As a sponsor of the 2013 CBI Risk-Based Monitoring (RBM) conference, CluePoints had the opportunity to connect with other industry thought leaders and present to a highly engaged audience in downtown Philadelphia. Dr Marc Buyse, founder of CluePoints, gave a presentation which covered site performance and quality through Key Risk Indicators (KRIs) as well as Central Statistical Monitoring (CSM), examined software available to enhance CSM and insight into case studies that utilize this approach. Several questions were raised by the audience after the talk and at the CluePoints booth, to which tentative answers are suggested below. We welcome feedback, comments and criticisms on these questions/answers.
How does Central Statistical Monitoring differ from Key Risk Indicators?
Central Statistical Monitoring is holistic, unsupervised and universal.
- Holistic means that all data are taken into consideration (clinical data, lab data, patient reported outcomes, etc.)
- Unsupervised means that the data are being checked for atypical patterns without any prior hypothesis in mind (so the data could show discrepancies in terms of means, variability, propagation of identical values, excess of visits on Sunday, etc.)
- Universal means that the software is identical for all clinical trials (only the data structure is specific to each trial).
In contrast, Key Risk Indicators are data-specific, supervised and study-specific.
- Data-specific means that only key risk data are taken into consideration (e.g., lab values or patient reported outcomes would not be seen as KRIs).
- Supervised means that KRIs look at pre-defined indicators (e.g., over- or under-reporting of SAEs).
- Study-specific means that some KRIs may be peculiar to a study (e.g., blood pressure may be used in a KRI in a cardiovascular trial but not in a cancer trial).
Do regulatory agencies recommend one approach over the other?
The agencies bring to our attention the fact that sponsors should be targeting their on-site monitoring activities, but they do not recommend a unique approach to risk-based monitoring. Rather, they mention a range of approaches as being potentially useful. Our view is that several complementary approaches are better than a single one. Subjective Key Risk Indicators based on operational metrics are certainly of value but there is no doubt that an objective assessment of data quality is crucial to managing regulatory risk and ultimately reducing costs. The FDA ‘s final guidance on Risk-Based Monitoring and Clinician Oversight further amplified it’s encouragement to use Central Statistical Monitoring in order to detect anomalous data, target SDV and de-risk studies as a consequence.
What are the benefits of CSM?
The benefit of CSM is to check for data consistency across all clinical sites of a trial, so as to achieve data consistency when possible, or to document where and why data inconsistencies appeared otherwise. This approach ensures that you are able to determine the quality and integrity of your data and ensure that you focus efforts on errant sites quickly and efficiently – often resulting in reduced regulatory submission risk and costs.
Bringing together the industry leaders of Risk-Based Monitoring in clinical trials, the RBM conference in Philadelphia was informative and thoughtful with much interaction between the speakers and the audience. This event has further confirmed that RBM is a major step forward for our industry and one that all sponsors should embrace. The opportunity to improve data quality and integrity whilst mitigating risk is the outcome that everyone wants. CluePoints is delighted to be at the forefront of this initiative and is available to consult with anyone who is interested in the practical application of these techniques based on proven and validated experience to date.