Ensuring Accuracy with Data in Oncology Research

In a recent video interview with Applied Clinical Trials, C.K. Wang, MD, chief medical officer, COTA, discussed a recent study he led on leveraging real-world data (RWD) to improve oncology reporting. Additionally, Wang highlighted recent trends and challenges in the oncology research space involving the use of RWD as well as real-world evidence (RWE).

ACT: What are your greatest takeaways from the study findings? How do you hope they impact the industry moving forward?

Wang: The takeaway from our study is that COTA’s composite mortality variable is very complete and very accurate. I would say that of the 21,500 records that we looked at across seven different cancers, and this includes acute myeloid leukemia, chronic lymphocytic leukemia, diffuse arch basil lymphoma, follicular lymphoma, marginal zone lymphoma, multiple myeloma, and myelodysplastic syndrome, we found our death data to carry a very high sensitivity and specificity at 87.8% and 95.7% as well as a very high positive predictive value and a very high negative predictive value as well at 90.9% and 94.1% respectively. I think, more importantly, what we found in our study is that mortality data, as documented in the medical record systems, on its own, has a relatively low sensitivity rate at 45.7%. Out of all this, it is my hope that our study findings will encourage everyone who's using real-world data to really scrutinize the completeness and the accuracy of the death data that they see, and for every provider of real-world data—whether it be another company, be a healthcare system, or oncology provider—to really step up to the plate and share the performance of their death information.