Bone mineral density (BMD) is measured with random error, but it is used as a surrogate definition of “osteoporosis,” and as an endpoint in clinical trials. A Bayesian model was formulated to address two questions: given an observed level of BMD for an individual, what is the individual's “true” level, and to what extent does an observed BMD change reflect a real change. For an individual, there is a good agreement between observed and “true” BMD values. In individuals with low measured BMD values, however, the 90% confidence interval for the true level is particularly wide, which leads to high (as high as 20%) false-positive and false-negative rates of diagnosis of osteoporosis. In a clinical trial with an overall average increase in BMD of 2%, no conclusion of significant change for an individual could be drawn until an observed increase of at least 5.5% or an observed decrease of at least 7.5%. It is proposed that the current practice of informing individuals about their t- and z-scores should be replaced with a report system in which their osteoporosis probability risk category is conveyed. Also, assessment of change in an individual should take into account the overall change in a population.