NHTSA AACN Activities Stephen Ridella Human Injury Research Division Office of Vehicle Safety Research Stephen Ridella Human Injury Research Division Office.

NHTSA AACN ActivitiesNHTSA AACN Activities

Stephen Ridella

Human Injury Research Division

Office of Vehicle Safety Research

Stephen Ridella

Human Injury Research Division

Office of Vehicle Safety Research

22

OutlineOutline

2. NHTSA and NHTSA/CDC AACN Work2. NHTSA and NHTSA/CDC AACN Work

3. EDR Rule and Analysis 3. EDR Rule and Analysis

5. AACN Scorecard/Next Steps5. AACN Scorecard/Next Steps

1. Overview1. Overview

4. Injury Prediction Algorithms 4. Injury Prediction Algorithms

33

OverviewOverview

44

?

OverviewOverview

55

If you are severely injured, care at a Level I trauma center lowers the risk of death by 25%.

McKenzie, et. al.NEJM, 2006

66

Getting from Crash to Trauma CenterGetting from Crash to Trauma Center

Finding the car

Notifying 9-1-1

Appropriate EMS response– Getting the right people there

Triage– Getting the right patient to the right hospital

Care and transport

Designated trauma centers

Consistent communication essential

Finding the car

Notifying 9-1-1

Appropriate EMS response– Getting the right people there

Triage– Getting the right patient to the right hospital

Care and transport

Designated trauma centers

Consistent communication essential

77

Getting from Crash to Trauma CenterGetting from Crash to Trauma Center

Early Notification

EMS and Trauma System

Designated trauma centers

Early Notification

EMS and Trauma System

Designated trauma centers

88

NHTSA Historical AACN WorkNHTSA Historical AACN Work

99

NHTSA Historical AACN WorkNHTSA Historical AACN Work

ACN Field Operational Test (1995-2000)– Purpose

• Establish EMS response times relative to time of crash

• Determine if ACN information can reduce those times

– System Description• Outfitted 700 vehicles with ACN system

• GPS, crash sensor and embedded cell phone

• Basic injury algorithm based on delta-V and PDOF

• Ran vehicles in Erie County, PA for nearly 3 years

– Results• 15 ACN crashes during study

• PSAP notified within 1 minute of crash

• Some system failures and false positives due to system issues

ACN Field Operational Test (1995-2000)– Purpose

• Establish EMS response times relative to time of crash

• Determine if ACN information can reduce those times

– System Description• Outfitted 700 vehicles with ACN system

• GPS, crash sensor and embedded cell phone

• Basic injury algorithm based on delta-V and PDOF

• Ran vehicles in Erie County, PA for nearly 3 years

– Results• 15 ACN crashes during study

• PSAP notified within 1 minute of crash

• Some system failures and false positives due to system issues

1010

NHTSA/CDC AACN WorkNHTSA/CDC AACN Work

1111

NHTSA/CDC AACN WorkNHTSA/CDC AACN Work

“Vehicle telematics consistent with high risk of injury”

Field Triage Decision Scheme

1212

NHTSA/CDC AACN WorkNHTSA/CDC AACN Work

1313

Advanced Automatic Crash NotificationAdvanced Automatic Crash Notification

Critical information helpful to dispatch, respond and triage an injured occupant to final destination– Velocity change of vehicle

– Principle Direction of Force

– Seat belt usage

– Crash with multiple impacts

– Vehicle type

– Voice (GCS proxy)

Information can be collected in vehicle EDR for transmission

Critical information helpful to dispatch, respond and triage an injured occupant to final destination– Velocity change of vehicle

– Principle Direction of Force

– Seat belt usage

– Crash with multiple impacts

– Vehicle type

– Voice (GCS proxy)

Information can be collected in vehicle EDR for transmission

1414

Event Data Recorders (EDRs)Event Data Recorders (EDRs)

NHTSA published a final rule on August 28, 2006 regulating Event Data Recorders (EDR) Did not mandate EDRs

Established uniform set of criteria for EDRs voluntarily installed in vehicles

Applied to vehicles with GVWR of 3,855 kg (8,500 lb) or less manufactured after September 1, 2010

Required standardized statement in owners manual to notify consumers that vehicle is equipped with EDR and outlining its functions

NHTSA published its response to petitions for reconsideration in a final rule on January 14, 2008

Effective date of rule is Sept. 1, 2012

NHTSA published a final rule on August 28, 2006 regulating Event Data Recorders (EDR) Did not mandate EDRs

Established uniform set of criteria for EDRs voluntarily installed in vehicles

Applied to vehicles with GVWR of 3,855 kg (8,500 lb) or less manufactured after September 1, 2010

Required standardized statement in owners manual to notify consumers that vehicle is equipped with EDR and outlining its functions

NHTSA published its response to petitions for reconsideration in a final rule on January 14, 2008

Effective date of rule is Sept. 1, 2012

1515

NHTSA EDR RuleNHTSA EDR Rule

1616

NHTSA EDR RuleNHTSA EDR Rule

1717

NHTSA EDR AnalysisNHTSA EDR Analysis NHTSA has over 3000 Cases (NASS-CDS) of crashes with

EDR data (mostly GM vehicles)

Analysis to date– Compared EDR data to standardized crashes of similar vehicles

– Compared EDR delta-V to delta-V from reconstructions

– New contract to investigate recent (2008-2009) crashes relative to injury prediction with various vehicle crash data attributes besides delta-V

NHTSA has over 3000 Cases (NASS-CDS) of crashes with EDR data (mostly GM vehicles)

Analysis to date– Compared EDR data to standardized crashes of similar vehicles

– Compared EDR delta-V to delta-V from reconstructions

– New contract to investigate recent (2008-2009) crashes relative to injury prediction with various vehicle crash data attributes besides delta-V

1818

Triage Application: DV to Predict ISSTriage Application: DV to Predict ISS

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 10 20 30 40 50 60 70 80 90 100

EDR Longitudinal DV (mph)

Pro

ba

bil

ity

of

ISS

>=

16

All Occupants

Lower 95% Confidence Bound

Upper 95% Confidence Bound

Source Data

1919

Injury Prediction AlgorithmsInjury Prediction Algorithms

Urgency Algorithm (NHTSA funded)– Champion et al. studied criteria for recognizing time critical injuries

at the crash scene

– Malliaris et al. (1997) developed relationships between crash attributes and injuries based on Champion data

– Augenstein (2001, 2005) made subsequent improvements to algorithm

– Urgency predicts probability of AIS 3+ or fatalities using vehicle crash and occupant data (delta-V, peak crush, occupant gender, occupant age, entrapment, vehicle weight, etc.)

It is worthwhile to note that Urgency is based on retrospective analysis of crash data

Urgency Algorithm (NHTSA funded)– Champion et al. studied criteria for recognizing time critical injuries

at the crash scene

– Malliaris et al. (1997) developed relationships between crash attributes and injuries based on Champion data

– Augenstein (2001, 2005) made subsequent improvements to algorithm

– Urgency predicts probability of AIS 3+ or fatalities using vehicle crash and occupant data (delta-V, peak crush, occupant gender, occupant age, entrapment, vehicle weight, etc.)

It is worthwhile to note that Urgency is based on retrospective analysis of crash data

2020

AACN ScorecardAACN Scorecard

ACN and AACN systems have been installed in vehicles since 1996. Other companies have expressed interest.

GM and BMW have announced that the have injury prediction algorithms to alert PSAP of probability of a serious injury. To date, the systems are:

ACN and AACN systems have been installed in vehicles since 1996. Other companies have expressed interest.

GM and BMW have announced that the have injury prediction algorithms to alert PSAP of probability of a serious injury. To date, the systems are:

Manufacturer System Number of Vehicles Offer/Cost Comments

General Motors On-star Several Million Free 1st year, $200/$300 per year after

Have assisted in over 100,000 airbag-deployed crashes

BMW BMW Assist 500,000 (est) Free 1st 4 years, $200 per year after

Toyota Safety Connect N/A N/A Summer 2009 introduction

Ford 911-Assist (Sync) 1,000,000 (est) Option package Direct phone link to PSAP

2121

AACN Scorecard : BMW exampleAACN Scorecard : BMW example

2222

AACN : Next StepsAACN : Next Steps

AACN data can be used to predict injury severity, conveyed to EMS services and trauma centers, and integrated into the field triage process.

CDC and NHTSA– Working together to create awareness

– Meeting with industry to solicit cooperation

– Determining course of action relative to Expert Panel Recommendations

– Determining benefits

AACN data can be used to predict injury severity, conveyed to EMS services and trauma centers, and integrated into the field triage process.

CDC and NHTSA– Working together to create awareness

– Meeting with industry to solicit cooperation

– Determining course of action relative to Expert Panel Recommendations

– Determining benefits

2323

AACNAACN

Shows promise in improving outcomes in severely injured crash patients by:– Predicting the likelihood of serious injury

– Decreasing response time

– Assisting with field triage decisions

– Decreasing time to trauma center

– Decreasing death and disability

Shows promise in improving outcomes in severely injured crash patients by:– Predicting the likelihood of serious injury

– Decreasing response time

– Assisting with field triage decisions

– Decreasing time to trauma center

– Decreasing death and disability

2424

TM

ENDEND