CRO

Electronic data capture systems not speeding clinical trials as expected: Tufts report

We were under the impression that electronic technologies have transformed clinical trials, but are we there yet, or are we still at an early stage? A recent study by the Tufts Center for the Study of Drug Development found that currently available electronic data capture systems still have a ways to go before study speed could live up to our expectations.

Although all 250 clinical data management professionals responding to Tufts’ recent 2017 eClinical Landscape Study use EDC for clinical studies, as many as 83% of them reported having previously released a study database after first patient, first visit (FPFV), and that probably caused further delays down the stream, the study found.

Those that deliver the database after FPFV take nearly twice as long (10 days vs. five days) to enter patient data throughout the study than those that deliver before FPFV. They also take about 75% longer to database lock.

“FPFV is the first key clinical milestone in a trial and signals the transition from start-up to execution of the study,” said Hugo Cervantes, VP of Vault EDC at Veeva Systems, in an emailed interview with FierceCRO. “In fast-enrolling studies, every day the database is delayed will result in a significant volume of data that is unavailable to study personnel, which makes data-driven decisions impossible to execute.” Veeva Systems sponsored the Tufts study.

When the study is off track, the data management team will deploy the electronic case report form ahead of edits checks, and this creates a situation where the backlog of data leads to delays in cleaning and eventually finalizing the database, Cervantes explained.

The study also found that it takes respondents 68 days on average to build and release a study database, and when asked about what caused the delays, 45% of them cited protocol changes. To Cervantes, that was far longer than the ideal 20 to 30 days.

Since protocol changes are unavoidable in a clinical study, and as database build is only getting more complex as the data captured grow and the complexity of trials increases, Cervantes said it is up to EDC providers to make their systems flexible. “The EDC provider, specifically, plays a dual role,” Cervantes explained. “First, as a technology provider, streamline the build process through functionality such as configurable workflows. Second, as a database designer, deliver fast, accurate database delivery services.”

Two main problems affect the flexibility of currently available systems, according to Cervantes. First, they do not create a database design environment that enables rapid build activities or automates tracking changes. If a modification to a CRF is needed, the system doesn’t allow users to only check the new or amended functionality. Today’s EDC also doesn’t make it easy to amend the database mid-development, and changes often require jettison of all data mid-testing.

“Many organizations now have standards libraries—templates for database design—which should promote reuse in building a study and drive down cycle time to days, not weeks,” Cervantes said. “If there is a need to create new CRFs, then EDC should provide simple user interfaces to design that much faster. The remaining cycle time is review and approval of specifications and testing of the database, where technology solutions should help with configurable workflows that manage this process and also point the user to only test what is new, not what was previously validated.”