← Back
THREADDCTs

We Need Descriptive and Predictive Metrics for Advanced Trials - The Time is Now

Cassidi Williams

The use of remote data collection and eSource technologies in clinical trials has exploded. Approximately a quarter of all clinical trials in the last year are estimated to have deployed decentralized approaches.1 Using technology and decentralized approaches to make clinical trials more accessible to more participants is, certainly, a step in the right direction of making trials more inclusive. Expanding the potential pool of participants beyond traditional geographies is a great benefit for sponsors. However, the performance within the majority of trials are being managed and measured using tools and methods developed with traditional, site-based studies in mind. It begs the questions; are we really maximizing the participants clinical trial experience in order to secure high quality data in our decentralized and hybrid studies (DCTs) and, how would we even know?

If we continue to measure and monitor performance of decentralized trials using traditional metrics that don’t fit, we can never be sure how effective these studies are. DCTs and Hybrid DCTs are simply too different from site-based trials for older study metrics to fit well. Traditional studies rely on clinic visits, where doctors and clinical staff collect various types of data derived from participant interviews, lab tests, and point-of-care instruments (e.g. blood pressure cuffs, pulse oximeters, etc.). A standard metric used in these types of trials is time from visit to data entry.

In DCTs, data collection is much more broad, coming from a variety of sources, such as:

  • ClinRO Data collected using telehealth/virtual visits
  • eConsent data
  • Electronic Patient Reported Outcome (ePRO) data
  • Data such as physical exam findings collected from home health providers
  • Data from in-home medical devices (eDRO, e.g. blood-pressure cuff)
  • Data from wearable sensors (smartwatch, mobile ECG, etc.)
  • Data from other electronic outcome reports (e.g. observers)

An important distinction between site-based studies and DCTs is the frequency of data collection. Traditional study visits are intermittent and often weeks apart. Remote technologies allow for much more frequent - and in some cases continuous - data collection. This is one of the key benefits of DCTs and an excellent example of how old-school metrics can’t begin to successfully account for DCT performance. Without a way to track compliance within studies in as close to real-time as possible, study teams are just guessing as to how their DCT is going. In a DCT, we need to know immediately if a participant is close to missing the data collection window in order to contact the participant and encourage them to remain compliant. We need to understand if participants are not engaging with the platform in a way that gives us the quality output we need for drug approval. We need real time analytics on data status including eConsent, ePRO, Survey, EDC and telehealth visit status. This is our goal at THREAD, to provide metrics that make a difference in data compliance and quality in our DCT platform.

It’s Time for DCT-specific Metrics

This is why THREAD has put years of expertise running DCTs to work in developing a descriptive, predictive and proprietary solution - THREAD Analytics Dashboards. Advanced algorithms and predictive modeling provide easy-to-digest THREAD scores, allowing study managers to immediately understand the status of enrollment, compliance, engagement, retention and data performance by site and for and the study overall. Viewers can see metrics on participant statuses, enrollment progress, and predicted study completion dates - all based on real-time data. The dashboards utilize a proprietary, technology-driven approach to measuring the success of DCTs in five key categories:

  1. Data management
  2. Participant compliance
  3. Participant engagement
  4. Enrollment and retention
  5. Study Performance

The ability to track and analyze data in real-time is a game-changer that promises to transform how conduct trials are conducted. An obvious reason for this is that study teams can more quickly identify potential issues as they occur and intervene as necessary. Less obvious, though, is that real-time data allows us to create highly accurate predictive metrics. Using up-to-the-minute data, THREAD dashboards visualize trends in enrollment, device usage, frequency of telehealth visits and more. Activity completion data feeds predictive analytics that show when individual activities, and the study itself, are likely to be completed. This allows study leaders to anticipate potential issues and intervene before they become real problems.

Conclusion

Necessity, convenience factors and cost savings are the main drivers of the increase we’re seeing in DCTs. While those are absolutely benefits of using decentralized and hybrid study approaches, we cannot know if we’re truly realizing those benefits if we’re still using old-school approaches for monitoring and measurement. The value of access to real-time data is lost without real-time monitoring and analysis. Therefore, as quickly as the industry has moved to begin implementing DCTs, we must move just as quickly to apply metrics that are purpose-built to help optimize these kinds of trials. By doing so, we can truly understand how DCTs can help to run faster, less expensive trials while improving the participant experience.

For more information visit: THREAD Analytics Dashboards.


1 Chancellor, Daniel. (2020, July 21). Decentralized and Hybrid Trials 2020 [Presentation]. PharmaIntelligence and Clinical Trials Europe. https://pharmaintelligence.informa.com/resources/product-content/2020/07/20/10/43/sitecore/shell/~/media/informa-shop-window/pharma/2020/covid-24-campaign/slides/decentralized-clinical-trials-in-2020.pdf