T-score discordance is observed when the difference between the hip and spine are sufficient to result in the two measurement sites falling into two different diagnoses as defined by the World Health Organization (WHO) classification system. This article examines the prevalence of patterns of T-score discordance between the total hip and PA L1-L4 total spine measurement sires in 5051 female patients undergoing dual X-ray absorptiometry (DXA) bone density testing at a community-based outpatient osteoporosis testing center between 1989 and 1999. Data for all patients were stored in a relational database that was searched for various combinations of T-scores reflecting the nine possible discordant and concordant permutations of normal, osteopenia, and osteoporosis. The WHO diagnostic classification system for osteoporosis categorizes patients into three diagnoses-normal, osteopenia, or osteoporosis-and is based on their T-score. This system was used to determine the patient diagnoses for the various T-score results in the PA L1-L4 lumbar spine and total hip. Concordance was defined as present when the spine and hip T-score placed the patient in the same diagnostic class. Minor discordance was defined as present when the difference between two sites is no more than one WHO diagnostic class. Major discordance was defined as present when one site is osteoporotic and the other site is normal. The results showed that in 56% of the cases the T-score concordance was present between the hip and the spine, 39% of the rime minor discordance was observed, and 5% of the time major discordance was found. Discordance was also classified into five etiologic types: physiologic, pathophysiologic, anatomic, artifactual, and technical. In summary, this data analysis showed that while T-score concordance is most commonly observed, minor discordance is seen in about two of every five patients tested with DXA between the spine and hip sites. Major discordance was uncommon, found in only 1 of every 20 patients tested.