Successful clinical drug development depends upon having accurate, reliable data from numerous, disparate sources to satisfy the demands of regulators, payers, sites, sponsors, CROs and increasingly, patients. Fortunately, countless technological innovations now provide clinical researchers the ability to capture and aggregate data throughout the entire development lifecycle; presenting enormous opportunities to answer even more complex research questions--and ultimately--accelerate access to novel drugs/treatments. Many of these technologies and models do indeed help to accelerate timelines and potentially reduce development costs, but we are seemingly now at an inflection point at which increased access to data solves problems in some areas while inadvertently creating new problems in other areas.

To fully leverage technology-enabled access to the patient, clinical, operational and outcome data required for market authorization applications, a clinical data strategy should be included as part of every life science organization’s infrastructure and become part of the clinical trial plan and strategy from the very early stages of development. The strategy should include specific methods and modalities for collecting, exchanging and analyzing all the data collected from the various technologies that will be used as part of the study, as well as a plan for capturing real-world evidence and other patient-centered outcome data to help fulfill regulator and payer evidentiary requirements. Creating a clinical data strategy and including it in the trial plan can help solve many of the challenges associated with harnessing the increasingly complicated flow of clinical research data from an ever-increasing volume of sources.

One of the biggest challenges in clinical trials is patient recruitment. Finding and enrolling eligible patients can be time-consuming and costly. However, there is a wealth of data available that can potentially help organizations identify eligible patients. Electronic health records, claims data and social media data can all be used to identify patients who may be eligible for a trial. Additionally, there is an increasing focus on diversity of the patient population included in a trial. A data strategy that leverages these data sources can help organizations streamline the recruitment process and improve trial timelines.

Data is Everywhere

The primary challenge with access to so much data from so many different areas and modalities is that there is a limited industry-wide standard or regulatory guidance for data connectivity. Interoperability issues continue to persist and pose barriers to seamless connectivity between EHRs, EMRs, EDC, eCOA, ePROs, labs and other operational data. Lacking such standardization, innovative technology vendors, sponsors and CRO’s are working collaboratively to solve many of these problems; with groups such as the Decentralized Trial Alliance (DTRA), the Society for Clinical Data Management (SCDM), and the Avoca Quality Consortium advancing standardization ideas and proposals for technology-enabled clinical studies. Regulators, too, are generally supportive of innovative patient-centric methods and models for clinical research, but they have yet to provide clarity around data connectivity among the larger healthcare ecosystems.

Despite the frenetic activity and regulatory uncertainty, new technologies are continually being utilized in clinical research to help address key considerations such as patient preference, convenience and engagement. Many applications of decentralized and other virtual technologies reduce patient burden by bringing the trial to the patient; however, introducing technologies into clinical trials also adds complexity. Further, we continue to see clinical research technologies that require manual efforts to connect with each other—or they require multiple data entries. Modalities like smart watch, sleep trackers, blood glucose monitors and other wearable sensors, for example, may produce so much data, so frequently, that their use can be overwhelming (at best) and no longer useful (at worse). Similarly, hybrid design studies that utilize pharmacies, labs and other community health providers add complexity to the process of aggregating all that data required for regulatory submissions.

Multiple data entries and inconsistency in how data is captured also hinders efficient development by denying researchers a single source of truth. From a site perspective, a single investigator working on five different trials from five different sponsors or CROs might be using a number of different technologies. This naturally increases the burden on the investigator and the site staff, which also elevates potential risk and cost.

Consider the process for source data verification, one of the most expensive and time-consuming elements of a clinical study. Data from each patient visit must be verified manually to ensure it was captured correctly, potentially elevating risk from inconsistent data transfer and manual errors. Harmonizing the various independent systems would provide researchers with a single source of truth, minimize the potential for errors and benefit both sponsors and site staff.

Plan Early, Plan Often

Including a clinical data strategy as part of the study’s protocol can help identify areas in which multiple data entries might prevent a single source of truth from being obtained. We continue to use a model in which clinical monitors visit research sites and audit whether the data was entered properly and accurately from the EHR or other sources into the EDC. This is a costly and inefficient process that could be eliminated entirely by capturing and validating data directly from the original source. Some companies have begun using artificial intelligence and machine learning applications to enable data to flow seamlessly between systems and potentially remove much of the human effort associated with basic trial activities. These efforts are new but should be encouraged, as doing so may help researchers harness the increasing flow of disparate data to uncover meaningful trial information more quickly, more affordably and with less potential for human error.

The COVID-19 pandemic has certainly accelerated the use of decentralized clinical trial technologies and patient centered approaches to help improve the patient experience in clinical research. We are now seeing complex clinical trials that include decentralized elements such as e-consent, in-home nursing, telemedicine visits and electronic patient-reported outcomes, among others. These new modalities for capturing and collecting clinical trial data directly from the patient are certainly enhancing the patient and experience—and increasing the volume and diversity of the patients that may be willing to participate—but they are also adding new complexities and changing the already complicated process.

To help manage this complexity, a clinical data strategy should be developed in the very early stages and included as part of the trial’s data management plan. It should include specific detail for how each data point will be captured, collected and analyzed, and it should be flexible enough to accommodate unforeseen situations or the introduction of new technologies that might be appropriate to add to the study down the line. Designed to solve complex problems around recruitment, diversity, speed and cost, a thorough clinical data strategy is no longer simply a nice-to-have feature. It is now a critical and requisite instrument for continuing to evolve the clinical trial process into one that is more patient-friendly, less prone to risk and more primed for success.