Ken Getz: Findings from 2017 eClinical Landscape Study

This presentation summarizes the key findings of the 2017 eClinical Landscape Study, a baseline assessment conducted by Tufts (Ken Getz) focusing on the current state of clinical data and data management practices across the pharmaceutical and life sciences industries. The study surveyed nearly 260 companies, primarily sponsors and Contract Research Organizations (CROs), providing an aggregate view of industry experiences and practices, highlighting areas where expectations were affirmed and where surprising challenges emerged regarding efficiency and data handling.

One of the study’s key affirming findings was the significant and growing diversity and heterogeneity of data elements being collected in clinical trials. Data is no longer confined to Electronic Case Report Forms (eCRF) via Electronic Data Capture (EDC) systems, but now includes inputs from social media, online communities, smartphones/mobile applications, and Real-World Evidence (RWE). This proliferation means clinical data must be integrated from numerous, often incompatible sources, creating substantial complexity and challenges for data management teams tasked with standardization and quality control.

Despite numerous industry efforts and technological advancements aimed at accelerating clinical development, the study revealed surprising insights regarding cycle times. The time required to manage and lock the data—the cycle time from the last patient visit to database lock (DBL)—remains stubbornly long and has not significantly decreased. Furthermore, the analysis uncovered significant downstream effects of data management challenges, particularly impacting site personnel. Delays in database build and release, often associated with protocol changes or functionality requirements, directly increase the time site staff must spend entering patient data, creating friction and inefficiency at the clinical site level. The study specifically measured and discriminated the impact of various causes of delay, finding that while protocol design changes are the most frequent, changes to the database design itself and its underlying functionality have the most profound negative impact on the final study closeout cycle time (last patient last visit to DBL).

The research also provided insight into operational advantages within the ecosystem, noting that CRO companies tend to offer a speed advantage across most data management timelines, including database build/release and DBL. This speed is attributed, in part, to the CRO’s more standardized and systematic approach, driven by the need to maintain efficiency against a predefined budget and plan. Ultimately, the study concludes that the industry is only at the very beginning of an evolutionary period regarding clinical data diversification. Organizations must anticipate the implications of this growing data complexity and prepare to fundamentally change their data management practices to cope with the influx of heterogeneous data sources.

Key Takeaways: • Data Heterogeneity is a Baseline Challenge: Clinical trials are increasingly reliant on diverse data sources beyond traditional EDC, including RWE, social media, and mobile applications, necessitating robust data integration strategies to handle incompatible formats and sources. • Stagnant Data Management Cycle Times: Despite industry focus on acceleration, the cycle time required for data management, particularly the duration from last patient last visit to database lock (DBL), has not improved, indicating fundamental inefficiencies remain unaddressed. • Database Design Impacts DBL Most Severely: While protocol changes are the most frequent cause of delays in database build, changes to the database design and functionality itself have the largest measurable impact on extending the critical DBL cycle time at the closeout of the study. • Downstream Effects on Site Personnel: Data management challenges, such as delays in database release, directly increase the burden on site personnel, forcing them to spend more time on data entry and potentially impacting data quality and site satisfaction. • CROs Offer a Speed Advantage: Contract Research Organizations (CROs) generally demonstrate faster timelines across key data management metrics (database build, DBL) compared to sponsors, likely due to their systematic, standardized operational approaches and efficiency focus. • Anticipate Evolutionary Change: The diversification of clinical data is viewed as being in its nascent stages; companies must proactively anticipate and adapt their data management practices to handle the coming wave of complex, non-traditional data. • Focus on Functionality Stability: To reduce cycle times, organizations should prioritize finalizing and stabilizing database design and functionality early in the study lifecycle, as changes here create the greatest drag on study closeout. • Integration of Incompatible Sources: The need to integrate data from incompatible sources presents a major technical hurdle, demanding advanced data engineering solutions and robust data pipelines to normalize and harmonize disparate datasets. • Need for Automation in Data Management: The persistent length of data management timelines suggests a critical need for automation and intelligent systems to streamline data cleaning, reconciliation, and quality checks, particularly given the increase in data volume and variety.

Key Concepts:

• eClinical Landscape: Refers to the technological ecosystem and practices surrounding the collection, management, and analysis of clinical trial data.
• Cycle Time (Last Patient Last Visit to DBL): A critical metric in clinical trials measuring the efficiency of data management and quality control processes from the end of patient treatment/observation until the final database is locked for analysis.
• Data Heterogeneity: The state of clinical data originating from many different sources (EDC, RWE, mobile apps) and existing in various formats, often requiring significant effort to integrate and standardize.
• Real-World Evidence (RWE): Data collected outside of traditional randomized controlled trials, such as electronic health records, claims data, and patient-generated data, increasingly used to support clinical decisions.