Results of a recent study suggest that advanced automatic crash notification (AACN) technology, which sends telemetry data in the event of a motor vehicle accident, can be used to accurately predict the injury severity of vehicle occupants.
The research by Stewart Wang, MD, PhD, director of the Program for Injury Research and Education at the University of Michigan and a trauma surgeon, analyzed data from 836 vehicles with OnStar AACN capabilities that were involved in crashes involving 924 occupants between January 2008 and August 2011.
Wang used an algorithm to analyze three types of data from each accident: crash factors, vehicle factors and occupant factors. Results of the algorithm were then used to predict whether each vehicle occupant met the 20% or higher risk of having an Injury Severity Score (ISS) of 15+, which is the threshold set by the National Expert Panel on Field Triage for urgent transport to a trauma center.
Accuracy of AACN Data
The Wang et al. study1 showed that when the gender and age of the vehicle occupants were known, the algorithm could use AACN data to predict which occupants had a high likelihood of severe injury with a sensitivity of 64% and a specificity of 96%. Without information about age and gender, the sensitivity was 45% while the specificity was 98%.
“Sixty-three percent sensitivity may not sound like an impressive number but when you look at the experience over the last four to five decades, it is a significant improvement,” says Wang. “Even trained personnel are no more than 40%–50% accurate in picking out patients with severe injuries.
“In the past, EMS did not usually receive AACN data or if they did, they were appropriately skeptical because it was a prediction that wasn’t proven. But with this new trial, we know we should be taking AACN very seriously. If the vehicle is reporting that there is a high risk of severe injury, you should see that report as at least as accurate or more accurate than your best colleague.”
Data Collected by AACN
A panel of experts on AACN technology and patient triage have made recommendations on what information a vehicle should transmit in the event of a crash, which include:
Whether the crash included multiple impacts,
The vehicle’s change in velocity,
The principle direction of force,
Whether or not seat belts were in use, and
The type of vehicle.
Right now, when an AACN-enabled car is in an accident, the vehicle automatically routes essential telematics information to a vendor-operated call center. The call center then processes the information and calls the nearest public safety answering point (PSAP).
“With next-generation 9-1-1 technology, however, there is the potential for the data to flow directly to the PSAP,” says Crystal McDuffie, communications center and 9-1-1 services manager at APCO International. “The PSAP can then pass this information to the responders, although current call centers would remain a vital link as the first contact.”
But not all of vehicle telemetry data needs to be relayed to first responders.
“EMS crews are busy taking care of things on the scene. They don’t need to be inundated with a bunch of complicated information, what they need is a simple indication of whether a patient is at high risk of severe injury,” says Wang, which was his aim in developing the injury severity prediction algorithm.
Wang goes on to say that a trauma center, however, might need other information: “The medical team at the trauma center will be very interested in the details of the crash because this information can help us narrow down the types of injuries we worry about the most.”
AACN Helping EMS
To help EMS, 9-1-1 and emergency medicine professionals learn more about AACN technology and the implications of AACN crash data, a new, one-hour online training program will be available this summer at http://www.aacnems.com. This course teaches EMS providers and medical directors about the biomechanics of crash injuries, the research that supports the validity of AACN predictions, how data can be used to predict injury severity and how to integrate this data into local EMS systems.
The online training program is supported by the American College of Emergency Physicians (ACEP) and the National Association of EMS Physicians (NAEMSP), with funding from the National Highway Traffic Safety Administration (NHTSA).
Scott Sasser, MD, associate director for International Programs for the Center for Injury Control at Emory University and the principal investigator on the ACEP/NAEMSP training course, says, “I think the key part that we, as EMS professionals, medical directors and 9-1-1 center directors need to learn about this new technology that is out there and growing in prevalence. We need to know how the information supplied by the vehicle can not only help identify injury patterns but can also help with everything from resource allocation to destination decision-making.”
Sasser, who is also an emergency room physician, points out that AACN data offers EMS and 9-1-1 personnel three main advantages when responding to a crash. AACN provides faster notification of the crash, the exact location information for the crash, and telemetry data, which can be used to predict injury severity and even injury patterns based on the engineering of the car and the mechanics of the crash.
“We will always need the expertise of the professionals in the field to interpret the data, get people out of the vehicles, safely assess them, and transport and treat them en route,” says Sasser. “But the data we can get from AACN adds important information to the data points we can get in field. This data can help us allocate the appropriate resources to the scene.”
Cory Richter, battalion chief at Indian River County Fire and Rescue in Vero Beach, FL, is a big proponent of the technology and thinks that its use and accuracy will only improve in the future.
“The amazing thing about this information is you can use it before you even get to the scene. You might not have even left the station and it can help change your response,” says Richter. “Where it specifically changes the EMS response is when we prelaunch the helicopter or call in additional units before we get on scene because we already know we need them.
“Sometimes it’s going to tell us it’s a trauma alert and it’s obvious because the patient is unconscious or unresponsive. But where it can really make a difference is if they [the patients] are up walking around or awake and alert but have potential internal injuries because of the mechanism of injury. In those cases, I tell my crews to do a really thorough evaluation and err on the side of transporting them to a trauma center.”
Future Directions of AACN Technology
Beyond the potential for AACN to direct a faster, more informed EMS response, this technology also has the potential to change trauma care, save more lives and help manufacturers build safer vehicles.
“The next step we want to look at is whether AACN can help us lower morbidity and mortality with motor vehicle crashes,” says Sasser.
Wang is particularly interested in how AACN can help with triage and diagnosis in the trauma center. “At a trauma center, we want to know things like if it was a driver involved in a left-sided crash. In that case, there might be high likelihood of a pneumothorax injury or a spleen injury because the spleen is on the left side. Knowing more about a crash can help us narrow down the possible diagnoses, which makes me a far better doctor.”
Wang, who does a lot of crash research, says, “Right now, we spend a lot of time going and looking at the car, trying to examine what is the principle direction of force, direction of the crash, and how severe it was. The sensors in the vehicles are far more precise. With time and enough data, we will definitely be able to use this information make cars safer.”
Widespread Adoption of AACN
Although AACN technology is not yet widespread in the U.S. automotive fleet, Sasser cautioned it is important for EMS, 9-1-1 and emergency medicine professionals to get ahead of this trend and understand the technology.
“The use of this technology in cars is going to be increasingly prevalent,” says Sasser. “As a profession, I would hate to see us get caught five or 10 years down the line and not be adequately prepared for it by developing the education, policies and protocols we need now. I think this technology will be well received by the EMS community because of its potential impact on our patients. We’ve got to keep supporting it, talking about it, getting educated about it and looking at the research on it.”
“I think EMS leadership should really be paying attention to AACN because it’s cutting-edge technology,” says Richter. “The future of EMS, in terms of motor vehicle crashes, is going to be notifications and information coming in from vehicles, sight unseen. We need to embrace this technology and use it now so the next generation of EMTs and paramedics are comfortable with it and can use it more as it gets more prevalent.”
AACN Course Overview
To help EMS, 9-1-1 and other medical professionals learn more about AACN technology and the implications of AACN crash data, a new one-hour online training program will be available this summer at http://www.aacnems.com. This course offers medical directors, EMS personnel and 9-1-1 directors an introduction to vehicle telemetry crash data and an understanding of what happens in a crash. The course also reviews the science behind the AACN injury severity predictive algorithms, how AACN fits into the CDC’s Field Trauma Triage Guidelines, and the future of AACN and how this technology is being implemented in local systems.
AACN Project Stakeholders
The Advanced Automatic Collision Notification project is a partnership between the American College of Emergency Physicians and the National Association of EMS Physicians and is funded by the National Highway Traffic Safety Administration.
Other stakeholders in the project are: American Academy of Pediatrics, American College of Surgeons/Committee on Trauma, Association of Public Safety Communications Officials, International Association of Fire Chiefs, National Academy of Emergency Medical Dispatch, National Association of EMS Educators, National Association of EMS Physicians, National Association of Emergency Medical Technicians, National Association of State EMS Officials, National Emergency Number Association, AACN vendor and automakers, including Onstar, Ford, SiriusXM and other major automakers, listed at aacnems.com.
New research is showing that automatic crash notification systems can be used to accurately predict the injury severity of vehicle occupants in motor vehicle crashes. These early notifications can help EMS crews know likely injury severity and the number of injured occupants before they even get on scene and direct a more informed EMS response.