The 2003 International Society for Clinical Densitometry consensus guidelines recommend exclusion of vertebral bodies for lack of increase in bone area (BA) or bone mineral content (BMC), or an unusual T-score discrepancy (>1 standard deviation [SD]) between adjacent vertebrae. It is unclear how often nonprogression in BA, BMC, and T-score discrepancies predicts abnormal vertebral morphology, such as compression fractures. We prospectively studied 101 individuals sent for clinical dual-energy X-ray absorptiometry (DXA) scanning, including 20.8% males and 79.2% females. The population was 85% Caucasian, 13% African-American, and 3% Hispanic. The mean age was 65.6 yr; 20.2% were currently on steroids and 22.7% were taking drugs for osteoporosis. All subjects underwent the usual posteroanterior (PA) spine DXA scan PA and lateral vertebral fracture analysis (VFA). The presence of vertebral compression fractures and/or scoliosis of the lumbar spine by VFA were correlated with nonprogression of area or BMC, and/or a difference of >1 SD in T-scores using Fisher's exact test. By VFA, we detected 22 lumbar compression fractures among 101 subjects, which was 16% of the population. Nonprogression of BA, BMC, and T-score discrepancy were not statistically associated with the presence of vertebral compression fracture as assessed by VFA. Thirty percent of subjects had lumbar spine scoliosis. The presence of scoliosis was significantly related to a T-score discrepancy at L1-L4.