DETERMINATION OF FRONTAL OFFSET TEST CONDITIONS BASED ON CRASH DATA

Sheldon L. Stucki
William T. Hollowell
NHTSA R&D
Osvaldo Fessahaie
ISSI, Inc.
United States
Paper Number 98-S1-O-02

ABSTRACT occupants with air bags. It appears that the oblique

impact with over 50 percent overlap produces the most
This paper reports on the test procedure development severe responses on the "target" car. Development of this
phase of the agency’s Improved Frontal Protection impact configuration into a potential frontal test
research program. It is anticipated that even after all cars procedure has been completed using a moving deformable
and light trucks have air bags for drivers and front seat barrier (MDB).
passengers there will remain over 8,000 fatalities a year
and over 100,000 moderate to severe injuries. This INTRODUCTION
research program will address these injuries/fatalities
through development of crash tests with impact In the United States, air bags with lap and shoulder
conditions not currently addressed by FMVSS No. 208, belts are specifically required by legislation (i.e., the
development of additional or more appropriate National Highway Traffic Safety Administration
instrumentation and injury criteria on the test surrogate, Authorization Act of 1991) for both front outboard
and evaluation of other sizes of test surrogates. seating positions in all passenger cars manufactured after
September 1, 1997. They are also required in all light
An analysis of crash data is presented using the trucks, multipurpose passenger vehicles (e.g., vans, utility
National Automotive Sampling System (NASS) and the and sport vehicles), and buses with a gross vehicle weight
Fatality Analysis Reporting System (FARS) for fatality rating of 3,846 kilograms (8,500 pounds) or less and an
counts. The population is drivers in frontal collisions unloaded vehicle weight of 2,489 kilograms (5,500
with air bag restraints. Using NASS, frontal impact pounds) or less manufactured after September 1, 1998.
modes are grouped into general "test" conditions which NHTSA’s "Third Report to Congress - Effectiveness of
will best represent the real world impact environment. Occupant Restraint Systems and Their Use", dated
These general test conditions include full barrier, left and December 1996, estimates that drivers protected by air
right offset, and other impact modes. Using these general bags experience a reduced fatality risk of 11 percent
groupings of impact conditions, the analysis further overall and 31 percent in pure frontal accidents.
assesses degree of overlap and impact direction to
determine more specifically which crash conditions result The detailed performance requirements for these
in highest injury/fatality to drivers with air bags. systems are contained in Federal Motor Vehicle Safety
Injury/fatality risk is also assessed by driver size and Standard (FMVSS) No. 208, Occupant Crash Protection.
body region, with a more detailed analysis of leg injuries. The standard has long specified a barrier test requirement
Finally, a preliminary benefits analysis is presented for a using both belted and unbelted dummies. Beginning in
future frontal, left, offset test procedure. March, 1997, Standard 208 has been temporarily
modified to allow for a 48 kmph sled test requirement for
A test procedure has been developed, and is reported unbelted dummies which made it easier for manufacturers
on in a separate paper [1]. Collinear and oblique, offset, to quickly introduce less aggressive, depowered air bags.
frontal crash testing, at different widths of overlap, has This temporary option expires in September, 2001 and
been conducted with several current model, "target" cars thereafter the full barrier test is again required. The main
into a standard "bullet" car at closing speeds of about 110 dynamic performance requirements in FMVSS No. 208,
kph. Dummy injury measurements and structural either sled or barrier test, involves successful testing with
responses provide a basis for determining which impact a 50th percentile adult dummy at all speeds up to 48
conditions produce the most severe environment for

Stucki, 1
kilometers per hour (30 miles per hour) at all angles 5].
between perpendicular and 30 degrees to either side of
perpendicular. The tests can be run both with the dummy The agency has been directed by Congress to develop
being unbelted and with the belts on. "Successful" crash a frontal, offset compliance test to complement the
testing requires that the dummy Head Injury Criterion current FMVSS No. 208 full frontal test. The agency is
(HIC) be 1,000 or less, the dummy chest deceleration be evaluating a 40 percent overlap, 60 Kmph full-fixed-
60 G’s or less, and the dummy femur loads be at or below deformable barrier test which has been adopted in
10,000 Newtons. The chest deflection on the Hybrid III Europe, but at a test speed of 56 kmph. This will
dummy must be less than 75 millimeters. determine whether benefits can be realized in the U.S.
from adopting this test procedure in the near future. The
The agency is currently in the process of proposing plan for making this assessment was presented in a report
further requirements for reducing air bag aggressiveness to Congress in April, 1997. The results of the FY 1997
which will lead to advanced air bags. Based on testing is presented in a proposed paper for the 16th ESV
assessment of technologies which will come available in Conference [6]. The oblique/offset test being developed
the next few years future air bag systems may include by NHTSA ‘s Research and Development office would
variable level deployments, or suppression, based on be considered a longer term project.
crash severity, and/or restraint use, pre-crash occupant
position and/or size. As previously noted the full barrier Defining the problem includes assessing crash data
test will again be required in September, 2001 to possibly and identifying general laboratory test conditions that can
"recapture" injuries/fatality savings in high severity be used to replicate the safety performance of air bag
vehicles in use. Then, evaluating the performance of a
crashes which may have been lost with depowered air
variety of production vehicles under those preliminary
bags. Part of the analysis in this paper is to look at
crash conditions, comparing their performance, and
crashes which may be represented by the 30 mph fixed
conducting potential benefits assessments to guide the
barrier test of FMVSS No. 208 in terms of frequency of
agency for the "final" selection of a test procedure(s).
involvement, and injuries.
Some general conclusions from the analysis are:
Even after full implementation of driver and
passenger air bags as required by FMVSS No. 208, it has
 For drivers in frontal collisions with air bags, the
been estimated that frontal impacts will still account for offset crash configurations with highest frequency
over 8,000 fatalities and 120,000 moderate-to-critical and risk of serious to fatal injuries is a left offset,
injuries (i.e., injuries of AIS 2). The fatality estimate is vehicle-to-vehicle impact with substantial overlap .
based on 1995 FARS figures adjusted to a baseline non-
air bag fleet and applying an air bag effectiveness  Drivers with air bags have a higher risk of leg injury
estimate of 11 percent (from the Agency’s "Third Report in left offset crashes than in other frontal crashes and,
to Congress - Effectiveness of Occupant Protection thus, reducing leg injuries should be a prime
Systems and Their Use.") to predict fatalities for an all air objective in development of a offset test procedure.
bag fleet. The number of fatalities in non-rollover frontal Leg injury should address tibia, knee and ankle
impacts is based on the proportion estimated by the measures, not addressed currently in the standard.
NASS analysis and the computations are shown as part of
Table 9. The estimates of annual numbers of moderate-  For left offset impacts, improvements to reduce
to-critical injuries are from the Agency’s "Final injury should address leg/instrument panel and floor
Regulatory Evaluation - Actions to Reduce the Adverse interaction and all regions with left side surfaces.
Effects of Air Bags - FMVSS No. 208 - Depowering."
The objective of this research program is to address these  The size grouping representing 50th percentile males
fatalities and injuries and provide a basis for the possible results in the highest crash exposure and number of
future improvements in frontal protection. This may injuries/fatalities for left offset impacts. However,
include upgrade of FMVSS No. 208 injury criteria and both smaller and larger drivers have a higher risk of
test devices, and the development of supplementary test AIS 2 injuries and larger drivers have higher risk of
procedures for the evaluation of occupant injury in AIS 3 injuries and fatalities even though their crash
crashes of higher severity and in different impact modes exposure is much lower than that for the 50th
than those addressed by the current FMVSS No. 208 [2- percentile grouping.

Stucki, 2
 7%
 Based on various assumptions, a requirement for a
left offset test procedure could save as many as 6%

Risk of MAIS>=3
5,100 AIS 3 injuries and over 20,000 AIS 2 5%
injuries each year. Leg injuries alone could be
4%
reduced annually by about 11,000 for AIS 2 and
about 2,000 for AIS 3. Although not estimated, it 3%
appears that substantial fatalities could be reduced. 2%
The European (EU) offset test procedure could
potentially address many of the leg injuries, while 1%
other recent or future vehicle improvements, such as 0%
Belted Unbelted As Used
FMVSS No. 201 head protection or future advanced
Belt Use
air bags will eliminate many of the other injuries and
fatalities. Air Bag No Air Bag

The analysis is based on relatively limited cases of Figure 1. Serious Injury Risk by Restraint, Drivers in
drivers with air bags in NASS and findings may change Frontals
with additional data.
1.4%
CRASH ENVIRONMENT 1.2%

Risk of Fatality
1.0%
The agency's National Automotive Sampling System
(NASS) files for the years 1988-96 were used to project 0.8%
the occupant injuries that will occur in an all air bag fleet. 0.6%
In the 1988-1996 NASS there are about 2700 vehicles 0.4%
with driver air bags in frontal crashes. The analysis will
0.2%
identify test conditions to simulate crashes with highest
risk and frequency of injury/fatality. These test 0.0%
Belted Unbelted As Used
conditions can be used to analyze the safety performance Belt Use
of baseline vehicles and to assess potential
Air Bag No Air Bag
countermeasures. The NASS is a statistical sample of
the United States accidents investigated in detail. About
Figure 2. Fatality Risk by Restraint, Drivers in Frontals
4,500 crashes per year are currently being investigated.
The NASS files for these years differ from those of
previous years in that only the more serious accidents
AIS>=3 Injury, Belts "As Used"
qualified for inclusion into the files. Crashes involving
air bag-equipped vehicles have been increasing along 1.4%
with the increasing installations. Between 1988 and
1996, the NASS teams investigated 44,368 crashes, 1.2%
representing an estimated 21 million crashes and 1.0%
Percent

12 million injured vehicle occupants nationwide. In these 0.8%

crashes, 2,891 driver and 378 right, front passenger air
0.6%
bag deployments were investigated, representing an
estimated 1,012,263 driver and 124,506 right, front 0.4%
passenger air bag deployments that occurred during that 0.2%
time frame. 0.0%
Arms Thorax Head Legs
When comparing drivers with air bags to those Body Region
without air bags serious injury risk is slightly lower with
air bags and belts and belts “as used”, i.e, no Air Bags No Air Bags
discrimination for whether belts were or were not used
(Figure 1, and Table 1.) However, for fatalities air bags Figure 3. Serious-to-Fatal Injury Risk by Body
have lower rates for all restraint conditions and Region, Belts “As Used”

Stucki, 3
 AIS>=3 Injury, Belts Used
1.0% AIS>=3 Injury, Belts Not Used
3.5%
0.8% 3.0%
2.5%
Percent

0.6%

Percent
2.0%
0.4%
1.5%
0.2% 1.0%
0.5%
0.0%
Arms Thorax Head Legs 0.0%
Body Region Arms Thorax Head Legs
Body Region
Air Bags No Air Bags
Air Bags No Air Bags
Figure 4. Serious-to-Fatal Injury Risk by Body Region,
Belts Used Figure 5. Serious-to-Fatal Injury Risk by Body Region,
Belts Not Used
substantially lower for belts “as used” (Figure 2.)
accident analysis has been coupled with offset crash
Figures 3 to 5 and Table 2 show risk of serious testing to determine which impact configurations produce
injury by body region in frontal crashes for drivers of air the highest likelihood and frequency for injury/fatality.
bag equipped vehicles in air bag and non-air bag cars with
and without belts. For serious-to-fatal injury and belts Drivers of all vehicles in 1988-1996 NASS were
“as used” head, thorax and leg injuries are substantially grouped by their general area of damage (GAD) and
lower with air bags. (Figure 3.) Arm injuries are principal direction of force (DOF1) into a frontal impact
somewhat higher with air bags. Since the majority of population. Drivers were considered to be in frontal
drivers in frontal impacts are belted (about 84 percent impacts if their vehicle sustained DOF1 between 11 and 1
with air bags and 68 percent without) the injury risks by o'clock or DOF1 was 10 or 2 and GAD1 was front or side
body region are similar when belts are used (Figure 4) as with damage forward of the A-pillar. The frontal impact
when “as used” . For unbelted drivers there is no population is then separated into specific crash modes to
apparent reduction in serious to fatal chest or head identify potential impact configurations with high
injuries with air bags (Figure 5.) frequency and risk of injury to be simulated by crash test
procedures. The frontal population was separated by
Traditionally, fatality reduction has been the direction of force (DOF) into collinear or oblique (left or
emphasis of the agency’s research program. More right), by damage distribution into offset (left or right) or
recently, however, attention has been focused toward distributed, and by object contacted into another vehicle
injury reduction, particularly for those injuries which lead or fixed object. Counts in the paper are weighted unless
to life long disabilities. This added focus includes the noted. DOF is used to delineate collinear (12 o'clock),
role of lower extremity and pelvic injuries in frontal left (10 & 11 o'clock) and right (1 & 2 o'clock) oblique
crashes. impacts. For frontal damage (GAD1=F), overlap is
defined by the crash "D" variable when known and after
Selection of Test Conditions Based on Crash Impact 1989; otherwise, the primary specific horizontal location
Modes (SHL1) is used, and is separated into distributed ("D"=0
or SHL1=D), left ("D"<0 or SHL1=Y or L) and right
An additional test procedure for increased frontal ("D">0 or SHL1=Z or R) offset impacts. For those
protection should simulate those impact modes in the impacts with left or right damage (GAD1) the location
“real-world” crash environment which result in highest must include the front corner of the vehicle (SHL1=F)
frequency and risk of injury/fatality. Since FMVSS No. and is entered as left or right 1/3 of the vehicle's front
208 sets performance requirements for full frontal (equivalent to SHL1=L or R for GAD1=F.)
impacts, the initial analysis focused on "offset", frontal
impacts as candidate accident modes for simulation. The Grouping Into Most Appropriate Test Procedure

Stucki, 4
 The exposure population for frontal impacts, i.e, car from -30 to +30 degrees. As shown in the figure, all
number of collisions, is based on 1988 to 1996 NASS to collinear, distributed damage impacts and oblique,
estimate the exposure for an all air bag fleet. Drivers in distributed damage, fixed object impacts with distributed
frontal collisions are grouped by impact conditions (DOF, damage are assumed to be best simulated by this test
damage distribution and crash partner) into the most condition. The left offset configuration, either collinear
appropriate test situation to simulate the type of collision. or oblique direction of impact is assumed that all left
offset impacts, either collinear or oblique, are best
For specifying impact conditions for a future frontal, represented by this test condition. Also, it is assumed that
offset test both crash pulse and intrusion are of the left oblique, vehicle-to-vehicle impact, with
comparable importance in occupant injury outcome. distributed damage is better simulated by the left offset
However, the agency is also currently addressing issues test than by the barrier test. Not only may less than full
of what are the appropriate conditions for a full frontal overlaps often produce distributed damage, but the
test procedure in FMVSS No. 208, i.e, the full barrier, the interaction of the vehicle and the propensity for higher
current sled test or some other simulation test. For these intrusion is well simulated by this test even though there
types of tests, intrusion is of secondary importance and may be near distributed damage.
crash pulse alone is the important crash factor in the
occupants injury outcome. To show a comparison of A right offset configuration would include right side
intrusion in full barrier type impacts and offset impacts, impacts in the same way as left side impacts are included
crash situations with intrusions of 6 inches or more into in the left offset test. About 9 percent of cars have
the vehicle compartment are assumed to compromise the offset, frontal damage which is opposite to the clock
compartment integrity and lead to serious or fatal injuries. direction, i.e., left and right oblique impacts with right
As shown in Figure 6 the incidence of 6 inches or more and left offset damage, respectively. Note that this
intrusion is much greater in the offset impact modes than would be the impact configuration for the "bullet" vehicle
in the full barrier type modes, especially for crash in a left or right oblique impact to the "target" vehicle, as
severities less than 30 mph. shown in Figure 6. Based on the assumed groupings of
vehicle impact conditions from above, the left offset test
Figure 7 shows the test situations, the impact would represent about 34 percent of cars with air bags in
conditions for that test and the percentage of all frontal "frontal" crashes, with right offset making up about 35
impacts represented. The first group of impact modes are percent and full barrier about 22 percent.
those in which the test must account for both crash pulse
and intrusion. The current frontal test is a full frontal If only crash pulse is considered, the full frontal
impact into a fixed rigid barrier with impact angles on the fixed barrier accounts for the majority of impact modes in
frontal crashes. Collinear, car-to-car crash tests at partial
overlaps of 50, 60 and 70 percent, and a 30 degree
7% oblique car-to-car impact with 50 percent overlap on a
Chevrolet Corsica using a Honda Accord as the striking
% of Vehicles with > 6"

6%
vehicle have been conducted. The car-to-car tests were
5% conducted with both cars moving at about 56 kmph.
4% Also, the agency has conducted an NCAP test using the
Corsica, i.e., a 56 kmph, full frontal, rigid barrier test.
3% The longitudinal compartment deceleration crash pulses
2% are shown in Figure 8. The collinear 60 and 70 percent
overlap crash tests appear to be well simulated by the full
1% barrier impact along with the oblique impact at 50 percent
0% overlap. However, for the collinear impact at 50 percent
Full Barrier Left Offset Total Offset overlap the crash pulse appears to deviate somewhat from
Crash Modes the full barrier pulse. Based on these comparisons, the
collinear impacts with overlaps ranging from somewhere
Under 30 mph All between 50 and 60 percent (say 55 percent) to full
overlap were classified as “full barrier-like” crashes.
Figure 6 - Percent of Vehicles with 6 inches or more Since exact overlap dimensions are not available in all
Intrusion by Crash Mode NASS cases, SHL1 also must be used. SHL1 is used to

Stucki, 5
Figure 7. Possible Frontal Test Conditions and Impact Modes Addressed (1988-1996 NASS)

define a crash as “full barrier-like” for values of E and compared to drivers of vehicles in all frontal crashes to
greater overlap. determine what proportion of all frontal impacts are
represented by “barrier-like” impact conditions. This
Oblique, car-to-car impact tests have been conducted comparison is made for:
only at nominally 50 percent overlap impact conditions.
As shown in Figure 8 the 50 percent overlap, oblique
5
crash test actually produces as severe compartment
deceleration crash pulse as the NCAP full barrier test, at 0
similar impact speeds for the Corsica. Thus, in the
Acceleration, G’s

absence of additional tests with varying proportions of -5

overlap, it is assumed that oblique impacts can be
-10
represented by the full barrier test at overlaps of D and
greater. The category of frontal impacts which qualify as -15
“full barrier” (FB) like crash pulses, based on this
analysis, include: -20

-25
 Collinear with 55 percent and greater overlap or
SHL1 E when exact overlap dimension is not -30
available 0 25 50 75 100 125 150
 Oblique (DOF1 12 o’clock) with SHL1 D Time (Milliseconds)
50% 60% 70%
The 1988 through 1996 NASS-CDS files are queried Full 50% Oblique
for impact conditions shown above. Drivers in crashes
Figure 8. Crash Pulses by Overlap, Chevrolet Corsica,
with “barrier-like” impact conditions on the vehicle are
hit by Honda Accord and NCAP, at 56 Kmph

Stucki, 6
 8.0%
 All drivers
 Drivers of vehicles with air bags, and MAIS 3 6.0%
injuries

The 1988 through 1996 NASS-CDS files are queried for 4.0%
impact conditions which produce crash pulses which may
be fairly well represented by the full barrier crash pulse, 2.0%
as discussed above (referred to subsequently as “barrier-
like.”) Drivers in crashes with “barrier-like” impact
conditions on the vehicle are compared to drivers of 0.0%
Full Barrier Left Oblique Right Oblique
vehicles in all frontal crashes. Left Offset Rightt Offset

For drivers in vehicles with air bags the proportion of MAIS>=2 MAIS>=3
driver in vehicles with “barrier-like” crashes as a percent
of all frontal crashes is: Figure 9. Injury Risk by Test Condition

 74 percent for all drivers

 83 percent for drivers with MAIS 3 injuries 0.5%
 73 percent for driver fatalities
0.4%
The remainder of the paper will consider those crash
configuration groupings which account for both crash 0.3%
pulse and intrusion as factors in occupant injury.
0.2%
Injury Risk by Test Configuration

Comparing injury risk shows that for moderate and 0.1%

more severe injuries (MAIS 2) the injury risk is
somewhat higher for vehicles in crashes fitting “left 0.0%
Full Barrier Left Oblique Right Oblique
offset” than those described by “full barrier” (7.6 Left Offset Rightt Offset
percent and 6.8 percent, respectively.) For serious and
higher injuries (MAIS 3), the "full barrier" groupings
has the highest injury rate (3.8 percent.) Left offset and
right offset groups both have much lower serious injury Figure 10. Fatality Risk by Test Condition
rates of about 2.1 percent and 1.3 percent, respectively
(Figure 9. And Table 3) Figure 10 shows fatality risk for offset impact modes would result in the highest number
the various impact modes grouped into appropriate test of drivers with MAIS 2 and fatal injuries (about 47,000
condition. These fatality rates are based on limited and 4,200, respectively.) Although full barrier type
observations: 10 for full barrier, 38 for left offset, and 10 impacts would account for the highest number of
for right offset. The left offset grouping has much higher MAIS 3 injuries (14,942) the left offset modes are only
fatality risk (0.43 percent) than full barrier (0.25 percent) slightly less (13,042.)
and almost four times that experienced by drivers with air
bags in right offset modes (0.11 percent.) Within the test groupings for left offset and right
offset the effect of overlap on injury rate was assessed.
An estimate of the annual injuries/fatalities which As a rough approximation of overlap percent, an average
might be expected with an all air bag fleet is computed in car width of 66 inches is assumed for "L" in the offset
Table 3 and shown in Figures 11 and 12. The estimates formula: Overlap = 1-(2*D/L), where “D” is the distance
are based on the injury/fatality risks, shown previously, from the vehicles center-line to the damage mid-point.
applied to the expected number of drivers with air bags in Overlap is then separated into 1/3 or less of the car width,
tow-away crashes in an average year (1988 through 1996 over 1/3 to 2/3 of the width and over 2/3 of the width. As
NASS divided by nine.) Based on these estimates the left

Stucki, 7
 50000 4%

40000
3%

30000
2%
20000

1%
10000

0 0%
Full Barrier Left Oblique Right Oblique 0-1/3 >1/3-2/3 >2/3 Total
Left Offset Rightt Offset
Left Oblique/Offset Right Oblique/Offset
MAIS>=2 MAIS>=3
Figure 13. Serious-to-Fatal Injury Risk by Overlap
Figure 11. Estimated Annual Injuries by Test Condition overlaps produce lower injury rates (1.8 percent for right
offset.) The left offset impact at over 2/3 overlap
5000 produces the highest MAIS 3 injury rate of all offset
impact modes considered (3.8 percent.)
4000

3000 Recommendation

2000 Based on analysis of the NASS crash data files of

drivers in frontal collisions with air bag restraints, the
1000 offset crash test which represents actual crash
configurations with the highest frequency and risk of
0 serious to fatal injuries is a left offset, vehicle-to-vehicle
Full Barrier Left Oblique Right Oblique
impact with substantial overlap (E or greater.) The
Left Offset Rightt Offset
specific recommendations for impact angle and overlap
percentage will be variables addressed in the crash test
development phase of the program. The remainder of the
Figure 12. Estimated Annual Fatalities by Test Condition paper assumes that this type of test condition will be
selected as the offset procedure for the future and the
analysis focuses on these crash modes.
discussed above, left and right damaged vehicles with
damage to the front corner were grouped into the 1/3
overlap category. By using these damage width Body Region Injury Assessment
groupings, the SHL1 parameter, which is separated into
damage width increments of one-third of the vehicle Injury measures, criteria and instrumentation and the
width, may be used when "D" is not known. The test surrogate itself should be selected based on the
relatively low injury risk for configurations grouped location and type of injuries experienced by the driver in
under a left offset test appears to be due to low frontal, left offset crashes.
occurrence of MAIS 3 injuries in narrow overlap
impacts. For left offset impacts the rate of MAIS 3 Injuries to specific body region are tallied by AIS
injuries is about 1.5% for 1/3 or less overlap (Figure 13 level counting only the single, most severe injury to each
and Table 4.) Overlaps in the 1/3 to 2/3 range, also, result individual body region which make up the general body
in fairly low injury rates for these configurations. At region group (head, chest, arms and legs.) The risk of
overlaps over 2/3, left offset impacts produce higher injury to a body region is the sum of injuries at the
MAIS 3 injury rates, increasing to about 3.6 percent for specific AIS level divided by all drivers in the crash
over 2/3 overlap while right offset impacts at larger mode. As shown in Figures 14 and 15 and Table 5, legs

Stucki, 8
 5% 50%

Proportion of Leg Injuries

4% 40%
Injury Risk %

3% 30%

2% 20%

1% 10%

0% 0%
Arm Thorax Head/Face Legs Knee Tibia Pelvis Ankle Thigh
Body Region
Left Obl./Offset All Frontal
Left Obl./Offset All Frontal
Figure 17. Proportion of AIS 3 Leg Injuries
Figure 14. AIS 2 Body Region Injury Risk, Left
have a higher risk of AIS 2 and AIS 3 injury in left
Offset and All Frontal Impacts
offset impacts than all frontals with other body regions
having similar rates in both crash modes. Thus, reducing
1.2%
leg injuries should be a prime objective in addressing left
1.0% offset crashes.
Injury Risk %

0.8% For drivers with air bags, AIS 2 leg injuries are
0.6% separated into specific injury location in Figure 16 and
Table 6. For these injuries, the ankle is most frequently
0.4% injured followed by the knee and tibia, regardless of
whether the impact is left offset or all frontals. Together
0.2%
these regions make up almost 90 percent of all AIS 2 leg
0.0% injuries in left offset crashes. The tibia and femur
Arm Thorax Head/Face Legs dominate the severe leg injuries, with about 45 percent of
Body Region
leg injuries to the tibia and almost 43 percent to the
Left Obl./Offset All Frontal
femur, again, regardless of impact mode (Figure 17.)
About E of moderate and serious leg injuries are
fractures. Thus, a test surrogate should have appropriate
Figure 15. AIS 3 Body Region Injury Risk, Left
hardware and be instrumented to assess AIS 2 ankle and
Offset and All Frontal Impacts
knee injuries and AIS 3 tibia and femur injuries with the
type of lesions listed.

40% Injury Assessment by Size

Proportion of Leg Injuries

30% The current frontal impact protection standard

(FMVSS No. 208) assesses vehicle performance with a
20% single size, 50th percentile, male dummy. An assessment
of the crash environment by driver size was conducted to
indicate whether there is a need to incorporate additional
10%
size dummies in future frontal test procedures. Drivers
were grouped into three categories based on height of test
0% dummies representing the 5th percentile female, 50th
Knee Tibia Pelvis Ankle Thigh
percentile male and 95th percentile male. The heights for
Left Obl./Offset All Frontal
each category are:
Figure 16. Proportion of AIS 2 Leg Injuries
• 5th % group - less than 164 cm

Stucki, 9
 60%
3%
50%

40%
2%
30%

20%
1%
10%

0%
5th% 50th% 95th%
0%
Left Oblique/Offset All Frontals 5th% 50th% 95th%
Left Oblique/Offset All Frontals
Figure 18. Crash Exposed Drivers by Size
Figure 19. MAIS 3 Injury Risk by Size
• 50th % group - 164 to 180 cm
• 95th % group - over 180 cm
0.7% 7
The distribution of drivers with air bags grouped by 27
0.6% 12
height is shown in Figure 18 and Table 7 for left offset
impacts and for all frontal impacts. The 50th % grouping 0.5%
represents about 58 percent of all involved drivers in left
0.4%
offset and all frontal impacts, the 5th % about 24 percent
and the 95th about 18 percent. 0.3% 40
7
0.2%
Figures 19 and 20 show MAIS 3 injury and fatality 2
risk, respectively, by the three size groupings. Previous 0.1%
analyses have shown that smaller drivers, generally
0.0%
females, tend to have lower severity crashes and thus may 5th% 50th% 95th%
have lower injury risk as a result. Because of the limited Left Oblique/Offset All Frontals
observations, as shown in Table 8, for assessing
injury/fatality risk, no attempt is made to consider severity Figure 20. Fatality Risk by Size
(deltaV.)
Reporting System (FARS, 1995) for fatalities and the
The fatality risk is based on limited numbers with NHTSA Final Regulatory Evaluation (FRE) on Air Bag
“raw” counts shown above each bar in Figure 20. The Depowering for MAIS 2 injuries.
5th percentile generally shows a lower injury/fatality rate
for the left offset crash modes; however, this group Table 9 presents the work sheet for computing the
experiences a higher MAIS 2 injury rate than the 50th estimates. The annual exposure by size group is from
percentile group. The 95th percentile shows highest risk 1988-1996 NASS for an average year which is multiplied
for MAIS 2 and MAIS 3 injuries and fatalities. The by the injury risk (Table 8) to give an estimate of annual
higher injury risk for the 95th % grouping is due, at least injuries/fatalities. Since the total driver fatalities based on
in part, to the higher risk of leg and head injury (Table 8.) NASS appears low, the estimates are adjusted by the
For all frontal impacts the 5th percentile exhibits similar computed number of driver fatalities in non-rollover,
injury risk as the other size groupings; however for frontal crashes with an all air bag fleet based on 1995
fatalities the risk is much lower, but is based on limited FARS as shown in the table. The fatalities for left offset
observations. crash modes are adjusted to be consistent with the
proportion of all frontal impact fatalities for these modes,
The number of injuries and fatalities which might be computed earlier. Likewise, the number of injuries are
expected annually for each size group with an all air bag adjusted based on the NHTSA Final Regulatory
fleet is estimated below for left offset impact modes. Evaluation (FRE) on Air Bag Depowering for MAIS 2
The estimates are based on current year Fatality Analysis injuries, which predicted 120,000 annually. The same

Stucki, 10
adjustment factor for MAIS 2 injuries is also applied to As shown in Figure 21 and Table 10 for speeds of 48
MAIS 3 injuries. As shown, based on the assumed size kph and less the MAIS 2 injury and fatality rates are
groups, i.e., based on the division of sizes by the mid- higher for the left offset crash modes (8.6 and 0.2
point of the difference in height between successive percent) compared to full barrier modes (5.2 and 0.0
dummy sizes, the 50th percentile group is the most percent.) No fatalities occurred at 48 kph and less in full
populous and thus experiences the most injuries/fatalities. barrier type impacts. For MAIS 3 injuries the full
However, the 95th percentile group, although the least barrier risk is actually higher than for the left offset for
populous, experiences substantially more fatalities and impacts at 48 kph and less (2.3 and 2.2 percent,
MAIS 3 injuries than the 5th percentile group and has respectively.)
the highest injury and fatality risks.
As shown in the previous sections, arm injuries in the
Based on the assumptions made and the limited data full barrier impact modes occur at a much higher rate than
on severe and fatal injuries, the 95th percentile group in impacts without air bags or even left offset impacts
experiences a substantial number of injuries/fatalities and with air bags. Arm fractures and other less severe
should be considered as an additional test surrogate to be laceration and contusion type injuries occur quite
used in a proposed left offset test procedure. The 5th frequently from aggressively deploying air bags. Another
percentile group, although experiencing less injuries, still NHTSA research program is vigorously addressing
has substantial numbers of moderate and severe to fatal problems associated with aggressive
injuries in this impact mode. A different definition of
5th, 50th and 95th percentile groupings, perhaps based on
statistical groupings or more narrow height ranges for the 10%
50th group would possibly lead to a different conclusion.
8%
Benefits Assessment for an Improved Test Procedure
6%
A preliminary method for estimating injury and
fatality reductions for a left offset test procedure is 4%
proposed. This method assumes that for under 48 kph
(the current FMVSS No. 208 test speed) the 2%
injury/fatality rates for drivers with air bags in left offset
crash modes will be reduced to levels similar to those for 0%
drivers in full barrier modes. In other words, drivers with MAIS>=2 MAIS>=3 Fatal
air bags in the proposed impact modes to be addressed by Full Barrier Left Obl./Off.
a left offset test would experience the same injury risk as
drivers with air bags in impact modes addressed by the Figure 21. Driver Injury/Fatality Risk for All
current requirement. Injury, 48 KPH

Estimated Annual Driver Injury/Fatalities in Left Offset Impacts With Air Bags

DRIVER SIZE GROUP

INJURY 5th Percent 50th Percent 95th

LEVEL Total Percent

MAIS 2 45,924 11,796 19,819 14,309

MAIS 3 11,520 1,261 7,307 2,953

Fatalities 4,243 224 3,004 1,015

deployment of air bags on out-of-position occupants and and not on resolving problems with aggressive air bag
arm injuries. This research program supported by this deployment. Since arm injuries are not the main concern
crash analysis is focused on improved frontal protection of an alternative frontal test the analysis was repeated to

Stucki, 11
compare injury risk when arm injuries are removed numerical estimate is made for fatality reduction, except
(Figure 22, Table 10.) Relative to the injury risk in full to say there appears to be potential for substantial
barrier type impacts this method shows a much higher reductions. Also, for drivers with air bags subjected to a
risk for left offset impacts (1.66 percent for MAIS 3 and left offset test procedure, an increase in MAIS 3 injuries
7.86 percent for MAIS 2) than for full barrier (0.69 in impacts is not expected and, thus, no change is
percent for MAIS 3 and 3.15 percent for MAIS 2.) predicted. The number of driver injuries and fatalities
expected in left offset impacts with an all air bag fleet is
The percent decrease in injury risk for the left offset shown in Table 10 and repeated below. The reduction in
impacts compared to the full barrier impacts is shown in injuries/fatalities is then the percent change applied to
the following table. Since the estimate of fatality risk in these expected injuries and fatalities.
impacts of 48 kph and less are based on few numbers and
a total reduction in fatalities is unreasonable no An analysis was also conducted to estimate the
number of leg injuries which might be eliminated by a
left offset test. Again, it is assumed that the benefit of
adopting a left offset test procedure is an injury rate
reduction for drivers with air bags below 48 kph to the
injury rate experienced in full barrier type crashes. Table
11 shows the risk of receiving a leg injury of AIS 2 and
of AIS 3 level for left offset and full barrier type impact
modes. Drivers with air bags in full barrier type impacts
8% below 48 kph have a lower risk of leg injury than those in
left offset impacts by the percentages shown in the Table
6% 11 worksheet. It is assumed that the number of injuries
in NASS are below the annual nationwide count by the
same factor as that used previously to estimate occupants.
4%
This factor is then applied to the NASS injury counts and
the proportion of leg injuries computed to yield an
2% estimate of leg injuries expected nationwide in one year.
This annual estimate is then multiplied by the reduction in
0% injury rate to give a rough approximation of number of
MAIS>=2 MAIS>=3 Fatal leg injuries, AIS 2 and AIS 3, which might be
Full Barrier Left Obl./Off. eliminated with a left offset test procedure, as shown
below. Based on this computation, over 11,400 AIS 2
Figure 22. Injury/Fatality Risk Without Arm and over 2,200 AIS 3 leg injuries could be saved.
Injuries, 48 KPH

Injury/Fatality Risk Difference: Left Offset Compared to Full Barrier, Under 48 KPH

Percent Change Number in Left Offset Reductions for Offset Test Procedure
All MAIS Arms Excluded All MAIS Arms Excluded All MAIS Arms Excluded
MAIS 2 - 40.0% -59.9% 40568 34611 16,227 20,732
MAIS 3 0% (+7.3%) -58.7% 10889 8689 0 5,100
Fatality Not Computed* 2664 2664 Not Computed*
AIS 2 AIS 3
Annual Leg Injuries 24,169 4,834
Reduction for Left Offset Test (Table 11)
Percent 47.2% 45.8%
Annual Leg Injuries 11,416 2,215
*
There were 13 fatalities (unweighted) to drivers in left offset impacts under 48 kmph with no fatalities in full
barrier type impacts.
REFERENCES

Stucki, 12
1. Ragland, Carl L., “Offset Test Procedure Development
and Comparison,” Sixteenth International Technical
Conference on Enhanced Safety of Vehicles, Windsor, 4. Stucki, Sheldon L., Ragland, Carl, Hennessey,
Canada, June, 1998. Barbara, and Hollowell, Tom, "NHTSA's Improved
Frontal Protection Research Program," 1995 SAE
2. Hollowell, William T., and Hitchcock, Ralph J., "The International Congress and Exposition, SAE 950497,
National Highway Traffic Safety Administration Program Detroit, Michigan, February 27-March 2, 1995.
to Improve Frontal Crash Protection," 26th International
Symposium on Automotive Technology and Automation, 5. Stucki, Sheldon L., Ragland, Carl, Hennessey,
Aachen, Germany, September 1993. Barbara, and Hollowell, Tom, "NHTSA's Improved
Frontal Protection Research Program," Fifteenth
3. Hollowell, William T., and Hitchcock, Ralph J., International Technical Conference on Enhanced Safety
"Improved Frontal Impact Crash Test Data," Fourteenth of Vehicles, Melbourne, Australia, May, 1996.
International Technical Conference on Enhanced Safety
of Vehicles, Munich, Germany. 6. Park, Brian T., Morgan, Richard M., Hackney, James
R., Lee, John, and Stucki, Sheldon L., “Frontal Offset
Crash Test Study using 50th Percentile Male and 5th
Percentile Female Dummies,” Sixteenth International
Technical Conference on Enhanced Safety of Vehicles,
Windsor, Canada, June, 1998.

TABLE 1 - Injury Risk by Restraint Condition, Drivers in Frontal

Impacts, 1988-1996 NASS
MAIS
RESTRAINT 2-6 3-6 Fatal Total
All Air Bag # 78845 24979 3488 1142704
(Belted+Unbelted) Row% 6.90% 2.19% 0.31%
Raw# 614 294 64 2555
Non-Air Bag # 934506 287563 55423 9428597
(Belted+Unbelted) Row% 9.91% 3.05% 0.59%
Raw# 6732 3031 1078 23730
# 52082 12355 1610 953887
Air Bag and Belts Row% 5.46% 1.30% 0.17%
Raw# 347 149 29 1884
# 397403 101323 12230 6107706
Belts Row% 6.51% 1.66% 0.20%
Raw# 2391 897 224 12568
# 24095 10612 1756 156198
Air Bags and No Belts Row% 15.43% 6.79% 1.12%
Raw# 246 132 32 598
# 480798 166496 38787 2806459
No Restraint Row% 17.13% 5.93% 1.38%
Raw# 3919 1892 752 9728

Stucki, 13
TABLE 2 - Injuries by Body Region for Drivers With and Without Air Bags
Serious and Greater Injuries by Body Region (Most Severe AIS>=3) Occupants
Arms Thorax Head Legs Known AIS
Air Bag Raw 66 110 64 142 2555
Belts "As Used" # 7369 9314 5472 10058 1161513
Risk% 0.63% 0.80% 0.47% 0.87%
Non Air Bag Raw 385 1433 1034 1236 23730
Belts "As Used" # 43300 120297 87340 115450 9566033
Risk% 0.45% 1.26% 0.91% 1.21%
Air Bag Raw 51 42 23 71 1884
Belts # 5840 4070 1979 4236 970219
Risk% 0.60% 0.42% 0.20% 0.44%
Non Air Bag Raw 137 406 265 338 12568
Belts # 14647 48205 23219 31261 6220408
Risk% 0.24% 0.77% 0.37% 0.50%
Air Bag Raw 13 57 40 64 598
No Belts # 1474 3327 3438 5423 158090
Risk% 0.93% 2.10% 2.17% 3.43%
No Restraint Raw 226 888 701 791 9728
# 27812 59930 59948 75979 2825620
Risk% 0.98% 2.12% 2.12% 2.69%
Moderate and Greater Injuries by Body Region (Most Severe AIS>=2)
Air Bag Raw 172 183 199 321 2555
Belts "As Used" # 21781 18701 20077 38286 1161513
Risk% 1.88% 1.61% 1.73% 3.30%
Non Air Bag Raw 1246 2578 3612 2699 23730
Belts "As Used" # 177592 296200 448255 309813 9566033
Risk% 1.86% 3.10% 4.69% 3.24%
Air Bag Raw 117 89 84 188 1884
Belts # 16371 9033 11603 27537 970219
Risk% 1.69% 0.93% 1.20% 2.84%
Non Air Bag Raw 459 976 1046 939 12568
Belts # 77453 148062 138056 125161 6220408
Risk% 1.25% 2.38% 2.22% 2.01%
Air Bag Raw 49 81 111 120 598
No Belts # 4875 7685 8199 10168 158090
Risk% 3.08% 4.86% 5.19% 6.43%
No Restraint Raw 721 1395 2385 1557 9728
# 90587 124360 283675 161361 2825621
Risk% 3.21% 4.40% 10.04% 5.71%

Stucki, 14
TABLE 3 - Injury/Fatality Rates by Test Condition
Left Obl./ Right Obl./ L&R Obl./
Full Barrier Left Off. Right Off. R&L Off. Other TOTAL
MAIS>=2 # 15100 32841 20449 10445 10 78845
Raw# 127 270 169 47 1 614
Row% 19.2% 41.7% 25.9% 13.2% 0.01%
Risk 6.79% 7.61% 4.58% 10.25% 0.33% 6.54%
Annual# 18175 31710 19882 11095 16 80879
Adjusted Injuries to FRE 26967 47048 29499 16462 24 120000
MAIS>=3 # 8367 9104 5870 1628 10 24979
Raw# 71 130 75 17 1 294
Row% 33.5% 36.4% 23.5% 6.5% 0.04%
Risk 3.76% 2.11% 1.31% 1.60% 0.33% 2.07%
Annual# 10071 8791 5707 1729 16 26315
Adjusted Injuries to FRE 14942 13042 8468 2566 24 39043
Fatality # 548 1875 506 548 10 3487
Raw# 10 38 10 5 1 64
Row% 15.7% 53.8% 14.5% 15.7% 0.29%
Risk 0.25% 0.43% 0.11% 0.54% 0.33% 0.29%
Annual# 660 1810 492 582 16 3561
Adjusted to FARS Annual 1546 4243 1153 1364 39 8345
Drivers with # 222419 431302 446804 101858 2993 1205376
Air Bags Raw# 457 1067 895 265 5 2689
Exposed Annual # 267716 416450 434416 108198 4921 1231701
Drivers Row% 21.7% 33.8% 35.3% 8.8% 0.40%

Stucki, 15
TABLE 4 - Injury Risk by Overlap Proportion for Left and Right Offset Impacts
Left Offset
0-1/3 >1/3-2/3 >2/3 Total
MAIS>=2 # 8017 6553 18273 32843
Raw# 68 76 126 270
Row% 24.4% 20.0% 55.6%
Risk 5.01% 5.34% 14.23% 7.99%
Annual# 11484 9387 26176 47048
MAIS>=3 # 2335 2136 4634 9105
Raw# 37 39 54 130
Row% 25.6% 23.5% 50.9%
Risk 1.46% 1.74% 3.61% 2.21%
Annual# 3345 3060 6638 13042
Total # 160134 122614 128445 411193
Raw# 315 279 418 1012
Row% 38.9% 29.8% 31.2%
Right Offset
MAIS>=2 # 5353 6650 8445 20448
Raw# 48 42 79 169
Row% 26.2% 32.5% 41.3%
Risk 3.66% 4.96% 5.91% 4.83%
Annual# 7722 9594 12183 29499
MAIS>=3 # 2152 1137 2580 5869
Raw# 20 15 40 75
Row% 36.7% 19.4% 44.0%
Risk 1.47% 0.85% 1.80% 1.39%
Annual# 3105 1641 3723 8468
Total # 146108 133960 142960 423028
Raw# 282 214 352 848
Row% 34.5% 31.7% 33.8%

Stucki, 16
TABLE 5 - Injury by Body Region for Left Offset and All Frontal
Left Offset TOTAL
Body Region AIS>=3 AIS>=2 AIS>=3 AIS>=2
# 2126 6669 6656 20929
Arm Raw# 32 75 66 173
Risk% 0.5% 1.6% 0.6% 1.8%
# 3220 8343 9329 18555
Thorax Raw# 51 84 111 184
Risk% 0.8% 2.0% 0.8% 1.6%
# 2077 7939 5472 20740
Head/Face Raw# 27 84 64 201
Risk% 0.5% 1.9% 0.5% 1.8%
# 4227 18423 9761 37269
Legs Raw# 69 155 142 321
Risk% 1.0% 4.5% 0.9% 3.3%
Total
Injuries # 11650 41374 31218 97493
Raw# 179 398 383 879
# 411192 411192 1142603 1142603
Total Drivers Raw# 1012 1012 2554 2554

TABLE 6 - Leg Injuries for Left Offset and All Frontal

Left Offset
Knee Tibia Pelvis Ankle Thigh Whole Total Drivers
AIS>=3 Raw# 1 27 6 2 42 2 80 1012
# 0 2153 247 268 2051 105 4824 411192
Row% 0.0% 44.6% 5.1% 5.6% 42.5% 2.2%
AIS>=2 Raw# 34 46 25 76 44 7 232 1012
# 7073 7189 1745 9220 2638 316 28181 411192
Row% 25.1% 25.5% 6.2% 32.7% 9.4% 1.1%
All Frontals
Knee Tibia Pelvis Ankle Thigh Whole Total Drivers
AIS>=3 Raw# 2 53 18 5 80 2 160 2554
# 26 4579 765 323 4706 105 10504 1142603
Row% 0.2% 43.6% 7.3% 3.1% 44.8% 1.0%
AIS>=2 Raw# 66 90 54 157 82 9 458 2554
# 11650 11106 3536 19034 5293 745 51364 1142603
Row% 22.7% 21.6% 6.9% 37.1% 10.3% 1.5%

Stucki, 17
 TABLE 7 - Driver Exposure by Size Groups
Test Driver Size
Condition 5th% 50th% 95th% Total
Raw# 194 512 169 875
Left Obl/Off # 86118 208241 66045 360404
Row% 23.9% 57.8% 18.3%
Raw# 502 1231 426 2159
All Frontals # 238928 571491 176787 987206
Row% 24.2% 57.9% 17.9%

TABLE 8 - Driver Injury by Size Group

Test Driver Size
Condition MAIS 5th% 50th% 95th% Total
Raw# 48 135 47 230
# 7111 11948 8626 27685
2-6 Row% 25.7% 43.2% 31.2%
Risk% 8.26% 5.74% 13.06% 7.68%
Raw# 21 65 21 107
Left Obl/Off # 760 4405 1780 6945
3-6 Row% 10.9% 63.4% 25.6%
Risk% 0.88% 2.12% 2.70% 1.93%
Raw# 2 27 7 36
# 96 1287 435 1818
Fatal Row% 5.3% 70.8% 23.9%
Risk% 0.11% 0.62% 0.66% 0.50%
Raw# 116 290 96 502
# 16821 37941 12236 66998
2-6 Row% 25.1% 56.6% 18.3%
Risk% 7.04% 6.64% 6.92% 6.79%
Raw# 49 145 48 242
All Frontals # 4176 11416 3948 19540
3-6 Row% 21.4% 58.4% 20.2%
Risk% 1.75% 2.00% 2.23% 1.98%
Raw# 7 40 12 59
# 557 1567 1008 3132
Fatal Row% 17.8% 50.0% 32.2%
Risk% 0.23% 0.27% 0.57% 0.32%

TABLE 9 - Worksheet for Estimating Annual Injuries/Fatalities

Stucki, 18
ANNUAL EXPOSURE
Total 5th% 50th% 95th%
Left Offset 416450 99510 240624 76316 88-95 NASS
All Frontal 1231701 298102 713029 220571 /8 Years
ANNUAL INJURIES/FATALITIES
Left Offset
MAIS>=2 31990 8217 13806 9967
MAIS>=3 8025 878 5090 2057
Fatalities 2101 111 1487 503
All Frontal
MAIS>=2 83591 20987 47338 15266
MAIS>=3 24379 5210 14243 4926
Fatalities 3908 695 1955 1258
FARS 1995 Fatalities 8345 1995 Driver Fatalities in Light Vehicles
FRE on DePowering 1995 FARS driver fatalities 31786
MAIS>=2 Injuries 120000 - Non-rollover 23995
- Rollover 7791
Adjustment Factors Total non-air bag fatalities 31952
All Frontal Fatals 2.14 - Non-rollover 24120
MAIS>-2 1.44 - Rollover 7832
MAIS>-3 1.44 Total air bag effectiveness 11%
Left Obl./Off. Fatals* 2.02 - Non-rollover 14.6%
*
Adjusted to agree with Table 3 - Rollover 0
estimate of fatalities in left Total fatalities with air bags 28437
offset crashes - Non-rollover 20606
- Rollover 7832
% Non-rollover fatalities in frontals 54%
1995 fatalities in frontals 11127
Frontal air bag effectiveness 25%
Frontal fatalities with air bags 8345

ANNUAL INJURIES/FATALITIES (Adjusted)

Left Offset
Total 5th% 50th%
95th%
MAIS>=2 45924 11796 19819 14309
MAIS>=3 11520 1261 7307 2953
Fatalities 4243 224 3004 1015
All Frontal
MAIS>=2 120000 30128 67956 21916
MAIS>=3 34998 7480 20447 7071
Fatalities 8345 1484 4175 2686

Stucki, 19
TABLE 10 - Comparison of Injury/Fatality Risk, Left Offset vs. Full Barrier
MAIS Without Arm Injury MAIS All Body Regions
DeltaV Known DeltaV Known
TEST <=30 >30 Unk. Total Total <=30 >30 Unk. Total Total
Raw# 0 6 4 6 10 0 6 4 6 10
Fatal # 0 410 138 410 548 0 410 138 410 548
Risk% 0.00% 20.60% 0.17% 0.33% 0.26% 0.00% 20.60% 0.17% 0.33% 0.26%
Raw# 15 23 24 38 62 20 24 27 44 71
Full Barrier MAIS>=3 # 838 969 4264 1807 6071 2840 1052 4475 3892 8367
Risk% 0.69% 48.69% 5.13% 1.46% 2.93% 2.32% 52.86% 5.38% 3.14% 4.04%
Raw# 37 31 42 68 110 46 33 48 79 127
MAIS>=2 # 3852 1455 6250 5307 11557 6290 1979 6831 8269 15100
Risk% 3.15% 73.12% 7.51% 4.27% 5.57% 5.15% 99.45% 8.21% 6.66% 7.28%
Raw# 223 34 180 257 437 223 34 180 257 437
Total # 122152 1990 83186 124142 207328 122152 1990 83186 124142 207328
Row% 98.40% 1.60% 40.12% 59.88% 98.40% 1.60% 40.12% 59.88%
Raw# 13 8 17 21 38 13 8 17 21 38
# 447 265 1164 712 1876 447 265 1164 712 1876
Fatal Annual# 2664 1579 4243 2664 1579 4243
Risk% 0.20% 7.93% 0.65% 0.31% 0.46% 0.20% 7.93% 0.65% 0.31% 0.46%
Raw# 55 18 38 73 111 66 20 44 86 130
Left Offset MAIS>=3 # 3779 902 2831 4681 7512 4930 975 3198 5905 9103
Annual# 8689 2074 10763 10889 2153 13042
Risk% 1.66% 27.01% 1.57% 2.03% 1.83% 2.17% 29.19% 1.77% 2.56% 2.21%
Raw# 134 31 69 165 234 154 32 84 186 270
MAIS>=2 # 17878 3091 7368 20969 28337 19526 3119 10196 22645 32841
Annual# 34611 5984 40596 40568 6480 47048
Risk% 7.86% 92.54% 4.09% 9.08% 6.89% 8.58% 93.38% 5.66% 9.81% 7.99%
Raw# 611 37 364 648 1012 611 37 364 648 1012
Total # 227573 3340 180280 230913 411193 227573 3340 180280 230913 411193

Stucki, 20
 Row% 98.55% 1.45% 43.84% 56.16% 98.55% 1.45%

TABLE 11 - Leg Injuries in Full Barrier and Left Offset Crashes

MAIS DeltaV Known
TEST Level <=30 >30 Unk. Total Total
Raw# 10 15 14 25 39
MAIS>=3 # 577 758 1082 1335 2417
Risk% 0.47% 38.09% 1.30% 1.08% 1.17%
Raw# 30 24 23 54 77
Full Barrier MAIS>=2 # 3427 1138 3214 4565 7779
Risk% 2.81% 57.19% 3.86% 3.68% 3.75%
Raw# 223 34 180 257 437
Total # 122152 1990 83186 124142 207328
Row% 98.40% 1.60% 40.12% 59.88%
Raw# 36 11 22 47 69
# 1984 478 1765 2462 4227
MAIS>=3 Row% 79.82% 20.18% 31.43% 68.57%
Annual# 4834 1222 6056
Risk% 0.87% 14.31% 0.98% 1.07% 1.03%
Raw# 89 21 45 110 155
# 12100 2590 3733 14690 18423
Left Offset MAIS>=2 Row% 91.57% 8.43% 28.99% 71.01%
Annual# 24169 2224 26393
Risk% 5.32% 77.54% 2.07% 6.36% 4.48%
Raw# 611 37 364 648 1012
Total # 227573 3340 180280 230913 411193
Row% 98.55% 1.45% 43.84% 56.16%
Percent of drivers with AIS>=2 leg injury of all drivers with MAIS>=2 injury 56.1%
Percent of drivers with AIS>=3 leg injury of all drivers with MAIS>=3 injury 46.4%
Annual drivers with AIS>=2 leg injuries = 56.1%*(all drivers with MAIS>=2) 26393
Annual drivers with AIS>=3 leg injuries = 46.4%*(all drivers with MAIS>=3) 6056

Driver Leg Injuries in Left Offset Impacts

MAIS or AIS>=2 MAIS or AIS>=3
<=30 >30 <=30 >30
Annual Leg Injuries 24169 2224 4834 1222
Reduction Over Full Barrier (Table 10)
Percent 47.2% 45.8%
Annual Leg Injuries 11416 2215

Stucki, 21