With increasing decentralisation of several aspects of trial conduct and management, are EDC systems still relevant? And if so, how can EDC systems address the evolving requirements of modern clinical trials? These were some of the key questions on the table at a recent Signant Health webinar, led by Dr Bill Byrom, the company’s VP Product Positioning and Intelligence, an eCOA scientist well known in the field.
From the discussion, these four takeaways emerged:
The use of direct data capture must improve
An audience poll during Signant’s webinar revealed that 67% of attendees use DDC or eSource in fewer than 10% of their trials. Sven Vogelgesang, a specialist in clinical data sciences and management, who has worked for pharmaceutical manufacturing group Merck for almost thirty years, said this aligned with his expectations but he expected progress in 2024. In most cases, he said, it is likely that it is currently seen as an inconvenience to enter the data twice – once directly into a template and then again in the EHR system.
“When electronic monitoring visit reports were first introduced about 25-to-28 years ago, no one wanted to carry around their heavy laptops, and they had to replicate the data after a visit,” Vogelgesang said. The combination of technological evolution, accessible tablets and lighter laptops, and working with sites to communicate the benefits of DDC, Vogelgesang predicts more trials will utilise it.
Resistance to change and lack of training are key barriers to DDC adoption, according to David Stein, an independent e-clinical consultant with more than 30 years of experience developing and implementing technology solutions for biopharma companies and vendors. He said current DDC solution implementations can be complex, but sites that do adopt are very positive about it.
Remote source data verification (SDV) will increase, but this raises regulatory question marks
Demand for SDV picked up during the pandemic, and Jan Breemans, Signant’s senior director of global solutions, predicts the use of remote SDV will continue to grow. While Covid-19 forced the world to conduct many activities remotely and showcased the potential of this approach in many industries, in terms of clinical trials, regulatory clarification is lagging post-pandemic.
Jan Breemans pointed out that some countries still won’t officially accept remote access to data source documents, while for others it’s become business as usual. For progress to be made in this area, Breemans said sites will need reassurances that the data remain safe, compliant, and private and that investigators maintain full control. “The only evidence that should be retained, in my opinion, is metadata, for no longer than needed,” said Breemans.
The number of data sources to integrate have exploded – but EDC should be seen as the key connection point
The audience was polled on which systems they prefer to use when integrating non-CRF data. Most said into a clinical data repository (CDR), followed by EDC and separate files.
Stein said the big challenge is that sites typically have software they prefer to use, which tend to be different for different tasks. For example, the way of recording televisits will typically be different for calendaring. The other challenge is following the draft FDA guidance on DCT, which means the PI is responsible for anything that’s taking place remotely. This could cause friction when setting up video conferences or scheduling when sites are trying to provide the same level of service.
“It’s a question of reconciliation rather than data cleaning or query management,” said Vogelgesang. “A model EDC system in the future will be intelligently connected to a clean data management system allowing queries to be raised within the CDR. For me, the EDC system is still the first connection point.”
Artificial intelligence (AI) and machine learning (ML) developments could make queries more efficient
The final poll of the discussion asked the audience if they were already using AI/ML to support data cleaning. Most of the audience said they weren’t, but a small portion said they were using it to create queries or detect issues in data.
Vogelgesang pointed out that no conversation is off-limits when it comes to efficiencies in data and AI could be used in detecting the importance of queries as well as database programming and query phrasing. “A lot of problems we have are around bad query text. AI can help to standardise the language used for the benefit of all stakeholders which will then reduce site burden and free up resources.”
Meet the panelists
David Stein
David is an independent e-clinical consultant with more than 30 years of experience developing and implementing technology solutions for biopharma companies and vendors.
Sven Vogelgesang
Sven is a specialist in clinical data sciences and clinical data management, working for more than 28 years in the pharmaceutical industry and leading global sponsor functions in knowledge management, clinical information management and clinical data sciences.
Jan Breemans
Jan Breemans is advising Signant Health and its customers on the value and use of data analytics in the oversight of their clinical development programmes. Prior to joining Signant Health, Jan worked on the sponsor side for more than 20 years and gained substantial experience in the field of data management and analytics, project management and clinical systems management.
Bill Byrom
Dr Byrom has worked in the pharmaceutical industry for more than 30 years. He has authored more than 80 publications and two industry textbooks on electronic patient-reported outcomes (ePRO). His recent scientific work includes the use of wearable technology and bring-your-own-device (BYOD) eCOA in clinical trials.
Access the full video here for more insights.
