The purpose of this investigation was to compare the ability of two different bone densitometry techniques (quantitative computed tomography and dual X-ray absorptiometry) to discriminate subjects without any vertebral fracture from those with definite vertebral fractures. Bone mineral density (BMD) was determined in 112 post-menopausal women. 83 women were classified free of vertebral fracture and 29 were considered to have definite vertebral fractures. Quantitative computed tomography (QCT) was performed using three different spinal regions of interest. Dual X-ray absorptiometry (DXA) was measured at the spine (both in posteroanterior (PA) and lateral (L-DXA) projection, and at hip (total, neck and Ward regions). An additional estimated volumetric BMD was derived from the PA and L-DXA spine scans. Individuals with definite vertebral fractures had lower bone mineral density for each of the methods and regions of interest (ROI). Spinal QCT results, total and Ward hip DXA results showed the greatest percentage decrement. Odds ratio estimates revealed a significant relationship between BMD and fracture with spinal QCT, hip, and volumetric spinal DXA. The greatest areas under the receiver operating characteristic (ROC) curves were found with results of spinal QCT, total and Ward hip DXA, and volumetric spinal DXA measurements. It is concluded that spinal QCT and femoral total DXA measurement may be preferred to PA and lateral L3 DXA measurements.