Background: Previously developed prediction models of fracture risk have limited
applications in clinical practice. We developed a simplified and convenient fracture risk
prediction model for elderly Korean women.
Methods: A total of 1,440,988 women who underwent a life-transition national examination at
66 years of age between 2011 and 2017 were included and followed up for 3 years. The
incidence of major osteoporotic fracture (MOF) and hip fracture at 1 and 3 years was analyzed.
Data on clinical risk factors such as age, body mass index, bone mineral density, history of falls
within 6 months, past fractures within 3 years, recent fractures within 1 year, and recurrent
fractures were collected. Multivariate logistic regression analysis was used to calculate the odds
ratio (OR) for each risk factor of osteoporotic fractures.
Results: Scores were calculated using the OR of each risk factor for osteoporotic fracture.
Scores were summed and classified into five groups: 0-4, 5-7, 8-10, 11-14, and 15-18 points.
In the subjects with the higher the score, the higher the risk of MOF and hip fracture at 1 and
3 years were observed: the incidence of MOF at 3 years were 3.4%, 5.8%, 12%, 21.3%, and
36.6%, and that of hip fracture at 3 years were 0.13%, 0.26%, 0.59%, 0.73%, and 1.82% in the
groups with SAFE scores of 0-4, 5-7, 8-10, 11-14, and 15-18 points, respectively.
Conclusion: In conclusion, the SAFE can predict imminent fracture risk within 1 or 3 years.