Modelling future bone mineral density: Simplicity or complexity?

BACKGROUND: Osteoporotic fractures are a major global public health issue, leading to patient suffering and death, and considerable healthcare costs. Bone mineral density (BMD) measurement is important to identify those with osteoporosis and assess their risk of fracture. Both the absolute BMD and the change in BMD over time contribute to fracture risk. Predicting future fracture in individual patients is challenging and impacts clinical decisions such as when to intervene or repeat BMD measurement. Although the importance of BMD change is recognised, an effective way to incorporate this marginal effect into clinical algorithms is lacking. METHODS: We compared two methods using longitudinal DXA data generated from subjects with two or more hip DXA scans on the same machine between 2000 and 2018. A simpler statistical method (ZBM) was used to predict an individual's future BMD based on the mean BMD and the standard deviation of the reference group and their BMD measured in the latest scan. A more complex deep learning (DL)-based method was developed to cope with multidimensional longitudinal data, variables extracted from patients' historical DXA scan(s), as well as features drawn from the ZBM method. Sensitivity analyses of several subgroups was conducted to evaluate the performance of the derived models. RESULTS: 2948 white adults aged 40-90 years met our study inclusion: 2652 (90 %) females and 296 (10 %) males. Our DL-based models performed significantly better than the ZBM models in women, particularly our Hybrid-DL model. In contrast, the ZBM-based models performed as well or better than DL-based models in men. CONCLUSIONS: Deep learning-based and statistical models have potential to forecast future BMD using longitudinal clinical data. These methods have the potential to augment clinical decisions regarding when to repeat BMD testing in the assessment of osteoporosis.

Bone mineral density and fractures in patients with rheumatoid arthritis: the DXA-HIP project.

Objectives: RA is a chronic disabling disease affecting 0.5-1% of adults worldwide. People with RA have a greater prevalence of multimorbidity, particularly osteoporosis and associated fractures. Recent studies suggest that fracture risk is related to both non-RA and RA factors, whose importance is heterogeneous across studies. This study seeks to compare baseline demographic and DXA data across three cohorts: healthy controls, RA patients and a non-RA cohort with major risk factors and/or prior major osteoporotic fracture (MOF). Methods: This is a cross-sectional study using data collected from three DXA centres in the west of Ireland from January 2000 to November 2018. Results: Data were available for 30 503 subjects who met our inclusion criteria: 9539 (31.3%) healthy controls, 1797 (5.9%) with RA and 19 167 (62.8%) others. Although age, BMI and BMD were similar between healthy controls, the RA cohort and the other cohort, 289 (16.1%) RA patients and 5419 (28.3%) of the non-RA cohort had prior MOF. In the RA and non-RA cohorts, patients with previous MOF were significantly older and had significantly lower BMD at the femoral neck, total hip and spine. Conclusion: Although age, BMI and BMD were similar between a healthy control cohort and RA patients and others with major fracture risk factors, those with a previous MOF were older and had significantly lower BMD at all three measured skeletal sites. Further studies are needed to address the importance of these and other factors for identifying those RA patients most likely to experience fractures.