4 Factors Changing The Way CRAs Monitor Trials
By Patrick Floody, Vice President, Global Clinical Trial Services, Regeneron
Clinical research has undergone a remarkable transformation over the past three decades, reshaping the role of the CRA from a primarily site-focused monitor to a data-driven strategist. Once defined by thick binders of case report forms and exhaustive source document verification, CRA responsibilities now intersect with advanced technologies, global regulatory frameworks, and sophisticated data management tools. As the industry embraces electronic data transfers, dynamic dashboards, and AI, the question is no longer whether the CRA role will evolve, but how quickly and to what extent these innovations will redefine the future of how CRAs oversee clinical trial sites.
CRAs are essential to the drug development process, ensuring clinical trials adhere to strict regulatory requirements, ethical guidelines, and study protocols. Their responsibilities may include identifying, onboarding, and monitoring trial sites, collecting and reviewing regulatory documents, negotiating budgets and contracts, and serving as a key point of contact with site staff and investigators.
CRAs play a vital role in maintaining high standards of data quality and compliance, ensuring that trials are conducted efficiently and ethically. During site visits, they meet with study coordinators, review medical records, verify drug accountability, and conduct source document verification (SDV), which involves comparing source documents against case report forms (CRFs) to ensure data accuracy.
While society has seen some major technological changes over the past 30 years, there have been relatively few technological advances in how we manage clinical trials and monitor sites. However, thanks to recent technological advances, there will be major changes in how a CRA gets the job done. Four key areas stand out as having the greatest potential for impact.
EMR To EDC Data Transfers
Electronic data transfer eliminates the need for CRAs to conduct SDV of any data that is electronically transferred and makes data instantly available to CRAs and study teams for instant analysis. Many sponsors, electronic medical records (EMR) companies, tech vendors, and sites are currently working to establish direct data transfers from the site EMR to the sponsor’s EDC systems. This is often referred to as E2E (EMR/EHR to EDC). There are a few companies that have an electronic source document that is created specifically for each study and collects only the information required by that protocol. This eSource then transfers data to the EDC system. While other companies have options to transfer data from the EMR directly to the EDC, this often requires detailed mapping of data to ensure accurate data transfers. However, several different options may likely be needed in any given study to allow all data to be collected through data transfers. To further compound this, some sites are developing their own proprietary system or have sole provider contracts with a tech vendor. This will greatly increase complexity and likely add to the workload of the data manager.
Availability And Use Of Data
The amount of data available to CRAs has grown significantly and will continue to grow. This compounds the preparation needed before a site visit. Historically — and for many today — most data assimilation is done manually or through spreadsheets. Dashboards or control towers have commonly been used to manage clinical studies for the past decade. Many present data in static displays, while only a few offer more advanced functionality. The static ones are not much better than spreadsheets and have a refresh period that may be weekly or monthly, which limits their impact. The more effective dashboards are those that are dynamic and drillable.
Because these dashboards are linked to either a data lake or directly to source systems, they are constantly refreshed and interactive. This enables CRAs to see the current status of the site of interest while also allowing them to compare one site to another. They also allow the study manager to quickly and easily see overall study progress and provide better CRA oversight, as they have access to the same information as the CRAs assigned to the study.
If built correctly, dynamic dashboards can be rolled up to a portfolio level, but this requires considerable planning in data nomenclature and strong data management. Once built, they take considerable time to maintain and govern, as there are constant requests for improvement. However, AI can update dashboards in real time and will remove the need for most of these tools.
The Role Of AI
Everywhere you look, AI is part of the conversation. It will be hard to identify any area of the ecosystem that will not be impacted. While I am not a data or computer scientist, I am quickly learning about AI. But before we get into the specific impact AI will have on clinical trials and monitoring, I need to clarify the difference between AI and automation. Most clinical trials have benefited from automation for years, such as filing certain types of documents in a TMF based on the document’s metadata or having a bot read an FDA Form 1572 and then checking the TMF to see if a CV or financial disclosure form (FDF) is filed for each individual listed.
AI, on the other hand, can assimilate information to understand a situation and then respond to the situation or make a recommendation based on the data.
Let’s say a CRA is preparing for an upcoming monitoring visit at a high-enrolling site. An AI agent can pull information from the last monitoring visit, any correspondence between visits, enrollment activity, and drug inventory based on inventory verification and drug reconciliation (IVDR) CRF completion and quality from the EDC system and identify any quality issues from a site audit or a significant quality event. The agent can then create a list of items for the CRA to investigate during the site visit based on the aggregate data. This has the potential to save the CRA several hours of prep work. Multiplied by the number of site visits performed each year, this can add up to days or even weeks of saved time. To take this a step further, by analyzing this data at all sites, the agent can then recommend which sites need a visit, which need a phone call, and which are doing fine on their own. The AI agent can then build and update dashboards based on its evaluation of each site, with those insights rolled up to a study dashboard and then to an asset- or portfolio-level view. This will help the industry truly achieve a risk-based quality management (RBQM) strategy.
The Impact Of Changing Regulatory Expectations
In some ways, regulatory bodies are encouraging the pharmaceutical industry to adopt strategies that will directly impact monitoring practices.
ICH E6 (R3) strongly encourages sponsors and investigators to adapt and adopt current practices that embrace RBQM methodologies. Their stance on using RBQM was also found in the prior revision of the guidelines, but the wording is much stronger in R3. This third revision also places greater responsibility on sponsors and investigators to justify and document the use of RBQM, establish safeguards, and define metrics to evaluate the effectiveness of those safeguards. While many sponsors have taken these steps in the past, R3 may drive broader adoption of risk-based practices. Regulatory bodies at the central agency level are generally aligned with RBQM; however, field-based inspectors are not always fully up to speed on RBQM and continue to report on issues that don’t materially impact trial quality, clinical outcomes, or patient safety.
The ICH, the U.S., and several European countries have issued guidance on the use of AI in drug development. The guidance across the U.S. and Europe is generally aligned; the introduction of country-level guidance is likely to lead to divergence in requirements as individual countries put their own stamp on implementation. Similar to R3, individual countries are beginning to add their own requirements. Although these measures are designed to make it easier to conduct clinical trials in that country, they also mean that sponsors may need to navigate subtle country-by-country differences in requirements.
The FDA developed its own AI tool, Elsa, and has been using it as part of clinical protocol reviews and for label comparisons. Additionally, the FDA has been encouraging sponsors to use AI in drug development, but leaving it up to the sponsors to decide where and how to use it, asking sponsors to carefully review what the AI is doing, have a clear audit trail of it, and document what analysis and recommendations come from AI.
Summary
The CRA of the future will stand at the crossroads of tradition and innovation, balancing interpersonal skills and regulatory expertise with fluency in data analytics and AI-driven tools. Emerging practices, such as end-to-end data transfers, risk-based quality management, and intelligent monitoring systems, promise to streamline processes and elevate trial quality, but they also demand new competencies and adaptability. Sponsors and regulators alike must support this transition, ensuring CRAs are equipped to thrive in an increasingly complex, technology-enabled environment. Ultimately, the evolution of the CRA role is not just about efficiency — it is about safeguarding patient safety and data integrity in a rapidly changing clinical landscape.
About The Author:
Patrick Floody, MBA, is the vice president, head of global clinical trial services, at Regeneron, where he oversees global site management, central monitoring, study feasibility, and site selection, as well as study/site start-up, eCOA, and patient technology initiatives. He joined Regeneron in 2020.
Before joining Regeneron, Patrick worked at Pfizer for 25 years in clinical program operations, GCP quality (level 1), and analytics. Patrick has extensive international drug development experience, as he was part of the team establishing Pfizer’s global development country offices in Latin America, India, and other parts of Asia. He later served as the head of Japan development operations and the development Japan portfolio & project management groups for 10 years in Pfizer’s Development Japan organization.