To study the effect of long-term treatment with raloxifene, compared with placebo, on the rate of new vertebral fractures in osteoporotic postmenopausal women with and without existing vertebral fractures.

Inclusion Criteria:

• Ambulatory postmenopausal women free of severe or chronically disabling conditions, have a life expectancy of at least 5 years, be expected to remain ambulatory throughout the entire study, and be expected to return for follow-up visits.
• Women who have had their last menstrual period at least 2 years before beginning the study.
• Women who have no language barrier, are cooperative, and who give informed consent before entering the study
• Substudy 1:Femoral neck or lumbar spine BMD measurements 2.5 or more standard deviations below normal peak bone mass for healthy, premenopausal women (T-score greater then or equal to 2.5).
• Substudy 2:Either at least one moderate or at least two mild vertebral fractures in the presence of low BMD (as specified above) or at least two moderate vertebral fractures, regardless of BMD.

Exclusion Criteria:

• Patients with known current bone disorders other than primary osteoporosis, such as hyperparathyroidism, Paget's disease, renal osteodystrophy, or osteomalacia
• Patients experiencing clinically severe postmenopausal symptoms at the beginning of the study that require estrogen-replacement therapy
• Patients with known, suspected, or history of carcinoma of the breast or estrogen-dependent neoplasia
• Patients who have had any history of cancer within the previous 5 years
• Patients with abnormal uterine bleeding
• Patients with a history of deep venous thrombosis, thromboembolic disorders, or cerebral vascular accident within the past 10 years except for patients with a history of deep venous thrombosis due to accidents
• Patients who have endocrine disorders requiring pharmacologic therapy except for type II diabetes
• Patients who are not biochemically euthyroid or who have had changes in thyroid replacement therapy in the 2 months before the start of the study.
• Patients with acute or chronic liver disease
• Patients who have impaired kidney function
• Patients with active renal lithiasis
• Patients with known, severe untreated malabsorption syndromes
• Patients with pathologic fractures (both substudies) or patients in Substudy II all of whose vertebral fractures are clearly a result of automobile accidents or other severely traumatic accidents
• Patients in whom satisfactory baseline thoracic and lumbar x-ray views cannot be obtained
• Patients with less than two lumbar and less than four thoracic vertebrae that are unfractured and evaluable for incident fractures
• Treatment with therapeutic doses of any of the following medications more recently than 6 months before beginning the study: Androgen, Calcitonin, Estrogen, Progestin
• Treatment with therapeutic doses of systemic corticosteroids for more than 1 month during the 12 months before beginning the study.
• Patients who have received therapeutic doses of fluorides
• Patients who have received bisphosphonate therapy for more than 14 days during the past 18 months or who have received any bisphosphonate therapy within the last 6 months before beginning the study.
• Patients requiring high-dose heparinization (>7500 U/day) at study entry for a total period of time that will presumably exceed 6 months
• Patients being treated with 50,000 IU or more of vitamin D once weekly more recently than 3 months before beginning the study will be excluded.
• Current systemic treatment with any of the following medications at the beginning of the study: Lithium, Anticonvulsants, regular use of phosphate-binding antacids.