Patients on long-term oral corticosteroids have an increased risk of low bone mass and fragility fractures. Fracture risk rises soon after commencement of corticosteroid therapy and it is possible that these agents adversely influence bone architecture disproportionately to their effect on bone mass. The best means of assessing bone status in patients using corticosteroids remains uncertain, but quantitative ultrasound of the calcaneus may provide evidence of microarchitectural changes not detected by dual-energy X-ray absorptiometry (DXA). Patients with Crohn's disease have an increased risk of low bone mineral density (BMD), the etiology of which is multifactorial but includes corticosteroid use. We studied 118 consecutive patients with Crohn's disease, 21 of whom used continuous oral corticosteroids, 70 of whom were intermittent users, and 27 who had never used the drug. All patients received DXA of the lumbar spine, hip and calcaneus and quantitative ultrasound (QUS) of the calcaneus. The different techniques were compared using a femoral neck T-score < or = -1.5 as the threshold of corticosteroid-induced osteoporosis. When compared with the femoral neck T-score, there were no significant differences between the predictive values of lumbar spine DXA, calcaneal DXA or calcaneal QUS to identify low femoral neck BMD. However, the absolute T-score required to give similar discriminatory capacity to femoral neck T-score varied substantially (T= -0.81 to -1.5) between the different measurement techniques and sites.