Data Management Strategies of Tomorrow: Bridging the Gap Between Retired Data Systems and Digital Innovation
How do data management processes need to evolve to facilitate changes in how the pharma industry engages with data? This was one of the key questions asked during Oxford Global’s annual Pharma Data and Smart Labs UK: In-Person event held on the 8th and 9th of September 2022 in London, UK.
During the event, senior representatives from leading pharmaceutical companies and research institutions came together to discuss the future of pharmaceutical IT and data management. One such instance came in the form of a panel discussion dedicated to exploring the data processing strategies of tomorrow.
Samantha Kanza, Senior Enterprise Fellow, at the University of Southampton, moderated the panel. She was joined by panellists Alfred Stefan, Manager of LU Informatics at AbbVie, and Jessica Justice, Digital Strategy Delivery Lead at Genentech/Roche.
Meet the Panel
Samanatha Kanza is a Senior Enterprise Fellow at the University of Southampton, working in the interdisciplinary research area of applying computer science techniques to the scientific domain, specifically using semantic web technologies and artificial intelligence. Her research includes looking at research data management, electronic lab notebooks and smart laboratories to improve the digitisation and knowledge management of scientific research using semantic web technologies; and using IoT devices in the laboratory. She has also worked on several interdisciplinary Semantic Web projects in different domains, including agriculture, chemistry, and the social sciences.
Jessica Justice, formally Chapter Lead of Computer Systems Operations (QC), is currently in the role of Digital Strategy Delivery Lead at Genentech/Roche. She is responsible for driving digital technology strategy through delivering key digital initiatives and uses cases working closely with the Digital & Data Backbone Capability Team to define the future PT’s Data, Digital and IT OT architecture. Justice focuses on the key partnerships with IT and Pharma functions to architect the planning, execution, and sustainment of digital solutions to achieve the PT Digital Aspirations.
Manager of LU Informatics at AbbVie, Alfred Stefan, studied Mathematics and Information Technology at the University of Kaiserslautern. He achieved his master’s degree in 1995 and, since 1996, has worked within the pharma industry in various roles. Starting as a SAS analysis programmer, Stefan then led the Discovery IT team at Abbott and is currently leading a team of Data Scientists at AbbVie, in Ludwigshafen. His current position enabled him to work across various aspects of the pharma R&D landscape.
Data Destruction: Understanding the Dos and the Don’ts
When it comes to data excess, the question of data destruction is somewhat challenging. Kanza professed: “whilst there’s definitely an argument for the requirement of data storage, I’m always a little cautious about the idea of destroying data.” This, she explained, is because “you never know when you might need it later.”
Kanza’s comments were in response to Stefan’s proclamation of “being an advocate for destroying data, especially old excess data.” To illustrate this, Stefano used the example of every-day data output that comes in the form of the weather forecast. “As I am sat in this conference room, indoors, the weather forecast is relatively irrelevant to me at this moment,” he began.
- Digitisation and Beyond: Understanding the Lab of the Future
- FAIR Data Management: What are the Best Approaches to Ensuring Information Findability?
- Real-World Data in Real-World Applications: Informing Future Approaches to Healthcare Provision and Drug Development
“However, tomorrow morning, when I want to find out whether to drive or cycle into work, such data is extremely relevant.” Once at work, and safely inside for the next eight hours, the data again becomes irrelevant until it is time to travel home. From this perspective, it thus becomes a case of having the right data at the right time.
When data becomes redundant, the longer it gets stored, the harder it can be to locate and retrieve. “If data is too old to use, it should be actively destroyed,” he continued.
Justice weighed in by suggesting that it might be necessary to select what data needs to be retained and what does not. Although under Genentech/Roche’s legal guidelines, all data must be held for at least 25 years; Justice notes some merit in “selecting and defining those data domains that will be essential to store.” Such measures can be implemented throughout the manufacturing, testing, or even research-in-the-lab processing stages.
Evaluating Data Management Strategies
From an academic perspective, espousing machine-readable data is essential. Kanza explained how often, within a doctoral lab setting, some students fail to manage and deposit their data sufficiently. She said, “we need to be pushing harder to make sure that people are capturing the right data and then are able to make that relevant data available.”
Insufficiently managed data has a knock-on effect during the review process, where the reviewers – who often work voluntarily – cannot locate the correct information and do not themselves have the time to try and do so. “If the data cannot be reproduced, then there is a serious problem, meaning there is an entire process we need to change,” Kanza confirmed.
Stefan agreed with Kanza’s point and iterated a similar occurrence in industry data management protocols. In particular, Stefan explained that there is often no way of interpreting old data that has been stored over 20 years. For example, the floppy disk, a popular storage device used from the 1970s to 90s, soon became obsolete and its data redundant once it fell out of favour by the late 1990s. “The way of storing data changes over time,” Stefan explained – “so too does the ability to interpret that data.”
What’s Next for Data Management in Pharma?
Stefan highlights the need for a more “flexible approach to data management and storage, where it is possible to retroactively add missing information back into the database.” Justice expressed a similar sentiment, deeming the next hurdle to overcome as “retaining data integrity in the face of digital innovation.” This is undoubtedly the case in the context of Roche’s modernised digitalised enterprise systems.
Roche’s Good Manufacturing Practice (GMP) space is undergoing several changes and updates in its data storage systems. It aims to introduce new single-lens LDS, ELM databases, and manufacturing systems in an effort to retire all outdated systems within the next two years.
The overhaul will require the utmost forethought and strategy when selecting what data needs to be retained. As Justice projects, what’s next for the data management strategies of tomorrow, will be the question of “how to retain data integrity in the transference of information from retired data systems to newer digital management systems.”
Want to find out more about the innovations happening in pharma data? Join Oxford Global’s annual Pharma Data & Digital Medicine event today. This 2-day conference brings together a panel of prominent leaders and scientists, sharing new case studies, innovative data, and exciting industry outlooks.
Related Resources
