This report describes how new injury risk curves for the knee/distal femur and the hip were developed through reanalyses of existing peak knee impact force data. New hip injury risk curves were developed using survival analysis with a lognormal distribution. This distribution was parameterized to account for the effect of stature, which was the only subject characteristic that had a statistically significant effect on the relationship between peak force applied to the hip and the risk of hip fracture. The empirically defined effects of hip flexion and abduction from a standardized seated driving posture on mean hip fracture force were also incorporated into the lognormal distribution as mean shifts. Injury risk curves for the midsize male crash test dummy were defined by applying the stature associated with this dummy and posture of 30° flexion and 15° abduction from a standard reference posture and the standard reference posture (0°flexion, 0° adduction) to the lognormal distribution. A new risk curve describing the relationship between peak force applied at the knee and the likelihood of knee/distal femur fracture was developed by applying survival analysis to an existing dataset in which there was uncensored, left censored, and right censored peak knee impact force data. This risk curve is similar to that currently used by the NHTSA to assess the risk of AIS 2+ KTH injury. Because the fracture forces in the dataset used to develop the new knee/distal femur risk curve were primarily from tests where rigid surfaces loaded the knees of elderly midsize male cadavers, the new risk curve only applies to rigid knee impacts and this segment of the occupant population. Future work should focus on developing knee/distal femur risk curves that apply to other segments of the driving population by characterizing and accounting for the effects of subject factors and impact surface rigidity on KTH fracture forces. (Author/publisher)
Abstract