The sub-segmental analysis of dual energy X-ray absorptiometry (DXA) scans from scoliotic vertebrae has established that there are differences in bone mass between the concave and convex sides of the vertebrae. Furthermore, these differences persisted in patients with low bone mass and were related to the geometry and applied loads, suggesting that this is a good model of bone adaptation in response to external stimuli. The goal of this study was to characterize the response of the human scoliotic spine to anti-resorptive treatments and estrogen withdrawal on the concave and convex sides of the spine. A total of 576 vertebrae (199 no treatment, 214 bisphosphonate, 69 estrogen and 94 estrogen withdrawal) were analyzed from 167 postmenopausal, Caucasian women. An analysis of variance (ANOVA) was used in conjunction with post-hoc Tukey tests to examine the effects of concavity, treatment group, and age. We found that the average change in BMD per year was greater than zero on the concave and convex sides with the exception of the estrogen withdrawal group. Discontinuing estrogen therapy caused patients to maintain bone mass on the concave side, but lose substantial bone density on the convex side. A differential response was also observed with respect to age. Patients younger than 60 exhibited a decrease in total BMD per year concomitant with a small degree of straightening, while those who were 60 or over demonstrated an increase in bone mass and a slight increase in the deformity. Based on these data, it is clear that the differences in BMD between the concave and convex sides of the vertebrae are not simply a result of the deformity, but more likely due to bone accretion. Further study is needed to elucidate the relationship between biomechanical forces and the adaptive response in the spine as a function of time.