Break-out Group Dissects the Development of a Risk-Based Approach To Clinical Trial Management
As SCDM's 2014 Chair I had the honor of presenting the keynote at the EU Leadership Forum held in The Netherlands, December 2014. Celebrating the organization's 20th Anniversary and all that's been accomplished, we took a light-hearted look at the Evolution of the Clinical Data Management Role.
The forum was interactive and informative, giving participants a chance to hear from regulators, industry veterans and innovators. We talked about a number of important topics, among them the very future of data management, as well as eSource and risk-based quality management within the context of the clinical data manager.
Expert panelists challenged us in thought-provoking roundtables. Afterward we were assigned to break-out groups where we drew on real-world scenarios from our own experiences. I was lucky to be assigned to Group 4, tasked with exploring development of a risk-based approach for clinical data management in conjunction with other organizational functions. As preparation we were asked to review core points of the EMA Reflection Paper on Risk-Based Quality Management in Clinical Trials. We were assigned a variety of questions and tasks to complete. A summary of key takeaways from my personal vantage point is shared below.
What are the challenges to taking a risk-based approach in CDM?
- Political risk and overall climate within an organization – This was identified as a common challenge we all face. While applying risk-based principles to clinical research is a welcome idea in some organizations, in others it's a more difficult transition to make, perhaps due to years of "this is the way we do it." Our group's advice is to know and understand your organizational climate and develop an approach that will best rise to the need.
- Agreement among data management and clinical people – Without agreement, little if anything can be expected to change with the exception of increased frustration. In order to improve how your studies are built, managed and closed, it's essential to get buy-in on new approaches. For study build, using a schedule of events as a starting point with iterative reviews might be an alternative to beginning with a specification. In the study conduct phase, one might explore the possibility of a joint data and clinical trial management plan that addresses data cleaning from both perspectives. The key takeaway here is the importance of being inclusive, recognizing the value different parties bring to the table.
Define CDM specific risk areas for the above.
Many different activities related to clinical data management can expose a clinical trial to risk. Here's where some of the key trouble spots are and what can be done to better manage risk.
Determine which of these risks are cross functional and the input / interaction required from these functions.
- Team Roles – As most project managers have come up through the ranks of clinical monitoring, they tend to be out of their comfort zone with biostatistics and data management. One recommendation of our group is have your different teams get together to help them better understand the others' role in putting a study in production, maintaining and closing it.
- Fragmentation – Many organizations function in silos despite our greatest efforts not to. Skilled project managers demonstrate an ability to bring different functional groups together to get the job done. They control the work and report upward to management
Additionally we noted organizations have their own unique definitions of the data management role. Any of the following may be included: reporting; programming; coding; provider management; eCOA; IRT; study management; back office oversight; and finally study data management. This further shows data management is one of the most dynamic roles in the industry!
Suggest how such an approach may be implemented.
- Quality Plans and Integrated Clinical/DM Plans – Among possible suggestions include development of a study quality plan or an integrated clinical and data management plan at trial outset. Such plans would help ensure stakeholders work together in an orchestrated manner.
- Risk Mitigation Plan – Just as IT groups have disaster recovery plans, a risk mitigation plan for data management may also be in order.
- Master Plan – Consider developing a plan covering all the moving parts of your study. Use this plan to get a holistic view of study risks. Consider data management early, rather than after data are entered.
- Risk-based Approach to Data Management – This will require having an Integrated Data Review Plan that looks at both clinical monitoring and data management. While the approach takes a change in mindset, it is critical.
- Database Lock – If you're closing in on database lock yet 250 discrepancies are open, it may be a good time to gather the team and determine how to resolve outstanding issues by asking the right questions.
- Study Management Goals – Hold quarterly reviews with your team to go over the plan and address any changes. Might financial incentives be part of the equation?
- Assessment of Risk – Risk is dynamic, meaning it will change over the course of your study. Be sure you know what might go wrong and operate on the principle ‘If something can go wrong, it will go wrong.' If risk level changes in your study, say from medium to high, or from low to medium, being upfront about it demonstrates you are keeping current. Identify your triggers and boundaries so no one is surprised when a level changes. Holding project review meetings on a regular basis can go a long way toward keeping everyone on the same page.
What are the challenges to this approach?
- Making sure there is adequate reporting around triggers for real-time insight into what's happening in the study.
- Meeting regularly and being proactive as opposed to reactive.
- Having checks built into the EDC and with controls that check legitimacy of edits; truly understanding what you're seeing; looking at the trends.
- Making sure you're not getting false positives, which tend to cause people to ignore future alarms and reports.
- Changing business processes or study expectations over the course of a study may require re-training and education to help keep everyone apprised of expectations.
From all of this discussion that took place at the European CDM Leadership Forum, one realizes information management, project management, and so many other aspects are important elements of our role as clinical data managers. This was a great beginning to a regular meeting that SCDM is planning on holding. The viewpoints expressed by roundtable participants, speakers and attendees provided a rich collection of perspectives and information that certainly helped me in my personal and organizational quest for understanding how we can apply risk-based principles and techniques to data management challenges.
Drop me a line and let me know what you're seeing from your vantage point. I look forward to hearing from you.