How to Improve Drug Diversion Investigation Workflows

As a former health system drug diversion investigator, I have experienced the ongoing responsibilities these professionals have to the organization. They manually comb through audit dispense transactions when there is suspected diversion. They also constantly look for ways to improve audit efficiency while working with organizational leadership who are responsible for ensuring a myriad of regulations, guidelines, and industry best practices for drug diversion monitoring and prevention, while keeping associated costs as low as possible. 

I understand the responsibilities and associated challenges, as I’ve spent a lot of time throughout my career working to solve these problems. It’s difficult, especially with all the other pharmacy operations obligations these professionals face. 

The mundane task of manually auditing for suspicious behavior  

When drug diversion investigators receive notification of suspicious medication handling practice from various hospital sources, they are obligated, per the institution’s policy, to quickly audit the surrounding three months of controlled substance activity to identify missing medication or abnormal dispensing behaviors. The investigation usually begins by exporting an activity report from the automated dispensing cabinet (ADC) data management system. Then Excel macros are run to format the activity export (hopefully you have macros), and a second monitor is necessary to show the organization’s electronic health records (EHR) at the same time as the audit report. By looking at these two systems together the investigator may be able to identify a noticeable trend, this user is holding onto IV opioids far too long after they are dispensed. All dispenses are accounted for so far, but the question still lingers: “Are they tampering with the medication or substituting medication and falsifying medical records?” 

There must be a better way 

Upon completion of the audit, which has taken several days to complete, it seems that all the dispenses are accounted for, but there are clearly practice patterns that aren't similar to other clinicians caring for the patients on which the audit was performed. 

A few hours or days later, after the findings have been reported, the suspected diverter’s nursing manager states they don’t suspect diversion and subsequently educated the clinician on proper handling practices and completed a corrective action plan. This manager has also requested that the drug diversion investigator keep a watchful eye on this user’s behavior over the coming days, weeks, and months. This means that the investigator will need to set calendar reminders to manually audit their behavior again in seven days and then once again at 30 days to verify the correction action plan is being followed.

This is for just one user suspected of diversion. How can investigators effectively manage this process for hundreds or even thousands of other users across their system? 

This is the reality for most healthcare facilities, large and small, across the country. Pharmacists, pharmacy technicians, nurses, compliance specialists, and other healthcare professionals perform these manual audits every day, ranging from expansive integrated delivery networks (IDN), large academic medical centers, to smaller rural hospitals.

Improve efficiency and investigation workflows

One of the questions drug diversion investigators constantly ask themselves is, “How can I make this process more efficient while increasing accuracy of detecting diversion and maintaining regulations?”     

Organizations could always hire more people, but that conflicts with the executives’ priority to keep costs low, especially during a global pandemic, and it doesn’t efficiently scale. Another option would be to perform a few analytic assessments on ADC volumes to identify high volume outliers. However, this is cumbersome considering the auditing process and there’s potential to miss critical diversion activity. This also conflicts with the manager’s need to maximize ROI on the valuable drug diversion investigator because this method traditionally experiences relatively high rates of false positives.  

Lastly, organizations can leverage artificial intelligence (AI) to monitor every transaction from various data systems such as ADCs, EHRs, HR, and time and attendance systems to build a comprehensive view of every user’s activity. This technology, which is behind the Protenus Healthcare Compliance Analytics platform, uses clinical context to understand what normal user behavior looks like and raises an alert when it deviates from the norm, like when medications are missing in the medication use cycle, abnormal medication handling practices, or when there are policy violations from dispense to disposition. 

The same analytics also proactively identifies users with behaviors most at risk for diversion, proactively surfacing the most suspicious users to the investigation team with a holistic view of their activity detailing why the alert has been sent. This technology also streamlines investigation workflows by providing on-the-spot education of organizational policy, and the AI learns from the investigation workflows, improving diversion incident identification. This technology also has the ability to provide insight into EVERY unaccounted medication, as well as other policy violations that have occurred with valuable insight that is necessary to bring incidents to resolution.  

Download our business use case to learn more about how Protenus Drug Diversion Surveillance can leverage AI to detect and prevent drug diversion within your organization.