Seatbelts Determine Location of Brain Injury

Frank Hillary, M.S., Drexel University University, Psychology Department
Philip Schatz, Ph.D., Saint Joseph's University, Psychology Department
Douglas L. Chute, Ph.D., Drexel University University, Psychology Department


This retrospective study examined the effects of seatbelt use on the location and severity of brain injury. CT, MRI and acute care hospital records were analyzed for 171 patients enrolled in the Pennsylvania Head Injury Program. Those utilizing a seatbelt experienced a shorter duration of loss of consciousness and endured shorter acute care hospital stays. Approximately 60% of both restrained and unrestrained occupants sustained frontal lobe lesions. However, unbelted passengers sustained a greater number of posterior cortical lesions (37%) than did belted passengers (13.5%). Belted passengers sustained significantly greater subcortical damage (27%) compared to unbelted passengers (12%). The number of brain stem lesions was not significantly different between belted (16%) and unbelted (17%) passengers. Seatbelt use apparently affects the location and severity of brain injury which accounts for part of the eventual rehabilitation outcomes.


In the United States, motor vehicle accidents (MVA) are the most common cause of traumatic brain injury (TBI), accounting for over half of all brain injuries annually (Jagger, 1992). In 1966, the US federal government required that seatbelts be placed into all new automobiles as standard safety equipment. Since 1966 there has been a steady decrease in the number of injuries sustained in MVA's each year. Nationally, seatbelt usage is on the rise (67%), and it is estimated that between 1985 and 1995, seatbelts have prevented 55,600 deaths (Popular Press, 1995).

Crash testing research reveals that wearing a seatbelt greatly reduces the chance of brain injury during roll-over and frontal impact crashes; the restrained passenger will experience less overall body movement and is less likely to be ejected from the vehicle (Strother, 1984). In a 35 mph frontal barrier collision, the average displacement of a restrained passenger is 56, 40, and 30 cm for the head, pelvic, and chest areas. While it is difficult to estimate the displacement of a unrestrained passenger, during the same collision, the overall displacement would be measure in meters (Moffat, 1984). Due to greater overall movement by unrestrained passengers, one would expect an unrestrained participant to sustain a greater number of impact points to the head, resulting in a greater number of cortical lesions.

The purpose of this study was to analyze the effect of seatbelt usage on type and severity of TBI. It is commonly accepted that using a restraint while operating a motor vehicle decreases the likelihood of serious injury. This notion was supported. For example, only 21.4% of the head injury patients in our study came from the presumed 67% of seat belt wearers, while 78.6% came from the presumed 23% of unrestrained drivers. In the current investigation, belted and unbelted subjects who sustained moderate to severe TBI were compared. Specific effects of seatbelt usage were analyzed on two levels: type of brain injury sustained (i.e., the anatomical structures involved), and head injury severity, as measured by Glasgow Coma Scale (GCS), loss of consciousness (LOC) and length of hospitalization.


Subjects: Subjects were 171 individuals drawn from a pool of over 500 individuals considered to be domiciled in the state of Pennsylvania at the time of injury who sustained a head injury and applied to the Pennsylvania Head Injury Program (PHIP) between 1985 and 1996.

Subjects were included in the study only if their file contained a completed PHIP application, acute care medical records, and evidence of head injury as determined by the PHIP criteria. Approximately 100 subjects were excluded due to prior head injury, congenital defects, organic or degenerative brain disorders, vascular disorders, psychiatric history, or other prior physical, emotional, or cognitive disabilities, (in accordance with Schatz, 1995). In addition, approximately 229 subjects were excluded from the study because they did not sustain TBI as a motor vehicle occupant.

Procedures: Lesions specified on CT and MRI reports were coded for 27 distinct locations. Cortical lesions were grouped into anterior (frontal lobe) and posterior (parietal/occipital). Subcortical lesions were divided into subcortical (diencephalon and basal ganglia) and brainstem (mesencephalon and metencephalon) groups.


Severity of Injury A two-tailed T-test revealed significant effect of seatbelt usage on acute care length of stay [F=11.25, p<.001]; subjects not wearing seatbelts spent significantly more days in acute care hospital (mean = 53) than did subjects wearing seatbelts (mean = 37). Chi2 analysis revealed a significant effect [X2=20.76, p<.03] of seatbelt usage on Glasgow Coma Scores when arriving in the emergency room (See Figure 1). Those utilizing a seatbelt on average had significantly higher scores (6.8) than unbelted passengers (5.3). Finally, a two tailed t-test revealed a significantly greater LOC in passengers not utilizing a seatbelt (F= 1.6, p< .03).

Location of Injury Incidence and percentage of lesions were calculated for cortical lobes and for subcortical regions for each group (belted vs. unbelted motor vehicle occupant).

The presence of lesions was collapsed into anterior and posterior regions. Chi2 analysis reveals a significant effect [X2(1)=7.10, p<.01] of seatbelt usage on posterior lesions; subjects not wearing seatbelts were significantly more likely to sustain damage to the posterior region of the brain. Anterior lesions were common to both groups, as 59% of all restrained drivers and 61% of all unrestrained drivers sustained frontal lobe damage

Because roughly 60% of all participants sustained frontal lobe damage, analysis of cortical damage excluded this region. A two tailed t-test revealed significant effects of seatbelt usage on total number of lesions sustained in the parietal and occipital lobes [F=1.98, p<.002]. Those utilizing a seatbelt sustained significantly fewer posterior cortical lesions.

In addition, subcortical lesions were counted for each subject, ranging from (0) lesions to (4) lesions per individual. The average number of lesions for each group is displayed in Table 1. A statistical analysis of brain injury to the hypothalamus-thalamus region, basal ganglia and internal capsule revealed marked differences between groups. Participants wearing a seatbelt were much more likely to have positive findings in this group of structures (labeled "Subcortical"), at [F= 4.44, p < .001]. However, an analysis of only brainstem structures (pons, medulla, cerebellum, and fourth ventricle) did not reveal significant differences between groups.

Table 1 Lesions to the Interior of the Cerebrum








No Seatbelt





*-hypothalamus/thalamus, basal ganglia, internal capsule

**- pons, medulla, cerebellum, fourth ventricle


As stated only 21% of the subjects in the study were wearing a seatbelt at the time of their TBI. This frequency is significantly disproportionate to the national statistics (67%). This discrepancy offers considerable support to the notion that seatbelts are an effective profilaxis to brain injury during a motor vehicle accident. Furthermore, when comparing injureies, belted motor vehicle occupants appeared to sustain less severe brain trauma. This is supported by the observation that several variables widely believed to describe brain injury severity (LOC, GCSER, and length of hospital stay) were less severe in the belted group.

Lesion analysis allowed for several generalizations to be drawn about the types of brain injuries sustained during MVA. Subjects in both groups sustained an equal percentage of anterior lesions, which can be explained by deceleration conditions commonly experienced by all vehicle occupants (restrained or unrestrained) during MVAs. The present results show that subjects not wearing a seatbelt at the time of their TBI sustained significantly more focal and posterior lesions. It is probable that this is due to the greater displacement experienced by unbelted passengers. Furthermore, belted passengers sustained a greater number of subcortical lesions. This may be the result of head and neck flexion at impact.

By reducing severity of TBI, seatbelt utilization may have significant implications for rehabilitation prognosis. In addition, decreased acute care hospital stays with seatbelt use may translate to significantly higher residual funding for rehabilitation in the current managed-care climate. Furthermore, understanding the fundamental differences in brain injury between these groups may ultimately be useful in the intervention and rehabilitative planning for the brain injured patient.

In gaining a better understanding of the physical process of brain injury, different methods to protect the central nervous system can be developed. As this study points out, the physical conditions that the human body endures during an MVA might be quite different, depending upon the use of a safety restraint. A multidisciplinary approach to studying brain injury could offer specific answers about TBI by bringing physiology face to face with physics. For example, the best prophilaxis to brain injury may be to allow for increased movement by the head and neck, while preventing head/obstacle contact. The answers are not simple, yet by understanding exactly how brain damage occurs, engineers may develop the next generation of safety systems with the brain in mind.

Future Directions:

A follow up study to the current investigation should be careful to control for several additional factors. First, collecting CT results in the current investigation was performed retrospectively from PHIP application forms and copies of acute care hospital records. This did not allow for consistent analysis of lesion size or type (hemorrhage vs. contusion). Furthermore, the PHIP database is composed of mostly of patients that have suffered moderate to severe brain injuries. It would be extremely useful to analyze the effects of seatbelt use on mild TBI (post concussion syndrome). In fact, anlaysis of such a group might provide a better indication of the type of TBI sustained by those wearing a seatbelt.

The seatbelt is only one factor accounting for the variance in brain injury severity. The authors of this investigation plan a prospective investigation into other factors believed to be contributing to patient injury and outcome. These factors include vehicle factors such as size and speed, external factors such as what is struck or what is striking the vehicle, as well within cab factors such as airbag deployment. There have been very few ecologically valid studies performed that investigate brain injury. As stated, most investigation into MVAs and passenger outcome focuses on survival rates and not specifically injuries.


Jagger, J., Levine, J. I., Jane, J.A., & Rimel, R.W. (1992). Epidemiologic features of head injury in a predominantly rural population. Journal of Trauma, 24 (1), 40-44.

Kallieris, D., Mellander, H., Schmidt, G., Barz, J., Mattern, R., "Comparison between Frontal Impact Tests with Cadavers and Dummies in a Simulated True Car Environment", 26th Stapp Car Crash Conference, SAE, 1982.

Moffat, C., Moffat, E., Weiman, T., "Diagnosis of Seat Belt Usage in Accidents", SAE, 840396, March 1984.

Popular Press, Motor Trend (1993). Peterson Publishing Company, October, Vol. 45, No. 10, pg. 114, ISSN: 0027-2094.

Schatz, P. (1995). Predicting level of Independence following moderate and severe traumatic brain injury. Doctoral Dissertation, Drexel University.

Strother, c., Smith, G., James M., Warner, C., "Injury and Intrusion in Side Impacts and Rollovers", SAE, 840403, March 1984.