NHTSA | NHTSA Task Order 1 | NHTSA Task Order 2 | NHTSA Task Order 3 | NHTSA Task Order 4

NHTSA Task Order 2: Anton’s Law: Safety Definition

Study Background

On December 4, 2002, the President of the United States signed Anton’s Law External Website to provide for the improvement of the safety of child restraints in passenger motor vehicles, and for other purposes. Under the new law, NHTSA was required to conduct a series of analyses in order to support the validity of the legislation and provide direction for further improvements in the federal regulation of child passenger safety.

In order to fulfill these requirements, NHTSA contracted with the Center for Injury Research and Prevention, based on its comprehensive data set of real-world crashes and its leadership in the field of Child Passenger Safety Research, through the PCPS Study.

Through an Indefinite Quantity Contract (IQC) between the Center and NHTSA, a unique partnership was created: one that enabled NHTSA to achieve its mandate by effectively sustaining and expanding the research capabilities of the PCPS team. Endeavors through the IQC not only promote further analysis of data within the surveillance system but also support research that assists in the interpretation and use of the findings from the PCPS study.

Study Goals

Center researchers were contracted to address this need with support from NHTSA. This project sought to assist NHTSA in achieving the following requirements laid out by Anton’s Law:

  • Evaluate injury performance criteria for child restraints, including booster seats and other products for use in passenger motor vehicles for the restraint of children weighing more than 50 pounds
  • Develop and adopt an anthropometric test device that simulates a 10-year-old child for use in testing child restraints used in motor vehicles (including injury criteria)
  • Provide further support for the requirement of lap-shoulder belt systems for all designated seating positions (as opposed to seating positions with lap-only belts)
  • Evaluate integrated or built-in child restraints and booster seats, in part, based on the availability of real-world crash data

Study Approach

Task Order Two analyses were based on data from the PCPS surveillance system according to the objectives above. The scope of work was divided into the following tasks:

Individual Project Details

The results of each portion of this research are summarized below.

Injury/Fatality Patterns for Children Weighing more than 50 Pounds
  • Objective

    As part of Anton’s Law, the NHTSA was directed to consider injury performance criteria for child restraints, including booster seats and other products for use in passenger motor vehicles for the restraint of children weighing more than 50 pounds.

  • Study Population

    The study sample of children in crashes was divided into groups based on the age and weight of the child occupant. All children between 6-12 years of age weighing between 50- 105 lbs were grouped together as the primary group of interest (target group). Children who were at least 12 months old, weighing between 20- 49 lbs. were grouped together as comparison group 1 (small comparison group). Note that all children under 12 months in a rear-facing child restraint were excluded from this comparison group as it was felt that their restraint use and injury risk characteristics would be markedly different from the target group and render them an inappropriate comparison. A second comparison group consisted of all children weighing over 105 lbs, regardless of their age (large comparison group). Specifically excluded from all groups were front-seated children in passenger airbag deployment crashes as there was a desire to exclude the potential effects of the deploying airbag as a source of injury for children in this study.

    Children in the target group accounted for 36% of all children in this study, thus making up a substantial proportion of children involved in automobile crashes.

  • Results

    1 out of every 8 children in the target group was restrained with a lap-only belt, the majority of whom were seated in the center rear seating position (other children in lap-only belts may have been seated in outboard positions equipped with lap/shoulder (L/S) belts but placed the shoulder portion of the belt either behind their back or under their arm).

    Children using lap-only belts were twice as likely to be injured (2.6% vs. 1.3%) as children using the lap and shoulder portions of the L/S belt. It is possible that the parent-reported nature of the data used to classify belt type underestimates the number of children who placed the shoulder portion of the L/S belt under their arm or behind their back due to the discomfort from the improperly fitting belt. Therefore, it is possible that the risk of injury to children using both the lap and shoulder belt is even lower than we determined.

    Nearly one quarter of injured children were seated in the right front seat, and an additional 18% were seated in the center rear. Frontal impacts accounted for nearly one half of all injured children, with rear impacts (21%), near side (13.5%) and far side (8.4%) accounting for the remainder of the crashes in which children were injured. Passenger cars account for the majority (57%) of vehicles in which injured children were riding.

  • Conclusions

    Although there have been improvements in restraint use since this study was first conducted, children of booster seat age and size are at significantly greater risk of injury than their younger counterparts who are more likely to be appropriately restrained in some form of child restraint system (CRS). Current CRS use, specifically BPB use, remains low in the target group, thus putting this group at particularly high risk for seat belt-associated injuries.

Development of Anthropomorphic Test Device Simulating a 10-year-old Child Study Sample
  • Objective

    As part of Anton’s Law, the NHTSA was directed to develop and adopt an anthropometric test device (ATD) and accompanying injury criteria that simulated a 10-year-old child for use in testing child restraints used in motor vehicles. Through this analysis, in part, heped NHTSA to determine how children of the appropriate age and size (i.e. with respect to the approximate size of an average 10 year-old) are being injured/killed in current restraints as used (including belt use/misuse, other devices, etc.), to determine any characteristic patterns of injuries based on restraint status and to correlate similar patterns of postures with associated injuries, of the 10-year-old ATD.

  • Study Sample

    The inclusion criteria for this analysis were based on the weight represented by the Anthropomorphic Test Device Simulating a 10-Year-Old Child (10-year-old ATD): 75 lbs. All 2,120 children between 9-12 years of age weighing between 60- 90 lbs in the study sample were grouped together as the primary group of interest. This weight group reflects approximately ± 25 % of the weight of the 10-year-old ATD and also reflects an average 9- and 12-year-old child respectively (based on most current CDC growth charts [9]). The study sample was further subdivided into two groups based on restraint status of the child occupant: Restrained (children who were restrained by either a seat belt, child restraint or after market products) and Unrestrained (children who were not restrained by a belt or child restraint device). The seat belt-restrained children were further categorized as those using only the lap portion (either due to availability or due to placing the shoulder portion of the belt behind the back) and those using both the lap and shoulder portion of the belt.

    Children between 9-12 years of age, weighing 60- 90 lbs., a representation of the 10 year-old ATD (the study sample), accounted for 10.7% of all children from the Partners for Child Passenger Safety study.

  • Results

    Overall, there was a high rate of reported restraint use in the overall sample (97%). 96% of the children were restrained by a seat belt, and 1% used some after-market product. Among the children restrained by seat belts, 86.5% used both portions of the lap/ shoulder belt while 12.2% were restrained only by the lap belt.

    Nearly one quarter of injured children were seated in the right front seat, and an additional 17.8% were seated in the center rear. These children are at significantly greater risk of injury when unrestrained. Given that an injury occurred, the restrained group appears to have relatively fewer head, face and lower extremity injuries and relatively more injuries to the neck/spine, abdomen and upper extremities compared to the unrestrained group. Head injuries are the most common consequential injury.

Lap/Shoulder Belts in the Center Rear: Occupancy Patterns, Use Patterns, and Performance
  • Objective

    As part of Anton’s Law, NHTSA was directed to require a lap and shoulder belt assembly for each rear designated seating position in a passenger motor vehicle with a gross vehicle weight rating of 10,000 pounds or less. In preparation for this change, NHTSA supported the Center in conducting an analysis to assess the performance and use of lap/shoulder belts in rear center seating positions and identify any trends in the data to support/negate the inclusion of lap/shoulder belts for all rear seating positions.

  • Study Sample

    The study sample for these analyses was children seated in vehicles equipped with a lap/shoulder belt in the center rear (2nd and/or 3rd row) (target group). Data on which vehicles were equipped with this restraint system were obtained from NHTSA via their Buying a Safer Car brochure. Children in vehicles equipped with only a lap belt in the center rear seat position were chosen as the comparison group. The comparison group was limited to model year 1998 and newer passenger cars, minivans, and SUVs to better match the potential sample of vehicles in the target group.

  • Results

    Overall, one-third of center rear-seated children used a seat belt. Among these children, those in vehicles equipped with a lap/shoulder belt are 81% less likely to sustain a serious injury than those seated in the center rear equipped with a lap-only belt system. This benefit may be directly due to the restraint but also may be attributed to improvements in vehicle crashworthiness and impact protection found in those vehicles offering a shoulder belt in this position. Over one-third of occupants in the center rear seating position are belt users.

    The presence of a shoulder belt in the center rear seating position is associated with the seating pattern for children in motor vehicles to a limited degree. This effect is seen only when there is a single child passenger.

    It is important to note that the injury risks measured in this analysis are injury risks for the restraints in an AS USED status. Not all of the belted children in the target group are using the shoulder portion of the belt.

Built-in Child Restraint Systems: Use Patterns and Performance
  • Objective

    As part of Anton’s Law, the NHTSA was directed to evaluate integrated or built- in child restraints and boosters seats, in part, based on the availability of real-world crash data. Built-in CRS are typically designed for use for children weighing 20 to 60 lbs.

  • Study Sample

    The study sample for these analyses was children less than 9 years of age restrained in the rear rows in forward-facing child safety systems (both built-in and add-on devices). Large vans, pickup trucks, and model years 1990-1992 and 2002 for other vehicle types were excluded from the analyses to better match the potential sample of vehicles with built- in CRS. All children in built-in CRS were grouped together as the primary group of interest (target group).

    Three comparison groups were chosen: children in add-on forward- facing child safety seats (FFCSS), belt-positioning booster seats (BPB), and shield. These groups were chosen to provide a robust comparison to other child restraint systems that children of the size appropriate for built-in CRS might use.

  • Results

    These analyses indicate that the built-in systems do not provide enhanced performance over add-on CRS.

    Within the study population, 1.7% of all children less than 9 years of age and 3.3% of those children less than 9 years of age in any child restraint system.

    The data show that the typical child using a built- in device is less than three years of age and less than 40 lbs which suggests tha t those using built- in restraints are more similar to those in add-on FFCSS than those in add-on BPB. Differences among the restraint types in seating position, vehicle type, and vehicle model year are directly related to availability and configuration of the built- in restraints.

    The injury rates among the four restraint types were low and not statistically different from one another. A review of the body regions with serious injury indicates that efforts to reduce injuries in all three restraint types should focus on the head and, for those in harness based systems, on the lower extremity as well.

    In order to understand why certain people chose to use this system over an add-on system, further research is needed.

Our Research