The Evolving Science of Crash Data Collection

July 15, 2014

Many of our readers involved in traffic safety research are aware that the National Highway Traffic Safety Administration (NHTSA) is moving forward with the modernization of the National Automotive Sampling System (NASS) for the first time since NASS’s inception over 40 years ago. NHTSA recently released an RFP for professional services to develop a new crash data system that meets the current and future data needs of NASS. NHTSA also announced name changes for the two data collection systems that encompass NASS; the Crash Investigation Sampling System (CISS) will replace the NASS Crashworthiness Data System (CDS) and the Crash Report Sampling System (CRSS) will replace the NASS General Estimates System (GES).


Cover of the National Child Occupant Special Study's recruitment materials.

Over the past 7 years, CIRP@CHOP has worked with NHTSA to develop the National Child Occupant Special Study (NCOSS), a series of pilot studies that developed, implemented and evaluated a system for collecting supplemental child-specific data that leveraged the NASS-GES. The CIRP@CHOP research team successfully adapted procedures commonly used in public health surveillance systems to collect supplemental data for traffic safety research. This body of work provides an effective strategy to enhance the value of police crash reports that will continue to be collected through NASS-CRSS to support traffic safety research. The methodology and results of the third and final NCOSS pilot, which tested enhanced methods of encouraging eligible subjects to participate in NCOSS, were recently published in Injury Prevention.

CIRP@CHOP has (and will continue to) partner with NHTSA to ensure that the unique safety needs of children are considered as NASS is updated. In order for NASS to be a useful tool for child traffic safety research, child-specific data on a sufficient number of children are required. Limitations of the current surveillance systems include:

  • NASS-GES: A nationally representative sample of police reported motor vehicle crashes of all types, from minor to fatal; This system has an adequate number of pediatric subjects, but does not collect sufficient child-specific data.
  • NASS-CDS: Detailed data on a representative, random sample of thousands of minor, serious, and fatal crashes; This system collects sufficient child-specific data, but does not have an adequate number of pediatric subjects.
  • NSUBS: Probability-based nationwide child restraint use survey that observes restraint use and interviews adult occupants to collect race, ethnicity and other data; This system collects observational data, but does not include crashes.

We are fortunate at CIRP@CHOP to have first-hand experience with the rigorous research that can be conducted with sufficient child-specific data through the Partners for Child Passenger Safety (PCPS) study. For over a decade, PCPS served as the nation’s only large scale child-focused crash surveillance data system, informing new product development, test protocols and federal motor vehicle safety standards, public education and curricula, policy, and medical practice to improve child restraint use and advance the safety of children in motor vehicle crashes.

Important parts of the PCPS legacy are the educational and advocacy tools that were developed as a result of the research. We are fortunate to have recently been able to redesign the charts (with support from the Association of Global Automakers) to give them an updated look and feel. Although the information is still relevant, new robust data need to be collected to reflect changes to restraint use, injury patterns, technological advances in vehicles and child restraint systems, and more.

Although PCPS ended data collection in 2007, the result was a more than 50% reduction in the number of children killed in crashes following the inception of PCPS in 1997. CIRP@CHOP looks forward to the continued modernization of NASS, and will continue to be a voice for efforts that will save the lives of children in the US.