Insights from a Data Scientist: eIQ Review Use Case Spotlight | Part 3

As discussed in Part 1 and Part 2 of this blog series, the integration of AI/ML can address the limitations of traditional data management approaches that are simply not scalable enough to support the pace at which clinical trials are evolving. With today’s clinical trials becoming increasingly complex and data intensive, leveraging innovative technologies will not only ensure that data management objectives are achieved, but can also provide opportunities to streamline processes and increase productivity across the clinical trial life cycle. Further driving the need to take an innovative approach to data management practices is the evolution of Clinical Data Management into Clinical Data Science. According to SCDM’s position paper on this topic “Clinical Data Science also expands the scope of Clinical Data Management beyond the study construct by requiring the ability to generate knowledge and insights from clinical data to support other clinical research activities which requires different expertise, approaches, and technologies.”

Summarized below is another example of an AI/ML-enabled use case deployed within elluminate^® IQ Review (eIQ Review) and highlights the data management efficiency gains as a result of leveraging these capabilities embedded within the elluminate Clinical Data Cloud while also supporting the evolution of clinical data management into clinical data science.

Concomitant Medication Consistency (CMCON)

Concomitant Medication datasets (CM) include the medications patients have taken during a clinical trial. Each row within a CM data set consists of the medication name, route, frequency, dose, unit, indication, and dates along with other variables – FDA submission requires that the data is consistent and recorded correctly in order to maintain data integrity. The identification of inconsistencies across CM data is traditionally performed manually by a data reviewer to ensure that unit, route, and frequency are consistent with the name of the medication across each row. Because there are oftentimes thousands of records within CM datasets, manual review is time consuming, resource intensive, and increases the opportunity for human errors to be made.

The Concomitant Medication Consistency (CMCON) Model within eIQ Review predicts the unit, route, and frequency of a concomitant medication given for each entry and compares it with the recorded value for consistency. The CMCON model uses a supervised CatBoost algorithm model to identify records that are not consistent and automatically surface these to the user on a record level within eIQ Review. This model is designed to assist the data reviewer when checking for consistency between medication, dose, unit, route and frequency. Leveraging this model significantly reduces the workload of the data reviewer by eliminating the need to review up to 97% of CM records.

As previously noted, the more abundant and complex clinical trial data becomes, the greater the need to adopt advanced technologies for managing data in a scalable manner. Additionally, the evolution of clinical data management into clinical data science serves as a catalyst to incorporate innovative data management approaches. As part of eClinical Solutions’ Innovation Series, Diane LaCroix, VP of Clinical Data Management and Venu Mallarapu, VP of Global Strategy & Operations, discuss how eClinical’s Biometrics Services team is leveraging the elluminate Clinical Data Cloud to transform their approach to align with this evolution. Listen to the full conversation here: Innovation Series: Evolution of CDM to CDS.

Authors

Deepthi Pabbisetty

Senior Data Scientist, eClinical Solutions

Alison Cronin

Director of Product Marketing, eClinical Solutions