acceleration of the headform to not more than 67 g, (2) absorb not less than 70-inch pounds of energy before the head form acceleration equals 25 g, and (3) have a surface area of contact with the head form in square inches not less than A/8, where A is the peak acceleration expressed as a multiple of the acceleration of gravity.
A subsequent May 1978 NPRM proposed somewhat different requirements relating specifically to the impact performance of a child restraint. Injury criteria (expressed in terms of limits on resultant acceleration) were proposed for both the head and chest of the 3-year-old test dummy to allow a quantitative evaluation of the dynamic performance of the child restraint. As such, the measurement of padding effectiveness would be accomplished during the dynamic test, eliminating any need for a separate test for that purpose and the costs associated with such a test. Child restraints recommended for children weighing 20 pounds or less could not be tested with the 3-year-old dummy at the time. As such, no quantifiable performance evaluation could be made of those restraints, because the 6-month-old dummy was not instrumented.
Therefore, the agency proposed that for those restraints for children weighing less than 20 pounds, surfaces that are contacted by the dummy head during dynamic testing must meet a padding requirement and static compression test. Specifically, the NPRM proposed that each system surface which is contactable by the dummy head when tested dynamically shall be covered with slow recovery, energy-absorbing material (1) at least ¾ inch thick, and (2) had a 25 percent compression-deflection resistance of not less than 5 and not more than 9 pounds per square inch when tested in accordance with ASTM Standard D1056-73.
Further, the agency noted that child restraint systems should be designed so that they did not present hostile or hard surfaces capable of injuring a child in a crash. To ensure this, the agency specified minimum surface area requirements for forward and side restraint surfaces and contour requirements for forward, side, and rear surfaces, which were provided in restraints other than car beds. The torso dimensions of the three-year-old dummy were the basis of the proposed minimum areas and contours of the side and forward support surfaces. Note that the May 1978 NPRM is also the point at which the agency did not repropose the rear- and side-impact dynamic tests for FMVSS No. 213.
The 1979 final rule that followed made some changes to these proposed requirements in response to comments from a number of organizations. In response to manufacturer comments, NHTSA reevaluated the materials used in child restraints and determined that those and other widely available materials could provide sufficient energy absorption if used with a specified thickness. The agency changed the proposed compression-deflection requirements to allow the use of a wider range of materials that would enable manufacturers to provide protective padding for children without having to increase the price of the restraint.
Several manufacturers raised objections to the proposed requirement that head restraints of child restraint systems have a width of not less than 8 inches. They pointed out that the minimum head restraint width requirement was intended to prevent a child's head from going beyond the width of a head restraint in a lateral or rear impact. They argued that restraints with side supports or "wings" should not have to meet the 8 inch width requirement since the side supports will prevent an occupant's head from moving laterally outside the restraint system. NHTSA agreed, and therefore exempted those restraints from the 8-inch minimum width requirement. However, to ensure that child restraints with side supports have sufficient width to accommodate the heads of the largest child using the restraint, the agency set a 6-inch minimum width for those restraints. In addition, to ensure that side supports are large enough to retain an occupant's head within the restraint, the agency set a minimum depth requirement of 4 inches for those supports. Anthropomorphic data showed that the head of a 50th percentile 5-year-old child measured 7 inches front to rear and 6 inches in breadth. Therefore, a 4-inch support should contact a sufficient area of the child's head to restrain it.
Several manufacturers objected to the proposed force distribution requirement set for the sides of child restraint systems. The specifications did not require manufacturers to incorporate side supports in their restraints; they only regulated the surfaces that the manufacturer decided to provide so that they distribute crash forces over the child's torso. The commenters requested that the agency define the term "torso" and explain the reason for setting different side support requirements for systems used by infants weighing less than 20 pounds than for systems used by children weighing 20 pounds or more.
In restraints for infants less than 20 pounds, the minimum side surface area requirements were based on anthropometric data for a 6-month-old 50th percentile infant to ensure maximum lateral body contact in a side impact. Since the skeletal structure of an infant is just beginning to develop, it is important to distribute impact forces over as large a surface area of the child as possible, rather than concentrating the potentially injurious forces over a small area. For restraints used by children weighing more than 20 pounds and, therefore, having a more developed skeletal structure, the minimum surface area requirement is based on anthropometric data for a 50th percentile 3-year-old child to provide restraint for the shoulder and hip areas of the child.
A. Petition Submitted on Padding Requirements
On April 1, 1980, General Motors (GM) filed a petition for rulemaking seeking changes in the padding requirements set forth in the December 1979 FMVSS No. 213 final rule. GM noted that the compression-deflection resistance of padding is sensitive to the rate at which deflection occurs during the test procedure. As the deflection rate increases during testing, so does the measured resistance of the material. GM said that the padding used in the head impact areas of its child seat has a maximum compression-deflection resistance of 3 psi. However, several permissible deflection rates are permitted in the padding tests by the ASTM test procedures incorporated in to FMVSS No. 213. GM reported that the measured 25 percent compression-deflection value of the padding it uses could be as low as 1.8 psi.
To accommodate the differences attributable to use of different deflection rates permitted in the testing, the agency proposed to permit padding with a compression-deflection resistance of 1.8 psi or more to have a minimum thickness of ½ inch (45 FR 68694, October 16, 1980). This proposal was adopted in a final rule published on December 15, 1980 (45 FR 82264).
B. ISO Plan for Future Test
Over the years, injuries to children in side impacts continued to be at an unacceptable level. The International Organization for Standardization (ISO) working group ISO TC22/SC12/WG1, "Child Restraint Systems," declared that the risk of side impacts to children in cars is an important working item, and an ad-hoc group was established in 1993 to analyze this area. The ISO working group has noted that, "From different accident research units, it was reported that critical or fatal injuries of child restraint-protected children in side collisions show about the same importance as in frontal collisions." Therefore, ISO noted that there is an interest in evaluating the risk of injuries to children in side impacts and in analyzing the side impact performance of child restraint systems. The ISO working group was given the task of developing an international standard of uniform test criteria for such evaluation. The ISO has developed a draft standard based on accident data. However, injury criteria and other aspects of the test have not been proposed. This work remains ongoing at this time.
C. EEC Regulation 44
A review of the E.C.E. Regulation 44, specifically with respect to side impact tests, padding requirements, and requirements for "wings," reveals the following:
E.C.E. does not require any side-impact testing to be performed on any child restraints.
Section 7.1.2 of E.C.E. Regulation 44 prescribes energy absorption requirements for materials used in child restraints as follows:
220.127.116.11. For all devices with backrests there shall be internal surfaces comprising material with a peak acceleration of less than 60 g when measured in accordance with annex 17 of this regulation. This requirement applies also to energy absorbing materials that cover areas of impact shields which are in the head strike zone.
The only requirements for "wings" in the E.C.E. Regulation 44 apply to rear-facing child restraints. These restraints must have side wings with a minimum depth of 90 mm measured from the median of the surface of the backrest. These side wings shall start at the horizontal plane passing through point "A" and continue to the top of the seat back. Starting from a point 90 mm below the top of the seat back, the depth of the side wing may be gradually reduced.
(See Data/Testing for a discussion of comparative testing of U.S. and European child restraint designs).
IV. THE ANPRM
The agency is requesting comments on two different tests to be performed on child safety seats. 1) A 20 mph sled test with no wall and, 2) A15 mph test with a rigid wall at end of the seat on the sled. There are no requirements in the world that specify a minimum level of performance in a dynamic side impact simulation. Worldwide efforts to mandate improved child restraint safety have concentrated on performance in frontal impacts, where injury risk exposure is the highest and the potential the greatest for countermeasure development. In side impact crashes, data are not widely available as to the sources of injuries to children; that is, to what degree injuries are caused by intrusion of an impacting vehicle or object. Also, the TNO P- and Q -series child dummies have been developed for side impact, but their robustness and suitability for such testing remains in question.
Several international bodies have focused attention on side impact safety by developing consumer information rating programs that assess child restraint performance in side impact tests. The European Economic Commission (ECE) impacts vehicles with a moving deformable barrier at 30 mph at a 90-degree angle. An 18-month-old dummy is placed on the vehicle's near-side in a rear-facing or forward-facing child restraint. The vehicle is rated on head containment, resultant head acceleration, and chest acceleration. A consortium of consumer advocates in Australia has a program that incorporates a lateral dynamic sled test of tethered child restraints with a 20 mph pulse at 45 and 90-degree angles. The Australians assess the dummy's lateral head excursion relative to a simulated vehicle door. Child restraints are ranked according to their ability to prevent the dummy's head from hitting the door.
The agency is considering two different child restraint sled tests. One would be a 90 degree lateral impact that simulates a 20 mph crash with no wall. The child restraint would be installed in the center seating position with the LATCH system or on an outboard seating position with a lap and shoulder belt with the tether attached. When tested in this fashion, each child restraint would have to limit head motion to 20 inches in each lateral test with the Hybrid III-3-year-old test dummy and limit head motion to 20 inches with the CRABI 12-month-old dummy, measured from the center line of the 6 mm diameter LATCH attachment bar located furthest from the impact, and to suffer no loss of structural integrity.
The second test is a 15 mph test with a rigid wall at the end of the seat on the sled. A more detailed explanation of the test can be found in the injury criteria section and the testing section.
Some countermeasures against side impact injuries are possible short-term endeavors while others are not. Among the countermeasures available are:
At this time, the agency has not designed or tested any of these potential countermeasures to determine their practicability, feasibility or potential benefits.
VI. DATA AND TESTING
NHTSA obtained six child restraints that currently meet the E.C.E. Regulation 44 with respect to padding and "wings" to examine what, if any differences would be seen when tested in identical test configurations. The Britax King and the Century Accel were tested using the same FMVSS No. 214 and NCAP pulses as the "U.S. seats" tested in Table 1 and with the Hybrid III 3-year-old dummy. Physical review of the "European" seats prior to testing did not reveal significant differences in the padding or size of the "wings." Again, a suitable instrumented side impact child dummy was not available for use, so evaluation of the seats focused primarily on the kinematic response of the dummy for near-side testing. In each case, the test dummy's head went out around the side of the child restraint and impacted the door frame of the sled buck.
From this, NHTSA has tentatively concluded that the U.S and the European restraints perform comparably in side impact crashes, and that any differences in padding and/or wings between the U.S. and European restraints will likely have little or no effect on the injury potential for occupants.
In an effort to evaluate the performance of U.S. child restraints in side impact crashes, VRTC performed a limited set of 90 degree side impact tests using a MY 1999 Pontiac Grand Am sled buck in December 2000 (Test Matrix is provided as Table 1). Tests were performed with both the FMVSS No. 214 side impact pulse at approximately 14.5 mph and 20.6 peak g's and with the NCAP side impact pulse at approximately 21 mph and 26 peak g's. Child restraints currently available in the U.S., both LATCH and non-LATCH configurations, were used in the testing. Child restraints were placed in both the near- and far-side seating positions in the rear seat. It must be emphasized that these tests were performed primarily to examine the kinematics of the child dummies, since there currently is no consensus in the biomechanical community regarding the biofidelity of child test dummies and injury criteria for side impact. However, the Hybrid III 3 year-old dummy and the injury criteria specified for frontal impact were used for comparative purposes. The sled testing demonstrated that the dummy positioned on the nearside hit its head on the door frame, resulting in high values for HIC of about 800-1100. By contrast, the dummy positioned on the far side did not appear to contact the vehicle and had low values for HIC, suggesting a low probability of injury. The side impact testing conducted at VRTC did not evaluate injury due to intrusion, which may or may not play a major role in fatality and injury for the nearside occupant in side-impact crashes. The ongoing ISO effort described above attempts to replicate the effects of intrusion, but this is not near completion at this time. The testing performed by VRTC, while admittedly limited, gives some indication of the kinematics of vehicle occupants that may be useful in improving side-impact crash protection for children.
In each instance, the near-side occupant test dummy hit its head on the door frame, while it appeared that the far-side occupant test dummy would not have suffered significant injuries.
Fourteen tests conducted on the Grand Am with the 3-year-old dummy, ten were near-side tests and four were far-side tests. Test results for these tests are shown in Table (1). Of the 14 child restraints tested six child restraints passed all the injury criteria. All four of the far-side tests passed all the injury criteria. There was no head excursion data available. Seven of the seats tested failed the HIC15 reference values, six seats failed the chest acceleration reference values, and one failed Nij; they were all near-side tests.
In support of the agency's rulemaking, tests were conducted on the FMVSS No. 213 sled test fixture. On the sled test fixture there is no side door/door frame for the dummy to hit its head on. In this test program (see results in Table 2), there were 15 child restraints tested, ten forward facing and five rear facing, the first six tests shown are 45 degree tests, the last nine tests shown are 90 degree tests. We have excursion values for all these tests. Two forward facing seats passed the excursion criteria. The seats tested without top tethers had the largest excursion limits. Two rear facing seats recorded the highest excursion limits. All of the seats tested passed the HIC 15, chest deflection and chest acceleration reference values. However, none of the 3-year-old dummies passed Nij when tested at 90 degrees.
Summary Results for Side Impact Child Restraint Systems in Grand Am Sled Tests
|In-Postion Critical Values|
|Test Type||HIC 15||HIC
|Nij ET||Nij EC||Nij FT||Nij FC||Nij Max.||Peak
|TRC327||3 yo||Far Side; Triad/LATCH @ 14.5 mph / 214 pulse||57||98||0.581||0.030||0.210||0.134||0.581||654||82||7.1||9.8||5.10||26.3|
|"||3 yo||Near Side; Touriva/L-S No tether @ 14.5 mph / 214 pulse||382||382||0.134||0.008||0.383||0.087||0.383||778||42||6.7||1.6||2.29||47.3|
|TRC328||3 yo||Far Side; Triad/LATCH @ 21.0 mph /SNCAP pulse||163||287||0.958||0.110||0.342||0.172||0.958||999||147||11.7||17.6||9.91||29.6|
|"||3 yo||Near Side; Touriva/L-S No tether @ 21.0 mph / SNCAP pulse||1085||1085||0.793||0.008||0.165||0.046||0.793||1143||46||10.0||13.4||3.56||65.9|
|TRC329||3 yo||Far Side; SafeEmb/LATCH @ 21.0 mph /SNCAP pulse||256||410||0.774||0.012||0.528||0.086||0.774||1216||68||13.8||11.1||9.40||51.5|
|"||3 yo||Near Side; SafeEmb/L-S No tether @ 21.0 mph / SNCAP pulse||796||796||0.774||0.034||0.139||0.018||0.774||1094||49||5.0||10.5||1.78||73.7|
|TRC330||3 yo||Far Side; Touriva/L-S No tether @ 21.0 mph / SNCAP pulse||88||218||0.487||0.021||0.294||0.152||0.487||652||116||7.4||7.5||6.60||27.9|
|"||3 yo||Near Side; Triad/LATCH @ 21.0 mph /SNCAP pulse||817||817||1.034||0.022||0.101||0.004||1.034||1528||16||3.9||11.6||12.95||45.7|
|"||3 yo||Near Side; Britax King/ L-S @ 15.0 mph /214 pulse||366||366||0.420||0.011||0.145||0.087||0.420||544||49||5.8||5.7||2.29||45.5|
|"||3 yo||Near Side; Century Accel/ L-S @ 15.0 mph /214 pulse||573||573||0.450||0.004||0.124||0.100||0.450||775||30||6.9||5.3||2.03||51.0|
|"||3 yo||Near Side; Britax King/ L-S @ 21.0 mph / SNCAP pulse||709||709||0.532||0.003||0.158||0.165||0.532||707||81||9.6||9.9||3.30||68.2|
|"||3 yo||Near Side; Century Accel/ L-S @ 21.0 mph / SNCAP pulse||1029||1029||0.976||0.003||0.136||0.095||0.976||1360||30||8.0||14.9||2.79||71.5|
|"||3 yo||Near Side; Britax King/ L-S w/ Tether @ 21.0 mph / SNCAP pulse||478||480||0.582||0.071||0.194||0.055||0.582||518||111||9.1||11.9||2.03||63.9|
|"||3 yo||Near Side; Century Accel/L-S w/Tether @ 21.0 mph/SNCAP pulse||899||899||0.874||0.008||0.221||0.006||0.874||1034||31||12.6||15.1||4.32||70.6|
Summary Results for Side Impact Child Restraint Systems @ 45 and 90 Deg., 20 mph Using 213 Seat Fixture
|Test #||Dummy Size||Test Type||Excursion (in) **||HIC 15||HIC unlimited||Peak Tension||Peak Compression||Peak Flexion (Y-axis)||Peak Extension (Y-axis)||Peak Flexion (X-axis)||Peak Extension (X-axis)||Chest Deflection (mm)||Chest Acceleration|
|TRC591||3 yo||Near Side, Cosco Triad-LATCH/ 45 deg.,1/2 Sine pulse||22.0||122||226||963||318||6.6||12.8||0.9||36.8||13.46||29.1|
|TRC591||12 mos.||Far Side, Cosco Touriva-lap only (rear-facing)/ 45 deg., 1/2 sine pulse||23.0||82||146||849||11||3.0||8.7||1.9||2.8||NA||30.5|
|TRC592||3 yo||Near Side, Century STE-LATCH/ 45 deg.,1/2 Sine pulse||23.0||150||255||419||950||5.1||15.3||2.2||35.5||11.68||31.8|
|TRC592||12 mos.||Far Side, Century STE-lap only (rear-facing)/ 45 deg., 1/2 sine pulse||26.0||126||163||591||8||2.4||8.5||2.6||3.0||NA||29.2|
|TRC593||3 yo||Near Side, Cosco Triad-LATCH (NO Tether)/ 45 deg.,1/2 Sine||27.0||122||268||493||436||3.6||15.7||1.7||37.2||13.00||26.4|
|TRC594||3 yo||Near Side, Century STE-LATCH (NO tether)/ 45 deg.,1/2 Sine pulse||26.0||131||240||430||698||2.8||15.6||1.2||33.8||11.18||27.9|
|TRC595||3 yo||Near Side, Cosco Triad-LATCH/ 90 deg.,1/2 Sine pulse||20.0||76||160||253||670||2.5||23.8||0.7||16.6||2.03||23.8|
|TRC595||12 mos.||Far Side, Cosco Touriva-lap only (rear-facing)/ 90 deg., 1/2 sine pulse||25.0||180||244||579||46||4.6||7.9||7.1||3.9||NA||24.1|
|TRC596||3 yo||Near Side, Century STE-LATCH/ 90 deg.,1/2 Sine pulse||19.0||107||159||600||985||1.9||25.9||1.1||17.4||3.05||23.3|
|TRC596||12 mos.||Far Side, Century STE-lap only (rear-facing)/ 90 deg., 1/2 sine pulse||28.0||247||248||580||7||2.1||3.8||6.7||7.9||NA||27.5|
|TRC597||3 yo||Near Side, Cosco Triad-LATCH (NO tether)/ 90 deg.,1/2 Sine pulse||21.0||99||200||113||643||2.2||30.6||0.1||13.0||3.81||23.8|
|TRC598||3 yo||Near Side, Century STE-LATCH (NO tether)/ 90 deg.,1/2 Sine pulse||22.0||93||170||316||861||3.4||24.3||0.1||12.6||5.59||25.9|
|TRC602||3 yo||Near Side, Century STE-LATCH/ 90 deg., 213 pulse||22.0||67||135||278||876||1.6||21.6||0.6||12.0||2.79||21.3|
|TRC602||12 mos.||Far Side, Century STE-lap only (rear-facing)/ 90 deg., 213 pulse||28.0||268||307||707||27||2.1||9.5||2.1||3.7||NA||23.1|
|TRC603||3 yo||Near Side, Century STE-LATCH (No tether) / 90 deg., 213 pulse||21.0||71||168||322||678||1.5||24.1||0.0||13.0||3.56||24.1|
|** 24.5" is proposed maximum allowed (corresponding to 19" in NPRM) for HIII 3 yr. old|
|19.5" is proposed maximum allowed (corresponding to 14" in NPRM) for CRABI-12|
Average Excursion in the Tests
|Test Dummy||Test Angle||Tethered?||Average Excursion||Number of Tests|
|12 mo. old
|3 yr. Old
Table 3 shows that there is a significant difference in excursion for the rear facing child restraints with the12-month-old dummy (27 inches of excursion at 90 degrees) compared to the forward facing restraints with the 3 year old dummy (averaging 20.3 inches of excursion at 90 degrees). Since the proposed test excursion limit is 20 inches in a 90 degree test, a significant improvement would be required for the rear facing child restraints and a small improvement for the forward facing child restraints. The data also shows that tethers reduce excursion in these tests.
Comparing the results in the 45 degree tests to the 90 degree tests shows that the 45 degree test would be less severe than the 90 degree test for the rear facing child restraints, but more severe for the forward facing child restraints. The agency has decided not to consider both a 45 degree and 90 degree test in this NPRM for several reasons: 1) 45 degrees is not a typical side impact resulting in injuries. Most severe injuries are the result of 60 to 90 degree impacts. 2) The agency does not require 45 degree testing for side impacts for adults. 3) The agency would like to limit testing burden. 4) With the tether attached, the difference in test results for the forward facing child restraints was 2.2 inches (22.5 inches at 45 degrees compared to 20.3 inches at 90 degree), and 90 degree testing was the worse case for rear facing child restraints.
Tables 4(a) and 4(b) give a summary of the head excursion test results and a summary of the injury measurements.
Test Results - Head Excursion
|Proposal||Tethered||Average Excursion||Percent Passing|
|12 month old rear facing||Yes||20"|
|3 year old forward facing||Yes||20.3"||67%|
Injury Measurements (Percent Passing)
|12 month old rear facing||100%||67%||100%|
|3 year old forward facing||100%||0%||100%|
There were 1,317 children between the ages of zero to twelve who were killed in motor vehicle crashes in 1999 (See Table 5a). Of this total, 31.89 percent were in side impact crashes. There were 91 fatalities that were restrained in child restraints. Children seated on the side nearest to the crash accounted for 55 percent of the fatalities. Fatalities from the mid- and far side seated individuals were 23 and 22 percent respectively. Although FARS does not have a properly restrained code, there is a child seat system use which indicates that a child seat is used "improperly." Based on this coding, we are assuming that approximately 19 percent of the fatalities were in restraints that were improperly used.
The annualized data for the period 1993 to 2000 (Tables 6(a) and 6(b)) show that approximately 16 percent of the nonfatal injuries to children were from side impact crashes. The side impact totals are different in the tables because the unknowns were distributed in one table and not distributed in the other table. Approximately 18 percent of the children with nonfatal injuries were in child restraints, approximately 48 percent were belted in the vehicle belt and approximately 34 percent were unbelted. Tables 6(c) and (d) show the breakout of injuries by MAIS levels.
1999 FARS Occupant Fatalities in Passenger Vehicles (Ages 0 -12)
|Type of Crash||Frontal Impact||Side Impact||Rear Impact||Other||Total|
1999 FARS Side-Impact Crashes Passenger Vehicles Occupant in Safety Seat (Ages 0 -12)
Estimated Non-Fatal Injured Children (ages 0-12) in Passenger
Vehicles Annual Average of 1993 to 2000 CDS Data
|Type of Crash||Frontal Impact||Side Impact||Rear Impact||Other||Total|
Estimated Non-Fatal Injured Children (ages 0-12) in Side Impacts
Annual Average of 1993 to 2000 CDS Data
|Type of Restraint||Forward Facing||Rear Facing||Belted||Unbelted||Total|
Estimated Non-Fatal Injured Children (ages 0-12) in Side Impacts In Rear Facing Child Safety Seats Annual Average of 1993 to 2000 CDS Data By MAIS Levels
|MAIS 1 Injuries||235|
|MAIS 2-5 Injuries||20|
Estimated Non-Fatal Injured Children (ages 0-12) in Side Impacts In Forward Facing Child Safety Seats Annual Average of 1993 to 2000 CDS Data By MAIS Levels
|MAIS 1 Injuries||3,080|
|MAIS 2-5 Injuries||295|
In Table 7 for the rear-facing seats, injury to the children was caused by contact with the child seat. For the forward-facing child restraints, injuries were caused by a variety of elements.
All Non-Fatal Injured Children (Ages 0-12) in Side Impacts In
Child Safety Seats by Source of 1993 to 2000 CDS Data
|Seat Type||Point of Contact
|Rear Facing||Child Safety Seat||100|
|Forward Facing||Child safety seat||30.44|
|Left side interior surface||0.67|
|Left side window glass/frame||2.59|
|Right side interior surface||3.60|
|Seat back support||1.79|
|Belt restraint webbing/buckle||19.96|
|Interior loose objects||1.06|
From the injuries with known contact sources in Table 6, approximately 9.4 percent of the injuries sustained by children in forward facing child seats are due to some contact with the left side or the right side of the vehicle.
The agency has made pre-test measurements of child restraint systems placed in 20 NCAP vehicles to determine seating and available lateral internal space, and found that the dummy head to the side window ranged from 311 mm (12 inches) to 566 mm (22 inches) with an average distance of 377 mm (14.8 inches).
The agency has looked at side impacts tests done on 15 vehicles at the 214 test criteria and 20 vehicles done at the NCAP test criteria. These tests measured the degree of intrusion at the mid-door level and at the window-sill level for both front and rear seats. Table 8 is a summary of the results.
Average Intrusion in Side Impact Testing
|FMVSS 214 Tests||NCAP Side Impact Tests|
|Front Door||Window Sill||179.9 mm (7.1 inches)||210.2 inches (8.3 inches)|
|Mid-Door||240.7 mm (9.5 inches)||289.6 mm (11.4 inches)|
|Rear Door||Window Sill||149 mm (5.9 inches)||186.8 mm (7.4 inches)|
|Mid-Door||208.8 mm (8.2 inches)||256 mm (10.1 inches)|
For the 214 side impact crash tests, the static crush of these vehicles at the rear seat window sill ranged from 52 mm (2.05 inches) to 227 mm (8.94 inches), with an average crush of 149 mm (5.9 inches). The static crush of these vehicles at the front seat window sill ranged from 91 mm (3.58 inches) to 234 mm (9.21 inches), with an average crush of 179.9 mm (7.1 inches). The static crush of these vehicles at the rear seat mid-door level ranged from 91 mm (3.58 inches) to 317 mm (12.5 inches), with an average crush of 208.8 mm (8.2 inches). The static crush of these vehicles at the front seat mid-door level ranged from 162 mm (6.38 inches) to 307 mm (12.09 inches), with an average crush of 240.7 mm (9.5 inches).
For the NCAP side impact side tests, the static crush of these vehicles, at the rear seat window sill level, ranged from 69 mm (2.7 inches) to 262 mm (10.32 inches), with an average crush of 186.8 mm (7.4 inches). The static crush of these vehicles, at the front seat window sill level, ranged from 93 mm (3.7 inches) to 269 mm (10.6 inches), with an average crush of 210.2 mm (8.3 inches). The static crush of these vehicles at the rear seat mid-door level ranged from 148 mm (5.8 inches) to 343 mm (13.5 inches), with an average crush of 256 mm (10.1 inches). The static crush of these vehicles at the front seat mid-door level ranged from 219 mm (8.6 inches) to 347 mm (13.7 inches), with an average crush of 289.6 mm (11.4 inches).
For the 214 side impact tests the average Delta V was 25.3 kph (15.18 mph), and for the NCAP tests the average Delta V was 28.67 kph (17.2 mph). Thus, the NCAP test is closer to the 20 mph sled test in terms of severity.
Table 8 provides several different intrusion measurements in two different test series with different impact severities. Since the child's head is close to the height of the window sill area, and the child's head appears in the sled tests to have the largest excursion, the window sill area appears to be the appropriate intrusion measurement for this analysis. The amount of intrusion is dependent upon where the child is seated compared to where the vehicle is struck. In the FMVSS 214 and NCAP test procedure, the front door is the primary target of the striking deformable moving barrier and the intrusion levels for the front door are larger than for the rear door. However, the intrusion at the window sill area is not that different between the front seat (210 mm - 8.3 inches) and the rear seat (187 mm - 7.4 inches). For this discussion, we will assume that when the point of contact of the striking vehicle is directed at the child, there will be 8.3 inches of intrusion.
The agency is considering using a measurement based on the far-side anchorage point of the LATCH system on the standard child restraint bench seat. Based on the above data we estimate that the side of the head of a child in a child restraint is 14.8 inches from the side of the average vehicle, based on the 3-year-old dummy. We estimate that this same side of the head is 8.0 inches from the far-side anchorage point. Thus, the anchorage point is 22.8 inches away from the side of the average vehicle. In a 20 mph delta V crash there is about 8.3 inches of intrusion. If we wanted to eliminate the chances of a near-side child's head hitting the side of the average vehicle in a 20 mph delta V crash given that the striking vehicle struck adjacent to the child restraint position, we would have to limit head excursion to 14.5 inches (22.8 - 8.3 inches) from the anchorage point. If the vehicle was impacted near the front or rear of the vehicle and did not intrude upon the child occupant space, then there is 22.8 inches of space before the child's head would hit the side of the vehicle.
At lower impact speeds there is less head excursion and less vehicle intrusion and at some speed the child's head will just miss the side of the vehicle, providing benefits to the child.
The following two charts show the pulse over time of the two sled tests under consideration:
TCR 327 is the Grand Am pulse at 15 mph
TCR 595 is the ½ sine pulse at 20 mph.
Figure 1 - Pulse for 15 mph Side Impact Sled Test
Figure 2 - Pulse for 20 mph Side Impact Sled Test
The target population in Table 4 is 91 fatalities. There were 50 (91x.55) near side fatalities. Based on an examination of side impact fatalities and Delta V, the agency believes that side impact crashes with a Delta V greater than 30 mph would be catastrophic. We found that nine percent of the cases would be catastrophic and this would reduce the target population to 46 cases. These 46 fatalities are further divided into 20 rear-facing fatalities and 26 forward-facing fatalities. The agency is considering two tests, the 90 degrees ½ sine pulse at 20 mph with no wall and the 15 mph rigid wall test. These tests would be run with the latch system and tether connected.
Rear-facing infant restraints
The 21 mph Grand Am forward-facing tests in Table 1 are used as a proxy measure for the 15 mph rigid wall test for rear-facing infants. This takes into account several factors:
It is assumed for rear-facing seats passing the test that there would be a reduction in fatalities by approximately 50 percent, and a reduction in injuries by 10 percent. This assumption is based on the fact that contact with the wall would result in very high HICs of greater than 1,000, and the potential benefit of bringing these high HICs back to the level of the proposed standard of 390 HIC for infants.
Benefits for infants are estimated to be:
20 x .5 = 10
MAIS 1 injuries,
235 x.1 = 24
MAIS 2-5 injuries,
20 x .1 = 2
Forward-facing child restraints
Forward-facing child restraints were also tested at 14.5 mph and 15 mph in the Grand Am. The Grand Am door was slightly closer than the 20 inch rigid wall proposed in the test. We assume that the 14.5/15 mph Grand Am tests are a proxy measure for the 15 mph test with the rigid wall.
Based on comparing these test results to the reference values, the agency has estimated that a countermeasure that passes this test would be 15.45 percent effective for fatalities and 0.85 percent effective for injuries.
Benefits for forward facing restraints are estimated to be:
26 x .1545 = 4
MAIS 1 injuries,
3,080 x .0085 = 26
MAIS 2-5 injuries,
295 x .0085 = 3
Table 9 shows a breakout of each countermeasure and the injuries and fatalities that can be prevented.
Estimated Benefits of the Various Countermeasures
VIII. INJURY REFERENCE VALUES
Currently, there are no injury criteria in FMVSS No. 213 for side impact. The reference values used for this analysis incorporate some aspects of FMVSS No. 208 into FMVSS No. 213 for side impact.
Standard No. 208 and Standard No. 213 address different sources of potential harm to the child occupant. The reference values in this analysis are taken from FMVSS 208 injury criteria for HIC15, chest g, chest deflection and part of the neck criteria. Currently, Standard No. 213 specifies a head injury criteria (HIC) of 1,000 and a maximum acceleration level for the chest of (60g.). These were based on the criteria that were specified for the adult male dummy in Standard No. 208. There were no injury criteria separately scaled from an adult dummy to reflect anatomical differences and differing injury tolerance of children. Table (6) shows the proposed reference values under consideration for Standard No. 213.
In the frontal test analysis the agency is proposing new limits on chest acceleration. Currently Standard No. 213 limits chest acceleration to 60 g's. There is no chest deflection limit in Standard No. 213 because the current Hybrid II dummies cannot measure chest deflection. Incorporating the 3-year-old dummy into Standard No. 213 would enable the agency to measure chest deformation-deflection.
HIC measures the injury that occurs, it does not matter if it is side or frontal impacts, it is designed to minimize skull fracture/brain injury due to contact with interior components of the vehicle. The reference values under consideration would replace the HIC 1000 limits in Standard No. 213 with the scaled HIC values adopted by the May 2000 air bag final rule i.e. 570 for the 3-year-old dummy; and 390 for the CRABI 12-month-old dummy. The agency would also calculate HIC over a 15 msec duration for Standard No. 213 based on recommendations from motor vehicle manufacturers that the duration for HIC computations should be limited to 15 msec with a limit of 700 for the 50th percentile adult male dummy. The agency has determined that the stringency of HIC 70015 was equivalent to HIC 100036 for long duration pulses, because HIC15 produces a lower numerical value for long duration events, its 700 lower failure threshold compensated for the reduction. The agency also believes that for pulse durations shorter than approximately 25 msec, the HIC15 700 requirement was more stringent then HIC36. 1000
Currently there are no neck injury criteria in Standard No. 213 because the current Hybrid II test dummy cannot measure forces on the neck. The Hybrid III 3-year old dummies and the 12-month-old CRABI are capable of measuring neck bending moments and forces in the fore and aft directions and axial compression and tension loads. The agency would incorporate an Nij criterion in Standard No. 213 and changes have been made to the critical limits to make them scalable from the 5th female in-position N15 limit without the axial load limits. The agency is also considering whether a different lateral measurement is needed in the neck area. Table 10 shows a summary of the reference values.
|Dummy||Chest Clip (g)||Chest Deflection (mm)||Head Excursion||HIC15||Nij|
|12 month old||50||30||20 inches||390||1|
|3 year old||55||34||20 inches||570||1|
IX. PERFORMANCE TEST
Researchers from the Roads and Traffic Authority (RTA) of New South Wales, Australia, found head strikes could be prevented in 90 degree tests depending in part on the depth of the side wings (1). In addition, these researchers found that a top tether had some ability to limit the lateral movement of the child restraint in side impacts, whereas it might not have a significant role in retaining the head within the confines of the child restraint.
The performance test that the agency is considering is based on an NPRM that the agency issued in 1974 (42 FR 7959; March 1, 1974). This test is very similar to the test that Australia uses in their child restraint rating's program, and is a test that researchers regularly conduct to evaluate side impact performance. Under the 1974 proposal a 90 degree lateral impact would have been conducted that simulated a 20 mph crash. When tested in this fashion, each child restraint would have had to retain the test dummy in the system, limit head motion to 19 inches in each lateral direction measured from the exterior surface of the dummy's head, and to suffer no loss of structural integrity. NHTSA subsequently withdrew the proposal after testing a number of restraints at a speed of 20 mph and at a horizontal angle of 60 degrees from the direction of the test platform travel. The research found that for outboard seating positions, only one of those restraints--one that required a tether--could meet the lateral head excursion limits that had been proposed in the 19974 NPRM. This was of concern because tethers were widely unused at that time. Further, the agency found that some restraints with impact shields, which performed well in frontal crashes and which were rarely misused, could not pass the lateral test even when placed in the center seating position. The agency decided not to pursue lateral testing of child restraints given the cost of the design changes that would be necessary to meet the lateral test, the problems with misuse of tethers, and the possible price sensitivity of child restraint sales. (43 FR 21470, 21474; May 18, 1978.)
The agency has revisited this issue in light of several developments in recent years. Forward-facing child restraints are now subject to a 28-inch head excursion limit that results in all of them having tethers. Vehicles are now required to have user-ready tether anchorages in rear seating positions, along with standardized child restraint anchorage systems, as part of the requirements of Standard No. 225. We expect that with user- ready anchorages in vehicles, tethers will generally be used, and thus there is a greater likelihood that countermeasures that depend on tether use will be effective.
Tethers alone, however, might not be able to limit head excursion to the prescribed limits. Design features such as deeper side wings and reinforced side structure may also need to be incorporated into child restraint system designs. Limited research (2) in recent years has found those measures to be promising in limiting head excursion. We also revisited this issue in light of TREAD's statutory mandate to consider improvements in side impact protection, and to do so by November 2001.
The dynamic side impact tests under consideration would limit head excursion and would have limits on other injury criteria. In addition, tested child restraints would be subject to system integrity requirements (S5. 1. 1) to reduce the likelihood that the child restraint system will disintegrate in a side impact. Belts and buckles would also have to meet the pre- and post-test requirements that apply to the frontal impact simulation.
Head excursion would be limited to 20 inches for the 3-year-old dummy and 20 inches for the 12-month-old dummy. We believe that the limit will help keep a child's head away from injurious door structure and from impacting objects if there is no intrusion. The 20 inch head excursion limit at 90 degrees is what we have chosen to be the excursion limit because in our testing with the 3-year-old dummy results tended to cluster around 20 inches. The average is 20.8 inches. The 20 mph velocity change was chosen because it is consistent with the speed used by New South Wales in its consumer ratings program, and with the pulses derived from Standard No. 214 (which have vehicle delta V's of 15 mph, but delta V in the door area above 20 mph.
However, there are other conditions where limiting excursion will be beneficial. First, if the dummy is sitting in the center seat position, there will be at least 30 inches between the dummy's head and the side door, in this case there will be enough distance to avoid striking the door even in this severe case considering eight inches of intrusion and 20 inches of excursion equals a total of 28 inches. Second, in lower speed crashes there will be less intrusion and less excursion. At some point the dummy would avoid the intruding door leading to benefits. The agency has not established this speed yet. Third, not all side impacts occur at the place where the child is sitting. In these cases, there would be no intrusion at that point. In essence reducing head contact with the side of the car will result in benefits, reducing head excursion will reduce the probability and severity of head contact. The agency has not been able to identify the speed at which these benefits will take place and has not been able to match them to the target population yet to enable itself to quantify the potential benefits.
The agency considered having the tethered child restraint oriented at 45 degrees and at 90 degrees to the direction of sled travel. Ninety degrees was selected to be consistent with the ECE and Australian ratings programs. Would the path of a child's head in a 90 degree impact be lateral? Would the path depend on factors such as the speed of the struck vehicle, and the point of impact to the struck vehicle (forward part, middle, rear)? The child restraint would be installed with the LATCH system. A door structure would not be present in the 20 mph test. If the child restraint has a top tether, it would be attached. Comments are requested on the whether the lower part of the child restraint should be attached to the standard seat assembly by way of the LATCH system, a lap belt, or a lap and shoulder belt. All passenger vehicles manufactured on or after September 1, 2002 will be equipped with LATCH systems, and all child restraints manufactured on or after September 1, 2002 will have components that attach to the LATCH anchors in vehicles.
The test procedure would use the CRABI and Hybrid III 3-year-old dummies in the tested child restraints. We are mindful that there is some question whether these dummies are appropriate for use in side impact testing. The Hybrid III 3-year-old has a rigid shoulder and a rigid torso, and probably would not fully replicate a child's kinematics in a side impact. The agency and the biomechanical community are developing more advanced side impact dummies, such as the Q series omni-directional 3-year-old (Q3) test dummy, which is the product of a European dummy manufacturer. We are also evaluating prototype heads and necks with side impact capabilities for the Hybrid III dummy. However, in the interim before the new dummies are available, we are testing with the existing Hybrid III 3-year-old.
The child restraint manufacturers have never had to design their child restraints with a side impact standard in mind. The fact that some of the child restraints can meet the reference values for Nij, gives the agency hope that design changes could be made to the forward-facing child restraint that would allow the child restraints to pass the test without any significant changes to the design of child restraints. The agency believes that for the rear-facing infant restraints the addition of one inch thick padding, covering an area of 96 square inches should be adequate in some cases. There are about 700,000 infant seats sold annually. It is estimated that the incremental cost of padding per seat is approximately $2.50 (in 2000 economics). Total cost of padding the infant seats is $1.750 million. Comments are requested about the feasibility of meeting the reference values.
NHTSA believes that a second potential countermeasure for side impact injuries would be to increase the sides (wings) of the top portion of the forward-facing child restraint. This is the area where, in the event of a side impact crash, the child's head might go past and contact the vehicle side. Increasing the sides of the restraint would help to contain the child's head within the restraint. Bigger sides would require more plastic (probably reinforced plastic) and additional padding. The agency estimates that on a convertible child restraint, which typically cost the consumer $70.00, the bigger sides and padding would cost an additional $15.00 for the forward facing child seat, a roughly 20 percent increase in price. This countermeasure does not appear feasible for infants in the rear-facing seat since the CRABI's head was retained with the sides of the child restraint.
There are approximately four million rear facing and convertible child restraints sold annually. Of this total, it is estimated that there are approximately 3.3 million convertible seats, and 700,000 infant seats. If the chosen countermeasure extended the side of convertible seats, then the annual consumer cost would be approximately $49.5 million (3,300,000 x $15). Comments are requested about the feasibility of producing larger wings and their ability to reduce injury.
A third potential countermeasure is to not allow the child restraint to turn toward the impact. This might possibly be achieved by adding a tether to the bottom of a forward facing child restraint and anchoring it to the floor. The cost of a tether is $4.14 (in year 2000 economics) per child restraint. For 4 million child restraints the cost would be $16.6 million(3). The cost to a motor vehicle to have two tether anchors attached to the floor in the rear seat would be $1.40 (2 x $0.70). Assuming 15.5 million vehicle sales per year, the cost would be $21.7 million (15.5 million x $1.40). Thus, the total cost for this countermeasure would be $38.3 million.
A fourth potential countermeasure is to have rigid LATCH connectors. The difference in cost between the rigid and the flexible LATCH connector is $25.15 per child restraint. For 4 million child restraints the total cost would be $100.6 million. (4)
Table 11 presents a summary of potential total costs for each countermeasure. Since the agency has not determined the feasibility of these countermeasures in a side impact test, there is a wide range of potential total costs at this time. If possible, the convertible seats would take the low cost option and make minor changes. While the potential costs range up to $100 million, this is not a likely scenario.
Cost Summary ($ millions)
|Countermeasure||Infant Seats||Convertible Seats||Vehicle Manufacturers|
|1. Design changes||$1.75||0||0|
|3. Added Tether||$2.9||$13.7||$21.7|
|4. Rigid Latch||$17.6||$83.0||0|
Table 12 is a breakout of the cost of the sled tests to the child restraint manufacturers. Since each sled test can accommodate two child restraints, there would be a total of 360 sled tests at a cost of $1,300 per test. Total cost of testing all the child restraint models is $468,000 (360 x $1,300).
Costs of the Sled Tests
|Child Restraint Systems||Number of Models||Required Sled Tests
for Each Models
|Total Number of
XI. Cost Effectiveness
This section combines costs and benefits to provide a comparison of the estimated injuries and lives saved per dollar spent. It should be noted that costs occur when the vehicle is purchased, but benefits accrue over the lifetime of the vehicle. Benefits must therefore be discounted to reflect their present value and to put them on a common basis with costs.
In some instances, costs may exceed economic benefits, and in these cases, it is necessary to derive a net cost per equivalent fatality prevented. An equivalent fatality is defined as the sum of fatalities and nonfatal injuries prevented converted into fatality equivalents. This conversion is accomplished using the relative values of fatalities and injuries measured using a "willingness-to-pay" approach. This approach measures individuals' willingness to pay to avoid the risk of death or injury based on societal behavioral measures, such as pay differentials for more risky jobs.
Table 13 presents the relative estimated rational investment level to prevent one injury, by maximum injury severity. The data represent average costs for crash victims of all ages. AIS is an anatomically based system that classifies individual injuries by body region on a six point ordinal scale of risk to life. Injuries are assumed to be valued based on the relative costs of MAIS injuries(5).
Comprehensive Fatality and Injury Relative Values
|Injury Severity||1994 Relative Value* per injury|
|*includes the economic cost components and valuation for reduced quality of life|
Table 13 shows the estimated equivalent fatalities for adding the various countermeasures to child restraint systems. Table 14 shows the equivalent fatalities for the two types of test configurations.
Child restraint Used In The Different Test Configurations
|Injury Benefits||Equivalent Fatalities|
|20 mph No Wall||15 mph Rigid Wall||20 mph No Wall||15 mph Rigid Wall|
The following is an example of the calculation of the cost per equivalent fatalities for adding the various countermeasures to the child seats before discounting.
Appendix V of the "Regulatory Program of the United States Government, April 1, 1990 - March 31,1991, sets out guidance for regulatory impact analyses. One of the guidelines deals with discounting the monetary values of benefits and costs occurring in different years to their present value so that they are comparable. Historically, the agency has discounted future benefits and costs when they were monetary in nature. For example, the agency has discounted future increases in fuel consumption due to the increased weight caused by safety countermeasures, or decreases in property damage crash costs when a crash avoidance standard reduced the incidence of crashes, such as with center high-mounted stop lamps. The agency has not assigned dollar values to the reduction in fatalities and injuries, thus those benefits have not been discounted. The agency performs a cost-effectiveness analysis resulting in an estimate of the cost per equivalent life saved, as shown on the previous pages. The guidelines state, "An attempt should be made to quantify all potential real incremental benefits to society in monetary terms of the maximum extent possible. For the purposes of the cost-effectiveness analysis, the Office of Management and Budget (0MB) has requested that the agency compound costs or discount the benefits to account for the different points in time that they occur.
There is general agreement within the economic community that the appropriate basis for determining discount rates is the marginal opportunity costs of lost or displaced funds. When these funds involve capital investment, the marginal, real rate of return on capital must be considered. However, when these funds represent lost consumption, the appropriate measure is the rate at which society is willing to trade off future for current consumption. This is referred to as the social rate of time preference, and it is generally assumed that the consumption rate of interest, i.e. the real, after- tax rate of return on widely available savings instruments or investment opportunities, is the appropriate measure of its value.
Estimates of the social rate of time preference have been made by a number of authors. Robert Lind(6) estimated that the social rate of time preference is between zero and 6 percent, reflecting the rates of return on Treasury bills and stock market portfolios. Kolb and Sheraga(7) put the rate at between one and five percent, based on returns to stocks and three month Treasury bills. Moore and Viscusi(8) calculated a two percent real time rate of time preference for health, which they characterize as being consistent with financial market rates for the period covered by their study. Moore and Viscusi's estimate was derived by estimating the implicit discount rate for deferred health benefits exhibited by workers in their choice of job risk.
Four different discount values are shown as a sensitivity analysis. The 2 and 4 percent rates represent different estimates of the social rate of time preference for health and consumption. The 10 percent figure was required by 0MB Circular A-94, until October 29,1992. The 7 percent figure is the current OMB requirement, which represents the marginal pretax rate of return on an average investment in the private sector in recent years.
Safety benefits occur when there is a crash severe enough to potentially result in occupant death and injury, which could be at any time during the safety seat's lifetime. Data on 15,785 child safety seats were collected at the SAFE KIDS BUCKLE UP events from 2/01 to 9/01. As seen in Table 15 (a), the majority of the child seats in use were manufactured within the most recent period. One characteristic of the data is child safety seat by year of manufacture. For this analysis, the agency assumes that the distribution of child safety seats for the period pre-1981 to 2001 is an appropriate proxy measures for the distribution of such crashes over the child safety seat's lifetime (see Tables 15(a and b)).
Table 15 (a)
|Child safety seats by year of manufacture*|
|*Data from 15,785 seats seen at SAFE KIDS BUCKLE UP Events from 2/01-9/01|
Child Safety Seats Age and Discount Factor
|Seat age in years||Frequency||Survival Probability||Weighted
|Percent of Frequency||Discount
Multiplying the percent of a child safety seat's total lifetime usage that occurs in each year by the discount factor and summing these percentages over the 21 years percent discount rate results in the factors shown at the top of Table 16(a). For example, at a 7 percent discount rate, the present discounted value factor is estimated to be 0.8954. These values are multiplied by the equivalent lives saved to determine their present value (e.g., Table 16(a) 10.154 x 0.8954 = 9.09). The costs per equivalent life saved for design changes and larger wings are then recomputed and shown in Table 16(b) e.g., ($1.75 million/9.83 = $0.18 million, $1.75 million/8.07 = $0.20 million).
Equivalent Lives Saved
|Base Equivalent||2 Percent||4 Percent||7 Percent||10 Percent|
|Child Restraints||X .9678||X .9375||X .8954||X .8568|
|Design Changes = 10.154||9.83||9.51||9.09||8.70|
|Larger Wings = 4.205||4.07||3.94||3.77||3.60|
|Add Tether = 14.359||13.90||13.46||12.86||12.30|
|Rigid Latch = 14.359||13.90||13.46||12.86||12.30|
Costs per Equivalent Life Saved (in millions)
|Child Restraint||Undiscounted||2 percent||4 percent||7 percent||10 percent|
|Design Changes = $1.75||$0.18||$ 0.18||$0.18||$0.19||$0.20|
|Larger Wings = $49.5||$11.772||$12.16||$12.56||$13.15||$13.74|
|Add Tether= $38.3||$2.667||$2.76||$2.85||$2.98||$3.11|
|Rigid Latch = $100.6||$7.006||$7.24||$7.47||$7.83||$8.18|
XII. Initial Regulatory Flexibility Analysis
The Regulatory Flexibility Act of 1980 (Public Law 96-354) requires agencies to evaluate the potential effects of their proposed and final rules on small businesses, small organizations and small governmental jurisdictions.
Section 603 of the Act requires agencies to prepare and make available for public comment an initial regulatory flexibility analysis (IRFA) describing the impact of proposed rules on small entities. Section 603(b) of the Act specifies the content of an IRFA. Each IRFA must contain:
1. Description of the reasons why action by the agency is being considered
NHTSA is considering this action to protect children involved in side impact crashes.
Better protection for children involved in side impact crashes is important because of the number of children killed and injured in vehicle accidents. In 1999, 420 children ages zero to twelve were killed in side impact crashes and approximately 16,800 were injured as occupants in motor vehicle side impact crashes.
While child seats are highly effective in reducing the likelihood of death or serious injury in motor vehicle crashes, the degree of their effectiveness in side impact crashes is still to be determined. The more biofidelic 12-month-old and Hybrid III family dummies equipped with greater instrumentation are available for crash tests. A more stringent and dummy size specific injury criteria have been developed. NHTSA believes that child safety would be improved with these new dummies and the more stringent injury criteria.
The ANPRM is also issued in response to the mandate in the Transportation Recall Enhancement, Accountability and Documentation Act (the TREAD Act) (November 1, 2000, Pub. L. 106-414, 114 Stat. 1800) to initiate a rulemaking for the purpose of improving the safety of child restraints. Section 14(a) of the TREAD Act mandates that the agency "initiate a rulemaking for the purpose of improving the safety of child restraints, including minimizing head injuries from side impact collisions."
2. Objectives of, and legal basis for, the proposed rule
This document proposes to require that motor vehicles and add-on child restraints be equipped with means to reduce the severity of the impact on the children during side impact crashes.
Children are not very well protected in nearside side impact crashes. Thus, it is difficult to design countermeasures for the vehicle or child restraints in side impact crashes. Children seated in the center and far-side seats are better protected because there is greater distance between the child and the door given vehicle intrusion.
NHTSA has issued this ANPRM under the authority of 49 U.S.C. 322, 30111, 30115, 30117 and 30166; delegation of authority at 49 CFR 1.50. The agency is authorized to issue Federal motor vehicle safety standards that meet the need for motor vehicle safety. This proposal is also issued under section 14 of the TREAD Act.
XIII. SMALL BUSINESS IMPACTS
Description and estimate of the number of small entities to which the proposed rule will apply
The proposed rule would affect portable child restraint manufacturers. NHTSA estimates there to be about 10 manufacturers of portable child restraints, four of which could be small businesses.
Business entities were generally defined as small businesses by Standard Industrial Classification (SIC) code, for the purposes of receiving Small Business Administration assistance. The SIC codes have changed. In the small business section of our analyses we have used 500 employees as the cut-off for small businesses for many years. Business entities are now defined as small businesses using the North American Industry Classification System (NAICS) code, for the purposes of receiving Small Business Administration assistance. One of the criteria for determining size, as stated in 13 CFR 121.201, is the number of employees in the firm. There is no separate NAICS code for child restraints. Possible categories include: a) To qualify as a small business in the Motor Vehicle Seating and Interior Trim Category (NAICS 336360), the firm must have fewer than 500 employees, b) In the "All Other Motor Vehicle Parts Manufacturing" category (NAICS 336399), the firm must have 750 employees, c) In the "All Other Transportation Equipment Manufacturing" category (NAICS 336999), the firm must have 500 employees. We believe child restraints fit better into category a) or c). Thus, we will continue to use 500 employees as the limit. Table 17 gives a listing of the child seat manufacturers.
The agency believes that a rule would have a significant impact on some of these businesses. A rule should have a positive effect on the manufacturers of test dummies and instrumentation for test dummies. In order to do the required tests, an increased number of test dummies would be needed. Currently, there are four manufacturers of dummies or parts of dummies: First Technology Safety Systems, Advance Safety Technology Corp., UTAMA, and GESAC. There are six manufacturers of instrumentation for test dummies. Four are manufacturers of load cells (P.A. Denton, First Technology Safety Systems, Sensor Developments, Inc., and Sensotec) and two are manufacturers of accelerometers (Endevco and Entran). All of these, except Endevco, are small businesses.
Employment of Child Restraint Manufacturers*
(less than 500 employees qualifies as a small business)
|Manufacturer||Number of Employees|
|Babyhood Manufacturing Co.||10|
|COSCO (Dorel Company)||1,000|
|Early Development Co. has less than 10 employees, however, it is partly owned and a joint venture with Takata of Japan||large company|
|Evenflo itself has 250 employees, but Evenflo is a division of Spalding & Evenflo Co. Inc.||2,600|
|Gerry is a product of GERICO, which has 900 employees, but GERICO is a subdivision of Huffy||11,000|
|Safeline Children's Products Co.||< 10|
|Little Cargo, Inc.||< 10|
*Source: Standard and Poor's Register of Corporation Directors- mid Executives, (1995 updated).
4. Description of the projected reporting, record keeping and other compliance requirements for small entities
A final rule would adopt new performance requirements that would enhance the safety of child restraints. Child restraint manufacturers would have to certify that their products comply with the final rule's requirements. Manufacturers could use any means to determine that their products comply, so long as they exercise due care in making their certification.
Some seats would only need to be redesigned, but some might need larger wings. NHTSA estimates that the wing requirement would add $15 to the current price of a seat. Adding a second tether to the seat would cost approximately $4.14. Another option would be the ISO hard attachment. This would cost an additional $25.15 to the cost of the child restraint.
The cost increase could significantly raise the price of child restraints, which may have a significant economic impact on a substantial number of small businesses. NHTSA does not know what the elasticity of demand is for child restraints. An increase in the price of a child restraint could lead to a decrease in demand for the product, notwithstanding the restraint use laws. While child restraint use is mandated by each State, there is significant nonuse of restraints.
Comments are requested on the effect that raising child restraint prices by $15 (larger wings) to possibly $25 would have on small businesses that manufacture child restraints. Would an across-the-board increase in price reduce small business sales? What is the magnitude of the impact?
An increase in child seat prices may also affect loaner and giveaway programs. The agency does not know how many programs exist and requests information on this issue. A cost increase could result in fewer seats being purchased by the program for loan or giveaway. Comments are requested on the impacts on not-for-profit programs. While such a program could have fewer seats available, NHTSA seeks information on the extent to which the number of seats a program makes available impacts on the organization itself. For example, do proceeds from loaner or giveaway programs (where a nominal fee might be charged) support the not-for-profit organization?
5. Duplication with other Federal rules
There are no relevant Federal rules that may duplicate, overlap or conflict with the proposed rule.
6. Description of any significant alternatives to the proposed rule
One alternative is adding bigger wings on the child restraint. Another alternative is adding a second tether to the child restraint to prevent lateral movement of the child restraint. A third alternative is the ISO hard anchors.
Comments are requested on these issues.
XIV. UNFUNDED MANDATES REFORM ACT
The Unfunded Mandates Reform Act of 1995 (Public Law 104-4) requires agencies to prepare a written assessment of the costs, benefits, and other effects of proposed or final rules that include a Federal mandate likely to result in the expenditures by State, local or tribal governments, in the aggregate, or by the private sector, of more than $100 million annually (adjusted annually for inflation with base year of 1995). Adjusting this amount by the implicit gross domestic product price deflator for the year 2000 results in $109 million (106.99/98.1 = 1.09). The assessment may be included in conjunction with other assessments, as it is here.
This ANPRM is not estimated to result in expenditures by State, local or tribal governments of more than $109 million annually. It is not going to result in the expenditure by child restraint system manufacturers of more than $109 million annually. The highest estimated annual cost is $100.6 million.
1. Kelly et al., "Child Restraint Performance in Side Impacts With and Without Top Tethers and With and Without Rigid Attachment (CANFIX)," 1995 International IROCOBI Conference on the Biomechanics of Impact, September 13 - 16, 1995, Brunnen, Switzerland.
3. FMVSS No.213 FMVSS No.225 Child Restraint Systems, Child Restraint Anchorage Systems Final Economic Assessment, Pages 38 -39. Office of Regulatory Analysis, Plans and Policy, February 1999.
5. The relative value of an MAIS 1 injury was estimated as follows: The quality of life portion of an MAIS 1 injury was computed by subtracting from MAIS 1 comprehensive costs the costs of travel delay and property damage. This calculation is:($10,840-$3,263-$203=$7,374), where $3,263 is property damage and $203 is travel delay. Dividing this by the total comprehensive costs for a fatality (2,854,500-9,591) gives .0026(7,374/2,844,909) to four decimal places. The MAIS figures are taken from The Economic Cost of Motor Vehicle Crashes 1994", pages 8, 59, and 66.
6. Lind, RC, A primer on the Major Issues Relating to the Discount Rate for Evaluating National Energy Options, in Discounting for Time and Risks in Energy Policy, 1982, (Washington, D.C. Resources for the Future, Inc.).
7. J. Kolb and J.D. Sheraga, A Suggested Approach for Discounting the Benefits and Costs of Environmental Regulations, unpublished working papers.
8. Moore, M.J., and Viscusi, W.K.,Discounting Environmental Health Risks: New Evidence and Policy Implications, Journal of Environmental Economics and Management, V.18, No. 2, March 1990, part 2 of 2.