By: Christian Schmidt, Associate Director, Phastar
Harnessing data to empower trusted, actionable insight is crucial to the success of any clinical trial. This is not about speed alone but having standardized, clinically meaningful, decision-ready data that supports timely action.
But with trials generating more data than ever, the industry is grappling with new management and analysis challenges. A major issue in clinical oversight is not only latency – manual work related to statisticians – but also heterogeneity across systems including different formats, identifiers, coding conventions and timing of updates. Despite the advantages of web-based, electronic data capture systems (EDCs), integrating disparate data sources continues to pose a major challenge to the maintenance of data quality.1 A further challenge lies in not only integrating this data but in making it consistent enough to trust.
To adapt to a changed data landscape, sponsors need to adopt unified, holistic approaches to data management and analytics. Data should be harmonized into a stable model which is continuously running and active and supports reusable, scalable oversight workflows.
Analytics platforms leverage new technologies alongside advanced statistical and data analysis expertise to accelerate drug development. Instead of relying on fragmented, resource-heavy processes which result in outdated insights, they provide access to real-time data – presented in a way which empowers users. This streamlines processes, improves decision-making and helps sponsors tackle the common pain points of clinical trials.
The Challenge – Delayed Data Access
The volume and variety of data collected in clinical trials is continuing to grow, with 5.9 million datapoints now collected on average per phase III protocol – up 11% annually since 2020.2 In this data-heavy environment, traditional data management approaches are struggling to keep pace. Periodic reviews which take place days, or even weeks, after the data has been collected can result in delayed detection of protocol deviations and adverse events. They are also labor and resource intensive, and subject to human error even when carried out by highly skilled staff.
Two other key challenges with traditional data management are data fragmentation and limited scalability. Four in ten sponsors say data harmonization is a key challenge.1 Different sites and systems rarely communicate seamlessly, resulting in data silos and further inefficiencies. At the same time, traditional site monitoring approaches lack the flexibility for the increasing adoption of decentralized and hybrid clinical trials.
What is needed is an approach that allows sponsors to immediately see when issues arise – and empowers them to take effective action to mitigate them.
The Solution – Real-Time Data Analytics
Access to real-time data empowers users to quickly identify trends and outliers and take effective action to tackle common clinical trial pain points. Advanced analytics platforms harness the power of new technology, including AI-powered data review and predictive modelling, to accurately assess vast volumes of operational, clinical and safety data, allowing us to delve deep into the data and generate meaningful insights more easily.
A key benefit is the automatic creation of data visualizations, allowing users to spend more time gaining a deeper understanding of the data and addressing anomalies, and less time trying to analyze data in sub-optimal formats. Interactive elements allow users to explore and render rich information from a single view. Various, unstructured data sets, including electronic data capture (EDC), can be integrated into user-friendly dashboards, aiding cross-functional review and understanding.
Centralized dashboards which aggregate and display critical data from multiple sites in real time, provide clinical teams with a holistic view of the entire trial. This allows seamless management across clinical trial phases, helping to unify previously siloed processes. At the same time, a shared suite of tools helps to increase collaboration across teams and raises wider awareness of the importance of the data management function.
Advanced analytics platforms provide a holistic approach to medical monitoring, clinical operations, safety and RBQM. This unified approach allows us to more effectively tackle common clinical trial pain points.
Pain Point One: Site Monitoring
Traditional monitoring approaches involve frequent on-site visits and 100% source data verification (SDV). However, 100% SDV is costly, resource intensive and predominantly detects random errors which have little impact on the results of clinical trials.3 In contrast, risk-based quality management (RBQM) approaches allow targeted monitoring based on site performance and risk levels. Recent Good Clinical Practice (GCP) guidance, including ICH E6(R3) has emphasized the importance of harnessing data analytics and basing monitoring activities on identified risks.4
A key feature of RBQM is its capacity for near real-time alerts. This can alert users to issues like protocol violations, persistent delays in data entry at a site or region, data quality inconsistencies or unusually low adverse event reporting versus comparable sites. Early identification of these issues enables immediate action to address underperforming or at-risk sites – before they lead to costly delays or compromise trial integrity.
Automated risk assessments and risk heatmaps quickly identify sites which deviate from expected performance for further investigation, for example if lab data inconsistencies are creating false outliers or signals. This supports adaptive site monitoring, allowing sponsors to focus resources on high-risk sites and ensuring teams can concentrate their efforts where they are needed most.
While access to real-time data is crucial, not all data in a clinical trial has the same level of importance. Recent analysis found a third of all data points collected in phase II and phase III clinical trials were not needed for the studies’ key analyses.2 RBQM applications allow clinical teams to focus on critical data which has the greatest impact on patient safety, data integrity, or overall trial success. This focus is particularly important in high-risk areas like oncology where there may be complex endpoints like overall survival (OS).
By incorporating an RBQM tool into a unified advanced analytics platform, we can help clinical operations teams focus on high-risk areas, enhance data integrity and improve patient safety.5
Pain Point Two: Recruitment
Recruitment is major pain point for clinical trials. Around 8 in 10 trials fail to meet their initial enrollment target, resulting in lost revenue of up to $8 million per day.6 Common challenges include an inability to quickly identify bottlenecks and a lack of necessary information to apply effective solutions.
A recruitment forecasting app using advanced statistical modeling can predict participant progression and treatment administration timelines, accounting for late site openings and real-time data updates. This can optimize site performance, reduce timelines and generate savings.
For example, Sponsor X had clinical trials which were failing to meet the target of administering the first treatment to the last participant within the planned time frame. Use of an advanced data analytics platform, including visualization software, provided new insights into recruitment efforts. The app predicted when the last participant would receive treatment, factored in non-started sites and proposed strategic reallocation scenarios. Integration of a novel Kaplan-Meier model enabled different countries to optimize sites based on participants in screening and their probabilities of progression, accelerating trial completion.
Staff were empowered with real-time data updates and customized forecast filtering. The integration of a write-back solution further empowered authorized users to view forecasts and actively reallocate participants based on real-time data and predictions. Together, these tools empowered data-driven decision making and proactive enrolment management.
By harnessing real-time data, we can dynamically adapt to recruitment challenges and optimize participant distribution for maximum efficiency.
A New Era of Real-Time Data
Real-time data access offers the opportunity to unlock new insights in clinical trials, leading to greater efficiency, improved patient safety and the reduction of historic pain points. However, we need the right tools to ensure the data is consistent, trustworthy and meaningful.
By combining new technologies with advanced statistical methodologies, unified analytics platforms offer the opportunity to optimize performance across clinical trial phases, unlock insights that support timely action and generate significant savings. It is vital these platforms make data accessible, consumable and actionable in real time for users – and that they are scalable across multiple phases and trials.
In a rapidly evolving data landscape, sponsors can no longer afford to rely on outdated manual processes. Adoption of effective tools which harness the power of real-time data and empower cross-clinical teams with new knowledge will be crucial to success.
About the Author
Christian Schmidt is an Associate Director at Phastar, leading Clinical Intelligence implementations across multiple oncology studies and clients. With over a decade of experience working with data insights, transformations, and architecture, he has developed deep expertise in turning complex data into actionable intelligence. Throughout his career, Christian has guided organizations through digital transformation, led high-performing consulting teams, and delivered innovative, data-driven solutions to support decision-making in challenging environments. At Phastar, he combines his technical knowledge with strategic leadership to deliver high-quality Clinical Intelligence solutions that drive efficiency and value for global sponsors.