A Four-Step Plan for Optimizing Clinical Data Reviews
Clinical teams across the life sciences realm contend with an endless cycle of report requests, clinical data reviews, and overburdened infrastructure. Here’s our Rx for a technology strategy that can rejuvenate the data monitoring process.
In November 2016, the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) issued a revised guidance for Good Clinical Practice (GCP): ICH E6 R2. This guidance laid out a set of guidelines that trial sponsors should consider as they adopt risk-based approaches to trial monitoring and quality management. The guidance called for routine review of data collected, identifying data anomalies, outliers, and deviations, using statistical methods to detect trends, risks, and inconsistencies in data and that all decisions and actions from a centralized monitoring process should be audited throughout the course of the study. Since this guidance, nearly all big and mid-size pharma have begun to embrace the importance of applying centralized data monitoring strategies and techniques in line with this guidance.
But supporting such strategies continues to be a challenge.
Clinical teams are still locked in a painful cycle of static report requests, lengthy clinical data review cycles, and trouble with database infrastructure too limited to handle the data load. Clinical teams also struggle to unify and analyze data across domains, including adverse events (AEs), labs, concomitant medications, demography, and exposure, among others. Incomplete clinical data access disables the ability to see “at risk” patients and make optimal decisions on safety and efficacy, early in the process.
So what can clinical teams do?
They must adopt not only data monitoring strategies, but also technology strategies to support monitoring. There are four actionable steps on the way:
Create a data foundation that effectively manages your data needs.
First, your data foundation should ultimately be a holistic data review solution. The primary requirement for success is that it should provide anyone challenged with analyzing clinical data (such as medical monitors, safety review teams, biostatisticians, data managers, and pharmacologists) access to the data.
Create a plan for visualizing data within your monitoring strategy.
Access to data is critical, but once you have it, can you interpret it? Can you quickly spot outliers? When you find one, can you dig deeper into the underlying data to better understand?
A strong data foundation is key, but once the data is accessible it must also be usable. There are many visualization tools on the market, and many ways to build your own visualization layers.
Since there are nearly too many options to consider, it may help to define outcomes. Either by implementing an analytics tool, or by creating your own visualization in Excel or something else, you should ensure your monitors can:
Easily see deviations, violations, and possible risks. This means, the types of charts, graphs, or graphics need to be built for the data’s purpose. No one-size-fits-all.
Quickly focus on just the critical data, based on rules you define.
See the latest data and detail. Real-time is usually unnecessary. Your data foundation should ensure regular refresh, but make sure your visualization layer does the same.
Access underlying data. Most people who monitor data are naturally inquisitive. When they spot an outlier, or a cluster of events, they want to understand the full picture. Don’t restrict them — make sure your technology can support depth.
Ensure strong methods for cross-functional collaboration & communication.
Since data reviewers are naturally inquisitive, your organization should prepare for lots of questions.
Many clinical teams seek better collaboration. We did a recent survey of some of our clinical contacts, and found that more than 80% of clinical teams want to improve in this area. So what does this mean for your organization?
While reviewing safety data, your reviewers may have questions relating to outliers or suspect results seen in a clinical trial. So, being able to communicate the actual case data with the individuals responsible for maintaining the data is critical to assessing potential issues with the study drug.
Using two different systems, such as one for identifying potential issues (like many visualization tools) and the other for recording the request for information or follow-up (like email/phone/wandering past an office), is inefficient. It can also potentially introduce time delays or an inaccurate translation of the data in question.
Your organization’s technology strategy needs to accommodate ways for clinical teams to review, and then interact together within the data sets. That could mean annotating data points, but better still it should enable reviewers to assign tasks and queries to others, provide notifications and reminders, and truly encourage working in the tool.
This is because performing actions inside of the data set, such as asking questions and responding to questions, means that every communication and thought is captured. This matters because as reviewers explore things they see and identify follow-ups, they are unwittingly creating a complete data audit trail — something that can protect them, your organization, and the entire trial.
Track everything along the way.
Data integrity is under a great deal of scrutiny today in the life sciences field — with an uptick in FDA warning letters over the past few years, and scandals over clinical research popping up. It’s a valid point to keep an infallible record of reviewers’ and monitors’ work.
We call this a safety profile. By providing this view on trial participants, (including a view into patient demographics, adverse events, labs, vitals results, concomitant meds, and other safety domains), medical monitors can feel assured that they have not missed a potential safety signal. Further, clinical teams can be sure that in an audit they’ve got an accurate picture of the decisions that were made — and why they were made.
Taking into account the audit trail is a step that brings us back to our original action: ensuring a data foundation that works for your clinical teams. Your overall data and technology goal should essentially be to merge data across domains in a usable visual environment where monitors and reviewers can easily identify “at risk” patients, then investigate their profiles across data domains.
A technology strategy that supports your organization’s monitoring strategy will ensure that your teams are better equipped to discover trends across study visits, and identify abnormalities as quickly as possible. It will also reduce the churn and rework from static reports and encourage self-service exploratory analysis.
With better analysis, team members can optimize the clinical trial process and focus their efforts on obtaining the insights and answers they need to bring drugs and devices to market faster.