Motorcycle Helmets Associated with Lower Risk of

Cervical Spine Injury: Debunking the Myth

Joseph G Crompton, MD, Curt Bone, BS, Tolulope Oyetunji, MD, MPH, Keshia M Pollack, PhD, MPH,
Oluwaseyi Bolorunduro, BS, Cassandra Villegas, BA, MPH, Kent Stevens, MD, MPH,
Edward E Cornwell III, MD, FACS, David T Efron, MD, FACS, Elliott R Haut, MD, FACS,
Adil H Haider, MD, MPH, FACS

BACKGROUND: There has been a repeal of the universal helmet law in several states despite definitive evidence
that helmets reduce mortality, traumatic brain injury, and hospital expenditures. Opponents of
the universal helmet law have successfully claimed that helmets should not be required because
of greater torque on the neck, which is thought to increase the likelihood of a cervical spine
injury. There is currently insufficient evidence to counter claims that helmets do not increase the
risk of cervical spine injury after a motorcycle collision. The objective of this study was to
determine the impact of motorcycle helmets on the likelihood of developing a cervical spine
injury after a motorcycle collision.
STUDY DESIGN: We reviewed cases in the National Trauma Databank (NTDB) v7.0 involving motorcycle
collisions. Multiple logistic regression was used to analyze the independent effect of helmets on
cervical spine injury. Cases were adjusted for age, race, sex, insurance status, anatomic (Injury
Severity Score) and physiologic injury severity (systolic blood pressure ⬍ 90 mmHg), and head
injury (Abbreviated Injury Score ⬎ 3).
RESULTS: Between 2002 and 2006, 62,840 cases of motorcycle collision were entered into the NTDB;
40,588 had complete data and were included in the adjusted analysis. Helmeted riders had a
lower adjusted odds (0.80 [CI 0.72 to 0.90]) and a lower proportion of cervical spine injury
(3.5% vs 4.4%, p ⬍ 0.05) compared with nonhelmeted riders.
CONCLUSIONS: Helmeted motorcyclists are less likely to suffer a cervical spine injury after a motorcycle colli-
sion. This finding challenges a long-standing objection to mandatory helmet use that claims
helmets are associated with cervical spine injury. Re-enactment of the universal helmet law
should be considered in states where it has been repealed. (J Am Coll Surg 2011;212:295–300.
© 2011 by the American College of Surgeons)

There has been a sharp rise in the number of motorcyclists incidence of motorcycle fatalities has nearly doubled since
on the road over the last 10 years in the United States and that time.1,2 In developing countries, traffic deaths are pro-
abroad. Motorcycle injuries in the United States have in- jected to be the third most important health problem by
creased by approximately 5,000 per year since 1997 and the 2020, and a large proportion of these deaths involve either
motorcycles or motor scooters.3 In Vietnam, for example,
Disclosure Information: Nothing to disclose. an estimated 60% of all road traffic deaths involved motor-
Abstract presented at the annual meeting of the Eastern Association for the cycle riders or their passengers.4
Surgery of Trauma, January 2010, Phoenix, AZ. Numerous studies have demonstrated that helmets re-
Received July 1, 2010; Revised September 1, 2010; Accepted September 1, duce mortality and traumatic brain injury after a motorcy-
2010. cle collision.5-8 The National Highway Traffic Safety Ad-
From the Departments of Surgery, University of California-Los Angeles,
School of Medicine, Los Angeles, CA (Crompton); The Johns Hopkins Uni- ministration estimates that helmets reduce mortality by
versity, School of Medicine, Baltimore, MD (Bone, Villegas, Stevens, Efron, 35% and traumatic brain injury by 67%.9 A recent Co-
Haut, Haider); and Howard University College of Medicine, Washington,
DC (Oyetunji, Bolorunduro, Cornwell); and Johns Hopkins Bloomberg
chrane review on the subject confirmed the findings show-
School of Public Health, Department of Health Policy and Management, ing the strong evidence of the benefit of motorcycle hel-
Center for Injury Research and Policy, Baltimore, MD (Pollack). mets in reducing mortality and traumatic brain injury.10
Correspondence address: Joseph G Crompton, MD, Department of Surgery,
UCLA School of Medicine, 75 Westwood Blvd, Los Angeles, CA 99500. Even with this preponderance of evidence that proves
email: [email protected] that helmets reduce traumatic brain injury and mortality,

© 2011 by the American College of Surgeons ISSN 1072-7515/11/$36.00

Published by Elsevier Inc. 295 doi:10.1016/j.jamcollsurg.2010.09.032
296 Crompton et al Motorcycle Helmets and Lower Risk of Spine Injury J Am Coll Surg

many states, including Florida, Pennsylvania, and Texas, pared between helmeted and nonhelmeted motorcyclists.
have repealed their universal helmet laws due to strong The Student’s t-test was used to compare continuous vari-
lobbying efforts of some motorcycle riders.11 Attempts to ables and chi-square was used to compare categorical vari-
legislate mandatory helmets laws in developing countries ables for bivariate analysis. All analysis was carried out us-
have also met significant resistance.12-14 ing STATA v10.18 The Johns Hopkins Institutional Review
Motorcyclists who lobby against these laws often claim Board approved this study.
helmets increase the risk of cervical spine injury due to the A multiple logistic regression was undertaken to assess
increased weight of the helmet on the head. A study by the independent effect of motorcycle helmets on the pri-
Goldstein15 is often cited, which suggests this may be true. mary dependent variable of injury to the cervical spine. The
Even through the study had many limitations and no other regression was adjusted with an extensive set of covariates
investigators have replicated the findings, subsequent anal- that have been shown to affect trauma outcomes including
yses have not conclusively refuted the Goldstein study by patient age, severity of injury, severity of head injury, race,
showing that helmets are not associated with an increased sex, and insurance status. To adequately adjust for injury
likelihood of cervical spine injury.10 severity, the Injury Severity Score (ISS) was used to measure
The objective of this study was to use the largest available the magnitude of anatomic injury, and the presence of
trauma database to determine the impact of motorcycle shock on arrival at the emergency department (systolic
helmets on the likelihood of developing a cervical spine blood pressure ⬍ 90 mmHg) was used to measure severity
injury after a motorcycle collision. of physiologic injury in each patient.19 Given the impor-
tance of severe head injury in this analysis, we also con-
trolled for severe head injury, which was defined as Abbre-
METHODS viated Injury Score (AIS) ⱖ 3 in the head region.
This was a retrospective analysis of all registered cases of Patients were also adjusted for demographics including
motorcycle collision in the National Trauma Data Bank age (years), sex, and race (black, white, Hispanic, and oth-
(NTDB) that occurred between 2002 and 2006. The ers). Patient race and ethnicity were included because evi-
NTDB is managed by the American College of Surgeons dence demonstrates racial disparities in medical treatment
and is a convenience sample in that it consists of data vol- and outcomes after trauma.20 Insurance status (uninsured,
untarily reported by trauma centers around the United commercial insurance, government insurance) was added
States and its territories. Although all participating institu- to control for differences in outcomes based on insurance
tions are encouraged to submit complete data, some cen- status.20 Finally, patient sex was included because there is
ters do not routinely enter data regarding safety equipment evidence that women have a survival advantage over men
for motorcyclists.16 A total of 680 of 712 (96%) reporting after severe trauma 21,22. It is also may be that men display
hospitals included information on motorcyclists and hel- greater risk-taking behavior than women, which may po-
met use. This study was restricted to patients from these tentially confound the association between helmet use and
hospitals that reported use of safety devices.17 Detailed in- cervical spine injury.
formation on the specific data collection procedures for the To account for missing data in the NTDB, a sensitivity
NTDB has been published elsewhere.16 analysis using multiple imputation was undertaken. The
Cases involving motorcycle collision as a rider or passen- dataset was imputed 5 times using Rubin’s Rules.23 A mul-
ger (E-Code 810.x to 825.x, series 0.2 or 0.3) were identi- tiple logistic regression was then performed on the imputed
fied and an electronic search of all International Classifica- dataset using the same covariates that were included in the
tion of Diseases (ICD version 9) external cause of injury original nonimputed (list-wise deletion) dataset. The out-
codes for motorcyclist-related collisions in the NTDB was comes of the imputed multiple logistic regression included
undertaken. Injury to the cervical spine included ICD-9 mortality and cervical spine injury in helmeted versus non-
codes (952.00–952.09; 952.16; 806.11, 16, 31, 36, and helmeted motorcyclists.
71) representing bone and nonbone injury, anterior and
central cord syndromes, open and closed cord injuries,
complete and incomplete spinal cord injury, and spinal RESULTS
cord injuries not otherwise specified (NOS). Between 2002 and 2006, 1,862,348 patient cases were en-
Motorcycle riders and any passengers on the motorcycle tered into the NTDB; 1.46 million of these cases were
that were 18 years of age or older were included in the entered at hospitals that report motorcycle-related events.
analysis. Demographics and characteristics of the collision, Of these, 59,274 were involved in motorcycle collisions.
including age, sex, race, crude mortality, injury severity and Among the total cases of motorcycle collision, 40,890
type, injury intent, and mechanism of injury were com- (69%) had complete records for analysis (Fig. 1). Approx-
Vol. 212, No. 3, March 2011 Crompton et al Motorcycle Helmets and Lower Risk of Spine Injury 297

Table 1. Patient Demographic and Injury Severity Character-

istics of Helmeted Motorcyclists Compared with Nonhel-
meted Motorcyclists Involved in Collision
Helmeted Nonhelmeted
Variable (n ⴝ 35,799) (n ⴝ 10,563) p Value
Male, % 87.3 85.9 ⬍0.001
Age, y, mean (SD) 38.9 (13.4) 38.4 (12.6) ⬍0.001
Insurance, % ⬍0.001
Commercial 51.3 39
Government 7.8 6.2
None 18.5 23.4
Race/ethnicity, % ⬍0.001
Black 8.8 7.5
White 82.1 84.6
Hispanic 4.2 5.2
Other 4.9 2.7
Injury Severity Score
(ISS)
ISS, mean 13.4 14.7 ⬍0.001
ISS ⬍ 9 34.3 32.2 ⬍0.001
ISS ⱖ 9 ⬍ 16 34.4 31.2
ISS ⱖ 16 ⬍ 25 16.8 18.3
ISS ⱖ 25 14.5 18.4
Shock, % (systolic
Figure 1. Patient selection in National Trauma Databank (NTDB). blood pressure
ⱕ90 mmHg) 4.8 5.8 ⬍0.001
imately 4% (n ⫽ 2,620) of riders involved in motorcycle Traumatic brain injury
collision died from their injuries. Helmets were worn by (Abbreviated Injury
Scale ⱖ 3), % 8.5 18.9 ⬍0.001
77% of the riders.
Crude mortality, % 3.5 6.2 ⬍0.001
Demographic information on helmeted and nonhel-
meted riders is presented in Table 1. Helmeted riders had
similar age and gender profile compared with nonhelmeted 0.88]) and 65% decreased odds of traumatic brain injury
riders. Nonhelmeted motorcyclists were more frequently (0.35 [95% CI 0.33 to 0.38]) after a motorcycle crash
Caucasian and more likely to be uninsured. when compared with nonhelmeted riders (Fig. 3). Hel-
Injury severity characteristics and crude mortality in meted riders also had a 37% decreased odds of death (0.63
nonhelmeted riders involved in a motorcycle collision [95% CI 0.55 to 0.73]) compared with their nonhelmeted
compared with helmeted riders in a motorcycle collision counterparts.
are presented in Table 2. Nonhelmeted riders had a greater After imputing the dataset for the sensitivity analysis,
proportion of severe head injury (18.9%, p ⬍ 0.001), helmeted riders had a 21% reduced odds of cervical injury
shock on admission (5.8%, p ⬍ 0.001), Injury Severity (0.79 [95% CI 0.69 to 8.89]) and a decreased odds of
Score (mean 14.7, p ⬍ 0.001), and crude mortality (6.2%, mortality (0.64 [95% CI 0.57 to 0.71]) compared with
p ⬍ 0.001) compared with their helmeted counterparts. nonhelmeted riders.
The unadjusted prevalence of cervical spine injury, mor-
tality, and traumatic brain injury among helmeted com-
pared with nonhelmeted riders is represented in Figure 2. A DISCUSSION
greater proportion of nonhelmeted riders had traumatic This analysis of the largest trauma database ever assembled
brain injury and died compared with helmeted riders. Un- demonstrates that injured motorcycle riders are signifi-
adjusted analyses revealed that cervical spine injury was cantly less likely to suffer from a cervical spine injury when
significantly less likely in helmeted riders compared with wearing a helmet compared with nonhelmeted riders, ef-
nonhelmeted riders (3.5% vs 5.4%, p ⫽ 0.001). fectively debunking the myth that motorcycle helmets are
After controlling for potential confounders in the mul- associated with higher risk of cervical spine injury after a
tiple regression model, helmeted riders had significant, crash. There have been several studies that showed that
22% reduced odds of cervical injury (0.78 [95% CI 0.68 to helmets reduce traumatic brain injury and mortality asso-
298 Crompton et al Motorcycle Helmets and Lower Risk of Spine Injury J Am Coll Surg

Table 2. Prevalence of Various Cervical Spine Injuries in Helmeted versus Nonhelmeted Motorcycle Riders
Nonhelmeted
Helmeted riders riders
(n ⴝ 35,799) (n ⴝ 10,563)
Variable n % n % p Value
All 1,245 3.48 565 5.35 ⬍0.00
Closed vertebral column fracture 1,076 3.01 510 4.83 ⬍0.00
Open vertebral column fracture 2 0.01 1 0.01 ⬍0.66
Vertebral column fracture with closed spinal cord injury 108 0.30 37 0.35 ⬍0.43
Vertebral column fracture with open spinal cord injury 5 0.01 2 0.02 ⬍0.72
Spinal cord injury without vertebral column fracture 122 0.34 35 0.33 ⬍0.88

ciated with a motorcycle crash, but this is the first large The finding that helmeted riders are 22% less likely to
analysis that demonstrates that motorcycle helmets also suffer from cervical spine injury than nonhelmeted riders is
protect against cervical spine injuries.5-8,24 Our results are inconsistent with results from previous literature. In fact,
particularly important considering the precipitous rise in previous research suggests that helmets have no effect or
motorcycle collisions and fatalities witnessed in the last may even increase the risk of developing a cervical spine
decade in the United States and abroad. injury after a motorcycle collision.15,11,27 Goldstein15 con-
This study has significant construct validity because it cluded, from a study of 644 riders, that the weight of the
corroborates previous findings that demonstrated that hel- helmet increases the torque on the neck of the rider and
mets reduce mortality and traumatic brain injury after mo- results in more cervical spine injuries, especially when ex-
torcycle collisions. The finding that helmets reduce mor- ceeding speeds of 13 miles per hour. The strength of Gold-
tality by 37% in this study is the same conclusion as that stein’s analysis is that it used a detailed dataset with more
reached by The National Highway Traffic Safety Adminis- than 1,045 data elements in an attempt to reconstruct the
tration in their analyses. It was also demonstrated in this motorcycle crash as completely as possible. However, its
study that helmets reduce traumatic brain injury by 65%, methods, which include a causal model based on regression
which is consistent with the findings of the Crash Outcome analysis, have been sharply criticized by several authors,
Data Evaluation System, in which it was concluded that including the National Highway Traffic Safety Administra-
motorcycle helmets are 67% effective in preventing brain tion, for flawed statistical reasoning.28 Even so, the impact
injury.25 Finally, Croce and colleagues,26 in their analysis of of the Goldstein study in the debate on mandatory helmet
the National Trauma Databank, found a similar reduced laws has been remarkable.
prevalence of cervical spine injury among helmeted riders Analyses subsequent to the Goldstein study have in-
compared with nonhelmeted riders (3.9% vs 5.9%), as cluded reviews of medical records, autopsy reports, analyses
demonstrated in this analysis. That these findings are so of national databases, and prospective studies, which have
consistent with previous well-conducted studies lends face produced equivocal results suggesting that motorcycle hel-
validity to this analysis and corroborates earlier studies. mets are neither a risk factor nor a protection against spinal

Figure 2. Unadjusted comparison of mortality, cervical spine injury, Figure 3. Regression analyses depicting the adjusted odds of mor-
and traumatic brain injury in helmeted compared with nonhelmeted tality (n ⫽ 34,919), cervical spine injury (n ⫽ 35,264), and trau-
motorcycle riders (n ⫽ 46,362 p ⬍ 0.001). Light bar, helmeted; matic brain injury (n ⫽ 35,264) among helmeted versus nonhel-
dark bar, nonhelmeted. meted motorcycle riders (reference).
Vol. 212, No. 3, March 2011 Crompton et al Motorcycle Helmets and Lower Risk of Spine Injury 299

injury. According to the Cochrane Review, these studies are tal or who died at the scene, were not included in the
generally limited by sample size or appropriate control for analysis. However, given that we are using this subset of
confounders. The case control study by O’Connor27 was data in a comparative study of helmeted versus nonhel-
the only study to adjust for confounders and found there meted injuries, the impact of this potential bias should be
was no difference in the risk of cervical spine injury be- minimal. Although we had information on several impor-
tween helmeted riders and their nonhelmeted counter- tant covariates, we lacked information on potential residual
parts. Of the 14 studies that did not adjust for confounders, confounders including location of death (eg, emergency
only 1 showed that helmets are protective against spine department, operating room, ICU), medical or surgical
injury.29 therapies implemented, involvement of drugs and alcohol,
Among the fundamental weaknesses of previous studies or patient comorbidities. This study was limited by its in-
are also limited sample sizes. The largest study included ability to collect information on important crash-related
data from 26,425 crashes and the next largest study in- factors such as the speed, force, severity of the crash, in-
cluded 5,328 patients. The majority of studies, however, volvement of other vehicles, the type of helmet that was
report data from less than 1,000 cases.10,26,30 In our study, a worn, weather, and time of day. Finally, the analysis was
population of 40,890 complete cases (likely more than all performed using a dataset with a significant amount of
other studies combined) of motorcycle collisions were an- missing data. A multiple logistic regression using list-wise
alyzed and adjusted for key confounders known to affect deletion was performed, as was as a robust sensitivity anal-
trauma outcomes. ysis using multiple imputation, and the results were quali-
The implication of these findings regarding the lower tatively similar.
risk of cervical spine injury with motorcycle helmet use Despite these limitations, this study builds on research
should be considered in the ongoing debate about the value in the literature by addressing two significant weaknesses of
of mandatory helmet laws. Although earlier studies have earlier research: inadequate adjustment for important con-
conclusively demonstrated that helmets reduce mortality, founders such as injury severity and insufficient sample
size. Using a large national sample, and after controlling for
traumatic brain injury, and hospital expenditures, this is
important covariates, our results indicate that helmets sig-
the only study of adequate sample size and adjustment for
nificantly reduce cervical spine injury after a motorcycle
confounders to show that helmets are also associated with a
collision. These findings have implications for legislative
reduced risk of cervical spine injury.
policy, particularly when research is evaluated during pol-
Due to the overwhelming epidemiologic evidence that
icy debates regarding whether to repeal or implement state
motorcycle helmets reduce morbidity and mortality, there
mandatory helmets laws.
has been a global movement toward legislating mandatory
helmet laws.31 In 1991, the World Health Organization
recognized that nonhelmeted riders represented a public Author Contributions
health crisis and launched a global helmet initiative to en- Study conception and design: Crompton, Bone, Oyetunji,
courage helmet use worldwide.31 As of 2003, 29 countries Haider
had adopted universal helmet laws.32 Forty years ago, Acquisition of data: Crompton, Bone, Oyetunji, Bolorun-
nearly all states required helmets for motorcyclists of any duro, Haider
age in the United States. Today, motorcycle helmets are Analysis and interpretation of data: Crompton, Bone, Oye-
mandatory for all riders in only 20 states, Puerto Rico, and tunji, Pollack, Bolorunduro, Stevens, Cornwell, Efron,
the District of Columbia. Another 27 states require that Haut, Haider
minors (defined as age younger than 18 years or 21 years Drafting of manuscript: Crompton, Bone, Oyetunji, Pollack,
depending on the state) wear helmets. Bolorunduro, Stevens, Cornwell, Efron, Haut, Haider
Although we used a large national sample of injured Critical revision: Crompton, Bone, Oyetunji, Pollack, Bolo-
patients to determine the effect of motorcycle helmets on runduro, Stevens, Cornwell, Efron, Haut, Haider
the likelihood of developing cervical spine injury after a
motorcycle collision, there are several limitations to this
study. Data is voluntarily reported to the NTDB, resulting REFERENCES
in possible selection bias. The NTDB also does not have 1. National Highway Traffic Safety Administration. Recent trends
cause of death data, so a causal relationship between mor- in fatal motorcycle crashes: an update. Washington, DC: US
Department of Transportation; 2006. Available at: http://www-
tality and helmet use cannot be determined. Similarly, per- nrd.nhtsa.dot.gov/Pubs/810606.pdf. Accessed June 1, 2010.
sons who were involved in crashes, but who did not sustain 2. National Highway Traffic Safety Administration. Motorcycle
injuries severe enough to be admitted to a reporting hospi- helmet use laws. Washington, DC: US Department of Transpor-
300 Crompton et al Motorcycle Helmets and Lower Risk of Spine Injury J Am Coll Surg

tation; 2008. Available at: http://www.dmv.ne.gov/highwaysafety/ 18. StataCorp. 2007. Stata Statistical Software: Release 10. College
pdf/TSFMCHelmetUseLaws2008.pdf. Accessed May 25, Station, TX: StataCorp LP.
2010. 19. Barancik JI, Chatterjee BF. Methodological considerations in
3. Murray C, Lopez, A. Alternative projections of mortality and the use of the Abbreviated Injury Scale in trauma epidemiology.
disability by cause 1990-2020: Global Burden of Disease Study. J Trauma 1981;21:627–631.
Lancet 1997;349:1498–1504. 20. Haider AH, Chang DC, Efron DT, et al. Race and insurance
4. Pervin A, Passmore J, Sidik M, et al. Vietnam’s mandatory mo- status as risk factors for trauma mortality: Arch Surg 2008;143:
torcycle helmet law and its impact on children. Bull World 945–949.
Health Organization 2009;87:369–373. 21. Haider AH, Crompton JG, Oyentunji TA, et al. Females have
5. Kraus JF, Peek C, McArthur DL, Williams A. The effect of the fewer complications and lower mortality following trauma than
1992 California motorcycle helmet use law on motorcycle crash
similarly injured males. Surgery 2009;146:308–315.
fatalities and injuries. JAMA 1994;272:1506–1511.
22. Deitch EA, Livingston DH, Lavery RF, et al. Hormonally active
6. Luna GK, Copass MK, Oreskovich MR, Carrico CJ. The role of
women tolerate shock-trauma better than do men: a prospective
helmets in reducing head injuries from motorcycle accidents: a
political or medical issue? West J Med 1981;135:89–92. study of over 4000 trauma patients. Ann Surg 2007;246:447–
7. Norvell DC, Cummings P. Association of helmet use with death 453.
in motorcycle crashes: a matched-pair cohort study. Am J Epi- 23. Rubin DB. Multiple imputation for non-response in surveys.
demiol 2002;156:483–487. New York, NY: John Wiley and Sons, Inc; 1987.
8. Orsay E, Holden JA, Williams J, Lumpkin JR. Motorcycle 24. Kraus JF, Riggins RS, Franti CE. Some epidemiologic features of
trauma in the state of Illinois: analysis of the Illinois Department motorcycle collision injuries. Factors associated with severity of
of Public Health Trauma Registry. Ann Emerg Med 1996;27: injuries. Am J Epidemiol 1975;102:99–109.
389–391. 25. National Highway Traffic Safety Administration. The Crash
9. Vaca F. National Highway Traffic Safety Administration. Na- Outcome Data Evaluation System (CODES) [Technical Re-
tional Highway Traffic Safety Administration notes. Evaluation port]. Washington, DC: US Department of Transportation;
of the repeal of the all-rider motorcycle helmet law in Florida. 1996. Available at: http://www-nrd.nhtsa.dot.gov/Pubs/808-
Ann Emerg Med 2006;47:203. 338.PDF. Accessed June 3,2010.
10. Liu BC, Ivers R, Norton R, et al. Helmets for preventing injury 26. Croce MA, Zaraur BL, Louis LJ, Fabian TC. Impact of motor-
in motorcycle riders (Review). The Cochrane Library; 2009. cycle helmets and state laws on society’s burden: a national
11. Houston DJ, Richardson LE. Motorcycle safety and the repeal of study. Ann Surg 2009;250:390–394.
universal helmet laws. Am J Publ Health 2007;97:2063–2069. 27. O’Connor P. Motorcycle helmets and spinal cord injury: Hel-
12. Chiu WT, Kuo CY, Hung CC, Chen M. The effect of the met usage and type. Traffic Inj Prevent 2005;6:60–66.
Taiwan motorcycle helmet use law on head injuries. Am J Publ 28. Lawrence BA, Max W, Miller TR. Costs of injuries resulting
Health 2000;90:793–796. from motorcycle crashes. Available at: http://www.nhtsa.gov/
13. Solagberu BA, Ofoegbu CKP, Nasir AA, et al. Motorcycle inju-
people/injury/pedbimot/motorcycle/motorcycle_html/appb.
ries in a developing country and the vulnerability of riders, pas-
html. Accessed June 3, 2010.
sengers, and pedestrians. Injury Prevention 2006;12:266–268.
14. Li-Ping Li, Gong-Li Li, Qi-En CA, et al. Improper motorcycle 29. Sarkar S, Peek C, Kraus JF. Fatal injuries in motorcycle riders
helmet use in provincial areas of a developing country. Accident according to helmet use. J Trauma-Inj Inf Crit Care 1995;38:
Anal Prevent 2008;40:1937–1942. 242–245.
15. Goldstein J. The effect of motorcycle helmet use on the proba- 30. Keng S. Helmet use and motorcycle fatalities in Taiwan. Acci-
bility of fatality and the severity of head and neck injury. Eval- dent Anal Prevent 2005;37:349–355.
uation Rev 1986;10:355–375. 31. Jones MM, Bayer R. Paternalism and its discontents: motorcycle
16. Committee on Trauma. American College of Surgeons. NTDB, helmet laws, libertarian values, and public health. Am J Pub
version 7. Chicago, IL: American College of Surgeons; 2008. Health 2007;97:208–217.
17. Kardooni S, Haut ER, Chang DC, et al. Hazards of benchmark- 32. World Health Organization Helmet Initiative: About Us. Avail-
ing complications with the National Trauma Data Bank: numer- able at: http://www.whohelmets.org/about.htm. Accessed Au-
ators in search denominators. J Trauma 2008;64:273–277. gust 13, 2009.