Repeat Bone Mineral Density Screening Measurement and Fracture Prediction in Older Men: A Prospective Cohort Study

Abstract

Context: Whether repeated bone mineral density (BMD) screening improves fracture prediction in men is uncertain.

Objective: We evaluated whether a second BMD 7 years after the initial BMD improves fracture prediction in older men.

Methods: Among 3651 community-dwelling men (mean age 79.1 years) with total hip BMD at baseline and Year 7 (Y7), self-reported fractures after Y7 were confirmed by radiographic reports. Fracture prediction assessed using Cox proportional hazards regression and logistic regression with receiver operating characteristic curves for models based on initial BMD, BMD change, and the combination of initial BMD and BMD change (combination model).

Results: During an average follow-up of 8.2 years after Y7, 793 men experienced ≥ 1 clinical fractures, including 426 men with major osteoporotic fractures (MOF) and 193 men with hip fractures. Both initial BMD and BMD change were associated with risk of fracture outcomes independent of each other, but the association was stronger for initial BMD. For example, the multivariable hazard ratio of MOF in the combination model per 1 SD decrement in BMD was 1.76 (95% CI 1.57-1.98) for initial BMD and 1.19 (95% CI 1.08-1.32) for BMD change. Discrimination of fracture outcomes with initial BMD models was somewhat better than with BMD change models and similar to combination models (AUC value for MOF 0.68 [95% CI 0.66-0.71] for initial BMD model, 0.63 [95% CI 0.61-0.66] for BMD change model, and 0.69 [95% CI 0.66-0.71] for combination model).

Conclusion: Repeating BMD after 7 years did not meaningfully improve fracture prediction at the population level in community-dwelling older men.

Keywords: bone mineral density; fracture risk; older men.

Figure 1.

Participant flow diagram.

Figure 2.

Receiver operating characteristic curves for models* predicting (A) major osteoporotic fracture (MOF), (B) any clinical fracture and (C) hip fracture. *Models based on initial total hip BMD, annualized percent total hip BMD change and the combination of initial total hip BMD and annualized percent total hip BMD change. All models are adjusted for age, race, study enrollment site, prior fracture, fall in past year, multimorbidity, physical activity, body mass index and weight change.