Differences between dual-energy X-ray absorptiometry (DEXA) normal ranges can lead to patients being characterized as osteoporotic using one range and normal using another. To investigate the diversity of normal ranges used within the UK, a survey of all DEXA sites was carried out with a 60.6% response rate. The effect of the different ranges was evaluated by translating each range to an equivalent range for a Norland XR-26 system and applying the ranges to stratify a representative sample of over 1000 patients into grades of bone density based on percentages of age-matched mean BMD, Z-scores and T-scores. The effect of femoral neck and P/A spine L2-L4 regions was considered both separately and jointly. Large differences between the normal ranges were apparent, which resulted in the classification of the number of patients with a Z-score of less than -2.0 varying by a factor of more than 20 for the femoral neck and more than 3 for the spine. The number of patients defined as osteoporotic by a T-score less than -2.5 varied from none to over one-third of patients for the hip and by a factor of almost 3 for the spine. The exclusion criteria used for construction of the normal ranges varied markedly with none constructed using population-based sampling. Smoothing of normal ranges was carried out by DEXA manufacturers, while local normal ranges made use of raw unprocessed data. There is reason to question the validity of such processing. We recommend the construction of a unified UK normal range applicable to all UK DEXA systems in order to harmonize patient management and care.