Medidata unveils centralized statistical monitoring

Medidata-unveils-centralized-statistical-monitoring.jpg
(ipopba/iStock via Getty Images Plus) (Getty Images/iStockphoto)

The company’s Medidata Detect centralized trial monitoring solution is designed to improve management of data quality, site monitoring and patient safety.

Medidata, a Dassault Systèmes company, has launched its Medidata Detect centralized statistical monitoring solution. According to Medidata, the technology helps improve data quality and promotes patient safety in clinical trials.

Lisa Moneymaker, vice president of product management with Medidata, told Outsourcing-Pharma that Detect uses advanced analytical technology to break down data.

“Medidata Detect uses machine learning to survey millions of data points, comparing every variable in the data set to every other variable, searching for and identifying statistical relationships between them. Detect screens thousands of patterns in the data to identify inconsistencies or outliers that do not fit the pattern or fall outside the acceptable ranges established,” she said.

Part of the company’s unified Medidata Rave Clinical Cloud platform, the technology reportedly helps users better manage data quality, monitor site performance and promote patient safety. It is designed to pinpoint errors, trends and anomalies in data through statistical algorithms and tests.

According to a report in the Journal of the American Medical Association (JAMA), nearly 25% of new drug submissions had to go through one or more resubmissions before receiving approval, with an average delay of approval of 435 days after the first rejected submission. Detect seeks to avoid such delays through use of a central system for aggregation and review of data sources, flagging errors, trends and anomalies in real time, reportedly leading to improved data quality and decision-making ability during a trial.

Detect is designed to locate both known and unknown risks, then trigger corrective actions to help proactively minimize study delays and submission failures in trials. The solution reportedly enables users to define and manage study risks, expose unanticipated data anomalies and cut down on programmers’ time and effort through proprietary machine learning algorithms that require no customer programming.

According to Moneymaker, Detect can be of particular use for COVID-19-related trials.

Detect is unique in its ability to analyze clinical data for tens of thousands of patients in near real time without the need for ongoing extracts, imports, and refreshes; this is proving critical to COVID-19 trial support, as the patient volumes, patient diaries, and interim assessment demands require an anomaly detection tool capable of keeping pace,” she said. “Detect is currently being deployed across COVID-19 trials to improve data quality and reduce risk.”