New clinical prediction tools for myeloma
Posted Sep 17, 2021 from phc.ox.ac.uk
- Dr Constantinos Koshiaris has developed clinical prediction models for use in primary care with the aim of accelerating myeloma diagnoses.
- Using the Clinical Practice Research Datalink (GOLD version), a primary care database containing electronic health records for more than 11 million patients in the UK, the team identified the most common symptoms and full blood count results recorded for patients with myeloma.
- The most predictive of these were included in the models they developed and the new tools were validated against a set of test data.
- Decisions made using their prediction models resulted in fewer false positives and more true positives when compared to single tests or symptoms alone.
- By identifying patients at highest risk of myeloma in primary care, these new prediction rules have the potential to reduce diagnostic delays by a substantial amount.
- Further research is now needed to understand more about the feasibility and implementation of this tool in the primary care setting and the impact it will have on the diagnostic pathway and patient outcomes.