In this webinar we will present a quality risk management approach for clinical trials that is based on use of modern software technology and data science - we call this approach "Risk Apps".
A clinical trial is a scientific experiment carried out over time over dozens, hundreds or thousands of subjects, often at multiple sites. Clinical trials often last for years and generate large amounts of data.
Monitoring is a quality control tool for determining whether the experimental activities are being carried out as planned, so that defects can be identified and corrected.
However - onsite monitoring cannot ensure quality on its own.
For example - onsite monitoring cannot detect experimental design flaws - such as not collecting a particular parameter that is critical to the analysis of the study end point.
Other - more subtle issues, such as accumulated dosage of a treatment may require statistical analysis during a study. So-called "soft data locks" of the EDC system that are performed during the study in order to enable the study statistician to perform an interim review, create blackout periods during which patients may be consistently under-dosed or over-dosed due to site non-compliance with the study protocol.
And, while data entry issues can be easily prevented by online edit checks in the EDC system; protocol violation and compliance issues cannot be rectified by re-entering data after comparison with source documents.
"Risk Apps" measure and assess risks to data quality and patient safety in real time using central statistical analysis. The "Risk Apps" approach recognizes that people are very poor at identifying and consistently tracking trends in data.
The saying - "Not seeing the forest for the trees" illustrates this problem graphically.
When we are too close to a situation, we need to be able step back and get a little perspective. When we focus on a large number of details (for example SDV in a clinical trial), we are guaranteed to miss the overall view of data quality, site compliance and critical design flaws.
Because people are poor at signal detection in large data sets, it is highly unlikely that routine visits to all clinical sites and 100% data verification will ensure patient safety and experimental data quality.
This claim is borne out in practice by industry studies that show that 100% SDV detects less than 2% study quality events. Conversely - research suggests that remote monitoring can identify the great majority of on-site monitoring findings. The review determined that centralized monitoring activities could have identified more than 90% of the findings identified during on-site monitoring visits.
Unlike people, computers are very good at signal detection over time in large data sets. Statistical analysis isn't restricted to a site visit once every 4-6 weeks by a CRA and unlike people; software doesn't tire and can run the same algorithm thousands of times every hour of the day.
We will discuss the process of implementing Risk Apps for a study and demonstrate how they work on a live EDC system. We will conclude by discussing exciting future directions of predicting issues using machine learning and applications of Risk Apps to ePRO - electronically patient-reported outcomes for helping improve patient safety.
Why should you attend: Are you concerned with control of your data during your clinical trials?
Are you frustrated by carpet-bombing techniques of 100% SDV (source document verification) and on-site monitoring that cost you 20-30% of total study cost yet often fail to detect critical protocol issues, design flaws and problematic sites?
Are you burning cash waiting for your statistical report while the clinical data management team is fixing problems with the data?
This webinar will show you how to reap the benefits of modern data science and dramatically improve your control and your return on investment in monitoring and shorten the time to statistical by up to 10X.
Areas Covered in the Session: