It had been a while since we checked in with Rhonda Facile—a renewable energy source who leads one of the most cutting-edge projects in the pharmaceutical industry. She’s the Clinical Data Acquisition Standards Harmonization (CDASH) project director at the Clinical Data Interchange Standards Consortium (CDISC).
CDASH is an effort to get clinical trial data standardized from the moment it is collected. The first iteration of the standard was released earlier this fall. It’s one of the major technical milestones for 2008. Shoulda been on CNN.
In the spirit of transparency, CDISC has also published a 290-page spreadsheet with the 1,800 public comments on the new standard, along with the organization’s response to each item. Addressing the comments required heavy lifting and (because of participants all over the globe) a few conference calls in the wee hours.
The CDASH project has a history that we sketched in an earlier article. The effort acquired new momentum with 2006 backing from Janet Woodcock and the FDA’s Critical Path program. That CDASH was finished just two years later is a miracle, considering the complexity involved in a key element of a $50 billion clinical research industry. It’s also a testament to the CDISC leadership, Facile, sixteen core volunteers and some 200 additional participants.
Some readers may now be a bit antsy or agitated at the prospect of being forced to jettison their favorite case report forms (CRF). Worry is good, in this case. The humble case report form could change dramatically in years to come.
The philosophy behind CDASH is simple. It is not an attempt to dictate the appearance, layout or any other cosmetic aspect of a CRF. That remains at a sponsor’s discretion. Rather, CDASH is an attempt to standardize the underlying structure of a trial’s data and meta-data, and to push that data into sixteen categories called “domains,” ranging from adverse events to demographics.
It turns out that despite the multiplicity of types of trials, and data in those trials, there are fundamental elements used in all of them. Facile’s team collected actual CRFs, studied the frequency of various data elements, and codified it all. Its final case report form standard, in short, reflects the industry’s actual practice and should help it spin its wheels less in designing new CRFs. “We have more in common than we actually think,” says Facile. “We’re on the very beginning of some really exciting things happening in how we collect data.”
Altering which data will be gathered on which form across the industry is a potentially transformative shift. The CDASH team, after much deliberation, decided it was superfluous to have the physical examination (PE) form record medical history information or baseline conditions. That data is more appropriately recorded on other forms. The PE form will only be used to record whether the exam was done at all. Another change applies to inclusion/exclusion data; that CDASH form will only record criteria that a patient did not meet.
If the new standard were adopted by a portion of the industry, a more unified, predictable structuring of CRFs could significantly expedite clinical trials and the exchange of data between sponsors and every other participant in the process—technology vendors, regulatory agencies and contract research organizations (CRO). Less time would be needed to fiddle with how data should be labeled or stored in a database, giving the people involved more time to worry about important stuff.
It’s worth noting that CDASH was designed to pay dividends outside the world of regulatory agencies and sponsor companies. From the start, Facile and her colleagues wanted to help clinical sites that are obliged to use a mind-numbing number of slightly different forms that demand almost (but not quite) the identical information recorded on some other form in some other trial.
Reducing site confusion, weariness and disenchantment—especially among clinical research associates—was an organizing principle. Says Facile, a former CRA herself: “We need to free up clinical sites. This is what they need. This is how eCRFs should be filled in. Just being able to do that will save a lot.”
End In Mind
Crucially, the CDASH data are also pre-configured for analysis and exporting to regulators in the study data tabulation model (SDTM) format, another CDISC project. It governs data when submitted for regulatory review. SDTM is not yet a statutory mandate but is on a short path to becoming one. “This will be a huge help to people who are setting up the database on the back end,” says Facile.
The bottom line? In organizations that adopt CDASH, there should be a smaller number of forms that have overlapping data that must be reconciled, cleaned or fussed over. “We can cut down on the cross-checking, the duplication on the CRFs,” says Facile.
So instead of convening a committee or calling a meeting to discuss how to record a patient’s sex (not gender, as Facile and her team point out) or concomitant medications—bam! It’s done. CDASH has already delineated which data should go where. In effect, it’s developed best practices for eCRFs that are acceptable to industry and regulatory agencies in Japan, Europe and the U.S.
FDA On Board
Indeed, regulatory insight was part of the process of formulating the new standard, she says. All three ICH regions supplied input. “The FDA was part of the collaborative group,” says Facile. “They have been part of it from the very beginning. They’ve also reviewed the documents in all four periods where we had a public review. They have been an active participant.”
Along the way, the CDASH team eliminated a few sacred cows like a general comment form. Some sponsors like such a blank box. But it can lead to unresolved questions and misplaced data as a trial progresses. Says Facile: “The general recommendation is to do away with that form.” There is a similar view of the limited usefulness of a protocol deviation form; such data can be better collected in other places, the CDASH team believes.
For CDISC partisans keen to participate in a more efficient future, designing new CRFs to fit the CDASH format is a no-brainer. For others, with emotional attachment to particular data fields or CRF designs, the migration will be more painful.
Wanted: Test Subjects
What would Facile say to a skeptic who was determined to use a cherished, company-specific CRF? “I would still say, [CDASH] is worth a look,” she says. “Chances are, the data collection fields we have in there should mesh with what they are collecting. We know sponsors will need to add additional data fields for specific therapeutic area requirements.”
Facile adds that Lilly, Cerner and Quintiles are already doing interesting things with the CDASH standard. She also takes pains to say that she and her team are recruiting early adopters who can provide more detailed feedback of their usage of CDASH in the real world.
In some circles, CDISC has a reputation as an idealistic organization. The characterization has always seemed peculiar. CDISC is about as idealistic as a bucket. CDASH in particular is a concrete example of pragmatic thinking about actual challenges and chronic blockages in the current clinical trial environment.
No single sponsor, working alone, can simplify the workload of a site that works with many sponsors. No single technology supplier or CRO, no matter how innovative or efficient, can dictate what is best for all its partners and customers.
So if the technology environment and work flow in clinical trials is ever to be become any simpler or faster, it will only be as a result of collective effort on the part of the industry as a whole. And there is no better example of collective effort right now than CDISC.d9A2t49mkex