The pedestrian head protection performance test was introduced to the Japan New Car Assessment Program (JNCAP) in 2003. Fifty-four car models were tested in 2005. The tests rated total pedestrian head protection performance of cars into levels 1 through 5. Also, the Japanese government began regulating pedestrian head protection for passenger cars in 2005. It is expected that cars are becoming less aggressive in pedestrian accidents.

In such situations, we are interested in the effectiveness of the pedestrian head protection tests introduced in JNCAP.

We will use Japanese national accident data between 2001 and 2005. The pedestrian fatality/severe-injury rate (the number of pedestrians killed or severely injured divided by the total number of pedestrians involved in the accidents) is an index of crash safety for pedestrians. The logistic regression method is applied to adjust for confounding factors (gender of pedestrian, age of pedestrian, guilt of pedestrian, day or night accident and travel speed of the car).

As a result of the study, we saw a correlation between the fatality/severe-injury rate and pedestrian head protection performance levels (1 to 4) in test results, suggesting that passenger cars with better test results protect pedestrians from severe injury in real-world accidents. Also, we observed that fatality/severe-injury rate of car models without pedestrian protection design are higher than that of car models with pedestrian head protection design, suggesting that passenger cars with pedestrian protection design are safer than those without pedestrian protection design in case of pedestrian accidents.