Season 2 Episode 9: Are We Too Risk Averse in Clinical Research?

This episode of the State of Digital Clinical Trials podcast features a discussion between Richard Young and Ken Getz, Executive Director of the Tufts Center for the Study of Drug Development (CSDD), exploring the evolution of clinical research, the industry’s pervasive risk aversion, and the transformative role of data and patient engagement. Getz provides a historical perspective spanning 35 years, categorizing industry shifts from the privatization of research in the late 80s/early 90s to the focus on operational feasibility and efficiency in the 2000s, culminating in the current era defined by personalized medicine and patient engagement. A central theme is the industry's conservatism, which Getz argues is reinforced by historically low success rates and worsening phase transition probabilities, creating inertia against adopting novel trial designs and executional models.

The discussion heavily emphasizes the changing role of data management, transitioning from siloed, reactive, paper-based data to centralized Electronic Data Capture (EDC) systems, and now moving toward integrated, cloud-based, and synchronized data environments leveraging data science and AI enablement for predictive qualities. A critical insight shared by Getz is the quantitative evidence demonstrating the increasing complexity and inefficiency of trials; his research shows that the average number of protocol amendments has increased significantly, often leading sites to bundle necessary scientific changes until they reach a threshold deemed worth the administrative burden, resulting in suboptimal study execution in the interim. The speakers agree that if clinical trials were redesigned today, they would look fundamentally different, driven by novel, decentralized models and greater site autonomy in technology selection (e.g., telemedicine solutions).

Furthermore, the conversation addresses the crucial need for improved collaboration and trust across the ecosystem—between sponsors, CROs, investigative sites, and patients. Getz, who also founded CISCRP (Center for Information and Study on Clinical Research Participation), highlights the industry's failure to establish sustainable public trust and perceived value in clinical research, noting that the awareness boost from the COVID-19 pandemic was temporary. He stresses that while the intent for patient centricity is widespread, execution is lacking, often resulting in decentralized patient engagement functions focused on short-term recruitment goals rather than long-term partnership. The ultimate aspirational goal is to remove the separation between clinical care and clinical research, moving toward a "learning health model" that leverages existing infrastructure and data within the clinical care environment to accelerate research and elevate public health.

Key Takeaways: • Rising Protocol Complexity and Amendments: Research confirms that protocol complexity is a persistent issue, with a significant increase in the number of protocol amendments over the last seven years. This inefficiency is compounded by the practice of delaying necessary scientific amendments until enough changes accumulate, leading to periods of suboptimal study execution. • Data Transition to Predictive AI: The evolution of data in clinical research has moved from siloed, reactive data to strategic assets. The current phase is defined by the promise of AI and machine learning, requiring greater focus on data compatibility, interoperability, harmonization, and centralized, synchronized, cloud-based access to leverage predictive analytics. • Risk Aversion Hinders Innovation: The industry remains fundamentally risk-averse, a mindset reinforced by historically low success rates in phase transitions. Overcoming this conservatism requires a cultural shift and a willingness to embrace entrepreneurial approaches—succeeding quickly but failing early—without compromising patient safety. • Site Enablement and Autonomy: The current "sponsor-flows-downstream" model is inefficient. Innovation should be sourced from investigative sites, which often solve problems locally (e.g., telemedicine). Sponsors and CROs should accommodate site-preferred technologies rather than imposing proprietary systems, fostering better efficiency and automation. • The Need for a Learning Health Model: A major transformative goal is to eliminate the separation between clinical care and clinical research. Leveraging the infrastructure, data, and personnel within the clinical care environment is essential for realizing a "learning health model" that benefits both patient treatment and scientific discovery. • Patient Centricity vs. Engagement: While patient centricity is a universally accepted goal, execution is often flawed. The focus should shift to "patient involvement" or "patient engagement," treating patients as active partners in the process, ensuring scientific rigor is maintained alongside relevance to patient outcomes. • Quantifying Diversity Disparities: Empirical research has confirmed that the race and ethnicity of investigative site personnel are correlated with and predictive of the race and ethnicity of enrolled patients, providing quantitative evidence that supports the need for greater diversity and inclusion efforts within research staff. • Eliminating Redundancy and Lack of Trust: Two major inefficiencies that should be eliminated include labor-intensive practices like 100% Source Data Verification (SDV) and the redundant oversight stemming from an inherent lack of trust between sponsors, CROs, and contracted providers. • Temporary Public Trust: The increased public awareness and trust in clinical research generated during the COVID-19 pandemic were temporary. Sustainable public support requires integrating clinical research into the mainstream societal mindset and providing continuous education on its value, rather than viewing it merely as an alternative for the desperately ill. • Impact of Technology on Data Management: The introduction of certain technological advances, specifically EDC, inadvertently broke the focus on the patient by centralizing data management and fragmenting the process. The industry is now circling back to personalized medicine, requiring data systems that support patient-level management and distributed access.

Tools/Resources Mentioned:

• Veeva Systems Inc: (Channel host/context) A leading platform in the pharmaceutical industry, particularly for CRM and clinical trial solutions.
• Tufts Center for the Study of Drug Development (CSDD): Research organization providing data and insights on drug development efficiency and trends.
• Center for Information and Study on Clinical Research Participation (CISCRP): A non-profit organization dedicated to educating the public and patients about the clinical research process.
• EDC (Electronic Data Capture): Mentioned as a technology that centralized data but inadvertently fragmented the patient focus.

Key Concepts:

• Risk Aversion/Conservatism: The deeply entrenched culture in the pharmaceutical industry that prioritizes safety and compliance over innovation, often leading to slow adoption of new technologies and methodologies.
• Protocol Amendments: Changes made to a clinical trial protocol after the study has begun, often cited as a major source of inefficiency and cost.
• Patient Engagement/Involvement: Moving beyond simply recruiting patients to actively partnering with them in trial design and execution, ensuring outcomes are meaningful to patients.
• Learning Health Model: A conceptual framework aiming to integrate clinical care and clinical research, allowing data generated during routine care to inform research and vice versa.
• Site Centricity: The concept that clinical trials should be designed and executed with the needs and capabilities of the investigative sites at the forefront, recognizing that sites are where research is localized.