Osteoporosis (weak bones) is one of the most common musculoskeletal conditions in older people. Osteoporotic vertebral fractures (broken bones in the back) are particularly important because they highlight people at high risk of breaking more bones. However, less than a third of people with vertebral fractures are correctly diagnosed and treated. The main reason for this is difficulty understanding who should have spinal X-rays.
The Vfrac tool is a simple questionnaire developed for healthcare professionals in GP practices to complete for adults with backpain. The output of Vfrac indicates whether or not they should have a spinal X-ray.
Vfrac has never been tested in a real-world situation. This real-world testing will be a large study involving many GP practices, and will look at whether Vfrac improves the treatment of older people with osteoporosis. Before we can design this study, we need information to plan the size and processes of the study, based on information from CPRD. We will use the CPRD to focus on older people with back pain and identify
1. the numbers attending their GP practice, and the variability across the UK
2. how they are treated, and the variability across the UK
3. how long it takes for them to be investigated and treated, and the variability in this across the UK
The outputs of this research will help understand ‘usual care’ of older people with back pain, and help plan the future testing of the Vfrac tool.
Osteoporotic vertebral fractures (OVF) are the most common fragility fracture in older adults. They are one of the strongest indicators of severe osteoporosis and should result in immediate interventions to reduce risk of future fracture. 12% of older adults have OVFs but 80% are not identified and managed. Identification of those with existing OVFs is one of the most pressing healthcare issues in the field of osteoporosis. To address one component of the complex pathway between occurrence of an OVF and management with medications, we have developed a clinical decision tool called Vfrac. Vfrac takes 5 minutes to complete and identifies those with back pain at high risk of OVF, thereby helping healthcare professionals decide if spinal radiograph is required.
The current plan for the future definitive evaluation of Vfrac is a parallel cluster-randomised trial with randomisation at general practice level. However, before we plan the future trial, we need more information.
The aim of this CPRD-based project is to quantify UK-wide variability in key trial parameters. Variability will be assessed at patient level and general practice level. The study population of interest is adult men and women. Patient level characteristics of interest include demographics, socio-economic status and various relevant comorbidities. Practice-level characteristics include region of the UK, deprivation, size and case-mix.
CPRD data with a single area linkage for general practice level IMD classification is required. Initially we will identify new consultations for back pain and will describe the UK-wide variation by patient-level characteristics. Healthcare utilisation will be calculated by including professional background of healthcare professional who led the initial consultation, subsequent referrals and medication prescription. Data will be described as proportion of consultations with a subsequent referral or prescription. Variability in consultations and subsequent management according to general practice-level characteristics will be calculated using the same techniques.
Primary outcome: Consultations for back pain in older people; professional background of healthcare professional who ran the consultation for back pain; medication prescribing rates for osteoporosis in older people (oral bisphosphonates, denosumab, raloxifene, zoledronate, teriparatide, romosozumab). Codes listed in Appendix A
Secondary outcomes – patient level characteristics within 2-years of index consultation: medication prescriptions (oral bisphosphonates, denosumab, raloxifene, zoledronate, teriparatide, analgesics); referrals to secondary care; referrals to radiology/rheumatology; referrals for spinal radiographs; referrals for DXA scan; referrals for physiotherapy; referrals for community services e.g. exercise or weight loss programmes. See Appendix A plus separate Appendices for analgesic codes.
Secondary outcomes – general practice-level characteristics at time of index consultation: size of general practice; deprivation level of general practice based on areal IMD classification; region of the UK; case-mix of patients in practice (based on age, gender, bisphosphonate prescribing rates, Charlson co-morbidity index of older patients in each practice). See Appendix A - Charlson co-morbidity codes to be provided later.
Emma Clark - Chief Investigator - University of Bristol
Emma Clark - Corresponding Applicant - University of Bristol
Jessica Harris - Collaborator - University of Bristol
Nicholas Turner - Collaborator - University of Bristol
Nicholas Turner - Collaborator - University of Bristol
Practice Level Index of Multiple Deprivation