With AI dominating healthcare headlines, it’s easy to dismiss this technology as being all hype that’s not ready for primetime. However, unlike more forward-looking applications for this technology, incorporating AI into drug diversion technology is no longer a “nice to have.” With increasingly complex medication workflows, siloed medication information across disparate systems, tighter compliance expectations and stretched clinical resources, it’s become an essential component of patient care, colleague safety and operational performance.
Across the broader healthcare system and at St. Jude in particular, we’ve seen firsthand how AI-enabled diversion detection can help transform what used to be a reactive process into a proactive strategy—supporting applications from compliance to billing and customized policy development.
The Perceived IT Burden
One of the biggest concerns around adopting this technology boils down to the perceived ongoing burden of adopting new IT solutions. Certainly, setting up new systems is an additional lift for informatics teams—but with AI-powered monitoring tools, the technology decreases the burden post-implementation. Rather than needing to pull reports on an ad hoc – and unpredictable – basis, IT teams can now invest in the successful implementation of this technology to reduce their workflow burden in the long-term.
These tools rely on the data feeds from multiple systems—automated dispensing cabinets, EHRs, wholesaler shipping feeds and reverse distributor systems—so they can monitor the full lifecycle of a medication. It can flag documentation gaps or discrepancies, such as Patient-Controlled Analgesia (PCA) pump entries that don’t align with administration records, surfacing potential manipulation or workflow breakdowns. IT remains a crucial piece of the puzzle, playing a vital role in setting up systems for success from the start.
If your data is incomplete or inaccurate, AI doesn’t fix the problem—it can amplify it. Algorithms will draw conclusions from whatever they’re given, and if that input is flawed or missing data, so is the output. Clean data and human oversight are non-negotiable. But when thoughtfully, carefully and strategically implemented, problems down the line can be significantly mitigated. At St. Jude, around 80 behavioral-pattern alerts help detect not just missing medications, but also suspicious activity—such as users back-charting, accessing drug cabinets or vaults when not on shift, or manipulating access levels.
Before using diversion detection technology, even identifying and accessing the right data required unpredictable and time-intensive requests to IT. Today, with upfront alignment, the system becomes repeatable, scalable, and far more efficient. That’s especially valuable in complex EHR environments, where crucial information can be buried in less-structured clinical entries rather than clearly defined data tables. Once live, the system becomes a reliable tool not only for the pharmacy and compliance teams—but also for reducing risk and limiting institutional liability.
Benefits That Go Far Beyond Diversion
In 2023, St. Jude had over 15 million controlled substances transactions – all of which need to be monitored in order to have a full picture of the supply chain. Implementing AI is like using a magnet to find a needle in a haystack. It still takes plenty of focus, but the task becomes far more manageable. And without it, you’re far more likely to miss something as it is not humanly possible to review that volume of data within a reasonable time to take action.
AI is helping hospitals more quickly detect diversion—and often uncover previously undetected diversion behavior within weeks of implementation. But the value doesn’t stop there: AI surfaces medical practice issues that give organizations the opportunity to adjust their internal policies and workflows, specific to their environment and patient populations.
Medication-transaction data that powers diversion detection also highlights mismatched NDC billing and flags potential false claims—shielding organizations from multimillion-dollar penalties. Dashboards then track issues ranging from missed waste documentation to clerical dosage errors surfaced in data tables, creating a solid foundation for accurate billing and stronger compliance. At St. Jude, unifying these data streams in one place has proved invaluable, enhancing staff and patient safety while streamlining operations and boosting audit-readiness.
What’s Next for AI in Drug Diversion Prevention
The next wave of diversion prevention will rely on generative, agentic AI that goes beyond spotting problems to orchestrating their resolution. AI-generated prompts can guide pharmacy and compliance teams through every step of an investigation, or produce a customized audit plan with a single click—eliminating the hours usually spent drafting emails and compiling reports. Right now, gathering all the data points needed to brief clinical staff can be a scavenger hunt, but AI can automatically assemble the relevant facts, charts, and context into a ready-to-send message, so everyone gets the information they need to act—without the administrative drag.
Building on that capability, the rapid rise of agentic AI—systems that can pursue goals autonomously, guided by real-time context and feedback—promises an even higher level of assistance. These tools can continuously monitor for anomalies, triage alerts, and launch preliminary fact-gathering the moment an issue appears, so pharmacy and compliance teams can devote their expertise to the most critical decisions rather than the groundwork.
AI isn’t a silver bullet. But when implemented carefully and leveraged strategically, it’s a powerful catalyst for better outcomes. AI in diversion prevention is not just about identifying problems—it’s about creating safer systems, using data more intelligently across the board and freeing up manpower for areas where people are truly needed.
About Alex Rodriguez
Alex Rodriguez, MHIIM, CPhT, is Lead Compliance Data Analyst at St. Jude Children’s Research Hospital, where he leverages data-driven insights and advanced technologies to improve medication safety, regulatory compliance, and operational efficiency across the pharmacy supply chain.
Mr. Rodriguez has over 15 years of pharmacy experience spanning retail, compounding, and hospital settings, with deep expertise in pharmacy automation, medication supply chain management, pharmacy regulations and project management. He holds a Master’s degree in Health Informatics and Information Management from The University of Tennessee Health Science Center.
About Karen Kobelski
Karen Kobelski is Vice President and General Manager of Clinical Surveillance Compliance & Data Solutions, Wolters Kluwer, Health, where she oversees solutions that provide clinical surveillance, risk detection and data normalization to improve the quality of patient care, regulatory compliance and operational performance of organizations in the healthcare industry.
She brings more than 25 years of leadership experience in software and services businesses to the healthcare industry. A Six Sigma Black Belt, Karen holds an MBA from Harvard Business School and a bachelor’s degree from Georgetown University.
