Insurance Institute for Highway Safety

Status Report

Insurance Institute for Highway Safety Highway Loss Data Institute

On foot,
at risk
Study highlights rising pedestrian
deaths, points toward solutions

Vol. 53, No. 3 May 8, 2018

4 Subaru crash avoidance system
cuts pedestrian crashes
4 Michigan sees higher insurance
payouts for motorcyclist injuries


he March crash of an Uber vehicle that killed a
woman in Tempe, Arizona, was unusual for involving
a self-driving vehicle. But in other ways, it was typical
of fatal pedestrian crashes: an SUV traveling on an urban
arterial road struck a person crossing midblock in the dark.
Pedestrian deaths have jumped 46 percent since reaching their lowest point in 2009, as pedestrian crashes have
become both deadlier and more frequent. The increase has
been mostly in urban or suburban areas, at nonintersections, on arterials a busy roads designed mainly to funnel
vehicle traffic toward freeways a and in the dark, a new
IIHS study shows. Crashes were increasingly likely to involve SUVs and high-horsepower vehicles.
aUnderstanding where, when and how these additional
pedestrian crashes are happening can point the way to solutions,a says IIHS President David Harkey. aThis analysis
tells us that improvements in road design, vehicle design
and lighting and speed limit enforcement all have a role to
play in addressing the issue.a
A total of 5,987 pedestrians were killed in crashes in
2016, accounting for 16 percent of all crash fatalities. The
number of pedestrians killed each year has declined 20 percent since 1975, but the 2016 toll was the highest since 1990.
For the new study, IIHS researchers looked at pedestrian
crash trends during 2009a16 to pinpoint the circumstances
under which the largest increases occurred. Using federal
fatality data and crash numbers, the researchers looked at
roadway, environmental, personal and vehicle factors to see
Pedestrian crashes have become both deadlier and more
frequent. The increase has been mostly in urban or suburban areas, away from intersections, on busy main roads
and in the dark. Crashes are increasingly likely to involve
SUVs and high-horsepower vehicles.
how they changed over the study period. They also looked
at changes in the number of pedestrian deaths relative to
the number of pedestrians involved in crashes.
The researchers found that not only did pedestrian crashes
increase, they also became deadlier. Deaths per 100 crash
involvements increased 29 percent from 2010, when they
reached their lowest point, to 2015, the most recent year that
data on all crashes, including nonfatal ones, were available.
From 2009 to 2016, the largest increases in pedestrian
deaths occurred under the circumstances that historically
have seen the highest numbers of pedestrian fatalities. Pedestrian deaths increased 54 percent in urban areas, which
include both cities and what most people consider suburbs.
They also increased 67 percent on arterials, 50 percent at
nonintersections and 56 percent in the dark.
Although pedestrian crashes most frequently involved
cars, fatal single-vehicle crashes involving SUVs increased 81
percent, more than any other type of vehicle. The power of
passenger vehicles involved in fatal single-vehicle pedestrian
crashes, as measured by the ratio of horsepower to weight,
also increased, with larger increases at the top of the scale.


Status Report a Vol. 53, No.3

Pedestrian deaths


By land use


rural +25%


By road type




collectors and local roads +9%


interstates and freeways +49%
2009 10 11 12 13 14 15 16


By location


By light condition



2009 10 11 12 13 14 15 16

intersections +35%


daylight +20%


2009 10 11 12 13 14 15 16

dawn or dusk +27%
2009 10 11 12 13 14 15 16

Among age groups, the largest increase in pedestrian fatalities
per population was for people ages 20a69.

Designing roads for pedestrians

The large increase in pedestrian deaths on arterials isnat surprising.
These roads often have a shortage of convenient and safe crossing
aWhen people are forced to walk long distances to the nearest signalized intersection, they are more likely to choose the riskier option
of sprinting across multiple lanes of traffic,a Harkey says. aCommunities can improve safety by providing more options to safely cross.a
But, Harkey warns, itas not enough to simply paint more crosswalks on the pavement. Midblock crossings need features that alert
drivers to stop, such as pedestrian-activated beacons.
One example is the pedestrian hybrid beacon, which stays dark
until a pedestrian pushes a button, at which point it flashes yellow,
and then moves to solid yellow before activating two solid red lights.
This type of beacon, formerly known as a HAWK, has been shown to
reduce crashes (see Status Report, Sept. 28, 2010, at
Other improvements, such as curb extensions or median crossing islands, can shorten the distance people must walk across or

May 8, 2018


limits is a proven solution. Institute research has shown that automated speed enforcement reduces speed limit violations
and injury crashes (see Status Report, Oct.
1, 2015).

Improving vehicles

A woman crosses at a pedestrian hybrid beacon in Arlington, Virginia. The beacon remains
dark until a pedestrian activates it. First, it
flashes yellow before moving to solid yellow
and then to solid double red.
allow them to traverse just a couple of lanes
and a single direction of traffic at a time.
Adding sidewalks is an obvious way
to reduce the risk to pedestrians walking
along a road.
These elements can be part of broader
reconfigurations known as road diets, in
which the number of travel lanes for vehicle traffic is reduced. In addition to reducing the number of lanes for pedestrians to
cross and sometimes providing room for
bike lanes, road diets have been shown to
lower vehicle speeds.
aGood design should prioritize the safety
of all road users,a Harkey says. aItas possible to improve streets for pedestrians while
still allowing vehicle traffic to get where it
needs to go.a


Status Report a Vol. 53, No.3

Reining in speed increases

Of course, allowing vehicles to get where
they need to go doesnat mean they need to
go quite as fast as people are used to.
Faster speeds make for more frequent and
deadlier crashes. The faster a car is moving,
the less time the driver has to see a pedestrian and slow or stop. Higher impact speeds
also result in more injurious crashes.
Reliable information on vehicle speeds is
not available in fatality data, but IIHS researchers did find that the vehicles involved
in fatal pedestrian crashes, like the overall vehicle fleet, are increasingly powerful.
Previous IIHS research has shown that vehicles with higher horsepower-to-weight
ratios tend to be driven faster and are more
likely to violate posted speed limits (see
Status Report, May 24, 2016).
Despite the dangers of high speeds, the
story of speed limits in recent decades has
been one of continual increases.
In addition to lower speed limits, broader
use of speed cameras to enforce existing

Some risks to pedestrians could be lessened
by making changes to vehicles.
A large majority of pedestrian fatalities occur in the dark, and that number increased much faster than the number of
pedestrians killed in other light conditions.
In 2016, 4,453 pedestrians were killed in
the dark, compared with 1,290 in daylight
and 205 at dawn or dusk.
Although better street lighting may be
needed in some locations, another obvious
solution is better headlights. IIHS has been
working to encourage improvements in
this area through its headlight rating program, launched in 2016. Headlights have
been gradually improving. In the 2016
model year, there were just two models
with available good-rated headlights. So far
for the 2018 model year, there are 26 good
headlight packages.
Vehicles with front crash prevention systems that recognize pedestrians also would
help a particularly if they are designed to
work in low light. A recent HLDI analysis
found that Subaru vehicles equipped with
pedestrian detection had claim rates for pedestrian injuries that were 35 percent lower
than the same vehicles without the system
(see adjacent sidebar).
Finally, vehicle design changes could help
lessen the severity of crashes, especially
when it comes to SUVs. These make up an
increasingly large percentage of registered
vehicles, and previous studies have found
that SUVs, pickups and vans are associated
with higher risks of pedestrian deaths or
severe injuries to pedestrians. Such vehicles
have higher and often more vertical front
ends than cars and are more likely to strike
a pedestrian in the head or chest. Changes
in the front-end design of these vehicles
could help lessen the severity of injuries
when they strike pedestrians (see Status
Report, Dec. 30, 2013).
For a copy of aAn examination of the
increases in pedestrian motor vehicle
crash fatalities during 2009a16a by W. Hu
and J.B. Cicchino, email publications@ n

Subaru crash avoidance system
cuts pedestrian crashes


ront crash prevention systems are
known to reduce crashes with other
vehicles (see Status Report, Jan. 28,
2016, at At least one system is also
preventing vehicles from striking pedestrians, a new HLDI analysis shows.
Subaruas EyeSight performs several functions, including forward collision warning,
automatic emergency braking, adaptive
cruise control, lane departure warning and
lead vehicle start alert. It also includes pedestrian detection, enabling the system to
brake automatically for pedestrians in addition to other vehicles. The system relies on
two cameras mounted to the interior roof
behind the windshield, a set-up that leads
to lower repair costs than other front crash
prevention systems that rely on equipment
embedded in the vehicle exterior.
The new study found that EyeSight cut
the rate of likely pedestrian-related insurance claims by 35 percent.
aThe data clearly show that EyeSight
is eliminating many crashes, including

pedestrian crashes,a says HLDI Senior Vice
President Matt Moore.
To study the systemas effect on pedestrian crashes, analysts looked at bodily injury
liability claims that lacked an associated
claim for vehicle damage. Past HLDI investigations have found that such claims tend
to represent injured pedestrians or cyclists.
They compared the rate of these claims per
insured vehicle year for Subaru vehicles with
EyeSight, compared with the rate for the
same models without the optional system.
The first generation of EyeSight, which
used black and white cameras, was available
in the U.S. on the 2013a14 Legacy and Outback and the 2014a16 Forester. The second
generation, introduced on the Legacy and
Outback in 2015 and on the Forester in 2017,
uses color cameras and has longer and wider
detection ranges and other improvements.
EyeSight was offered for the first time on
the Crosstrek and the Impreza sedan and
hatchback in 2015. Only the second-generation system was offered on these vehicles.

A HLDI analysis shows that Subaru EyeSight
cuts the rate of likely pedestrian-related
insurance claims by 35 percent.
Today, all models except the BRZ are
available with EyeSight.
Looking at the Legacy, Outback, Forester, Crosstrek and Impreza individually,
HLDI found reductions in claim frequency for each of them, though only the results
for the Legacy and Outback were statistically significant.
HLDI also separated out first-generation and second-generation results for the
Legacy, Outback and Forester. The firstgeneration system reduced claim frequency
33 percent, while the second-generation
system lowered it 41 percent.
aSubaru has taken a good system and
made it even better,a Moore says. aItas great
to see the company moving quickly to
deploy the technology through its fleet.a
For a copy of the HLDI Bulletin Vol. 34,
No. 39, email n

May 8, 2018


Insurance payouts still rising for motorcyclist
injuries under Michiganas weak helmet law


tas springtime in Michigan, and that
means motorcyclists will be pulling
their bikes out of storage and, in many
cases, hitting the road without a helmet.
Six years after the state weakened its helmet
use law to exempt most riders, a new HLDI
analysis indicates that the average insurance payment for injuries to motorcyclists
in crashes has risen by 40 percent, compared with losses in nearby control states.
May marks the start of the seventh
riding season in Michigan since lawmakers relaxed the motorcycle helmet law to
cover only riders younger than 21. Motorcyclists 21 and older may ride without a
helmet if they have either passed a motorcycle safety course or have held the motorcycle endorsement on their driveras license
for at least two years. In addition, riders
who choose not to wear helmets must have
at least $20,000 in medical payment coverage and higher coverage for any passengers
who ride unhelmeted, too. More motorcyclists are opting for the higher policy limits
since the law change, HLDI has found.


Status Report a Vol. 53, No.3

This is HLDIas third look at the effects
of Michiganas partial helmet law repeal. A
2013 HLDI analysis found that the average
insurance payment on a motorcycle injury
claim rose 22 percent in Michigan after the
helmet law change took effect (see Status
Report, May 30, 2013, at The analysis controlled for policy limits to account
for the new medical payment insurance
requirement. HLDI updated the study in
2016 to add three more years of loss data
and found a 37 percent increase in insurance losses. The latest study adds a fifth
year of data to cover the 2010a16 May-toSeptember riding seasons.
HLDI examines motorcycle insurance
loss data under collision and medical payment, or MedPay, coverages. Motorcycle
collision coverage insures against physical
damage to a motorcycle in a crash when the
rider is at fault. MedPay covers injuries sustained by the motorcycle operator.
Insurance losses are measured as claim
frequency, claim severity and overall losses.
Claim frequency is the number of claims

for a group of vehicles divided by the exposure for that group, expressed in the study
as claims per 1,000 insured vehicle years.
An insured vehicle year is one vehicle insured for one year, two vehicles insured for
six months each. Claim severity is the average loss payment per claim.
For all three analyses, Illinois, Indiana,
Ohio and Wisconsin were used as control states because their laws on helmet use
didnat change during the period. Analysts
controlled for motorcycle age and class,
rider demographic factors, geographic factors and weather. They also controlled for insurance policy limits for MedPay coverage.
A separate analysis that didnat take into
account policy limits found that MedPay
claim severity was 68 percent higher in
Michigan after the law change, compared
with the control states.
aWith each year, the evidence against
Michiganas weakened motorcycle helmet
use law continues to mount,a says Matt
Moore, senior vice president of HLDI. aIf
lawmakers in Lansing are committed to the

Estimated increase in medical
payment claim severity after
Michigan helmet law change
Michigan vs. control states, 2010a16





without policy limits

These motorcyclists traveling I-75 near Gaylord, Michigan, opted for the protection
of helmets. Effective April 2012, Michigan implemented a partial helmet law
that requires riders younger than 21 to wear helmets but makes helmets
optional for riders 21 and older who meet certain criteria.

Wolverine stateas aToward Zero Deathsa goal,
requiring all motorcyclists to wear helmets
is one proven way to save lives.a
HLDI data donat include information on
the type of injury or where a crash occurred.
In this analysis, Michigan crashes are crashes of motorcycles insured and garaged in the
state. Likewise, the control-state crashes are

only crashes of motorcycles insured and garaged in those states. There also is no way to
know how many of the claims involved unhelmeted motorcyclists.
Weakening Michiganas helmet law also
has been associated with increases in the
number of head injuries among hospitalized trauma patients and the proportion of

with policy limits

Michigan motorcyclists 21 and older
who ride bareheaded must carry at least
$20,000 in medical payment coverage. The
average payout under MedPay rose 68 percent after the law change, compared with
the control states. Adjusting for the policy
limits, the average payout rose 40 percent.
injured riders with skull fractures, a 2016
study by IIHS and the University of Michigan found (see Status Report, Sept. 1, 2016).
A separate study published in The American Journal of Surgery in 2016 found that
the average acute care cost of unhelmeted
riders at a single Michigan trauma center
was nearly $28,000, 32 percent higher than
for helmeted riders. What is more, the Spectrum Health Butterworth Hospital study
found that 10 percent of riders involved in
a crash who werenat wearing helmets died,
compared with 3 percent of riders involved
in a crash who wore helmets.
Michigan is one of 28 states that have
helmet laws covering only some riders, usually those under 18. Illinois, Iowa and New
Hampshire have no helmet requirements.
Only 19 states and the District of Columbia
require helmets for all motorcyclists.
The National Highway Traffic Safety Administration estimates that helmets cut the
risk of a motorcycle fatality by 37 percent.
For a copy of the HLDI Bulletin Vol. 34,
No. 36, email n

May 8, 2018


Insurance Institute for Highway Safety
Highway Loss Data Institute

Status Report
Study of pedestrian fatalities points
to ways to stem increase42

IIHS is an independent, nonprofit scientific and educational organization dedicated to reducing the losses a deaths, injuries and
property damage a from motor vehicle crashes.

Subaru pedestrian detection
system cuts crashes45

resulting from the ownership and operation of different types of vehicles and by publishing insurance loss results by vehicle make
and model.

Payouts for motorcyclist injuries rise under
Michiganas weak helmet law46

Vol. 53, No. 3
May 8, 2018

Inquiries/print subscriptions:
Copy may be republished with attribution.
Images require permission to use.
Editor: Kim Stewart
Writer: Sarah Karush
Art Director: Steve Ewens
Photographers: Steve Ewens,
Craig Garrett, Dan Purdy, Kim Stewart


This publication is printed on recycled paper.

HLDI shares and supports this mission through scientific studies of insurance data representing the human and economic losses

Both organizations are wholly supported by the following auto insurers and funding associations:
AAA Carolinas
Acceptance Insurance
Alfa Alliance Insurance Corporation
Alfa Insurance
Allstate Insurance Group
American Agricultural Insurance Company
American Family Insurance
American National
Ameriprise Auto & Home
Amica Mutual Insurance Company
Auto Club Enterprises
Auto Club Group
Auto-Owners Insurance
Bitco Insurance Companies
California Casualty Group
Celina Insurance Group
Censtat Casualty Company
Colorado Farm Bureau Mutual Insurance Company
Concord Group Insurance Companies
COUNTRY Financial
CSAA Insurance Group
Desjardins General Insurance Group
ECM Insurance Group
Elephant Insurance Company
EMC Insurance Companies
Erie Insurance Group
Farm Bureau Financial Services
Farm Bureau Insurance of Michigan
Farm Bureau Mutual Insurance Company of Idaho
Farmers Insurance Group
Farmers Mutual of Nebraska
Florida Farm Bureau Insurance Companies
Frankenmuth Insurance
Gainsco Insurance
GEICO Corporation
The General Insurance
Georgia Farm Bureau Mutual Insurance Company
Goodville Mutual Casualty Company
Grange Insurance
Grinnell Mutual
Hallmark Financial Services, Inc.
The Hanover Insurance Group
The Hartford
Haulers Insurance Company, Inc.
Horace Mann Insurance Companies
Imperial Fire & Casualty Insurance Company
Indiana Farm Bureau Insurance
Indiana Farmers Insurance
Infinity Property & Casualty
Kemper Corporation
Kentucky Farm Bureau Mutual Insurance Companies
Liberty Mutual Insurance Company
Louisiana Farm Bureau Mutual Insurance Company

The Main Street America Group
Mercury Insurance Group
MetLife Auto & Home
Mississippi Farm Bureau Casualty Insurance Company
MMG Insurance
Munich Reinsurance America, Inc.
Mutual Benefit Group
Mutual of Enumclaw Insurance Company
New Jersey Manufacturers Insurance Group
NYCM Insurance
Nodak Insurance Company
Norfolk & Dedham Group
North Carolina Farm Bureau Mutual Insurance Company
Northern Neck Insurance Company
Ohio Mutual Insurance Group
Oregon Mutual Insurance Company
Paramount Insurance Company
Pekin Insurance
PEMCO Mutual Insurance Company
Plymouth Rock Assurance
Progressive Insurance
PURE Insurance
Qualitas Insurance Company
Redpoint County Mutual Insurance Company
The Responsive Auto Insurance Company
Rider Insurance
Rockingham Insurance
RSA Canada
Safe Auto Insurance Company
Safeco Insurance
Samsung Fire & Marine Insurance Company
SECURA Insurance
Selective Insurance Company of America
Sentry Insurance
Shelter Insurance Companies
Sompo America
South Carolina Farm Bureau Mutual Insurance Company
Southern Farm Bureau Casualty Insurance Company
State Farm Insurance Companies
Stillwater Insurance Group
Tennessee Farmers Mutual Insurance Company
Texas Farm Bureau Insurance Companies
The Travelers Companies
United Educators
Utica National Insurance Group
Virginia Farm Bureau Mutual Insurance
West Bend Mutual Insurance Company
Western National Insurance Group
Westfield Insurance
American Insurance Association
National Association of Mutual Insurance Companies
Property Casualty Insurers Association of America