In reply
I thank Drs Pieramici and Kuhn for their comments and in particular their enthusiasm for the field of ocular trauma. Although it is true that the National Automotive Sampling System (NASS) is not a random sample, this study1 was carefully developed to draw comparisons only within the NASS samples. In particular, crash severity was controlled between groups with and without air bags by limiting the cases to only frontal crashes, which ensures similar occupant kinematics, and by comparing the crash velocity changes. In this study, there was not a significant statistical difference between the average velocity changes of the cases with air bags vs the cases without air bags because during the sample years of 1993 through 1999, approximately 50% of the vehicle fleet did not have air bags.1