7 Tech enablers of new age clinical data management
Vice President,
Statistical Programming,
Navitas Life Sciences
(a TAKE Solutions Enterprise), Bengaluru
Advances in technology coupled with complexity in clinical trials and acceleration of decentralised trials have necessitated the need to move from traditional clinical data management (CDM) practices into much more advanced clinical data science processes.
We have seen clinical trial designs evolving from time to time to adapt and increase effectiveness, for example: Adaptive trials, Virtual trials etc. CDM processes have accordingly evolved from the last two decades from paper based clinical data collection systems to Electronic Data Capture (EDCs) and many other changes. Now, we are seeing further transformation required to support non-EDC centric approaches involving multiple data collection instruments, eSources, mobile technologies etc.
Sensors and wearables generate high volumes of data (millions to trillions of times more than EDC) at high velocity (i.e., generated continuously multiple times per second). In this context, traditional CDM processes would not be viable.
Key technology enablers in new age Clinical Data Management:
1. Reduction of Electronic Data Capture (EDC) - centric approach and increased adoption of eCOA (electronic clinical outcome assessments), wearables, sensors and other eSource solutions.
2. Adoption of Artificial Intelligence (AI)based solutions like CDM Chatbots (intelligent Virtual assistants) in Clinical Data Management. Based on Machine Learning (ML), Natural Language Processing (NLP), Voice recognition CDM chatbots can provide updates on study activities. Example: Ask a CDM Chatbot to “provide the counts of sites with more than 10 Serious Adverse events (SAE’s) reported on Study X”.
3. Robotic Process Automation (RPA) – RPA enables virtual robots to do predictable and repetitive human activities. Example: Virtual Robot could be fed with details of external data reconciliation errors. The Virtual Robot could then login in EDC with its own account (e.g., Login: CDM Robot) and post corresponding queries. The same technology could be applied to other simple CDM tasks. Virtual robots can become an unlimited virtual CDM workforce.
4. Blockchain technology - Blockchain is a chain of interconnected blocks (i.e., data). Each block contains a time stamped and non-modifiable version of data. Blockchain can help establish a chain of digital trust between the patient, healthcare providers and connected medical devices. Blockchain facilitates secure, real-time sharing of information within a trusted framework, preventing any misuse of medical or patients’ personal data.
5. Intelligent Clinical Data Management Systems (CDMS) – In order to be future proof, CDM needs a source and technology-agnostic data collection, consolidation and management strategy. This demands a new generation of CDMS including data platforms, workbenches, reporting framework etc. Intelligent CDMS must enable real-time, datadriven and confident decision making from active data (data from Clinical Research) & passive data (data from medical care and personal health devices).
6. Maximising value of EDC – EDC still has a critical role to play and its use must be maximised through a fit-for-purpose EDC Strategy while investing in future-proof solutions. EDC were not designed to be a central study data repository & should not typically be viewed as the place to load all external data. We need to investigate future-proof integration and reporting platforms that are compatible with current & future data streams, including Sensors and wearables. We need to establish more flexible and efficient EDC build processes to accommodate multiple study designs, foster direct data capture, efficient post-production eCRF changes etc.
7. Managing Big Data - Data credibility, reliability plays an important role and thus focus is on what matters (i.e., critical to quality factor), risk-based data strategies, AI-driven automation of issue detection and resolution and Fit-for-purpose solutions (scientifically plausible and strong enough to support reliability of trial results). RESTful Application Programming Interfaces (APIs) will provide interoperability between computer systems.
We could foresee a Clinical Data Management world built on a working model that includes virtual Clinical Data Managers working alongside Human Clinical Data Scientists. This has also resulted in greater need for upskilling of Clinical Data Management professionals to deal with new technology and adapt to new processes.