Decentralizing clinical trials in the interest of costs, accessibility and inclusiveness. Capturing real-world evidence and data from wearable or implantable devices. Harnessing the power of synthetic control arms and digital twins to expedite trial outcomes.
In these and other ways, hybrid and agile models are not only reshaping the clinical trial landscape but creating an evolving body of best practices, alongside new risks and pitfalls, for clinical data management. Here are five trends — and their attendant challenges — that should be top of mind for leaders in the life sciences space.
five emerging technology trends in clinical data management
A full 95 percent of life sciences companies say they’re poised to leverage decentralized clinical trials going forward (compared to just 28 percent prior to the pandemic), but where do they expect to see the most value? And what are the implications for clinical data management?
Let’s look at five tech-driven trends in clinical data management for insights and answers.
standardizing data — and expediting study setup — with AI and ML tools
New AI- and ML-powered tools can standardize the formatting of disparate information — everything from digitized protocols to case report form (CRF) design, electronic data capture (EDC) setup and more — enabling it to be stored centrally and passed to systems through application programming interfaces (APIs). As an added benefit, this information can then be reused throughout the entire clinical development lifecycle, and it allows you to leverage knowledge extracted from previous studies, including data scenarios, root causes, data discrepancies and more.
replacing EDC with DDC to reduce the burden on sites
EDC remains a significant pain point in many clinical trials today for reasons that are fairly obvious. For one, it forces site personnel to enter handwritten data. It also means that even as investigators are in the act of assessing patients, they’re also unfortunately tied up with and potentially distracted by computer screens. Swapping EDC for direct data capture (DDC) helps solve both issues, reducing the burden on site personnel in turn.
implementing cloud-based elastic data lakes for improved data availability and a single source of truth
Elastic data lakes are centralized repositories that allow you to store structured and unstructured data at virtually any scale — and they come with flexible data modeling and clustering capabilities, too. From the standpoint of clinical trials, that helps ensure real-time data availability. Plus, it has the advantage of giving you a single source of truth across EDC, ePRO and more.
leveraging smart chatbots for data discrepancy management
Today’s advanced chatbots allow you to run simple-to-complex data queries and refine the results as you go. Say you wanted to review serious adverse events related to cardiac disorders, for example: “Hi there, please show me AEs related to cardiac disorders for study XYZ,” you might type — and then, having parsed the results, add, “Can you filter out only SAEs?” The deeply intuitive user experience is one the the primary advantages of these chatbots. Since running a query is about as easy as conducting a Google search, it won’t take long to train your people or get them up to speed.
automatically connecting the dots through smart data reconciliation
As with end-to-end standardization, new technology tools can help you automatically connect the dots with data reconciliation processes, too. For example, AI and ML can be applied to drive data harmonization and map raw data to CDISC SDTM — the required standard for data submissions to the FDA in the U.S. These same tools can be used to enhance safety signal detection, too.
new hazards, risks and pitfalls for clinical data management
There’s no denying the potential benefits or upside of new technology for clinical trials, as we have seen — benefits which, notably, extend to pharma companies and patients alike. But at the same time, new hybrid and agile models come with distinct, and often unseen, challenges that companies will need to be prepared for in order to navigate.
Crucially, many of these challenges crystallize around data management, and data privacy, collection and reporting perhaps most of all.
What should organizational leaders be looking out for? Problems related to processes, technology and organizational structure are one piece of the puzzle, to be sure. Siloed data storage, for example, can easily exacerbate existing challenges. Ditto not having real-time access to raw and harmonized data or key visualization functionalities.
But at core, a lot of this boils down to human capital constraints in the following three areas:
- Expertise: Internally or via partners, do you have the requisite expertise in the therapeutic area being investigated or the criteria being used to assess safety and efficacy in order to drive the desired outcomes?
- Soft skills: Do key staff members have the soft skills to efficiently and proactively collaborate with the study team, as well as stakeholders upstream and down?
- Project management: Are you equipped with recognized project management best practices, particularly around study milestones, which can help keep you from compromising data quality?
If the answer is “no” to any of the above, you’re probably a long way from achieving audit readiness, and you should consider partnering with the clinical trial experts at Randstad Life Sciences in response. Get in touch with us today to learn more about clinical data management best practices — and all of the other ways we can help power your success.