Car Crash Prevention Research
The Teen Driver Safety Research team at the Center for Injury Research and Prevention (CIRP) at The CHOP Research Institute has been conducting car crash prevention research for more than a decade. We use several methodological approaches to help reduce the frequency and severity of crashes with teens behind the wheel. These approaches to car crash prevention research include:
Evidence-based Intervention Design and Evaluation
At CIRP we employ a systematic approach towards the development and evaluation of behavior change programs in our car crash research. Key ingredients of successful programs include a well-articulated theory of why the program will be effective, a clear understanding of how the program will be delivered, and complementary evaluation plans for each. Drawing upon existing frameworks we developed a common vernacular for describing the fundamental steps in the intervention development process in a simplified way:
- Key Outcome – The overall and long-term goal of your program; it may not be impacted by any one specific program.
- Behavioral Objectives – Behaviors that, if performed, will lead to your Key Outcome.
- Target Constructs – Factors (often psychological and environmental) predictive of the Behavioral Objectives. In order to be a Target Construct the factor must:
- lead to the behavior
- be feasible to change via intervention
- have room to change (i.e., cannot be at ceiling)
Target Constructs are identified using behavior change theories and formative research with the target population.
- Intervention Content -- Content needs to map directly onto the Target Constructs. Depending on the nature of the intervention and how it will be delivered, Intervention Content can be as big as a comprehensive 12-week curriculum or as small as an image or phrase.
Here is an example of this approach in the context of the teen driver and teen passenger safety program, Ride Like A Friend. Drive Like You Care (RLAF), which was developed at CIRP.
Analysis of Naturalistic and Simulated Driving Data
Using naturalistic driving data and a high fidelity driving simulator located at CIRP, researchers are conducting a number of studies to help us understand how to best assess driving skills, how various driving scenarios affect drivers’ behaviors and emotions, and how interventions affect driving behavior and skill level. Driving simulators have made huge technological advances over the past several years and now achieve realistic reproductions of driving experiences and conditions. Controlled scenarios typical in the driving environment, such as dynamic traffic behaviors, pedestrians, time of day, and weather, can be programmed into the simulator to fit the needs of our studies.
The systematic development of the SDA was grounded in epidemiologic research of how teens and adults crash and in validated metrics for safe and unsafe driving in real world situations. Research published in Injury Prevention demonstrated the validity of the SDA for safe evaluation of novice teen driver skill in high-risk driving scenarios
An example of this approach is the Simulated Driving Assessment (SDA).
Linkage and Analysis of Existing Data Sources
Analyses of these existing sources of data often yield important population-level insights and trends in teen driving safety, as well as the ability to monitor important metrics of success in reducing the number of teens killed or injured in motor vehicle crashes.
This approach involves the linkage, management, and analysis of existing driving-related sources of data. Typically, we conduct novel epidemiologic analyses of these data or identify unique subgroups of interest from larger populations. Data analyzed in recent CIRP studies include state-level crash and licensing databases (New Jersey, Pennsylvania), federal government data sources from the National Highway Traffic Safety Administration (including the Fatality Analysis Reporting System, the National Automotive Sampling System, and National Motor Vehicle Crash Causation Study), and data reported by the Centers for Disease Control and Prevention (including the Youth Risk Behavior Surveillance System and the Web-based Injury Statistics Query and Reporting System).
Examples of this approach are Young Drivers' License and Crash Patterns in New Jersey and Miles to go: Focusing on Risks for Teen Driver Safety.