Osteoporosis can be a major debilitating, expensive, long term, irreversible condition, and breast cancer survivors are particularly at risk for osteoporosis. As a result of some chemotherapy agents, many of these women experience premature ovarian failure. At least 60% of women with breast cancer have estrogen receptor positive tumor status; thus, they are not candidates for hormone replacement therapy. Without estrogen, bone loss occurs rapidly the first five years of menopause and continues over time but at a slower rate. The purpose of this study is to test whether strength/weight training exercises enhance the effectiveness of risedronate (5 mg/day), calcium (1200 mg/day), and vitamin D (400 IU/day) in improving bone mineral density (BMD) in post-menopausal breast cancer survivors. The sample will be 218 subjects recruited within a 100 mile radius of r sites across Nebraska (Omaha, Lincoln, Kearney and Scottsbluff). Post-menopausal women with a history of stage O (in situ), stage I or II breast cancer, with a BMD DEXA T-score of -1.0 SD or lower at any of 3 sites (hip, spine, forearm) will be stratified by time since menopause (5 yrs or less: > 5 years) and randomly assigned to one of two treatment groups (G1 and G2), with approximately 109 per group. Differences in tamoxifen, smoking, intake of calcium, and body mass index (BMI) between the two groups at baseline will be examined; if differences exist they will be controlled statistically. Both groups will receive risedronate, calcium and vitamin D; G1 also will receive strength/weight training exercises for upper and lower extremities and spine. Facilitative strategies based on Bandura's (1997) Self-Efficacy Theory are designed and used to encourage long-term adherence for both groups. The multi-component intervention is 24 months with follow-up at 30 and 36 months. The primary outcome measure is BMD of the hip, spine and forearm (via DEXA): secondary outcome measures are muscle strength and fractures. The impact of the multi-component intervention on outcomes will be assessed using General Estimating Equation methodology Measurement of outcomes will occur at 6 and/or 12 month intervals through 36 months. In addition, relationships will be examined between level of confidence in goal accomplishment and adherence to intervention components over time. This study may provide evidence of an effective alternative to HRT for treatment of osteoporosis in breast cancer survivors who are not candidates for HRT.