It has been well established that raloxifene (RLX) improves bone turnover, increases bone mineral density (BMD), and reduces the risk of fractures. However, it remains obscure how to monitor the therapeutic effects of RLX, while numerous clinical trials for other antiresorptive agents have suggested that greater short-term reductions of bone turnover markers (BTMs) can predict greater increases in BMD and greater reduction in risk of future fractures. The purpose of this study was to investigate associations between short-term reductions of BTMs and subsequent changes of BMD after 1-year treatment with RLX. Seventy-three Japanese postmenopausal women with untreated osteoporosis were selected for this study. Reductions in BTMs [bone-specific alkaline phosphatase (BAP) or serum N-terminal telopeptide of type I collagen (NTx)] after 3 months did not correlate with increases of BMD at any major sites (lumbar spine, femoral neck, total neck, and distal 1/3 radius) either after 6 months or after 12 months. Our results suggest that short-term reductions or 3-month reductions of BTMs with RLX treatment cannot be used to predict increases of BMD. However, this does not directly mean that short-term reductions or 3-month reductions of BTMs with RLX treatment cannot be used to predict risk reduction of future fractures or the ultimate effects of RLX on bone. Further studies with fracture endpoints in longer observation and larger number of patients are warranted to establish how to monitor the therapeutic effects of RLX or early identification of responders or nonresponders to RLX treatment.