OBJECTIVES: To investigate the cost-effectiveness of universal bone densitometry in women aged 65 and older combined with alendronate treatment for those diagnosed with osteoporosis (femoral neck T-score < or = -2.5). DESIGN: A Markov model with a lifetime time horizon and eight health states (no fracture, distal forearm fracture, radiographic (but clinically inapparent) vertebral fracture, clinical vertebral fracture, hip fracture, hip and vertebral fracture, and other fractures), using the societal perspective. SETTING: Women living independently and those in nursing homes. PARTICIPANTS: Caucasian women aged 65, 75, 85, or 95. INTERVENTION: Bone densitometry of the hip, with 5 years of alendronate therapy for those found to have osteoporosis versus no intervention (densitometry or drug therapy). MEASUREMENTS: Lifetime accumulated quality adjusted life years (QALYs), costs, and incremental cost-effectiveness ratios. RESULTS: The cost per QALY gained for the screen-and-treat strategy was 43,000 dollars per QALY gained for 65-year-old women and 5,600 dollars per QALY gained for 75-year-old women. For 85- and 95-year-old women, the screen-and-treat strategy was cost saving. Sensitivity analyses showed that the screen-and-treat strategy was cost-effective even under assumptions of reduced adherence to drug therapy, reduced fracture reduction benefit from alendronate therapy, or reduced QALYs saved by preventing fracture. CONCLUSION: Universal bone densitometry combined with alendronate therapy for those found to have osteoporosis is highly cost-effective for women aged 65 and older and may be cost saving for ambulatory women aged 85 and older (whether independently living or residing in nursing homes).