Bradley Gordon Hammill

I believe that we can and should leverage electronic health data for clinical research and quality assessment. While not the answer to all questions, real-world data (RWD) like health care claims data and electronic health record (EHR) data contain a wealth of information.

I’m currently working to bring RWD to many different projects:
- FDA Sentinel Initiative: I am responsible for the team that transforms and manages the 100% Medicare and 100% Medicaid claims data, within the CMS Virtual Research Data Center (VRDC), for the Sentinel distributed database. FDA uses this database to monitor its regulated medical products.

- Pragmatic Clinical Trials: I am working with the Data Coordinating Centers for multiple trials to obtain and incorporate RWD for participants into clinical trial databases. This is one way we hope to make trials more pragmatic and lower cost. Good examples of this work include the ADAPTABLE [ClinicalTrials.gov Identifier: NCT02697916] and PREVENTABLE [NCT04262206] trials. Both are pragmatic mega-trials relying on RWD for outcomes. Other trials leveraging real-world data that I am or have been involved with include HARMONY [NCT02465515], CEPIO-D [NCT03749629], and PRECIDENTD.

- NIH Cohort Studies: I have partnered with multiple investigators at other universities who run cohort studies sponsored by the National Heart, Lung, and Blood Institute to link Medicare claims data for their study participants. These studies include the Framingham Heart Study, the Strong Heart Study, the Jackson Heart Study, the Cardiovascular Health Study (CHS), and MESA (Multi-Ethnic Study of Atherosclerosis). 

- Health System Performance: I am working closely with Duke University Health System quality leadership to explore and benchmark performance using the same claims data that CMS, US News & World Report, and other ratings groups rely on. We have a keen interest in understanding the potential role of neighborhood deprivation on clinical outcomes and how this may affect hospital performance estimates.

Of course, for healthcare research, it is not easy to extract value out of RWD. Meaningful, actionable results require rigorous study design, a deep knowledge of data provenance, and appropriate statistical methods. I have a commitment to quality in all of these areas.