Only one or two in 10,000 compounds that enter pre-clinical trials will ever make it to approval. That journey will take on average 12 years and cost $2.6 billion or more. Adding these facts together, we’ve reached a situation where increasing pressure on R&D has seen return on investment shrink from over 10% in 2010 to under 2% in 2019. Everyone in the Life Sciences industry is aware of these figures so anything that can help improve them is to be enthusiastically embraced. Perhaps, one of the easiest wins lies in the intelligent and automated classification of clinical trial data. Let me explain why.
During COVID-19, it was estimated that the cost of trial delays was as much as $8 million per day and that almost 95% of trials were over one month late. The time and cost pressures placed on the sector by the pandemic have sharpened the focus on digital transformation, moving from buzzword to strategic imperative for the Life Sciences sector. However, some commentators believe the pace of change is still too slow in areas such as R&D and clinical trials.
More data, more silos
Personally, I think it’s a little harsh to say the sector isn’t digitizing. There have been major strides towards more digital operations and customer experience. Over the last decade, we’ve witnessed the widespread adoption of digital applications such as electronic trial master file (eTMF), electronic case report form (eCRF), Electronic Data Capture (EDC) and Remote Data Capture (RDC) systems, to name just a few.
In addition, we’re seeing the rise in importance of Real-World Evidence. It’s now becoming pivotal to expedite and improve the quality of both the clinical trial and submissions processes. Today, 94% of R&D professionals believe it will become vital by 2022. Then, there’s telehealth that saw up to a 175 times increase during the pandemic.
There’s no doubting that these new data and channels will result in greater trial efficiencies and cost saving. But they may still fall short of the goal for innovation transformation that’s meant to digitize and rationalize processes end-to-end to improve effectiveness, performance and innovation. Too often, elements of a process become automated but other parts remains stubbornly manual.
For example, I’ve met with many clinical trial organizations where an EDC ingests the data, but there is then a lengthy and potentially error-prone manual classification and validation process before that data can be passed onto other downstream clinical trial systems. The EDC captures information from some channels, but those channels are growing and it lacks the intelligence to gain insight from the volume and variety of data it’s faced with.
In addition, the EDC is only one way that the organization’s systems are receiving information. It’s coming in from healthcare professionals, contract researchers and manufacturers, external labs and patients. It’s in the form of clinical trial documents, phone calls, online forms, IoT data from sensors and devices, faxes from physicians and we’re all still getting snail mail!
Every clinical trial is faced with more data from more sources and has, over time, created more business and information silos that make the data a challenge to manage and exploit.
Unlocking the value in data
We live in the world of Big Data and, as far back as 2017, it was estimated healthcare data was growing by 48% each year. While a great deal of focus has been placed of the technologies that can derive insight from that data, we’re really putting the cart before the horse.
Gaining the proper insight means integrating, analyzing, and presenting datasets in ways that allow you to draw meaningful and actionable intelligence. But first you need to efficiently capture data from all available sources, and add classification and categorization so it can be effectively stored, searched, and retrieved by your clinical trial data management team.
It’s important that this takes place at each stage of the clinical trial process and supports multiple languages to meet regulatory requirements in each of the regions where your trials are happening.
EDC systems can perform the capture element of some documentation for clinical trial data but, in most instances, lack the intelligence to automate the classification and verification processes. This builds time, cost and inefficiency into the process that can see data not properly stored, siloed away from other data points and unavailable to the people and systems that need it, diminishing the usefulness and value of the data.
A new approach to intelligent classification
The Life Science sector is currently losing a great deal of time and money in the data management of clinical trials – and potentially putting the viability of new drugs at risk. OpenText is introducing an end-to-end solution for the intelligent classification of clinical trial data that delivers:
- AI-augmented capture of data and documentation from any source
- Automation & intelligent classification of documents and data
- AI capabilities to continuously improve quality of auto-classification
- Classification of documents to a sentence and paragraph level
- Classification of all data types including IoT and real-world evidence
- Automated QA and verification processes
Clinical Data Intelligence for Life Sciences from OpenText™ increases the speed and quality of data entering your clinical trial systems. It intelligently helps identify and surface all relevant metadata and documents, helps you comply with regulations in different regions and helps ensure your final submission is based on complete and correct information.
To learn more about the services that OpenText delivers to the Life Science industry, please visit our website. To learn more about why intelligent classification of clinical trial data is important, download the whitepaper Solving the challenges of data in clinical trial supply chains for rare diseases.