To explore whether there are ethnic differences in calculated hip strength that might explain the low incidence of hip fracture in China, we used Lunar DPX 'beta' version of hip strength analysis (HAS) and hip axis length (HAL) programs to compare hip geometry, calculated strength and densitometric values from Chinese subjects in Shenyang to those of Caucasian subjects in Oslo and Leuven participating in the European Prospective Osteoporosis Study (EPOS). Subjects were 210 Chinese and 403 Caucasian men and women aged 53-77 years. Parameters investigated included bone mineral density (BMD), bone mineral content (BMC), bone area (BA), cross-sectional moment of inertia (CSMI) and section modulus (both indicating strength and rigidity of the femoral neck), HAL, neck length (NL), neck diameter, tensile stress (Tstress) and compressive stress (Cstress) (indicating the stress in the femoral neck at its weakest cross section arising from walking or a standard fall, respectively), safety factor (SF, indicating the resistance to fracture for forces generated during walking) and fall index (FI, indicating the resistance to fracture from force generated during a fall in the greater trochanter). The Chinese men and women were significantly shorter and lighter than their Caucasian counterparts (P<0.01) and had significantly lower BMD, BMC and BA of the femoral neck (P<0.01). After adjusting for BA, weight and height, there was no significant ethnic difference in either gender in BMC. CSMI and section modulus were significantly lower, and HAL, NL and neck diameter were significantly shorter in the Chinese men and women (P<0.01). These differences all remained after adjusting for weight and height. There were no significant differences in Tstress, Cstress, SF and FI between ethnic groups in either gender. Most of the parameters of calculated hip strength in the Chinese subjects were similar to or poorer than those in the Caucasian subjects. There was no evidence to indicate that Shenyang Chinese have superior BMD or BMC or better calculated hip strength. The short HAL and NL of the population, however, could be an independent factor contributing to the low incidence of hip fracture.