Modeling fracture risk using bone density, age, and years since menopause

Introduction: Preventive strategies are essential for reducing the incidence of osteoporosis and its consequences. However, simple algorithms that predict an individual's future risk of fractures are scarce. The purpose of the study was to define a clinical decision aid that enables physicians to project an individual's lifetime fracture risk and therefore institute preventive therapies. Methods: A predictor equation for bone loss was developed using bone mineral density (BMD), age, years since menopause, and weight. This was applied to normal and osteoporotic women, ages 40-80 years (n = 117) screened for osteoporosis studies. Results: A spinal BMD cutoff of 0.86 gm/cm(2) had a sensitivity of 90% and a specificity of 60% for detecting subjects with vertebral fractures and was therefore defined as a high-risk BMD.Using the parameter estimates from the above equation and an individual's clinical data, we derived prediction curves-to forecast the age at which that individual would reach the above defined high-risk BMD, and therefore that person's expected number of remaining life-years at high risk for fractures. Conclusions: This study proposes a conceptual framework for the development of a clinical decision aid to provide guidelines for the prevention of osteoporosis. A longitudinal study that incorporates other variables such as prevalent fractures and biochemical markers of bone turnover would further validate this model and enhance its application.