The last week of September, I had the opportunity to be a chair on the EHR/MDR session at the 2016 CDISC International Interchange in Maryland. I always enjoy being involved with the CDISC Interchange; it is inspiring to see all the people who are passionate about submitting quality standardized data to the FDA to make better lives for us all.
The CDISC 2015 Annual Report has the message, "When Clinical Research is Done in Silos, We Can't Harvest Cures." When we don't share findings in clinical research, it's as if billions of dollars get buried. When that happens, we all lose out on finding cures to diseases that affect the people we love.
Ever since the FDA stated that for studies starting after December 16, 2016, sponsors and CROs are required to submit data in compliance with CDISC standards, the CDISC Interchange has grown and is now the leading industry meeting for clinical data standards.
Here's what excited me at the Interchange this year:
The FDA Commissioner's Update
FDA Commissioner Dr. Robert Califf's messages were solid, implementable directives for all of us:
- Previously, investigators had the data and wrote the results. We need a network where the data is shared, similar to the FDA National Evaluation System for health Technology (NEST) project that collects data on device performance and device adverse events.
- The biggest problem in clinical research isn't the data; it's getting doctors to find time to talk to their patients about research options.
- It doesn't make sense for monitors to fly around from site to site; we have to have automation. Providing centralized monitoring or minimizing site visits via risk-based monitoring are paths to this automation.
- We've gone from no electronic health records (EHRs) to everybody having an EHR.
- Genome editing is coming, and we need to know how to ensure it is done safely.
- mHealth is going from "clunky" to "real," so we need to figure out how to assess the value of this data. Using all of the data collected via mHealth systems doesn't make sense, so we need to learn how to utilize collection to obtain the data needed for research and determine the steps needed to ensure this is quality data.
- The FDA now has data on 200 million Americans in Sentinel, its national electronic system for monitoring the safety of medical products after they have reached the market, and can mine this data to look for subsets of interest.
- Regarding a data warehousing effort for post-marketing device studies, the FDA says "The Food and Drug Administration Safety and Innovation Act of 2012 explicitly requires expansion of FDA's Sentinel System to include medical devices. The FDA envisions using similar distributed data sources, which would include EHRs and registries, for the medical device Sentinel System to help maintain patient privacy."
Metadata Repositories (MDR)
Metadata, or "data that describes other data," are stored in metadata repositories. These repositories are finally maturing and being used in useful ways. They can store data, metadata and metrics for reuse and standardization. MDRs provide a way to automate the building of CDISC datasets, validating SDTM datasets and storing additional company standards.
Companies like Roche/Genentech shared how they are using their repositories to let their data managers easily design their EDC build by presenting them with the standards first, then letting them add additional fields after those additional fields are approved by the standards committee. Brilliant!
CDISC's Shared Health And Research Electronic library (SHARE) is a solution "for developing, integrating and accessing CDISC standards metadata electronically." CDISC allowed us to be really nerdy and query the SHARE metadata repository library using their RESTful API.
To translate, we typed queries in the URL portion of the web browser, and our browser page displayed results from SHARE. We could ask to see variables in a domain, like adverse events, or look up valid values for a codelist. It was super fun and easy. I can just imagine the benefits of having this much information available in a structured way. Many companies are already using SHARE to build CRF forms, validate datasets and populate metadata tables.
eSource Stakeholders Meeting
On the last day of the interchange, there was a Healthcare Link workshop and eSource Stakeholders Meeting. The eSource Stakeholders Meeting had many of the team leads, and we were able to plot a plan forward to publish a whitepaper, finish the primer we have been working on and incorporate some real demo projects into our landscape. For more details, see my blog post "CDISC eSource Stakeholders Plan to Supply Pharma with Path Forward for Collection of EHR and Other eSource Data."
The CDISC Interchange is the highlight of my year. The path forward to provide better and safer healthcare lies within the data; we just need to be able to get it all in one place and make use of it. Every year at CDISC, we get a step closer to being able to determine unique optimal treatment plans by reviewing the data of the millions of patients, searching for subsets that match our individual scenarios and determining the most successful course of action.