Digitization of clinical trials (Part 1)
Clinical trials are the engine of healthcare, a process through which almost all new treatments go before being approved, adopted, and used by patients. Given their importance, even small improvements to the conduct of clinical trials can have significant implications for patient care. It’s unsurprising, therefore, that a large segment of digital health is focused on leveraging digital technology to improve clinical trials. Today, we’re going to take a look at what’s been done already and what’s to come.
At their core, most changes that are made to the clinical trial process have one of three goals in mind: increasing the “power” of the study (the ability to derive conclusions from the data collected), decreasing costs, or saving time. We’ll use this framework to evaluate the impact of “digitizing” clinical trials.
We can look at the digitization of clinical trials in two phases — first through incremental change, and second, through paradigm change. Incremental change uses digital technology to supplement current pieces in the clinical trial process that can be done virtually, without fully replacing in-person visits. Paradigm change uses digital technology to completely supplant in-person visits, enabling much more dramatic improvements to how clinical trials are conducted. The former is taking place as we speak, while the latter is just starting to happen.
Part 1: Incremental Change
Increasing Power through Improved Compliance
Compliance is a stubbornly difficult problem across healthcare, and this is true in clinical trials as well. In new drug studies, 4% of participants never initiate the drug being evaluated, and in long-term studies, almost 40% of participants stop taking their medication within the first year (source). Compliance is especially important in clinical trials because poor compliance confounds safety and efficacy results, leading to messier data and reduced study power.
Historically, clinical trials have relied on patients to self-report compliance, which is notoriously inaccurate. Digital technology has allowed researchers to track compliance more accurately, using devices like smart pill dispensers and smart inhaler add-ons (source). Furthermore, these devices also give researchers the ability to encourage compliance, through timely reminders and notifications. Some of these tools have even received FDA approval as drug-device combinations, validating their clinical utility.
In the long-term, monitoring and improving compliance using digital technology is a win-win for everyone, making further innovation and adoption inevitable. Study sponsors want to ensure compliance in order to get what they paid for — a proper evaluation of investigational treatments. All else equal, poor compliance means that more participants will be needed to obtain the same conclusions, increasing the cost and time required. A similar logic can be applied to insurance companies, who certainly don’t want to be paying for medications that sit around in cabinets unused. For regulatory authorities, the stakes are even higher, as noisy data can lead to incorrect approval decisions and undue risk to patients. Perhaps there’s a future where the FDA and insurance companies actually require digital compliance monitoring across the board.
Increasing Power through More Data
Clinical trials are all about collecting data, and digital platforms provide a new and more convenient channel for doing so. This can take many forms, from wearables that automatically track metrics like body temperature, heart rate, and movement, to online Patient Reported Outcome (PRO) forms. Some of these metrics are already typically captured in-person, through interviews and paper forms, while others may have previously been unavailable to researchers. Either way, the convenience of digital platforms enables higher response rates (or even fully automated data capture), more time points, and less variability. As such, digital data collection allows researchers to broaden the pool of data they can capture, while also increasing the power of the study. Given all this, it’s unsurprising that the use of digital technology for capturing data in clinical trials has ballooned in recent years, growing from only 8 studies in 2000 to 1,170 in 2018.
Decreasing Costs through Operational Efficiency
Digital technology has a role to play behind-the-scenes in improving operational efficiency for clinical trials as well. Traditionally, data in clinical trials is first captured using paper source documents, and then transferred into an electronic database later. This extra step introduces the potential for mistakes, which results in a need for source data verification (manually checking the data on paper source documents with the electronic database). In this antiquated system, source data verification is estimated to cost a full 25% of clinical trial budgets.
To cut costs and improve data reliability, the field has begun experimenting with electronic source data — in other words, capturing data using electronic databases upfront, and therefore removing the need for source data verification. The FDA has accelerated this trend by issuing guidance around the controls that need to be put in place for electronic source data to be accepted by the agency.
Another key line item in clinical trial budgets is on-site monitoring, a practice where the study sponsor sends people to each clinical site to ensure data reliability and compliance with the study Protocol and Good Clinical Practice. On-site monitoring is estimated to take up about 25-30% of clinical trial costs (including some overlap with source data verification costs). Over the past few years, in an attempt to control these costs, the field has shifted towards risk-based monitoring, in which monitoring is only focused around critical data, processes, and risks. This trend is supported by the fact that study sponsors can now use digital technology (e.g. video conferencing) to remotely monitor non-critical aspects of trials remotely (and more affordably).
Wrapping up
It’s clear that there are a number of ways in which digital technology is already being used to improve the conduct of clinical trials, by decreasing costs while increasing the ability to derive conclusions from the data being collected. But none of the trends we’ve discussed are truly transformative; none of them are a 10x improvement on any dimensions for clinical trials, and none of them save time.
As we discussed last time with Teladoc’s vision, the reason for this is the online-offline gap in healthcare. Digital technology has been able to transform countless industries because of its scalability and decentralization, but when it encounters a service that must be delivered in-person, its impact will be limited. This is a major hindrance for the impact of digital technology in healthcare, due to the number of in-person activities. To fully evaluate the impact digital technology can have on clinical trials, we should hone in on truly digital clinical trials, where screening, treatment, and follow-up can all be done 100% remotely. We’ll look at this in Part 2.