Little is known about differences in the therapeutic efficacy of denosumab in subjects with and without rheumatoid arthritis (RA). This study compares the changes in bone mineral density (BMD) between RA patients and controls without RA who had been treated with denosumab for 2 years for postmenopausal osteoporosis. A total of 82 RA patients and 64 controls were enrolled, who were refractory to selective estrogen receptor modulators (SERMs) or bisphosphonates and completed the treatment of denosumab 60 mg for 2 years. The efficacy of denosumab in RA patients and controls was assessed using areal BMD (aBMD) and T-score of the lumbar spine, femur neck, and total hip. A general linear model with repeated measures analysis of variance was used to determine differences in aBMD and T-score between 2 study groups. No significant differences in percent changes in aBMD and T-scores by denosumab treatment for 2 years at the lumbar spine, femur neck, and total hip were evident between RA patients and controls (P > .05 of all), except T-score of the total hip (P = .034). Denosumab treatment equally increased aBMD at the lumbar spine and T-scores at the lumbar spine and total hip between RA patients and controls without statistical differences, but RA patients showed less improvement in aBMD at the femur neck (ptime*group = 0.032) and T-scores at the femur neck and total hip than controls (ptime*group = 0.004 of both). Changes in aBMD and T-scores after denosumab treatment in RA patients were not affected by previous use of bisphosphonates or SERMs. Differences of T-score at the femur neck among previous bisphosphonate users and aBMD and T-score at the femur neck and T-scores at the total hip were evident. This study revealed that 2 years of denosumab treatment in female RA patients achieved comparable efficacy on BMD to controls at the lumbar spine, but showed somewhat insufficient improvement at the femur neck and total hip.