Association Between NCAP Ratings and Real-world Rear Seat Occupant Risk of Injury

Principal Investigator: Allison Curry, PhD, MPH, Children's Hospital of Philadelphia

Below is an executive summary of this project. Please note that this summary describes results and interpretation that may not be final. Final interpretation of results will be in the peer-reviewed literature.

Risk of AIS≥2 Injury among Restrained Rear Row Occupants in All Crashes, by 2004-2010 Crash Test Star Ratings, (NASS-CDS, 2003-2012)

The risk of injury in real world crashes for restrained rear row occupants was lower in vehicles with 5-star driver and passenger ratings in frontal crash tests. There is no association, however, among rear row occupant injury risk and vehicles’ star ratings in side-barrier crash tests.

Many consumers utilize the National Highway Traffic Safety Administration’s US New Car Assessment Program (NCAP) “star ratings” to compare the safety of new vehicles. The ratings are based on evaluations of injury risk as measured by anthropomorphic test devices (ATDs), or crash test dummies, and range from 1 star (greater than average risk) to 5 stars (less than average risk).

Little is known about whether NCAP ratings predict real-world risk of injury to restrained rear seat occupants. Currently, only the side-impact barrier and pole tests include an ATD in the rear seat and no existing full vehicle crash test includes a rear-seated child-size ATD. This project examined how NCAP’s current rating system predicts risk of serious injury among restrained rear-seat occupants in real-world crashes.

Crash-involved vehicles, model years 2004-2013, were identified in the National Automotive Sampling System – Crashworthiness Data System (NASS-CDS) database. Using data on make, model year, body type, and other identifying information, 95 percent of the vehicles in NASS-CDS were matched to a specific vehicle in NCAP. The resultant database was analyzed to examine associations between vehicle ratings and rear-seat occupant injury risk. Data were limited to restrained occupants, and injuries were defined as injuries with an Abbreviated Injury Scale (AIS) score of 2 or greater.

The 18,218 vehicles represented an estimated 6 million vehicles with over 9 million occupants. The injury risk in all crashes for restrained rear-seat occupants was:

  • Lower in vehicles with a 5-star driver rating in frontal-impact tests (1.4 percent) than with 4- or fewer-stars (2.6 percent)
  • Lower in vehicles with a 5-star passenger rating in frontal-impact tests (1.3 percent) than with 4- or fewer-stars (2.4 percent)

Conversely, side-impact driver and passenger crash tests were not associated with rear-seat occupant injury risk (driver test: 1.7 percent for 5 star vs. 1.8 percent for 1 to 4 stars; passenger test: 1.6 percent for 5 stars vs 1.8 percent for 1 to 4 stars).

The findings indicate current frontal-impact test procedures provide some degree of discrimination in real-world rear-seat injury risk among vehicles with a 5-star rating compared to fewer than 5 stars. However, there is no evidence that vehicles with a 5-star side-impact passenger rating, the only crash test to include a rear-seat ATD, demonstrate lower risks of rear-seat injury than vehicles with fewer than 5 stars. These results support prioritizing modifications to NCAP’s program that specifically evaluate rear-seat injury risk to all restrained occupants.

Project Team Members

Dennis R. Durbin, MD, MSCE, Children’s Hospital of Philadelphia; Siobhan Gruschow, MPH, MEd, Children’s Hospital of Philadelphia; Kristina B. Metzger, PhD, MPH, Independent Consultant


Sayaka Ogawa, Drexel University; Kassandra Pickel, Drexel University

IAB Mentors:

Michelle Tsai, Consumer Reports; Lan Xu, FCA US LLC; Agnes Kim, Ford Motor Company; Steve Rouhana, Ford Motor Company; Steve Roberson, State Farm Mutual Automobile Insurance Company; Hiromasa Tanji, TK Holdings Inc.; Schuyler St. Lawrence, Toyota USA