Unifying observational research with a next-generation virtual approach
Observational research is core to drug development, helping researchers understand the natural history of diseases and the real-world impact of treatments on patients. Observational studies frequently default to traditional site-based methods designed for interventional trials. This creates unnecessary challenges and inefficiency, particularly for long-term research. In the extreme, cell and gene therapies require patients to travel to sites for 15 years of long-term follow-up.
Early decentralised clinical trials (DCTs) promised to reduce dependency on overburdened sites and improve patient participation. However, they struggled to effectively gather study data and often complicated processes for research teams, undermining their intended benefits.
Now, recent technology advancements, including AI, are enabling more sophisticated virtual research models that unify important steps of the research process and bring together comprehensive medical record data from anywhere patients receive care.
This unification is a major differentiator from compartmentalised and competing DCT approaches. As next-generation virtual models continue to take hold in observational research, research efficiency and data completeness will improve dramatically and ultimately lead to more impactful research.
Unifying technologies increases study efficiency
The site-based model remains the gold standard for interventional clinical trials, with proven, well-established processes that work effectively in conducting research. When the pandemic hit, researchers rushed to implement DCTs out of necessity, but these failed to deliver smooth, integrated end-to-end trials.
DCT models tried to adapt proven interventional research processes to a remote approach. Technology immaturity and a lack of hybrid-native solutions posed major challenges. Implementations felt “heavy”, requiring extensive workarounds, especially for studies needing both site-based and remote components. Initial DCT approaches didn’t accommodate this reality.
Most DCT vendors subsequently narrowed their focus to electronic solutions that could fit more easily into existing site-based, interventional trial designs. DCTs have become a collection of disparate, fragmented digital technologies and point solutions that have led to many of the same challenges and inefficiencies.
This approach doesn’t fully realise the transformative potential that modern, unified, virtual platforms can now deliver. New integrated platforms that leverage advanced technologies like AI enable sophisticated coordination among researchers across all components of a study, including patient consent, enrolment, medical record collection, and patient-reported outcomes collection. This solves execution challenges of cobbling together standalone components and represents a new model for virtual and hybrid trial designs.
Observational research and low-interventional studies particularly benefit from this modern virtual approach, allowing research teams to streamline study processes, instead of managing multiple solutions.
Unifying data collection provides a more complete picture of patient health
Comprehensive EHR collection that integrates patient records across all providers is critical to unlocking effective virtual models. Neither site-based approaches or traditional DCTs adequately collect the full patient experience that observational research requires.
Data gaps in observational research are the norm. Patient recall during site visits is limited, and data is fragmented in many silos and across many sites, making it difficult to collect the evidence needed to understand ongoing safety and efficacy of therapeutics. Post-marketing safety studies, for example, can be incredibly burdensome in terms of the time and cost it takes to collect data and extract insights.
Traditional DCTs primarily focus on prospective data collection, adapting interventional research processes to virtual approaches and providing an incomplete picture of patients. This approach offers a limited view of the patient journey and fails to capture full medical histories and real-world data critical for observational research.
Next-generation virtual models can now collect and integrate the patient's full retrospective and prospective medical history from all places a patient receives care, in addition to the patient-reported data. This allows researchers to gain a complete, 360-degree understanding of each patient’s journey.
This advanced virtual approach benefits from key advancements, such as EHR integration across healthcare providers (not to mention, enhanced remote engagement tools that make patient participation more convenient). And now, with AI, researchers can quickly collect and find patterns in patient data that would traditionally take years of cumulative clinical abstractors’ time to discover.
Looking forward, there is significant untapped potential for AI to easily capture data from more sources, helping researchers better understand the safety of treatments and more effectively monitor and improve patient outcomes.
A new era of observation research is underway
The industry is shifting from traditional site models and inflexible DCT solutions to embracing unified platforms that seamlessly integrate technologies and provide comprehensive EHR collection, establishing the foundation for fully virtual or hybrid observational studies.
This transition democratises research participation and improves study representativeness by eliminating systemic barriers for patients. Enhanced patient engagement and streamlined processes improve data quality and study efficiency, benefitting both participants and sponsors.
With passive data collection and reduced burden on sites, virtual-first approaches will usher in a new era of observational research that is more efficient, effective, and patient-centric.
About the author
Troy Astorino’s journey in healthcare started in the offices of his parents, who were both doctors. He saw firsthand the importance of patient-centred care, and the problems that come from records siloed across EHR systems. With deep AI expertise from MIT and SpaceX, he co-founded PicnicHealth to make it easier to capture patient-centred data and improve healthcare. He is PicnicHealth’s CTO.
