Drug development professionals have more data at their disposal than ever before—the question is, how can they use that information wisely and effectively? Updesh Dosanjh, practice leader for technology solutions with IQVIA
OSP: Could you please provide some perspective on how the use of technology to handle AEs and safety info have (or haven’t evolved in recent years?
UD: Adverse event (AE) case processing involves three main steps: receiving information, transforming information, and analyzing information. Case processing automation today focuses mainly on the second step: transformation.
Life science organizations largely felt that automating data transformation would solve downstream and upstream issues in receiving and analyzing information. However, there’s a lot of work that must be done before transformation can be executed.
Case sources are increasing as a result of the proliferation of social media, online forums, and other means of acquiring information on consumer reactions to medications, and the complexity of receiving information from multiple sources is increasing at an equally rapid rate. The cutting edge of technology is focused on finding and preparing the reactions to then be automatically taken in the processing step.
We are now also seeing the beginning of a focus on the analysis part. With the availability of big data tools and techniques, it is now possible to do novel analysis in real-time that would have not even been possible before. Both these contribute to reducing costs while giving deeper insight into data.
OSP: What are some of the problems with how teams currently handle AE and safety info gathering, and data processing?
UD: The current strategy of tackling increasing caseloads and complexity is to simply throw more workers at case processing. Unfortunately, this strategy is unsustainable as there is limited availability of human workers, and life science companies are already running out of humans to throw at the issue. This is why case processing automation needs to be deployed throughout the life science safety cycle, to improve all aspects of case processing and analysis.
The limited focus on case processing simply exposed gaps everywhere else. The technology now exists to tackle problems holistically and support every part of the lifecycle with effective technology.
OSP: Specifically, could you please talk about the silos, and how keeping all that important information in separate piles might harm the companies that rely on that info, and even patients?
UD: In the past, organizations used separate systems to cover each step in the AE case processing cycle: information reception, information transformation, and information analysis. Often, those systems are unconnected. If they are connected, they typically leverage multiple sub-systems to manage the integrations, which can be costly to maintain.
Separated systems also prevent a singular view of a case’s status in the processing cycle. The person who initially received information on the case doesn’t know what’s going on with the transformation or analysis of that case’s data.
In addition, multiple siloed systems create more sources of information that must be reviewed for reports. This requires collaboration between multiple team members to develop reports, and complex review cycles to ensure accuracy. This is why organizations need a single platform for case processing that provides a holistic view of a case’s status and processing. So, companies suffer from multiple hard-to-connect and maintain systems.
For patients, the risk is that as information flows between tools, and gets transformed each time to be usable in the particular technology, something will get lost or missed. Disparate information in multiple systems can only ever create a risk that can’t be easily found or mitigated
OSP: You’d mentioned AE reporting is expected to increase 15% annually—could you please share some details about that phenomenon, and how that increases challenges with collecting and processing AEs and safety info?
UD: The 15% increase in AE reporting is a trend that arose over the past few years. However, due to COVID, that rate is likely to increase in 2021 and beyond due to heightened public awareness of citizens’ ability to report AEs through online mediums such as social media and online forums. This will inevitably generate more data surrounding AEs, meaning that today’s case processing challenges are going to get worse in coming months and years, rather than improving.
OSP: Please share why artificial intelligence (AI) might help eradicate some of the problems and obstacles shared above.
UD: AI technology can support and optimize activities in each of the three main steps of case processing: information reception, information transformation, and information analysis, reducing the time and effort by 50% or more. AI supports information reception by automating the normalization of data and ensuring that all relevant data is complete and ready for transformation.
AI optimizes information transformation by immediately extracting and translating data to identify any gaps in data that will negatively impact analysis. Finally, AI streamlines information analysis by providing the ability to quickly and interactively mine data.
OSP: Could you please share some of the ways in which IQVIA can help organizations better manage AE and safety info? Please feel free to mention any technology tools or programs available.
UD: IQVIA’s vision for life science case processing is a zero-touch approach to safety throughout the entire cycle – not just a single department or aspect of safety operations. IQVIA’s Vigilance Platform brings enterprise-level automation and AI to companies of any size, reducing effort, allowing companies to focus on the data analysis, and eliminates the risks caused by silos by providing a single source of truth for case processing.
OSP: Anything to add?
UD: Natural language processing (NLP) technology allows case information to be translated and normalized for smoother transformation and analysis. NLP is critical for organizations that are operating around the globe and processing AEs in multiple languages.
NLP is also useful when it comes to normalizing case language and removing regional language differences. For example, American English and British English speakers may use different terms or spelling for certain words, such as “drug store/chemist” or “color/colour.” These terms must be standardized for accurate analysis. NLP is one of the key technologies that is underpinning the AI revolution in safety.
