Transforming Real-world evidence with Unstructured and Structured data to advance Tailored therapy (TRUST)

项目总结

Project Summary There is a national desire to implement real-world evidence (RWE) within regulatory and clinical pathways as a step toward personalized medicine, improved care, and more efficient care. This will accelerate use of routinely collected data to refine care pathways. By influencing what is approved, reimbursed, and selected for care, RWE will adjust the standard of care. But, adjusting the standard of care can have unintended and dangerous consequences. Bad data allowed into a patient’s electronic health record (EHR) has the potential to hurt one patient. Bad data allowed into regulatory or reimbursement pathways can harm a nation. RWE is often used to support trial recruitment, trial design, and marketing insight. As it is increasingly used to make clinical assertions, there is reason to believe that current approaches may benefit from greater rigor. Claims data often have accuracy below 50%. EHR problem lists often have accuracy below 60%. It is believed that low sensitivity incorporates skew since sicker patients with more touch points in the health system have more complete documentation. This program seeks to study data quality in the context of a potential drug launch. Leaders in the space intend to study data quality while testing a novel and highly rigorous approach to RWE. To achieve the goal of understanding how data quality influences RWE assertions, the proposed project includes innovations in phenotyping, gold standard, accuracy measurement, and enhanced privacy and security. This effort comes at a critical time, as regulators, payers, and providers are increasingly incorporating RWE insights into their decision-making processes. By studying data quality and demonstrating safe approaches to RWE, the country can move forward on solid footing.