Theses Doctoral

Improving osteoporotic fracture prediction and identification of high risk individuals

Yang, Jingyan

As the population ages, osteoporosis-related fractures represent a major and costly public health concern that is associated with increased morbidity and mortality in the United States, particularly in postmenopausal women [1]. A Surgeon General’s Reported has pointed out the importance of early diagnosis and appropriate treatment of bone diseases [2]. Treatment is typically indicated based on a bone mineral density (BMD) value of osteoporosis or a prior fragility fracture. Of note, many fragility fractures occur in postmenopausal women with non-osteoporotic BMD values. More significantly, a prior fragility fracture, particularly a prevalent vertebral fracture (VF), is a strong predictor for the elevated risk of subsequent fractures [3-7]. Hence, early identification of VFs is of great importance for initiating pharmacological therapy in women who may not otherwise be treated in order to prevent future fractures.
VFs are often subclinical which require additional efforts to identify these fractures [8]. Lateral dual-energy X-ray absorptiometry (DXA) scanning of the entire spine for vertebral fracture assessment (VFA) has been proposed by the International Society for Clinical Densitometry (ISCD) as an alternative of x-ray for the diagnosis of VFs [9]. Also, the National Osteoporosis Foundation (NOF) has provided guidelines when VFA should be performed [10]. However, the effectiveness of VFA as a screening tool for the identification of prevalent VFs is unclear and the cost-effectiveness of VFA is unknown, both limiting the implementation of VFA into routine care. Therefore, I conducted a systematic review and meta-analysis, the results of which have shown that the weighted pooled prevalence of VFA-detected VFs in asymptomatic women was 28%. Given that VFA is effective, I further evaluated the cost-effectiveness of VFA as a screening tool to reduce future osteoporotic fracture risk in U.S. postmenopausal women. The reference-case analysis has shown that VFA has the greatest cost-saving when the screening is initiated at age 65 years and with follow-up screening every 5 years. These findings support the NOF guidelines for the diagnostic use of VFA.
There are some women with increased risk for secondary osteoporosis who may not be eligible for BMD or VFA screening due to their younger age, for example, women with human immunodeficiency virus (HIV) infection. Therefore, an accurate fracture risk assessment tool is an important component in the management of bone health in HIV-infected women. The interests of validating the predictive accuracy of FRAX® (a widely accepted fracture risk prediction tool in general population [11]) arose from the reported poor performance of FRAX in older HIV-infected men [12]. I validated FRAX performance in HIV-infected women using the Women’s Interagency HIV study (WIHS), suggesting that FRAX also underestimated fracture risk in HIV-infected women, but improved with the addition of DXA parameters.
The results of the above studies demonstrate the potential role of VFA in reducing future fracture risk in women with prevalent VFs and the cost-effectiveness of incorporating VFA into routine screening for osteoporosis in postmenopausal women. Data were also provided for improving the fracture prediction in people with secondary osteoporosis using HIV infection as a model. These data may inform clinicians, policy makers and insurers on the benefit of including disease specific risk factors for fracture prediction and VF identification tools in the fight to prevent osteoporosis related fractures.


  • thumnail for Yang_cumc.columbia_0054E_10049.pdf Yang_cumc.columbia_0054E_10049.pdf application/pdf 1.28 MB Download File

More About This Work

Academic Units
Thesis Advisors
Nieves, Jeri W.
Dr.P.H., Mailman School of Public Health, Columbia University
Published Here
April 5, 2019