Faculty Bibliography

INTRODUCTION: While osteoporosis is common among women of Chinese descent, a readily available bone mineral density (BMD) referent database for Chinese American women does not exist. Fracture risk among this population is currently assessed using a Caucasian reference as well as diagnostic criteria for osteoporosis developed for postmenopausal Caucasian women. Many studies indicate that there are important racial differences in skeletal health and fracture risk, an observation that makes the application of Caucasian data to all groups problematical. This study was undertaken to establish a BMD referent database in Chinese American women and to compare it with a Caucasian female database. It is expected that a race-specific database will be useful in the assessment of bone health for Chinese American women. METHODS: Healthy Chinese American women (n=359), ages 20-90, were recruited. Along with dual-energy X-ray absorptiometry (DXA) of the total hip and lumbar spine, demographic, medical, familial, nutritional, and behavioral data were obtained. The mean and standard deviation for BMD at each site was calculated for each 10-year age group and compared to mean BMD values for Caucasian women supplied as found in the Hologic DXA instrument. Osteoporosis diagnosis rates for this cohort, calculated with the Caucasian and newly established Chinese American BMD referent values, were compared with each other. RESULTS: Compared with Caucasian women, Chinese American women have significantly lower BMD at the lumbar spine, total hip, and femoral neck across a wide spectrum of age groups. As a consequence, more than one-half of Chinese American women>or=50 years of age, who would be characterized as osteoporotic using a Caucasian referent, would not be diagnosed as such if a Chinese American referent were utilized. CONCLUSION: Chinese American reference BMD values are significantly lower than those for Caucasian women. Future studies relating Chinese American BMD values to fracture risk are necessary in order to determine if ethnic database-derived T-scores would be more predictive of fracture risk and to develop meaningful diagnostic criteria for this population.