Innovative Approaches to Detecting Fraud in Clinical Research: Leveraging Big Data, AI, and Forensic Accounting Techniques

The prevalence of fraud in clinical research has long been an industry challenge, but one we have collectively avoided. However, recent high-profile cases have underscored its existence and regulatory bodies like the US Department of Justice has put the industry on notice that they will scrutinize how companies address this critical risk to patient safety. Traditional methods for ensuring data integrity—such as edit checks, monitoring, and data reviews primarily focus on errors, omissions, and internal consistency, which leaves deliberately falsified data undetected, as it often appears deceptively "normal."

To tackle this issue, Novartis and ZS have pioneered a groundbreaking method that transforms the detection of fraudulent data in clinical trials. By harnessing the power of big data analytics, artificial intelligence (AI), and machine learning (ML), we continuously monitor patient data across entire portfolios, identifying anomalies that could signify data manipulation. Our approach uses advanced algorithms, drawing on mathematical techniques from forensic accounting, to automatically flag data deviations at the site level—deviations that may indicate deliberate tampering when compared to broader datasets within the study, therapeutic area, or region.

This automated system not only enhances the detection of fraud but also promises a more efficient and scalable alternative to traditional monitoring methods. Beyond fraud detection, our approach offers significant potential for improving quality, safety, and medical-scientific review processes across clinical research, setting a new standard for data integrity and patient safety.