Addition of symptoms to traditional risk factors improves prediction of those at high risk of vertebral fractures: Results of the Vfrac study

Clark, Emma; Khera, Tarnjit; Hunt, Linda; Gooberman-Hill, Rachael; Paskins, Zoe; Peters, Tim; Tobias, Jon; Davis, Sarah

Abstract

Less than a third of osteoporotic vertebral fractures (OVFs) come to clinical attention. This is due to a variety of reasons including inadequate understanding of the clinical triggers necessary to refer high risk individuals for spinal radiographs. Self-reported descriptions of back pain are increasingly recognised as useful clinical information. Purpose: To develop a simple clinical tool to help decide which older women with back pain should have a spinal radiograph, and to identify the additional clinical benefit of including self-reported back pain symptoms. Methods: 1634 women aged 65+ with back pain in the previous four months were recruited from primary care in the United Kingdom. Data were collected through self-completion questionnaires, physical examination and spinal radiographs. Exposure data included descriptions of back pain, traditional risk factors for osteoporosis, basic anthro-pometry and reported height loss. The outcome was the presence/absence of OVFs identified using the Algorithm-Based Qualitative method. Logistic regression models identified inde-pendent predictors of OVFs. AUC for the final model was calculated. AUC was recalculated after removal of self-reported pain descriptors, and compared to that for the final model. Proportions of those identified with OVFs with and without the use of back pain symptoms were identified. Results: Mean age was 73.9 years (range 65.4 to 96.8), and 209 (12.8%) had OVFs. The final Vfrac model comprised 15 independent predictors of OVF, with an AUC of 0.802 (95%CI 0.764-0.840). Removal of self-reported back pain symptoms reduced the AUC to 0.742 (95%CI 0.696-0.788). Without inclusion of back pain symptoms, the Vfrac tool identifies 66.5% of those with OVFs (53.7% with one and 92.5% with more than one). Adding back pain symptoms identifies 72.6% of those with OVFs (60.8% with one OVF and 96.1% with more than one). Sensitivity is increased from 66.5% to 72.5%. Conclusion: The Vfrac clinical tool appears valid and is improved by the addition of self-reported back pain symptoms. It now requires testing to establish real-world clinical and cost-effectiveness. Vfrac takes approximately 5 minutes to complete. The intention of Vfrac will be to help healthcare practitioners in primary care decide if an older woman with back pain is at high risk of an OVF and therefore requires a spinal radiograph to confirm the diagnosis.

Citation

Clark, E., Khera, T., Hunt, L., Gooberman-Hill, R., Paskins, Z., Peters, T., Tobias, J., & Davis, S. Addition of symptoms to traditional risk factors improves prediction of those at high risk of vertebral fractures: Results of the Vfrac study

• Presentation Conference Type: Conference Paper (published)
• Acceptance Date: Feb 9, 2022
• Online Publication Date: Feb 9, 2022
• Publication Date: Feb 9, 2022
• Deposit Date: Jun 20, 2023
• Journal: Journal of Bone and Mineral Research
• Print ISSN: 0884-0431
• Electronic ISSN: 1523-4681
• Publisher: American Society for Bone and Mineral Research
• Peer Reviewed: Peer Reviewed
• Volume: 37
• Issue: S1
• Public URL: https://keele-repository.worktribe.com/output/490091