Bringing drugs to market is notoriously time-consuming and expensive. Small companies are uniquely positioned to change that.
In order to support the approval of vital new therapies, clinical trials should ideally be fast, reliable and cost-effective. Unfortunately, today’s clinical trial processes are slow, error-prone and extremely expensive. The COVID-19 pandemic highlighted the importance of improving this speed: slow processes can cost lives, and longer trials result in higher costs that are ultimately passed down to the consumer.
One of the main reasons for the current inefficiencies in clinical trials is the process of data collection and management. Due to the lack of connectivity between medical centers’ electronic health records (EHR) and clinical trial databases, patient data is copied manually—literally by hand as clinical research coordinators or data entry clerks sit at computers copying data from one system to another. This takes time, costs money and inevitably results in mistakes. In 2024, when businesses in every other industry are using automated data input and analysis, it’s truly shocking that manual data transfer still rules supreme in clinical trials.
I saw this firsthand during the two decades I spent working in clinical testing roles at prominent pharmaceutical firms and CROs. That experience led me to establish Yonalink, which provides automated data streaming for clinical trials.
With several other companies now offering similar services, it’s safe to say that technology is no longer the issue. The real barrier is Big Pharma’s hesitation to implement technological solutions in specific areas—a hesitation that is holding the entire industry back. Despite this challenge, I believe that small pharma and biotech companies can change the industry’s course to enable the widespread adoption of data streaming, revolutionizing how researchers complete clinical trials.
Clinical Trials Have a Data Problem
Clinical trial data management is failing by every measure of efficiency. According to McKinsey, the average length of a Phase II clinical trial grew from 37 months to 41 months between 2011 and 2021, and from 41 months to 44 months for Phase III trials over the same period. A U.S. Congressional Budget Office report notes that each approved drug often runs in the $1- $2 billion dollar range to develop, with additional costs mounting because the FDA conducts inspections of manual data verification due to the prevalence of mistakes. These in-person data inspections by the FDA cost U.S. taxpayers millions of dollars, yet there is still no guarantee that the data being used to approve drugs are 100% accurate. We wouldn’t accept any mistakes on our bank account statements, so why should we accept them in clinical trials when so much is at stake?
The main cause of wasted time, increased costs and the volume of mistakes is manual data management. Clinical trials involve a massive amount of data, with a typical Phase III oncology study requiring 3.6 million data points, for example. The vast majority of data points are copied by hand from patients’ EHRs into the clinical trial database, known as an EDC (electronic data capture), by employees.
It takes an average of 6–8 weeks from when data are entered into a patient’s EHR in a medical center for it to reach the clinical trial’s EDC, ready for analysis. Manual data transfer results in errors that hinder the entire trial process. It also costs a lot of money for individuals to copy data from one screen to another, verify it manually and then verify the verification itself.
The Sticking Point: A Traditional Mindset
The dual need to speed up clinical trials and obtain more accurate data motivated us to develop an AI-based solution that streams patient data from EHRs directly to the EDC, producing reliable, accurate records within a day and at a fraction of the cost of manual entry. Competitors like Ignite Data and Flatiron have also developed solutions with different approaches to the problem.
Although data transfer for clinical trials is a complex and challenging problem, the lingering barrier is not the technology but the mindset of pharma leaders. The industry embraces AI and innovation at many stages of the clinical trial process but not for data capture and transfer.
It’s understandable that trial managers prefer techniques that have been proven to work in the past, even if they’re not perfect. Trial sponsors invest an average of $100 million before reaching the trial phase so the risks of changing traditional processes are high. Having been a trial manager myself, I know firsthand how hard it is to effect change when it comes to data management.
Unlike in other areas of drug development, in data capture AI isn’t simply improving the existing process—it’s transforming it entirely. This prospect is scary for trial managers, especially because data are the foundation of every successful drug. This is the mud in which pharma companies are currently bogged down: the need for change vs. the fear of change.
The Changemakers: Small Pharma and Biotech
Usually, innovation in the pharma world begins with the big companies and then trickles through to the smaller ones. But in this case, with Big Pharma hobbled by the fear of trying something new, it’s time for small biopharma startups to pick up the ball and run with it. Smaller companies suffer disproportionately from the burden of current data management methods. They don’t have the funds to waste on manual data entry and can’t risk waiting weeks, sometimes months, before catching mistakes. They need data management to be reliable, low-cost and fast, and they should be willing to blaze a trail in order to achieve it.
We are starting to see positive movement in this direction as small biopharma companies increasingly realize the benefits of innovation. Companies such as SAS, Nucleai and Datacubed have recently announced partnerships with biopharma companies to integrate their technologies into the clinical trial process.
The only obstacle to the widespread use of this technology is the mindset of big pharma companies. Small biopharma can’t afford to resist the benefits of streaming data transfer, so they are emerging as the first to embrace these tools. In this way, small companies will be pivotal in leading the industry to real change in the data domain, but I predict that within five years, all pharma companies will be using automated data streaming solutions.
