ARTICLE Computer Crash in the Development of Child Occupant Safety Policies

Flaura Koplin Winston, MD, PhD; Kristy B. Arbogast, PhD; Lois A. Lee, MD; Rajiv A. Menon, PhD

Objective: To address the predictability of injury from restrained male adults exposed to air bags or for child air bag activation by use of crash software. passengers restrained in the rear seat for the crash sce- narios simulated. Methods: Using current, validated crash simulation soft- ware, the effect of air bag activation on injury risk was Conclusions: Using current crash simulation software, this assessed for the 6-year-old child, both restrained and un- study demonstrated that the risk of air bags to school- restrained. Results were compared with those for adult aged children could be predicted. Our results confirmed occupants in similar crash scenarios. the previously identified risks to unrestrained children and provided the first evidence that air bags, in their current Results: For the unrestrained child passenger, crash design, are not beneficial to restrained children. This study simulations predicted serious head, neck, and chest in- illustrates that computer crash simulations should be used juries with air bag activation, regardless of crash sever- proactively to identify injury risks to child occupants, par- ity. For the restrained child passenger, crash simula- ticularly when limited real-world data are available. tions predicted similar severe injuries for high-severity crashes only. No serious injuries were predicted for un- Arch Pediatr Adolesc Med. 2000;154:276-280

of crash scenarios. They have been used Editor’s Note: This study can have significant impact on the man- as a standard, validated tool by vehicle and ner in which automobile crashes would cause air bag injuries. And restraint manufacturers to evaluate the safety of new designs for occupants. With it costs less than the dummy way. Catherine D. DeAngelis, MD computer crash simulations, the forces and experienced by occupants during different crash scenarios can be N THE AREA of motor vehicle occu- measured. Through recognized math- pant injury prevention, computer ematical relationships that were deter- simulation holds vast, yet unreal- mined through extensive biomechanical ized, potential to supplement ve- testing,2 these engineering measure- hiclesafety()test- ments can be translated into a predicted Iing. Crashworthiness testing has been used risk of biological and structural injuries for for decades to evaluate the safety of motor distinct body regions of either children or vehicles and, as a result, vehicles are safer adults. Realizing the invaluable contribu- and people are surviving crashes once tion of computer simulations as predic- thoughtunsurvivable.Crashworthinesstest- tors in other areas of health, the Institute ing uses lifelike surrogates (“dummies”) in of Medicine (Washington, DC) has re- actual crash scenarios to predict injury risk. cently called for the increased use of com- But the expense of crashworthiness testing puter simulation as an important tool for has limited federally required test proce- understanding injury causation to pre- dures to adult surrogates and to a meager dict injury risk.3 number of crash scenarios. Although au- One current controversy in injury tomobile manufacturers recommended the control is the safety of air bags for chil- incorporation of child surrogates in crash- dren. It is well accepted that air bags pose worthiness test procedures in 1996 and fed- a substantial risk to young, unrestrained, eral regulations for this testing have been school-age children (aged 5 to 7 years), but proposed,currentregulationsdonotrequire it is unclear whether restrained children From The Children’s Hospital 1 of Philadelphia (Drs Koplin their inclusion. in this age group are at a similar high risk. Winston, Arbogast, Lee, and Computer crash simulations pro- These children have recently outgrown Menon), and the University vide an efficient and economical ap- their child safety seats (the upper weight of Pennsylvania (Dr Koplin proach to extending safety testing to in- limit for most child safety seats is 18 kg) Winston), Philadelphia. clude children and a far greater number and many parents have begun using seat

(REPRINTED) ARCH PEDIATR ADOLESC MED/ VOL 154, MAR 2000 WWW.ARCHPEDIATRICS.COM 276

©2000 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/25/2021 MATERIALS AND METHODS (the young, school-aged child).13 The Hybrid III 50th per- centile adult male dummy model was chosen because cur- COMPUTER CRASH SIMULATIONS rent vehicle safety standards require protection of an oc- cupant of this size against serious injuries. For clarity, this MADYMO (Mathematical Dynamic Model; TNO, Delft, the article henceforth refers to the P6 child dummy model as Netherlands) is an engineering software tool that allows us- the “child passenger” and to the Hybrid III 50th percentile ers to develop, design, and optimize safety systems and to adult dummy model as the “adult male passenger.” assess injury risk to vehicle occupants. The validity of In addition to size of the occupant, 4 other indepen- MADYMO has been demonstrated by verification studies dent variables were simulated: air bag activation (air bag using experimental test data.6,7 MADYMO is currently used activated or air bag deactivated), occupant position (front by vehicle and restraint manufacturers to predict the risk passenger seat or rear seat), restraint usage (unrestrained, of occupant injury under specified crash dynamics and lap-only belt, or lap-shoulder belt), and crash severity occupant parameters, including air bag activation, impact (high or low). The low-severity crash was specified by a speed, preimpact braking, occupant size, occupant posi- speed change (also known as Delta V14) of 5.28 m/s (19 tion, and restraint usage. The effects of these different km/h) and the high-severity crash was specified by a speed crash parameters are measured in terms of forces and change of 10.28 m/s (37 km/h), as defined in a previous accelerations experienced by different body parts of the report.15 occupants. In this study, a model of a leading midsized sedan with OUTCOME MEASURES a top-mounted, passenger air bag was developed as our test vehicle. Because air bags are designed to deploy in frontal From the computer crash simulations, the risk of injury was crashes, a frontal crash into a rigid barrier was simulated. assessed using measurements of crash forces and accelera- In addition, preimpact braking was included in the simula- tions for the head, chest, and neck of vehicle occupants. Head tion because braking has been shown to be a common char- injury risk was measured using the Head Injury Criterion acteristic of air bag crashes with occupant fatalities.8 The rate (HIC), a nondimensional parameter calculated from the oc- of braking was simulated at 6.8 m/s2, previously shown to cupant head and over which the accelera- be characteristic of average driver behavior.9 For all simu- tion occurs. When the HIC exceeds the accepted injury tol- lations, the vehicle seat was set at the rearmost position. erance limit of 1000, a risk of concussion for frontal impact Two sizes of vehicle occupant surrogates were se- is predicted.16 A chest acceleration above the accepted tol- lected from standard databases supplied with the MADYMO erance limit of 588.72 m/s2 (60 g) sustained for 3 millisec- program10: a P6 child dummy model (“aged” 6 years)11 and onds predicts severe chest injury.17 Injury tolerance limits a Hybrid III 50th percentile adult male dummy model.12 for neck tensile and shear loads have not been universally These surrogates are mathematical representations of the accepted. Since limits of 1490 to 2900 N in tension and 1200 state-of-the-art in dummies that can be used in N in shear for the P6 child dummy model surrogate are cur- computer simulations to represent occupants. The P6 child rently proposed by the National Highway Traffic Safety Ad- dummy model was chosen because this surrogate most ministration (Washington, DC) for crashworthiness test- closely approximated the age of children killed in actual ing,2 measurements exceeding these proposed limits were air bag–related crashes and the age of interest for this study used in this study to predict neck injury.

belts for these children. In addition, parents may be ques- simulation software. Using standard, validated crash simu- tioning the continued need to place the child in the rear lation software, we evaluated the effect of front passen- seat. Current research indicates that 30% of the chil- ger air bags on child passengers—both restrained and un- dren in this age group are restrained in the right front restrained—in crashes of high and low severity and in seat at the time of a crash.4 Is this a safe practice in pas- crashes where children were placed in the front passen- senger air bag–equipped vehicles? ger seat vs the rear seat. Most safety recommendations are based on crash test results and early injury experience using the device. There RESULTS were very limited crash test data regarding children and air bags and, as a result, there was no initial recommen- COMPUTER CRASH SIMULATIONS dation to place young, school-aged children in the rear seat of passenger air bag–equipped vehicles. Now, after The computer crash simulations predicted that air bag the report of numerous child deaths due to air bags, activation propels an unrestrained front-seated child double-paired comparison methods using Fatal Analy- passenger rearward and upward (Figure). The simula- sis Reporting System data has provided some evidence tions showed a complex relationship between indepen- for the risk of injury to children from front passenger air dent variables (the size of the occupant, air bag activa- bags.5 But children had to die before these analyses could tion, occupant position, restraint usage, and crash be conducted. The prediction of injury risk using crash severity) and injury risk, as described below for the simulation software affords the possibility that proac- head, chest, and neck. tive safety policy might be set before children sustain in- The effects of air bag activation on injury risk are jury from new technology. summarized in Table 1. For a child passenger, injury The objective of this study was to address the pre- risk to the head, neck, and chest increased or remained dictability of injury from air bag activation by use of crash the same with air bag activation, independent of re-

(REPRINTED) ARCH PEDIATR ADOLESC MED/ VOL 154, MAR 2000 WWW.ARCHPEDIATRICS.COM 277

©2000 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/25/2021 Table 1. Effect of Air Bag Deployment on Injury Risk for Right Front–Seated Occupants as Predicted by the Computer Crash Simulations*

Passenger and Neck Neck Crash Types Head Chest Tension Shear

z Unrestrained child x y Low-severity crash I† I† I† I† High-severity crash I† I† I† I† Computer crash simulation of the unrestrained child passenger in the right Lap-belted child front passenger seat in a typical midsize vehicle moving at 38 km/h, with Low-severity crash I U U I preimpact braking, into a rigid barrier. This figure is one frame of the High-severity crash I† U I† I† simulation (at 776 milliseconds), at which time the air bag had activated, Lap-shoulder–belted child which propelled the child rearward and upward with extreme force. Low-severity crash I U U D High-severity crash I† U I† U† Unrestrained adult straint use and crash severity. For the unrestrained adult Low-severity crash I I I I male passenger, air bag activation decreased or did not High-severity crash D U U D change injury risk for all body regions in high-severity crashes; however, it increased injury risk (although not *I indicates increased injury risk; D, decreased injury risk; and U, unchanged injury risk (Ͻ10%). above the accepted tolerance limits) for low crash se- †Above suggested tolerance limits. verities. Of importance, rear seat placement was benefi- cial for all children as evidenced by reductions in injury parameters. It must be noted that these simulations were Table 2. Head Injury Criterion* Values for Child Passenger run with preimpact braking and an unrestrained child and Adult Male Passenger as Predicted in the rear seat during the braking phase slides up closer by the Computer Crash Simulations to the cushioned padding of the front seat back. This re- duces the momentum/forces from the impact. The ef- Front Seat Front Seat fect of air bag activation and rear seat placement on each Passenger Passenger body region is displayed in Tables 2, 3, 4, and 5. Crash and With Without Rear Seat Passenger Types Air Bag Air Bag Passenger HEAD INJURY RISK Low-severity crash† Lap-shoulder–belted child 240 21 22 Lap-only belted child 420 24 23 The simulations revealed that air bag activation in- Unrestrained child 3831‡ 223 31 creased the risk of head injury for a child passenger, as Unrestrained adult male 206 12 NA§ predicted by the HIC values. For low-severity crash with High-severity crash࿣ air bag activation, only the unrestrained child passenger Lap-shoulder–belted child 1216‡ 285 244 experienced HIC values above the suggested tolerance Lap-only belted child 1876‡ 655 431 limit. For high-severity crash, all child passengers ex- Unrestrained child 5915‡ 521 142 Unrestrained adult male 394 961 NA§ posed to air bags produced HIC above the threshold for serious injury. The rear-seated child passenger, regard- *Head Injury Criterion is a nondimensional parameter used to assess head less of restraint use or crash severity, experienced HIC injury risk that is calculated from the occupant head acceleration and time values well below the accepted tolerance limit (Table 2). over which the acceleration occurs. A value above 1000 indicates a substantial risk of serious head injury for both adults and children. In all scenarios, the unrestrained adult male pas- †The low-severity crash was specified by a speed change of 5.28 m/s senger displayed HIC values below the accepted toler- (19 km/h) as defined in a previous report.16 ance limit. For low-severity crash, the risk of head ‡Above tolerance level. injury for adult males increased marginally with air bag §NA indicates not applicable. ࿣The high-severity crash was specified by a speed change of 10.28 m/s activation. For high-severity crash, on the other hand, (37 km/h) as defined in a previous report.16 the risk of head injury decreased from near threshold lim- its without air bag activation to lower measures with air bag activation. celerations were below the accepted tolerance limits in low- severity crash simulations regardless of air bag deploy- CHEST INJURY RISK ment. In the high-severity crash simulations, the air bag reduced the chest injury risk for the adult to below the tol- For the restrained child passenger, air bag activation did erance limit (Table 3). not influence chest accelerations, regardless of restraint use or crash severity, and all chest accelerations were below the NECK INJURY RISK threshold. For the unrestrained child passenger, air bag ac- tivation produced chest accelerations above the accepted For low-severity crashes, the restrained child passenger tolerance level, regardless of crash severity. The rear- experienced neck tensile forces (elongation) (Table 4) seated child passenger, regardless of restraint use or crash and shear forces (sliding) (Table 5) below the proposed severity, experienced chest accelerations below the injury tolerance limit with air bag activation. For high-severity limit. For the unrestrained adult male passenger, chest ac- crashes, the forces increased substantially, exceeding pro-

(REPRINTED) ARCH PEDIATR ADOLESC MED/ VOL 154, MAR 2000 WWW.ARCHPEDIATRICS.COM 278

©2000 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/25/2021 Table 3. Chest Accceleration* for the Child Passenger Table 4. Neck Tensile (Elongation) Loads* and Adult Male Passenger as Predicted by the Computer for Child Passenger and Adult Male Passenger Crash Simulations as Predicted by the Computer Crash Simulations

Front Seat Front Seat Front Seat Front Seat Passenger Passenger Passenger Passenger Crash and With Without Rear Seat Crash and With Without Rear Seat Passenger Types Air Bag Air Bag Passenger Passenger Types Air Bag Air Bag Passenger Low-severity crash† Low-severity crash† Lap-shoulder–belted child 28 27 23 Lap-shoulder–belted child 500 400 384 Lap-only belted child 19 16 16 Lap-only belted child 760 490 500 Unrestrained child 183‡ 24 22 Unrestrained child 9850‡ 1139 130 Unrestrained adult male 36 8 NA§ Unrestrained adult male 3697 1548 NA§ High-severity crash࿣ High-severity crash࿣ Lap-shoulder–belted child 50 44 46 Lap-shoulder–belted child 1800‡ 1400 1187 Lap-only belted child 35 35 34 Lap-only belted child 2300‡ 1300 1100 Unrestrained child 252‡ 113‡ 47 Unrestrained child 10 700‡ 3825‡ 185 Unrestrained adult male 60 72‡ NA§ Unrestrained adult male 3410 4109 NA§

*Chest acceleration is a measure of acceleration due to gravity that is used *Neck tensile (elongation) loads is a measurement in newtons used to to assess chest injury risk. A value above 60 g indicates a significant risk of assess risk of distraction injuries to the neck. A value higher than 1490 to serious chest injury for both adults and children. 2900 N indicates a significant risk of serious neck distraction injury for †The low-severity crash was specified by a speed change of 5.28 m/s children and a value higher than 6500 N indicates a substantial risk for (19 km/h) as defined in a previous report.16 adults. ‡Above tolerance level. †The low-severity crash was specified by a speed change of 5.28 m/s §NA indicates not applicable. (19 km/h) as defined in a previous report.16 ࿣The high-severity crash was specified by a speed change of 10.28 m/s ‡Above tolerance level. (37 km/h) as defined in a previous report.16 §NA indicates not applicable. ࿣The high-severity crash was specified by a speed change of 10.28 m/s (37 km/h) as defined in a previous report.16 posed tolerance limits. The unrestrained child passen- ger experienced substantial neck tensile and shear forces with air bag activation, regardless of crash severity. In ture), and thoracic injuries (interatrial septum hematoma, most cases, placing a child passenger in the rear seat re- pulmonary hemorrhage, pulmonary contusions, inferior duced the risk of neck injury, except in the case of the vena cava laceration, and sternal fracture). Although very unrestrained child in the rear seat who hit the back of limited data exist for restrained children injured by air bags, the front seat and experienced high neck forces. serious, nonfatal injuries have been reported to the head, For the unrestrained adult male passenger, neck ten- neck, and chest.25,26 sile forces were below the proposed tolerance limits with When air bag–equipped vehicles first appeared on air bag activation. Shear forces were below the proposed the market, no recommendation existed regarding the tolerance limits in low-severity crash simulations, re- placement of school-aged children in these vehicles. It gardless of air bag activation. For the high-severity crash, was not until the first deaths of unrestrained children from air bag activation reduced the neck injury risk. air bags in 1993 that the current recommendation, seat- ing children in the rear, was made by the National High- COMMENT way Traffic Safety Administration and other safety ad- vocates.5,23,27-29 Initially, this recommendation included Using current computer crash simulation software, this only infants in rear-facing infant seats.20 Rear seat place- study demonstrated that the risk of air bags to school- ment for all children less than 12 years of age was not aged children could be predicted. Our results confirmed recommended until 1996 after several toddler and school- the previously identified risks to unrestrained children and aged children sustained fatal injuries from air bag inter- provided the first evidence that front passenger-side air action.21 If simulations similar to ours had been avail- bags, particularly those that are fully powered, are not ben- able when air bags were introduced, the risk of air bags eficial to restrained children. This study illustrates that com- to children could have been assessed sooner and appro- puter crash simulations should be used proactively to iden- priate policy set proactively. Safety decisions based on tify injury risks to child occupants, particularly when limited information from sources other than real-world crashes real-world data are available. are made every day when consumers choose new ve- For unrestrained children, our simulations pre- hicles based on laboratory tests from the federal govern- dicted occupant movement and resultant injuries that agreed ment, the Insurance Institute for Highway Safety (Ar- with investigations of child fatalities due to air bags.8,18-24 lington, Va), and the Consumer Union (Younkers, NY). Our model predicted serious head, neck, and chest inju- The current simulations involved one vehicle ries. The reported injuries sustained by children as a with the P6 child dummy model. The advantage of result of air bag exposure included head injuries (extra- computer simulations is that these parameters are cranial hemorrhage, parenchymal brain injury, cerebral readily modifiable. Future simulation enhancements edema, brain swelling, intracranial hemorrhage, brain her- should include additional classes of vehicles and chil- niation, and skull fracture), neck injuries (cervical cord dren of other sizes to generalize our findings to other injury, cervical spine dislocation, subluxation, and frac- configurations.

(REPRINTED) ARCH PEDIATR ADOLESC MED/ VOL 154, MAR 2000 WWW.ARCHPEDIATRICS.COM 279

©2000 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/25/2021 of Philadelphia, Philadelphia, PA 19104 (e-mail: flaura Table 5. Neck Shear (Sliding) Loads* for Child @mail.med.upenn.edu). Passengers and Adult Male Passengers as Predicted by the Computer Crash Simulations REFERENCES Front Seat Front Seat Passenger Passenger Crash and With Without Rear Seat 1. National Highway Traffic Safety Administration. Advanced air bags. Available at: http://www.nhtsa.dot.gov/airbag/. Accessed November 8, 1999. Passenger Types Air Bag Air Bag Passenger 2. Kleinberger M, Yoganandan N, Kumaresan S. Biomechanical considerations for Low-severity crash† child occupant protection. In: Proceedings of the 42nd annual meeting of the Lap-shoulder–belted child 370 1900‡ 351 Association for the Advancement of Automotive Medicine; October 5-7, 1998; Charlottesville, Va. Lap-only belted child 970 680 300 3. Committee on Injury Prevention and Control. Reducing the Burden of Injury: Ad- Unrestrained child 5760‡ 369 600 vancing Prevention and Treatment. Washington, DC: Institute of Medicine; 1999. Unrestrained adult male 1786 1359 NA§ 4. Glass R, Graham J. Where American children sit in passenger vehicles. J Safety High-severity crash࿣ Res. 1999;30:17-24. Lap-shoulder–belted child 2200‡ 2300‡ 933 5. Graham JD, Goldie SJ, Segui-Gomez M, et al. Reducing risks to children in ve- hicles with passenger . Pediatrics. 1998;102:e3. Lap-only belted child 3400‡ 1100 900 6. Lupker H, Coo Pd, Neiboer J, Wismans J. Advances in MADYMO Crash Simu- Unrestrained child 9360‡ 2165‡ 800 lations. Paper presented at: the 87th annual Society of Automotive Engineers In- Unrestrained adult male 1595 3556‡ NA§ ternational Congress and Exposition; February 25-28, 1991; Detroit, Mich. 7. Mathematical Dynamic Model. What is MADYMO? Available at: http:// *Neck shear (sliding) loads is a measurement in newtons used to assess www.madymo.com/html/mframe.htm. Accessed November 8, 1999. risk of dislocation injuries to the neck. A value higher than 1400 N indicates a 8. Winston F, Reed R. Airbags and children: results of a National Highway Traffic Safety Administration special investigation into actual crashes. J Passenger . substantial risk of serious neck dislocation injury for children and a value 1996;105:2008-2014. higher than 3100 N indicates a substantial risk for adults. 9. Czernakowski W, Otte D. The effect of pre-impact braking on the performance of †The low-severity crash was specified by a speed change of 5.28 m/s child restraint systems in real life accidents and under varying test conditions. 16 (19 km/h) as defined in a previous report. In: Proceedings of the 1st Child Occupant Protection Symposium; November 7, ‡Above tolerance level. 1993; San Antonio, Tex. §NA indicates not applicable. 10. TNO Road Vehicles Research Institute. MADYMO Database Manual. Vol 5. Delft, ࿣The high-severity crash was specified by a speed change of 10.28 m/s the Netherlands: TNO Road Vehicles Research Institute; 1997. (37 km/h) as defined in a previous report.16 11. Deter T, Hellkamp U. Development of MADYMO P6 child dummy model. Paper presented at: 15th International Technical Conference on the Enhanced Safety of Vehicles; May 13-16, 1996; Melbourne, Australia. In all areas of health research, the unique develop- 12. Backaitis S, Mertz H. Hybrid III: The First Human-Like Crash Dummy. Warren- dale, Pa: Society of Automotive Engineers; 1993. mental characteristics of children are being recognized. 13. National Highway Traffic Safety Administration. Special crash investigations. Avail- In particular, the National Institutes of Health, Bethesda, able at: http://www.nhtsa.dot.gov/people/ncsa/sci.html. Accessed November 8, Md, now requires consideration of children in all re- 1999. 30 14. Baker J, Fricke L. The Traffic-Accident Investigation Manual. Evanston, Ill: North- search study protocols. Recently, the National High- west University Traffic Institute; 1986. way Traffic Safety Administration has proposed the use 15. Weinstein E, Sweeney M, Garber M, Eastwood M, Osterman J, Roberts J. The effect of size-appropriate and proper restraint use on injury severity of children. of child dummies in one subset of crashworthiness tests: Paper presented at: Second Child Occupant Protection Symposium; November those studying occupant safety related to air bags in fron- 12, 1997; Orlando, Fla. tal impacts.1 The high cost of crashworthiness testing pre- 16. Society of Automotive Engineers. Human Tolerance to Impact Conditions as Related to Motor Vehicle Design. Warrendale, Pa: Society of Automotive Engineers; 1986. cludes the inclusion of children in all test protocols. 17. TNO Road Vehicles Research Institute. MADYMO Theory Manual. Vol 5. Delft, In a novel application of computer crash simula- the Netherlands: TNO Road Vehicles Research Institute; 1997. tions, we determined that current simulation software could 18. Hollands CM, Winston FK, Stafford PW, Shochat SJ. Severe head injury caused by deployment. J Trauma. 1996;41:920-922. predict the high risk of air bag–related injuries to school- 19. Willis BK, Smith JL, Falkner LD, Vernon DD, Walker ML. Fatal air bag mediated aged children. The findings suggest that computer crash craniocervical trauma in a child. Pediatr Neurosurg. 1996;24:323-327. 20. Air-bag-associated fatal injuries to infants and children riding in front passenger simulations should be employed more proactively in con- seats: United States. MMWR Morb Mortal Wkly Rep. 1995;44:845-847. junction with crashworthiness testing to develop vehicle 21. Update: fatal air bag-related injuries to children: United States, 1993-1996 [pub- safety regulatory policies for children. In the case of air lished correction appears in MMWR Morb Mortal Wkly Rep. 1997;46:40]. MMWR Morb Mortal Wkly Rep. 1996;45:1073-1076. bags, had such sophisticated modeling been available, stron- 22. Cooper JT, Balding LE, Jordan FB. Airbag mediated death of a two-year-old child ger warnings or other measures could have been issued wearing a shoulder/lap belt. J Forensic Sci. 1998;43:1077-1081. to reduce the placement of children in the front seat. 23. Giguere JF, St-Vil D, Turmel A, et al. Airbags and children: a spectrum of C- spine injuries. J Pediatr Surg. 1998;33:811-816. 24. Huff GF, Bagwell SP, Bachman D. Airbag injuries in infants and children: a case Accepted for publication July 26, 1999. report and review of the literature. Pediatrics. 1998;33:811-816. 25. Gotschall C, Eichelberger M, Morrissey J, Better A, Reardon J, Bents F. Nonfatal We acknowledge the financial support of State Farm air bag deployments involving child passengers. Paper presented at: Second Child Insurance Companies (Bloomington, Ill) and the National Occupant Protection Symposium; November 12, 1997; Orlando, Fla. Highway Traffic Safety Administration (Washington, DC). 26. National Transportation Safety Board. The Performance and Use of Child Re- straint Systems, Seatbelts, and Air Bags for Children in Passenger Vehicles. Wash- This work was presented by Dr Lee at the annual ington, DC: National Transportation Safety Board; 1996. conference of the American Academy of Pediatrics, New 27. Braver ER, Ferguson SA, Greene MA, Lund AK. Reductions in deaths in frontal Orleans, La, May 2-4, 1998. crashes among right front passengers in vehicles equipped with passenger air bags. JAMA. 1997;278:1437-1439. We thank John Werner, MS, of State Farm Insurance 28. National Child Passenger Safety Week: February 8-14, 1998. MMWR Morb Mor- Companies, Steve Ridella, MS, of TRW Inc (Cleveland, Ohio), tal Wkly Rep. 1998;47:59-60. 29. Graham JD, Thompson KM, Goldie SJ, Segui-Gomez M, Weinstein MC. The cost- Esha Bhatia, MA, of The Children’s Hospital of Philadel- effectiveness of air bags by seating position [see comments]. JAMA. 1997;278: phia (Philadelphia, Pa), and Miriam Davis, PhD, for their 1418-1425. assistance in developing and editing this manuscript. 30. National Center for Statistics and Analysis. NIH policy and guidelines on the in- clusion of children as participants in research involving human subjects. Avail- Reprints: Flaura Koplin Winston, MD, PhD, Trauma- able at: http://www.nih.gov/grants/guide/notice-files/not98-024.html. Ac- Link, 3535 Market St, 10th Floor, The Children’s Hospital cessed November 8, 1999.

(REPRINTED) ARCH PEDIATR ADOLESC MED/ VOL 154, MAR 2000 WWW.ARCHPEDIATRICS.COM 280

©2000 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/25/2021