Investigators examined whether various risk assessment tools were effective at predicting hip fractures in patients aged 80 years and older.

This demographic accounts for 70% of hip fractures in the United States.

In the recent study, JAMA Network Open, the investigators compared the performance of the Fracture Risk Assessment Tool (FRAX), the Garvan Fracture Risk Calculator, and femoral neck bone mineral density (FNBMD) tool in predicting 5-year hip fracture risk in 8,890 patients aged 80 years and older. They noted that the patients had a mean age of 82.6 years and were predominantly White (88.1%) and female (55.2%).

After a follow-up of 5 years, the investigators found that 6.5% (n = 321) of the female patients and 3.1% (n = 123) of the male patients experienced hip fractures. Additionally, 16.7% (n = 818) of the female patients and 23.1% (n = 921) of the male patients died before experiencing hip fractures.

Among the female patients, the area under the receiver operating characteristic curve (AUC) was 0.69 for FRAX, 0.69 for Garvan, and 0.72 for FNBMD; whereas among the male patients, the AUC was 0.71 for FRAX, 0.76 for Garvan, and 0.77 for FNBMD.

The investigators indicated that the FRAX tool modestly underestimated the risk of hip fractures in intermediate-risk patients, and the Garvan tool significantly overestimated the risk of hip fractures in high-risk patients.

The FRAX and Garvan tools both incorporated FNBMD and various clinical risk factors to estimate hip fracture probability. The investigators discovered that these tools did not significantly improve hip fracture discrimination compared with FNBMD in this patient population. Further, measurement of the femoral neck bone mineral density using FNBMD provided comparable, if not better, discrimination of hip fracture risk in the patients.

Additional analyses were conducted, excluding participants taking bisphosphonates and those with missing data on parental hip fracture history. The findings remained consistent.

The investigators emphasized that the large cohort size and comprehensive follow-up for hip fractures and mortality may have enhanced the study's reliability. The use of radiographic reports to confirm hip fractures also may have improved the accuracy of the data. However, the predominantly White study population may have limited the generalizability of the findings to other racial and ethnic groups. Additionally, the study did not include institutionalized elderly adults, who may have had different risk profiles.

The findings suggested that in elderly patients, femoral neck bone mineral density may offer a sufficient method for hip fracture risk assessment. Further research may be needed to confirm the results of the study in more diverse populations and settings. 

Conflict of interest disclosures can be found in the study.