Many randomized controlled trials (RCTs) have investigated drug treatment for women at high risk of fracture, with a reduction in fracture risk as their end point. There has also been progress in identifying women at the highest risk of fractures. The most important clinical determinant contributing to the clinical decision of initiating and choosing drug therapy for fracture prevention is a woman's fracture risk, which, in RCTs, was determined by menopausal state, age, bone mineral density, fracture history, fall risks and glucocorticoid use. Women with secondary osteoporosis were excluded, except in studies of glucocorticoid use. A second determinant of drug therapy is the evidence for fracture prevention in terms of spectrum (vertebral, nonvertebral and/or hip fractures), size and speed of effect. In the absence of head-to-head RCTs with fracture risk as the end point, however, the efficacy of antifracture drugs cannot be directly compared. Other determinants include the potential extraskeletal benefits and safety concerns of the drug, patient preferences and reimbursement issues.