Abstract
Efforts to improve child restraint designs would benefit from more detailed information on how child occupants are dying in crashes. Detailed reports involving 92 children (ages 5 and younger) in child restraints who died in crashes in 2000 were obtained from police departments. Cases were reviewed to obtain basic crash information and determine the factor most responsible for the fatality. Half of the crashes were considered unsurvivable for the child, and 12 percent of fatalities were judged to result from gross misuse of the child restraint. Forty percent of all of the crashes were side impacts, and in all fatal side impact crashes there was intrusion at the child’s seating position.
A total of 769 children ages 5 and younger died in motor vehicle crashes in 2001, 569 (74 percent) as passenger vehicle occupants. Among the fatally injured child occupants, 39 percent reportedly were unrestrained, 17 percent were reported in adult belts, and 37 percent were in child restraints. Although observational data suggest child restraints often are misused, restraints still are very effective in reducing fatality risk in crashes. These reductions are as much as 71 percent among infants and 54 percent among toddlers [National Highway Traffic Safety Administration (NHTSA), 2001].
Federal Motor Vehicle Safety Standard (FMVSS) 213 requires child restraints to meet certain specifications when subjected to frontal sled tests simulating 48 km/h (30 mi/h) crash tests. Limits are placed on head and knee excursions to limit forward movement in the event of a frontal crash, and the risk of injury to the head and chest also are measured. The standard was last modified in the late 1990s and is under review for modification in the near future. FMVSS 213 currently certifies child restraint performance only in frontal crashes, but NHTSA is considering incorporating a side impact test and has solicited comments about various possible side impact test configurations. However, little is known about how children are being seriously injured or dying in crashes, so it is difficult to make clear recommendations about what changes should be made to the standard to improve child occupant protection.
Restraint misuse also is an issue. Child restraints are complicated to install and use for a number of reasons including changing requirements as children grow, incompatibility between the restraints and vehicle seat designs, and variations among vehicle seat belts that require different methods to attach a restraint securely to a vehicle seat. The result is that many parents, despite their best efforts, are not using or installing child restraints correctly [National Safe Kids Campaign, 1999]. It is unknown, however, whether the most common kinds of misuse are much of a factor in the deaths of children who are in restraints.
To shed light on these issues, a more thorough understanding is needed about how children in restraints are dying in serious crashes. The current study attempts a more in-depth analysis than can be routinely undertaken using Fatality Analysis Reporting System (FARS) data. Local police departments were contacted to obtain more complete reports of the crashes that occurred in 2000 in which a child reported to be in a child restraint was fatally injured. In many cases, detailed crash reports were available for these motor vehicle fatalities and often included photographs. Researchers analyzed the available evidence and tried to make a judgment about how the fatality occurred and whether the crash potentially was survivable. A review of these crashes may provide evidence on how to improve the effectiveness of child restraints in crashes.
METHODS
FARS data were obtained for the year 2000 [NHTSA, 2001]. FARS is a database compiled by NHTSA. It is a census of fatal crashes that occur on public roads in the United States. There were 621 occupant fatalities involving children younger than 6, and 143 of these fatally injured children reportedly were in child restraints. State police, local police, and sheriff’s offices were contacted to request information on the crashes. Nineteen of the crashes were in three states (California, Montana, and Pennsylvania) that would not supply cases because of laws prohibiting release of information about them. For an additional 23 crashes, only basic police crash reports were available. Information on nine additional crashes could not be obtained. The remaining 92 crashes included a total of 96 fatally injured children in child restraints. However, after receiving the police reports, it was determined that four cases were incorrectly coded in FARS. Three children were using lap/shoulder belts (not riding in child restraints), and one child was not fatally injured. This left a total of 92 child fatalities.
All of the cases included more information than the basic police report, but the majority of reports was detailed and included numerous photos. In 15 cases, the investigating officers were contacted for more information. All available information was examined to catalogue basic crash factors (child age, seating position, child restraint, crash type, case vehicle, crash partner, etc.).
The cases also were reviewed to determine the factor that was most responsible for the fatality. A group of four reviewers, including three crash investigation specialists, reviewed each case and reached consensus. Six principal contributing factors were identified: unsurvivable, restraint misuse, ejection, contact with occupants or cargo, non-catastrophic frontal impact, and non-catastrophic side impact. Cases were considered to be unsurvivable if there was no survival space left at the child’s seating position or death was due to fire or drowning; no other contributing factors were considered in these crashes (note that this typically could be done only if photos were available, but occasionally the police report narrative would indicate massive crush, e.g., in one case a tractor-trailer ended up on top of a car). The fatality was classified as occurring because of restraint misuse if the crash was judged to be potentially survivable and there was clear evidence that significant misuse of the child restraint was responsible for the fatality — e.g., restraint not attached or very loosely attached or internal harness not used. Other potentially survivable crashes were classified as fatal due to ejection — i.e., a child was ejected from a restraint that was correctly attached to the vehicle and there was no evidence of gross misuse. In each case of ejection, the child restraint was equipped with a three-point harness. It is possible that the shoulder harness clip may have been misused or not used at all, and the flexible shoulders of the children may have slid out from the harnesses in the predominantly rollover crashes. The principal factor was classified as involving contact with occupants or cargo if there was evidence of the same and it was considered unlikely that the child would have struck other vehicle structures. All other potentially survivable crashes in which none of these factors was clearly present were categorized according to the crash type (e.g., non-catastrophic side impact and non-catastrophic frontal impact). Although these crashes were classified as survivable, this does not mean that an improved child restraint or vehicle countermeasures necessarily would have prevented the fatality. As reported here, many of these crashes were very severe and had intrusion directly at the child’s seating position. However, they did not involve complete destruction of the child’s survival space.
RESULTS
Table A1 in the appendix outlines each of the cases included in the study. Details are provided regarding child age, seating position, child restraint type, crash type, case vehicle and crash partner, as well as the factor deemed most important in the cause of death.
About 70 percent of the children were age 2 or younger, and fewer than 10 percent were age 4 or older. This is not surprising because child restraint use declines rapidly after age 4 [Decina and Knoebel, 1997; Winston et al., 1999]. There were fewer infants (9 percent) than expected in the sample. This may be because of reporting bias. Many police reports give age in 1-year increments, and some children reported as 1 year-olds might have been younger. In two cases in which the investigator was contacted, the reported age of 1 was closer to 6 months.
Only 15 percent of children were in front seats, and 3 percent were in middle front seats. Most (68 percent) of the children were in the outboard positions of back seats, and 15 percent were in middle rear positions. An additional 2 percent were in other seats.
Table 1 shows child restraint types. In 32 percent of the deaths, the type of child restraint was unknown. Use of forward- facing restraints was verified in 47 percent of the crashes. Shield booster use was reported in 11 percent of the crashes, followed closely by rear-facing restraints (8 percent). Belt-positioning booster seat use was reported in only two cases.
Table 1.
Child Restraint Types
| Child Restraint | Number | Percent |
|---|---|---|
| All forward facing | 43 | 47 |
| Shield with three-point harness | 18 | 20 |
| Unspecified forward facing | 15 | 16 |
| Five-point harness | 8 | 9 |
| T-shield | 2 | 2 |
| Shield booster | 10 | 11 |
| Rear facing | 7 | 8 |
| Belt-positioning booster | 2 | 2 |
| Not approved | 1 | 1 |
| Unknown | 29 | 32 |
CHILD INJURIES
The fatal injury location was unknown in 38 percent of the cases. Table 2 shows the numbers and percentages of reported injuries by injury location. In more than 80 percent of the cases with known injuries, head trauma contributed to the fatality. Neck injuries were indicated in 16 percent of the cases.
Table 2.
Known Injury Information
| Fatal Injury | Number | Percent |
|---|---|---|
| Head | 35 | 61 |
| Head/chest | 6 | 11 |
| Head/neck | 5 | 9 |
| Neck | 4 | 7 |
| Blunt trauma | 3 | 5 |
| Fire | 3 | 5 |
| Drowning | 1 | 2 |
| Total | 57 | 100 |
CASE VEHICLE
The majority (72 percent) of children who died were riding in passenger cars. The remaining fatalities occurred in sport utility vehicles (SUVs) (11 percent), pickup trucks and vans combined (10 percent), and minivans (8 percent).
CRASH PARTNER OR OBJECT
Seventeen percent of the child fatalities occurred when the vehicle collided with a stationary object (embankment, bridge, guard rail, tree, pole). About a third of the fatalities occurred in crashes in which the crash partner was a heavy vehicle (tractor-trailer, commercial truck, train, farm tractor) (Table 3). The majority (79 percent) of the heavy vehicles were tractor-trailers. About a third of the cases involved impacts with SUVs, pickups, or large vans. By comparison, among all passenger vehicle occupant fatalities in 2000 involving children younger than 6 who were not seated in child restraints, only 15 percent involved heavy vehicles and 21 percent involved SUVs, pickups, or large vans. Thus, fatalities in which children in child restraints were killed were much more likely to involve crashes with vehicles of greater mass.
Table 3.
Struck or Striking Vehicle or Object
| Percent | |
|---|---|
| SUV/pickup truck/van | 33 |
| Heavy vehicle | 32 |
| Stationary object | 17 |
| Passenger car/minivan | 13 |
| Rollover | 5 |
There were 69 vehicle-to-vehicle crashes (excluding cases in which fire was a factor). In 41 percent of these crashes the partner vehicle was a heavy vehicle (Table 4). The next most common other vehicle was a pickup/SUV/van. None of the fatalities to restrained children in SUVs or pickups occurred from collisions with cars.
Table 4.
Vehicle Types in Vehicle-to-Vehicle Crashes
| Case Vehicle | Companion Vehicle | Percent |
|---|---|---|
| Passenger car | Passenger car | 12 |
| Minivan | 1 | |
| Pickup/SUV/van | 39 | |
| Heavy vehicle | 29 | |
| Minivan | Passenger car | 3 |
| Heavy vehicle | 3 | |
| Pickup truck | Pickup/SUV/van | 3 |
| Heavy vehicle | 3 | |
| SUV | Pickup/SUV/van | 1 |
| Heavy vehicle | 6 |
PRINCIPAL CONTRIBUTING FACTOR
Half of the fatal crashes were considered unsurvivable, although the percentage varied by crash type. About 45 percent of side and frontal impacts were considered unsurvivable, as were 14 percent of rollovers and 82 percent of rear impacts. Confirmation that these crashes were very severe is corroborated by the other deaths in the vehicle. In crashes considered unsurvivable, 80 percent had at least one other occupant death, compared with 37 percent in the potentially survivable crashes. Significant child restraint misuse accounted for 12 percent of the fatalities. Ejections in three-point harness restraints and contact with occupants or cargo each accounted for 4 percent. Cases in which none of the above factors were present accounted for 29 percent — non-catastrophic side impacts (16 percent) and non-catastrophic frontal impacts (13 percent).
CRASH TYPE
Side impacts accounted for the largest number (40 percent) of fatalities; in 70 percent of these crashes, the child was seated on the struck side. About a third were frontal impacts, 12 percent were rear impacts, 8 percent were rollovers, and 6 percent involved roof crush.
Side Impacts (37 deaths)
Almost half (n=17) of the side impacts were deemed unsurvivable. Figures 1 and 2 show examples of side impacts that were judged unsurvivable for the child. In Case 71, the 1-year, 10-month-old child was seated in the rear outboard seat on the struck side of a passenger car that was hit by an SUV. In Case 72, the infant was seated in a rear-facing restraint in the rear outboard seat on the struck side. The passenger car was hit by a tractor-trailer with such energy that the car was split into three pieces. The driver also died.
Fig. 1.
Unsurvivable side impact with SUV, Case 71.
Fig. 2.
Unsurvivable side impact with tractor-trailer, Case 72.
Thirty-eight percent (n=14) of the crashes in which no other factor was present were deemed potentially survivable (non-catastrophic side impact). That is, vehicle intrusion did not extend completely into the child’s seating position. Some of these crashes were very severe, and it is possible that even the most sophisticated countermeasures could not have prevented the fatalities. However, there also were some crashes without excessive amounts of intrusion. Each of these 14 crashes had the primary intrusion located at the child’s seating position, and it is possible that the head contacted an external object through the window. In addition, at least 11 of these crashes involved a forward crash component. In the majority of deaths, the fatal injury location was unknown; but in the six cases of known injury, the fatality was due to head trauma.
Figures 3 and 4 show examples of side impacts in which there was some remaining survival space. In Case 24, the 2-year-old child was seated in a forward-facing restraint in the rear outboard seat. Both the child and adult driver were killed. In Case 70, the 3-year, 6-month-old male was seated in a shield booster directly at the site of impact with a tree. The child died from severe head trauma.
Fig. 3.
Side impact with passenger car, Case 24.
Fig. 4.
Side impact with tree, Case 70.
Three side impact crashes were deemed fatal due to misuse of the child restraint. In one case, the child was restrained in a device that was not approved for vehicle use. In another case, the child restraint was not attached to the vehicle. In the third case, the internal harness was not used.
In two cases, the child was judged to have been killed by contact with other occupants. In each case, the child was seated away from the intrusion, but adult occupants seated next to the intrusion were forced into the restrained child. Finally, one case involved a child who appeared to be correctly restrained in a three-point harness with shield seated in the second row on the struck side of the vehicle. Both the child and unbelted adult occupant on the far side of the second row were ejected out of the near side window. It is possible that contact from the adult occupant played a role in the child’s ejection because the adult would have struck the child and provided additional side loading.
Frontal Impacts (31 deaths)
Thirty-four percent of the child fatalities occurred in frontal crashes. Almost half (45 percent, n=14) of these crashes were considered unsurvivable. Restraint misuse was judged to be a factor in 16 percent (n=5) of the frontal crashes. Thirty-nine percent (n=12) of the fatalities occurred in potentially survivable frontal crashes in which no other factors played a clear role (non-catastrophic frontal impact).
Two of the crashes judged to be unsurvivable are shown in Figures 5 and 6. In Case 21, the 3-year-old child was seated in a shield booster in the left rear seating position. The passenger car was struck in the left front by a pickup truck, and the severe crush eliminated the survival space. The driver and 5-year-old child in the middle seat of the second row also were killed. In Case 46, the 6-month-old child was seated in a rear-facing restraint in the front passenger seat. Both the child and driver were killed due to significant intrusion of the instrument panel after collision with a large commercial truck.
Fig. 5.
Unsurvivable frontal impact with pickup truck, Case 21.
Fig. 6.
Unsurvivable frontal impact with commercial truck, Case 46.
Two crashes in which intrusion did not completely compromise the child’s occupant space are shown in Figures 7 and 8. In Case 36, the 2-year, 9-month-old child was seated in a forward-facing restraint. The vehicle sustained some intrusion of the instrument panel, but the entire survival space was not eliminated. In Case 38, the 1-year, 4-month-old child was seated in a forward-facing restraint with shield and three-point harness. In this crash, there was almost no intrusion of the child’s occupant space. The child died from an upper cervical spine dislocation, possibly caused by inertial tension from the head mass, but head contact also may have contributed.
Fig. 7.
Frontal impact with some instrument panel intrusion, Case 36.
Fig. 8.
Frontal impact with no vehicle intrusion, Case 38.
The remaining frontal crash fatalities (five cases) were due to significant child restraint misuse. Three of the cases involved children who were not buckled into the internal harness, and two of the cases involved child restraints that were loosely attached to the vehicle. In one of these cases, the child restraint was found upside down in the vehicle seat. In another case, the vehicle lap belt had a large amount of slack.
Rear Impacts (11 deaths)
Rear impacts accounted for 12 percent of the child fatalities. Nine of these 11 cases had massive amounts of vehicle intrusion and were deemed unsurvivable. In Case 84, the passenger car was struck in the rear by a tractor-trailer (Figure 9). The left rear seating position of the 1-year-old child suffered catastrophic intrusion.
Fig. 9.
Rear impact with extreme crush, Case 84.
Two of the rear impact fatalities were caused by contact with occupants or cargo. In one case, the relatively minor rear impact propelled an air tank from the rear of a pickup truck, through the rear glass, and into the child in the front seat of the pickup. In the second case, it was judged that the force of the rear impact caused the driver to rotate rearward and strike the child sitting behind the driver.
Rollover
Seven child fatalities (8 percent) occurred in rollover crashes. Three crashes involved ejection of children seated in restraints with a shield and three-point harness (two tray shields, one T-shield); none involved five-point harnesses. Evidence suggested overall correct restraint use, but much more detailed investigations would be required to definitively determine correct use or misuse of the restraint.
Three rollover crash fatalities were caused by significant restraint misuse. Two involved restraints that were attached very loosely and one was completely unattached, allowing partial ejections. Lastly, one child drowned when the vehicle had a low-energy rollover into a shallow creek, completely submerging the restrained child.
Crashes Involving Roof Crush
Five crashes (6 percent) involved roof crush caused by another object (unlike roof crush during a rollover). Four vehicles struck trees or poles, whereas one vehicle had a minor quarter-turn roll before being struck by a pickup truck in the roof. Each of these crashes caused excessive vehicle intrusion, which made the crash unsurvivable for the child.
DISCUSSION
FARS is the main source of information about the number and type of child occupant deaths in the United States. However, only very limited data are available regarding crash severity and vehicle damage location and intrusion. This study of fatal crashes of children in child restraints is based on a sample of more detailed crash reports obtained from police departments. Although formal crash investigations typically were not conducted, photographic information and details from the police reports allowed researchers to gain a better understanding of these fatal crashes.
Results of the current study indicate that many of the crashes in which children in child restraints died were very severe. Moreover, crashes often involved impacts with very heavy or unyielding partners (large trucks, trees, bridges, etc.). Half of the crashes were judged to be unsurvivable, involving catastrophic intrusion of the occupant compartment. An additional 12 percent of the fatalities likely involved gross misuse of the restraint. Thus, for almost two-thirds of the deaths it was virtually certain that child restraint performance was not a factor. That is, improvements in child restraint performance would have made no difference in the outcome.
Beginning with the 2003 model year, all passenger vehicles must be equipped with the LATCH system. This system uses universal attachments (lower anchorages and upper tether anchor) to secure child restraints in vehicles. The goal is to make child seats much easier to use, hopefully reducing the rates of child restraint misuse and possibly increasing restraint use as a whole. The upper tether, which was not found to be in use in any of these crashes, should help to reduce head accelerations and excursion amounts [Legault et al., 1997; Lumley, 1997].
About half of all side impact crashes were judged unsurvivable; however, even among the crashes that were deemed survivable, intrusion occurred at the child’s seating position. FMVSS 213 currently does not include a side impact crash condition, although one is being considered. This study suggests that any such standard should recognize that intrusion is a major factor in child deaths in side impacts. Injury information often was unavailable, but head injury occurred in each case for which injury was known. This confirms other studies that have indicated the majority of injuries to children in crashes are to the head [Arbogast et al., 2000; Henderson et al., 1994; Kelleher et al., 1993; Khaewpong et al., 1995]. Without detailed on-scene investigation, it is difficult to determine the source of injury, although it appeared that there were many cases where the child’s head contacted the intruding door and window sill, which are areas not covered by FMVSS 201. Brown et al. [1995] found that larger head side wings on restraints could help prevent head contact in purely lateral impacts. However, all but two of the potentially survivable side impacts had a frontal component. Any effort to modify child restraint designs to protect children in side impacts should take into account longitudinal and lateral forces and impact directions. Changes to vehicle crashworthiness also should help reduce these deaths. For example, side impacts from SUVs and pickups contributed largely to child deaths in this study, and a new consumer information test program (Insurance Institute for Highway Safety, 2001) simulating side impacts from these vehicles should encourage better protection in these crash events.
Less than half of the frontal crashes were judged potentially survivable. However, it often was not known whether injury was due to contact with intruding structure or some other factor. Research has shown that facial fractures sustained by children in frontal crashes often occur because of impact with the instrument panel when seated in the front, or the back of front seats when seated in the rear [Arbogast et al., 2002; Winston et al., 2002]. Further research is needed to examine whether current excursion distances are appropriate [Sullivan et al., 1993]. Some of the fatal head and/or neck injuries may have been due to inertia. Note that researchers in Australia claim to have no evidence of neck injury without direct head contact [Paine et al., 2003]. Consideration should be given to keeping children in rear-facing restraints longer than their first birthday to help prevent such injuries in frontal crashes.
There were a few cases in which children who appeared to be restrained correctly in the child restraint were ejected. Three of these cases involved children in restraints with three-point shields and one with a three-point T-shield. In at least two cases the harness clip that connects the two shoulder harnesses to help keep the harness in place was not attached after the crash. Arbogast [2002] has shown that there is a high rate of nonuse of harness clips. If the clip is necessary to reduce ejection risk, children in three-point restraint systems may be at risk of ejection. Further research is needed to establish whether such systems do increase the risk of ejection and whether five-point harnesses reduce that risk.
APPENDIX
Table A1.
Crash Information
| Case | Child Age | Seata | CRb | Crash Type | Case Vehicle | Crash Partner | Principal Contributing Factor |
|---|---|---|---|---|---|---|---|
| 46 | 6mo | 13 | RF | Frontal | Pass car | Comm truck | Unsurvivable |
| 31 | 1yr 2mo | 23 | FFS | Frontal | SUV | Farm Tractor | Unsurvivable |
| 09 | 1yr 2mo | 53 | UNK | Frontal | Van | Pass car | Unsurvivable |
| 58 | 1yr 11mo | 21 | UNK | Frontal | Pass car | Pickup | Unsurvivable |
| 21 | 3yr | 21 | SB | Frontal | Pass car | Pickup | Unsurvivable |
| 35 | 4yr | 21 | SB | Frontal | Pass car | Pickup | Unsurvivable |
| 57 | 4yr 0mo | 22 | FF | Frontal | Minivan | Tractor Trailer | Unsurvivable |
| 37 | 1yr | 23 | FF | Frontal | Pass car | Tractor Trailer | Unsurvivable |
| 85 | 1yr 2mo | 21 | FF | Frontal | Pass car | Tractor Trailer | Unsurvivable |
| 54 | 3yr 11mo | 23 | SB | Frontal | Pass car | Tractor Trailer | Unsurvivable |
| 68 | 3yr 11mo | 23 | UNK | Frontal | SUV | Tree | Unsurvivable |
| 91 | 2yr | 13 | UNK | Frontal | Pass car | Van | Unsurvivable |
| 87 | 2yr 6mo | 21 | UNK | Frontal | Pass car | Van | Unsurvivable |
| 88 | 5yr | 23 | UNK | Frontal | Pass car | Van | Unsurvivable |
| 38 | 1yr 4mo | 23 | FFS | Frontal | Pass car | Bridge | Frontal |
| 36 | 2yr 9mo | 13 | FF5 | Frontal | Pickup | Bridge | Frontal |
| 08 | 2yr 11mo | 21 | UNK | Frontal | Pickup | Embankment | Frontal |
| 25 | 1yr 11mo | 22 | SB | Frontal | Minivan | Pass car | Frontal |
| 19 | 1yr 2mo | 22 | FF5 | Frontal | Pass car | Pass car | Frontal |
| 74 | 2yr 2mo | 23 | FFS | Frontal | Pass car | Pass car | Frontal |
| 62 | 3yr 7mo | 13 | SB | Frontal | Pass car | Pass car | Frontal |
| 41 | 4yr | 22 | FF5 | Frontal | Pass car | Pickup | Frontal |
| 43 | 6mo | 12 | RF | Frontal | Pickup | Pickup | Frontal |
| 15 | 3yr | 21 | FFS | Frontal | SUV | Pickup | Frontal |
| 22 | 1yr | 21 | FF5 | Frontal | Pass car | Van | Frontal |
| 92 | 5mo | 21 | UNK | Frontal | Pass car | Van | Frontal |
| 12 | 3mo | 13 | UNK | Frontal | Pass car | Pass car | Misuse |
| 50 | 1yr 8mo | 21 | FF5 | Frontal | Pass car | Pickup | Misuse |
| 76 | 3yr | 21 | FFS | Frontal | Pass car | Pickup | Misuse |
| 42 | 2yr | 23 | UNK | Frontal | SUV | Tractor Trailer | Misuse |
| 40 | 1yr 6mo | 22 | FFT | Frontal | SUV | Tree | Misuse |
| 16 | 2yr | 23 | FF | Rear | Pass car | Pickup | Unsurvivable |
| 83 | 1yr 6mo | 23 | UNK | Rear | Pass car | SUV | Unsurvivable |
| 82 | 3yr | 21 | SB | Rear | Pass car | SUV | Unsurvivable |
| 67 | 2yr 8mo | 22 | UNK | Rear | Minivan | Tractor Trailer | Unsurvivable |
| 49 | 1yr | 23 | UNK | Rear | Pass car | Tractor Trailer | Unsurvivable |
| 84 | 1yr | 21 | UNK | Rear | Pass car | Tractor Trailer | Unsurvivable |
| 20 | 2yr 0mo | 23 | FF | Rear | Pass car | Tractor Trailer | Unsurvivable |
| 65 | 2yr | 23 | UNK | Rear | Pass car | Tractor Trailer | Unsurvivable |
| 64 | 3yr | 21 | Booster | Rear | Pass car | Tractor Trailer | Unsurvivable |
| 48 | 10mo | 13 | RF | Rear | Pickup | Comm truck | Contact |
| 34 | 1yr | 21 | FFS | Rear | Minivan | Pass car | Contact |
| 80 | 1yr 1mo | 21 | FF5 | Rollover | Pass car | Embankment | Unsurvivable |
| 18 | 1yr 0mo | 23 | FFS | Rollover | Pass car | NA | Misuse |
| 81 | 1yr 1mo | 21 | UNK | Rollover | Pass car | NA | Misuse |
| 07 | 5yr 9mo | 33 | FFS | Rollover | Minivan | Snow Bank | Misuse |
| 27 | 3yr 2mo | 21 | FFS | Rollover | Minivan | NA | Ejection |
| 59 | 2yr | 13 | FFS | Rollover | SUV | NA | Ejection |
| 33 | 1yr 6mo | 23 | FFT | Rollover | SUV | NA | Ejection |
| 17 | 2yr 11mo | 21 | FFS | Roof | Pass car | Barrier/Tree | Unsurvivable |
| 55 | 2yr | 12 | SB | Roof | Pickup | Pickup | Unsurvivable |
| 75 | 4yr 0mo | 21 | FF | Roof | Pass car | Pole | Unsurvivable |
| 05 | 1yr | 13 | FFS | Roof | Pickup | Tree | Unsurvivable |
| 03 | 2yr | 22 | FF | Roof | Pickup | Tree | Unsurvivable |
| 29 | 1yr | 23 | RF | Side | SUV | Tractor Trailer | Unsurvivable |
| 14 | 2yr 0mo | 21 | FFS | Side | Pass car | Comm truck | Unsurvivable |
| 66 | 3yr | 13 | UNK | Side | SUV | Comm truck | Unsurvivable |
| 63 | 11mo | 23 | FFS | Side | Pass car | Pickup | Unsurvivable |
| 73 | 2yr 6mo | 23 | UNK | Side | Pass car | Pickup | Unsurvivable |
| 89 | 2yr 6mo | 23 | FFS | Side | Pass car | Pickup | Unsurvivable |
| 71 | 1yr 10mo | 23 | UNK | Side | Pass car | SUV | Unsurvivable |
| 39 | 2yr | 23 | FFS | Side | Pass car | SUV | Unsurvivable |
| 30 | 7mo | 21 | RF | Side | Pass car | SUV | Unsurvivable |
| 72 | < 1 yr | 23 | RF | Side | Pass car | Tractor Trailer | Unsurvivable |
| 52 | 1yr 1mo | 21 | UNK | Side | Pass car | Tractor Trailer | Unsurvivable |
| 26 | 2yr | 21 | FF | Side | Pass car | Tractor Trailer | Unsurvivable |
| 11 | 3yr 3mo | 13 | UNK | Side | Pass car | Tractor Trailer | Unsurvivable |
| 60 | 2yr | 22 | UNK | Side | Pass car | Guard rail | Unsurvivable |
| 47 | 1yr 4mo | 22 | FF | Side | Pass car | Tractor Trailer | Unsurvivable |
| 01 | 1yr | 12 | FF | Side | Pickup | Tractor Trailer | Unsurvivable |
| 10 | 2yr 3mo | 22 | UNK | Side | Pass car | Train | Unsurvivable |
| 53 | 3yr 9mo | 23 | FF | Side | Pass car | Guard rail | Side |
| 24 | 2yr 0mo | 21 | FF | Side | Pass car | Pass car | Side |
| 56 | 2yr 10mo | 23 | UNK | Side | Pass car | Pass car | Side |
| 02 | 2yr | 23 | UNK | Side | Pass car | Pickup | Side |
| 79 | 1yr 2mo | 23 | FF | Side | Pass car | SUV | Side |
| 69 | 3yr 6mo | 23 | UNK | Side | Pass car | SUV | Side |
| 23 | 4yr | 21 | HBB | Side | Pass car | SUV | Side |
| 45 | 1yr | 23 | FF | Side | Pass car | Tractor Trailer | Side |
| 04 | 3yr | 23 | UNK | Side | Pass car | Tractor Trailer | Side |
| 32 | 4yr | 23 | SB | Side | Pass car | Tractor Trailer | Side |
| 70 | 3yr 6mo | 21 | SB | Side | Minivan | Tree | Side |
| 06 | 1yr | 21 | UNK | Side | SUV | Tree | Side |
| 61 | 3yr | 22 | FF | Side | Pass car | Minivan | Side |
| 51 | 1yr 9mo | 21 | FF5 | Side | Pass car | Pickup | Side |
| 90 | 2yr | 23 | SB | Side | Pass car | Pickup | Side |
| 13 | 1yr 6mo | 21 | NACR | Side | Pass car | Pass car | Misuse |
| 78 | 4yr | 23 | FF5 | Side | Pass car | SUV | Misuse |
| 86 | 1yr 2mo | 22 | FFS | Side | Pass car | Tree | Misuse |
| 77 | 1yr 9mo | 22 | FFS | Side | Pass car | Pass car | Ejection |
| 44 | 1yr | 13 | RF | Side | Pass car | Tractor Trailer | Contact |
| 28 | 3yr | 22 | UNK | Side | Pass car | Tractor Trailer | Contact |
Notes
Seat - First number denotes row, second number denotes seating position in that row – e.g., driver is position 11, middle front row passenger seat is position 12, etc.
CR (Child restraint type) - FF: unspecified forward facing, FFS: FF with shield, FFT: FF with T-shield, FF5: FF with 5-point harness, SB: shield booster, RF: rear facing, UNK: unknown.
REFERENCES
- Arbogast KB, Morris SD, Durbin DR, Winston FK. 46th Annual AAAM Conference. Des Plaines, IL: AAAM; 2002. The Influence of Harness Type on Child Restraint System Misuse; pp. 261–272. [PubMed] [Google Scholar]
- Arbogast KB, Moll EK, Morris SD, Anderko RL, Durbin DR, Winston FK. 44th Annual AAAM Conference. Des Plaines, IL: AAAM; 2000. Factors Influencing Pediatric Injury in Side Impact Collisions; pp. 407–428. [PMC free article] [PubMed] [Google Scholar]
- Brown J, Kelly P, Griffiths M. 1994 IRCOBI Conference. Bron, France: IRCOBI; 1994. The Performance of Tethered and Untethered Forward Facing Child Restraints; pp. 61–74. [Google Scholar]
- Decina LE, Knoebel KY. Child Safety Seat Misuse Patterns in Four States. Accid Anal Prev. 1997;29:125–132. doi: 10.1016/s0001-4575(96)00051-6. [DOI] [PubMed] [Google Scholar]
- Henderson M, Brown J, Paine M. 38th Annual AAAM Conference. Des Plaines, IL: AAAM; 1994. Injuries to Restrained Children; pp. 75–87. [Google Scholar]
- Insurance Institute for Highway Safety. Status Report. 1. Vol. 36. Arlington, VA: 2001. [Google Scholar]
- Kelleher-Walsh B, Walsh MJ, States JD, Duffy LC. Child Occupant Protection Proceedings(SP-986) Warrendale, PA: SAE; 1993. Trauma to Children in Forward-Facing Car Seats; pp. 127–132. [Google Scholar]
- Kelly P, Brown J, Griffiths M. 1995 IRCOBI Conference. Bron, France: IRCOBI; 1995. Child Restraint Performance in Side Impacts With and Without Top Tethers and With and Without Rigid Attachment (CANFIX) pp. 75–90. [Google Scholar]
- Khaewpong N, Nguyen TT, Bents FD, Eichelberger MR, Gotschall CS, Morrissey R. 39th Stapp Conference Proceedings. Warrendale, PA: SAE; 1995. Injury Severity in Restrained Children in Motor Vehicle Crashes; pp. 403–423. [Google Scholar]
- Legault F, Gardner B, Vincent A. Child Occupant Protection 2nd Symposium Proceedings (P-316) Warrendale, PA: SAE; 1997. The Effect of Top Tether Strap Configurations on Child Restraint Performance; pp. 93–122. [Google Scholar]
- Lumley MA. Child Occupant Protection 2nd Symposium Proceedings (P-316) Warrendale, PA: SAE; 1997. Child Restraint Tether Straps – A Simple Method of Increasing Safety for Children; pp. 123–135. [Google Scholar]
- National Highway Traffic Safety Administration. Fatality Analysis Reporting System, 2001. Washington, DC: U.S. DOT; 2001. [Google Scholar]
- National Safe Kids Campaign. Child Passengers at Risk in America: A National Study of Car Seat Misuse. Washington, DC: 1999. [Google Scholar]
- Paine M, Griffiths M, Brown J, Case M, Johnstone O. 18th International ESV Conference Proceedings (CD-ROM) Washington, DC: NHTSA; 2003. Protecting Children in Car Crashes: The Australian Experience. [Google Scholar]
- Slavik DH. Child Occupant Protection 2nd Symposium Proceedings (P-316) Warrendale, PA: SAE; 1997. Cervical Distraction Injuries to Children; pp. 137–145. [Google Scholar]
- Sullivan LK, Mouchahoir G, Stucki L, Howe JG, Chambers FK. Child Occupant Protection Proceedings (SP-986) Warrendale, PA: SAE; 1993. Assessment of Dynamic Testing Environment of Child Restraint Systems; pp. 275–292. [Google Scholar]
- Winston FK, Kallan MJ, Elliott MR, Menon RA, Durbin DR. Risk of Injury to Child Passengers in Compact Extended-Cab Pickup Trucks. JAMA. 2002;287:1147–1152. doi: 10.1001/jama.287.9.1147. [DOI] [PubMed] [Google Scholar]
- Winston FK, Durbin DR, Bhatia E, Werner J, Sorenson W. Child Occupant Protection in Motor Vehicle Crashes Proceedings. London, UK: PEP; 1999. Patterns of Inappropriate Restraint for Children in Crashes; pp. 59–69. [Google Scholar]