public and pedestrian safety
advocates. Given that re-search
takes many years to
be implemented into practice,
we wanted to share the find-ings
with the public in order
to help other City agencies in
their efforts to improve pe-destrian
safety.
We believe that making an
economic case for prevention
can help save lives and im-prove
the state of pedestrian
injury in San Francisco.
This study was approved by
the Committee on Human
Research at the University of
California, San Francisco.
Auto- versus- Pedestrian
( AVP) Collisions have re-mained
high in San Francisco.
The age- adjusted rate was
92/ 100,000 in 2008, com-pared
to 63/ 100,000 in the
United States. San Francisco’s
rate is also 4.5 times greater
than the Healthy People 2010
goal of 19/ 100,000.1 Since
2003, approximately 650 to
750 AVP collisions have re-sulted
in visits to San Fran-cisco
General Hospital
( SFGH) each year. SFGH is
the only Level 1 Trauma Cen-ter
serving San Francisco. As
such, we receive 98% of all
traumas that occur in the
City. As such, we are confi-dent
that most, if not all, AVP
injuries were treated at
SFGH.
Given the public’s concern
for pedestrian safety in San
Francisco and the current
economic state, we decided
to conduct an analysis of the
direct medical cost resulting
from AVP collisions in San
Francisco. The overall goals
of the study were to provide
tangible “ bottom line” data to
stakeholders working on pe-destrian
safety issues
throughout the City and to
provide direct feedback to
the Board of Supervisors
about the cost in their dis-trict.
A more technical report has
been submitted for publica-tion
to the Journal of Trauma.
The target audience for the
present report is the general
Using diagnostic codes as-signed
by physicians, we se-lected
all patients visiting
SFGH for an AVP injury be-tween
January 2004 and De-cember
of 2008. Any injuries
treated at SFGH and known
to have occurred in San
Mateo County ( the neighbor-ing
county to the South)
were excluded from the cost
analysis ( n= 20). Any AVP
injuries occurring as a result
of “ assault using a vehicle”
were also excluded be-cause
these injuries were
intentional. After these
exclusions, 3,598 AVP
cases remained and were
included in our analysis.
( cont. on page 2)
Introduction
CONTENTS
Introduction 1
Method 1- 2
Terminology 2
Findings 3- 4
About the
Injury Center
4
Contact 4
Method
Cost of Auto- versus- Pedestrian Injuries
San Francisco, 2004 - 2008
M A R C H 2 0 1 0
R O C H E L L E D I C K E R , M D , F A C S
D A H I A N N A L O P E Z , R N , M S N , M P H
M A R C I P E P P E R , B A
I A N C R A N E
M . M A R G A R E T K N U D S O N , M D , F A C S
San Francisco Injury Center W E N D Y M A X , P H D
Method ( cont.)
R O C H E L L E D I C K E R , M D , F A C S P A G E 2
Two groups of patients natu-rally
emerged: 1) Patients seen
in the Emergency Department
and released within 24 hours,
and 2) Patients admitted to the
hospital for further observation
or treatment, such as surgery.
These groups will be referred
to as “ nonadmitted” and
“ admitted” patients, respec-tively.
As we were collecting
the data, we found that the
admitted patients accounted for
the majority of the cost even
though they accounted for one
fourth of the total number of
AVP collisions. For this reason,
we decided to geocode the
more expensive patients ac-cording
to Supervisorial district.
We wanted to know if injuries
were more expensive in one
area of the City compared to
another.
Using data from police reports,
also known as
“ SWITRS” ( Statewide Inte-grated
Traffic Record System)
data, we geocoded the location
of injury for admitted patients.
We were successfully able to
match 77% of our admitted
patients to the SWITRS data-base
in order to map the inter-section
or address where the
injury occurred. Using the San
Francisco Geographic Informa-tion
System ( SFGIS), we coded
each intersection by its corre-sponding
Supervisorial
District..
vices, nursing and allied health
care, medical care, administra-tive
services, overhead, and
ambulance transportation. In
order to compare cost by year,
we used Consumer Price Indi-ces
published by the U. S. Bu-reau
of Labor to express all
costs in 2008 dollars.
Payer refers to the funding
source to which charges for
Charge refers to the amount
billed for all health care services
provided. Charges can be lik-ened
to the “ retail” price for
health care. Charges will not be
described in detail in this re-port.
Cost refers to the actual
amount paid for healthcare,
including room fees, medica-tions,
medical supplies and de-medical
care were billed.
Supervisorial District refers to
the geographic area assigned to
a San Francisco City Supervisor.
Countermeasures refers to
safety measures that are avail-able
for reducing the likelihood
of collisions. These may include
engineering countermeasures,
enforcement efforts, and educa-tional
interventions.
UCSF guided the methodology for
calculating costs. The detailed
methodology is summarized in the
article submitted for publication.
On a conceptual level, cost of di-rect
medical care was calculated
by summing three components of
care: 1) hospital costs, 2) profes-sional
fees for specialty care, such
as surgery, and 3) cost of trans-porting
a patient via ambulance.
Data were analyzed using expert
statistical consultants and ad-vanced
statistical software.
Five databases were used, in-cluding
the hospital’s trauma
database, the San Francisco De-partment
of Public Health
( SFDPH) Billing Information
System ( BIS), the Emergency
Medical Record ( EMR), the
Statewide Integrated Traffic
Record System ( SWITRS), and
the San Francisco Geographic
Information System ( SFGIS). 2
Dr. Wendy Max, a health eco-nomics
expert and Professor at
Data Sources and Data Analysis
Terminology
“ One of the unique
features of our study
is that it focuses on
cost — or what was
actually paid out—
rather than charges—
or what was
actually billed” -
Dr. Rochelle
Dicker, Trauma
Surgeon and
Principal
Investigator
General Findings– All Injured Pedestrians, SFGH ( 2004- 2008)
R O C H E L L E D I C K E R , M D , F A C S P A G E 3
• 3,598 pedestrians included
in the study
• 931 ( 26%) were admitted
and 2,667 were nonadmit-ted
( 74%)
• Age ranged from 0 to 94
years.
• Children ( 0 and 19 year)
accounted for 14%
• Adults ( 20 to 64 years)
accounted for 72%
• Elderly patients ( over 65
years) accounted for 13%
• Over 50% of the sample
consisted of Caucasians
( 33%) and Asians ( 25%)
• 98% lived in California at
the time of the injury
• 74% lived in the City of San
Francisco at the time of
the injury
• 0.6% were visiting San
Francisco from a foreign
country at the time of in-jury
• Homeless people ac-counted
for 7%
• 74% were released from
the Emergency Depart-ment
within 24 hours
( nonadmitted)
• 26% were admitted to the
hospital
• The total cost of Pedestrian
Injury was $ 74.3 million for all
5 years
• Admitted patients ( 26%)
accounted for 82% of the total
cost
• The strongest predictors of
high cost were the number of
ICU days, hospital days, and
age.
required mechanical venti-lation
for about 2.1 days.
• 55% of admitted patients
were discharged home
• 7% were transferred to
another acute care facility
• 7% were discharged to an
acute rehabilitation center
• 19% were discharged to a
Skilled Nursing Facility
• 9% died
• 931 ( 26%) injured pedes-trians
were admitted
over the 5- year period
• On average, their hospi-tal
length of stay was
11.6 days
• 30% of admitted patients
required a stay in the
Intensive Care Unit
( ICU) for an average of
2.8 days
• Of those in the ICU, 56%
General Findings— All Admitted Injured Pedestrians ( 2004- 2008)
Total Cost of Injury ( 2004- 2008)
Collision Year Total Cost ( 2008
Dollars )
2008 Pop Cost Per Capita
2004 $ 11,257,143.03 840,462 $ 13.39
2005 $ 13,480,653.08 840,462 $ 16.04
2006 $ 16,574,112.85 840,462 $ 19.72
2007 $ 17,673,296.91 840,462 $ 21.03
2008 $ 15,358,023.35 840,462 $ 18.27
All Years $ 74,343,229.22 840,462 $ 88.46
Mean Cost per Pedestrian for Admitted and Nonadmitted Patients
R O C H E L L E D I C K E R , M D , P A G E 4
Approximately 76% ($ 56.7 million) of the total cost was
paid for by public funding, including Medicare, MediCal, and
patients themselves. Conversely, 24% ($ 17.6 million) of the
cost was paid for by private insurance. The minimum
amount billed directly to an uninsured patient was $ 5,143
and the maximum was $ 505,952.
By mapping the admitted
patients ( most costly) ac-cording
to Supervisorial
District, we were able to
highlight “ hotspots”
where an economic case
can be made for imple-menting
and evaluating
sustainable countermea-sures.
Total Cost for Admitted Patients by Supervisorial District
Who is paying for this?
Admitted Nonadmitted
Year Mean Confidence Interval Frequency Mean Confidence Interval Frequency
2004 $ 47,303 ( $ 38,739, $ 57,760 ) 200 $ 3,798 ( $ 2,401, $ 6,006 ) 473
2005 $ 55,989 ( $ 46,467, $ 67,461 ) 194 $ 5,165 ( $ 3,533, $ 7,550 ) 507
2006 $ 76,440 ( $ 64,803, $ 90,168 ) 181 $ 4,881 ( $ 3,367, $ 7,075 ) 561
2007 $ 77,679 ( $ 66,336, $ 90,961 ) 195 $ 4,534 ( $ 3,081, $ 6,674 ) 557
2008 $ 72,754 ( $ 60,799, $ 87,058 ) 161 $ 6,405 ( $ 4,643, $ 8,837 ) 569
• We plan to share our data with other
agencies to help improve pedestrian
safety.
• We plan to calculate the indirect cost
of productivity losses and disability
from pedestrian injury, as well as
longer- term costs including the costs
incurred from rehabilitation and long-term
care.
• We plan to replicate this study with
auto- versus- bicycle collisions in San
Francisco.
Next steps… Acknowledgements
M. Margaret Knudson, MD
Lou Fannon
Peggy Skaj
Mary Nelson, RN, MPA
Valerie Inouye
Grace Fernandez
Kathleen Acosta
Cony Artigapinto
Maria Esguerra
Josephine Hermoso
Jim Paolucci
Mark Erdmann
Matthew Cuenot
Oliver Gajda
Ricardo Olea
The San Francisco Injury Center for Research and Prevention ( SFIC) is one of 13 Injury Control Research Centers
funded by the National Center for Injury Prevention and Control at the Centers for Disease Control and Prevention
( CDC). The SFIC was established in 1989 and is located at the San Francisco General Hospital campus of the Uni-versity
of California, San Francisco, School of Medicine.
The SFIC is a center without walls - bringing together multidisciplinary faculty investigators from throughout the
UCSF campus and beyond. The resulting collaborative efforts in laboratory research, clinical trials and injury preven-tion
research have the potential to improve outcomes for victims of trauma in our region and to influence the field
of injury control on a global basis.
About the SF Injury Center
Conclusion
Compelling arguments must be put forth to policymakers to invest in changes to enhance pedestrian safety. Pedes-trian
injury carries the intangible price tag of human life and the tangible price tag of health care expenditures. Pro-viding
not only an account of where collisions occur but also the monetary cost of the injuries incurred at those sites
gives credence to instituting life- saving and cost- saving measures targeting specific locations and specific road traffic
issues.
Interagency collaboration as well as support from the City Supervisors and Mayor will be crucial to the improvement
of pedestrian safety. Several sustainable and cost- effective countermeasures in the areas of urban planning, engineer-ing,
enforcement, and education have shown promise both in the City of San Francisco and in the scientific litera-ture.
3- 5 We hope that the cost of pedestrian injury will be strongly considered when making decisions to allocate
funds for injury prevention strategies in San Francisco.
References
1. Healthy People 2020 Public Meet-ings.
http:// www. healthypeople. gov/
hp2020/ Objectives/ files/
Draft2009Objectives. pdf. 2009
2. City and County of San Francisco Depart-ment
of Technology; San Francisco Enter-prise
GIS Program. [ cited 2010 January
15]; Available from: http:// www. sfgov. org/
site/ gis_ index. asp? id= 368
3. Pedestrian Safety: Report to Congress. 2008,
Federal Highway Administration, US
Department of Transportation.
4. Retting, R., Ferguson, SA, McCartt, AT, A
review of evidence- based traffic engineering
measures designed to reduce pedestrian-motor
vehicle crashes. American Journal of
Public Health, 2003. 93( 9): p. 1456- 1463.
5. Davis, J., Bennink, LD, Pepper, DR, Parks,
SN, Lemaster, DM, Townsend, RN,
Agressive traffic enforcement: A simple and
effective injury prevention program. Journal
of Trauma, 2006. 60( 5): p. 972- 7.
Contact:
Dahianna Lopez, RN, MSN, MPH
Prevention Director
1001 Potrero Ave
Box 0807 ( Ward 3A)
San Francisco, CA 94113
lopezd@ sfghsurg. ucsf. edu