A randomized placebo controlled study was undertaken to evaluate the effect of ipriflavone (IP) against the bone loss in premenopausal ovariectomized women and postmenopausal women. Thirty-seven Japanese women who underwent premenopausal bilateral ovariectomy within 3 months (early stage group) and 52 Japanese women who were ovariectomized or who had undergone menopause more than 3 years before the start of the study (late stage group) were enrolled. The patients were randomly allocated into two groups: those who received IP (600 mg/day) and those who received placebo. The bone mineral density (BMD) of the lumbar vertebrae was measured by dual energy X-ray absorptiometry, and the markers of bone metabolism were measured at the same time that BMD was measured. In the early stage group, the IP group showed a 6.7% decrease in BMD from baseline levels, whereas the placebo group showed a 10.7% decrease (P < 0.01) at 12 months of treatment, and 7.1% and 12.6% decrease at 24 months of treatment, respectively (P < 0.01). In the late stage group, there was a 0.3% increase in BMD in the IP group and a 2.3% decrease in the placebo group at 6 months of treatment (P < 0.01), and similar changes were seen at 18 months (1.4% increase and 3.9% decrease; P < 0.01). IP suppressed bone loss compared with placebo, however, did not prevent acute bone loss in the early stage following ovariectomy. The effect of IP alone on bone loss in the early stage is not sufficient to reduce the risk of osteoporosis in later life.