menu
target

Advanced analytics methods including machine learning, alongside our expanding access to more and better healthcare data sets, are helping to solve many of the problems that have plagued clinical research for decades. When we integrate disparate healthcare data sources, and apply cutting-edge analytics, we can design and conduct faster, more predictable studies with fewer amendments and higher quality results.

At least that is the promise.

Pharma leaders have recognized the potential benefits of analytics for at least a decade, but like many industries, harnessing the value of big data has been difficult to achieve. Data generated in the real-world by the healthcare system are being used for a different purpose than they were captured for. They are heterogeneous, complex, and disconnected. The effective use of analytics for clinical research requires effective data curation and management as well as advanced infrastructure and specialized analytic solutions—combined with deep clinical and scientific expertise to ensure the analysis is relevant and actionable.

At this week's DIA meeting in Chicago, we will present a session entitled Changing Trials through Analytics: How it Works. The presentation will explore the difference between traditional clinical development and our next generation, analytics-driven approach. We also plan to share several real-world examples of how the effective application of patient-centered big data analytics is already bringing new levels of precision and predictability to trials.

Data 101

Beyond traditional data used to plan clinical trials such as experience, performance, and competition, patient-centered data brings a new dimension to develop a more comprehensive view into the optimization of study design, planning and execution. However, no one single data type or source is relevant for all trials—it depends on the specific disease and application at hand. To inform this, data can be thought of in two broad categories, and both are vital to achieving our goals for the next generation of clinical development:

  1. Broad, nationally-representative data. These data sets, which include pharmacy records and insurance claims, provide insights into vast populations of patients without a lot of detail. They can help researchers understand standards of care in a region or population, locate patient pools and treating physicians, and more accurately choose trial sites for maximum recruiting opportunities. 
     
  2. Deep, clinically-rich data. These data sets, which include electronic health records and genomics data, offer clinically-rich and longitudinal information about anonymized patient records, but on a smaller scale. This information provides great insight into the patient’s journey. They can be used to answer important questions about patient experience, and the outcomes that are important to them. It can be used to refine protocols, support research assumptions, and locate hard to find patients, which is becoming increasingly important in an age of precision medicine.

When companies have the ability to access, analyze, and apply the right data as part of their clinical research approach, they can overcome long-standing barriers to clinical research, and make evidence-based decisions for more predictable results.

We are already seeing the profound affect these insights can have on research results. For example, these tools can also be used to pinpoint patients who meet inclusion/exclusion criteria—before a single site is opened. For example, in a recent Inflammatory Bowel Disease study, we initially found two sites that each had access to more than 350 potential patients. But practice patterns vary considerably, and after a detailed analysis of the patient level data, we found one site had more than 100 patients who met specific prior treatment criteria, while the other had just 16. Being able to conduct that more detailed analysis meant we could avoid choosing a site that had little chance of meeting its enrollment goals. In the presentation at DIA we will describe several of these analytics-driven success stories that underscore how data can be used to meet the strategic goals of a trial.

It’s important to remember that such accomplishments can be achieved only when talent, technology and data creatively combined to address the unique needs of each trial. Working with a partner that does this routinely is critical to achieving these goals.

  

We look forward to seeing you at our session. If you are unable to attend DIA, you can watch the presentation livestreamed from our website Tuesday, June 20 at noon Central time.