RBM: The Heart of Data Quality

November 11, 2015
Lisa Henderson

Lisa Henderson is Editor-in-Chief of Applied Clinical Trials and Pharm Exec. She can be reached at lhenderson@mmhgroup.com.

Applied Clinical Trials

CBI’s conference proves RBM was just the beginning. Now companies are honing in on the risks and mitigations related to clinical trial data reporting.

CBI’s conference proves RBM was just the beginning. Now companies are honing in on the risks and mitigations related to clinical trial data reporting.

 

CBI’s “Risk-Based Trial Management, Is It Only About Monitoring?” was held last week, and in answer to the question, basically, no, it’s not only about monitoring.

Discussion this year at the RBM conference has our editorial staff now wondering how we should be defining the acronym RBM. Is it Risk-Based Monitoring or Risk-Based Management? This year’s conference proves the discussion has really moved from implementation and monitoring to next steps in RBM: what it means to the quality of the data, and the trial, as a whole.

Not to say that monitoring is not central. Many speakers discussed monitoring as a triad-on-site, remote and centralized-the misconceptions of those legs and how they are evolving, or in some cases, not evolving.

For example, remote monitoring to some sponsors has meant extra legwork for the sites. It means not sending an on-site monitor to perform Source Document Verification (SDV), but for the site to fax in every piece of documentation to a remote location for review. “That was never the goal or intention of RBM,” said Jules Mitchel, President of Target Health. Other speakers shared how their entities-sites, CROs, and sponsors-have accepted the change in on-site monitoring as one that empowers the monitor and strengthens the data that sites produce.

Roger DeRaad, Director of Black Hills Cardiovascular Research, expressed his center’s acceptance of the change in monitoring. “But we do want to know, just like sponsors do, what the metrics are. We struggle to collect metrics on the things that matter,” he shared. Going back, he found a 46% decrease in on-site monitoring activity in a 2.25-year period 2014 to current. DeRaad said the monitoring change is based on perception: “When a monitor is there it feels like they are there for us. But when it’s on the phone, it feels like they are taking our time.”

Interestingly, the move away from 100% SDV, while proven in many instances to not be the most efficient or effective use of a monitors’ time or for translation into data quality, is not yet fully embraced. Examples included presentations of parallel implementations of RBM vs. SDV (or traditional monitoring approaches) to show improvements, as well as anecdotal information that sponsors and monitors alike still cling to the SDV and static on-site monitoring review plans. Applied Clinical Trials spoke to one small CRO executive who was attending to get more information and background on RBM. “We are getting ready that [our clients] will want to move to RBM, it’s inevitable, but now they still want 100% SDV.”

However, many presentations showed a clear path to RBM, and the use of technology to forge that path. Joanne Benedict, Senior Advisor at Roche, who presented a workshop on TransCelerate’s view of RBM, said, “eSource will make the SDV and SDR discussion moot. Anything to make the process more automated will be accepted.” She added that she was amazed to see what the technology companies have developed in just 18 to 24 months to facilitate that.

Some companies shared that they have implemented or tested RBM tools. Craig Serra, Senior Director and Business Process Owner, Data Management (Conduct and Closeout), Worldwide R&D for Pfizer, described its pilot use of CluePoints' centralized statistical monitoring product. “We wanted to take the CluePoints SMART™Engine for a test drive. The questions were: what use cases do we see for this software and are the statistically significant signals actionable in order to increase data quality?” Serra noted. With the successful pilot behind them, Serra said, “We are actively engaged in how we can include CSM as part of our development operations ecosystem moving forward, as the usage of CSM shows tremendous value in both RBM and data quality oversight frameworks.”

Duncan Hall, CEO of Triumph and TRI, an RBM consultancy with RBM solution, noted that smaller firms are struggling with what to measure, their Key Risk Indicators (KRIs). He is offering companies a free list of the Top 10 KRIs to start with. “How many indicators? It’s the Goldilocks’ question. You could have not enough, which isn’t helpful, or too many, which just creates noise.” Hall notes that the KRI discussion should be a cross functional one, and one of ongoing refinement.

That sentiment rang true with the other presenters. While there are basic KRIs for every study, there will be indicators that are just predicated on a single protocol or therapeutic area. In addition, risk indicators will change as the trial progresses. What is a signal for a site early in a trial may not be a problem after the signal is detected and the site re-trained on the issue.

Which leads to the protocol. If you think narrowing down the number of Key Risk Indicators is a problem, reducing protocol complexity is an exercise in “the restraint of data exuberance,” as explained by Sabrina Comic-Savic, Senior Director of GCP Compliance at The Medicines Company. She presented on the need to reduce complexity in design, setup up, analysis and implementation to achieve Quality by Design in a trial.

For protocol design, that means aligning the design with the practice ie., patient population, medication and timing of the procedures for a site, and reductions in trial-specific interventions. Comic-Savic said that each data variable counts as a data acquisition cost, which goes up exponentially with the number of patients and variables. She said that they aim for 200 to 250 data points to collect as their goal.

Mitchel too is a believer in eSource and of the simple protocol. He said that regulators want to know that informed consent is properly obtained, the protocol is followed and monitored, the primary endpoint is measured and documented properly, that all significant safety events are captured and reported and drug/device supply is properly managed. Mitchel also recommends that both FDA and EMA be involved early and often in an RBM-designed trial.

Keep up-to-date with the latest RBM information on our web site.