Hospitals Can Leverage Data From Second-Generation Technology for More Effective Drug Diversion Prevention

A cornerstone of health care is ensuring that the proper medications reach the patients who need them. Medication diversion, which occurs when drugs are misused, shared, or sold by employees during the supply chain process—poses a significant threat to the integrity of patient care. Medication diversion often prevents patients from receiving necessary treatments, leading to inadequate pain relief, prolonged suffering, and risk of exposure to infectious diseases from contaminated needles. Further, diversion poses a threat not only to patients but also to employees and the health care organization. If regulators detect diversion, it can severely damage a hospital's reputation and financial stability, with fines exceeding $500,000 per incident.

As pharmacy professionals, we understand the complexity of this issue. All hospital staff can be implicated drug diversion; however, employees with the most access pose the most risk. The handling and administration of medications involves numerous individuals and touchpoints, making it easier to divert small, concealable items. Hospital staffing shortages and limited resources further exacerbate the problem, with insufficient personnel and resources leading to delays in processing and managing diversion cases, resulting in inadequate monitoring, increased opportunities for diversion, and a more significant burden on existing staff, ultimately compromising the effectiveness and integrity of diversion programs.

Hospitals should leverage data from second-generation diversion monitoring technologies that utilize machine learning and analytics to combat these issues. Technologies and data analytics enhance diversion prevention programs' visibility, efficiency, and oversight. Coupled with guidelines from the American Society of Health-System Pharmacists (ASHP), this approach can significantly improve the effectiveness of drug diversion monitoring and prevention efforts.

Technologies and data analytics enhance diversion prevention programs' visibility, efficiency, and oversight. Image Credit: © Oulaphone - stock.adobe.com

Challenges in Detecting Diversion

Increased awareness of the opioid crisis and regulatory pressures have spurred many pharmacies, hospitals, and health systems to invest heavily, both in resources and finances, to implement measures to detect diversion, with modest results to date. Since medication handling is a people-intensive business, traditional technology solutions cannot ensure drug traceability and accountability throughout the hospital supply chain due to siloed data between several different source systems.

Until recently, diversion detection has primarily relied on manual and labor-intensive methods. These efforts often include inventory checks, paper logs, routine and random audits, incident reporting, and education efforts. Designated teams of employees meet frequently and pore over paper records or spreadsheets displaying rows of medications, quantities, departments, employees, patients, prescriptions, and diagnoses. These manual methods are time-consuming, prone to human error and bias, and are less effective at detecting subtle patterns of diversion, ultimately creating inefficiencies and limitations in oversight and results.

As diversion monitoring solutions advanced, hospital systems intermittently have run analytics on data to reveal deviations (though imprecise) from average volume or activity levels. While this was a positive step forward in diversion monitoring, challenges were still present in how and when data would become available. Further, standard deviation measures overall variability and not necessarily the presence of outliers or anomalies indicative of diversion.

Diversion involves complex and multifaceted behaviors that are subtle and irregular in pattern that standard deviation cannot always highlight effectively. Additionally, reports showing anomalies can take weeks or months to detect and investigate, leaning heavily into a narrow scope of practice for health care professionals. Although egregious diversion activities were sometimes detected, most needles remained buried in haystacks. Such was a common shortcoming for those reliant on first-generation technologies.

Benchmarking With Second-Generation Technologies

Artificial intelligence (AI) is revolutionizing diversion monitoring by enhancing detection capabilities, improving accuracy, and enabling proactive management through sophisticated data analysis and pattern recognition. Second-generation technologies, powered by AI and machine learning, offer a more efficient and accurate approach to automatically detect anomalies in drug handling to reduce risk and increase compliance. These technologies enable hospitals to automatically detect anomalies in dispensing patterns, usage volumes, waste volumes, action times, employee movement around the facility, changing work schedules, and more.

In the most advanced applications, near real-time analysis drives closed-loop reconciliation on a per-dose basis, offering enhanced visibility and adherence to regulatory and compliance standards. Data follows each medication from the wholesaler order to the narcotics vault, the automatic dispensing cabinet (ADC), and the electronic medical record (EMR), yielding insights into the hospital system, facility, and employees down to the unit level for controlled substance accountability.

By utilizing these advanced tools, hospitals can continuously improve their diversion programs by benchmarking them against industry standards and best practices. Benchmarking involves comparing current performance metrics with those of similar institutions to identify areas for improvement and ensure the effectiveness of diversion prevention programs. Using second-generation technologies allows hospitals to make these comparisons within the technology itself, enabling convenient and real-time benchmarking.

Focused Investigative Efforts

Utilizing AI-driven analytics for real-time monitoring enhances hospital investigative efforts through several advanced features. On-the-fly analysis can drive closed-loop reconciliation per dose, enabling immediate identification and investigation of suspicious activities. Implementing an AI-powered risk score for employees based on their actions and behaviors helps prioritize investigations. The technology uses numerical scores and stoplight icons (eg, red, yellow, green), intelligently prioritizing the review list. Additionally, a high-risk score doesn't necessarily confirm an employee is diverting; instead, it signals an anomalous behavior patterns shown in practice to indicate diversion activity. Whether an employee exhibits poor practice, a policy violation, or a risk of diversion, reviews are targeted and standardized, driving efficiency and impact.

Leveraging advanced data analytics to analyze medication usage patterns and trends by generating detailed reports and visualizations helps hospitals identify areas requiring closer investigation, focusing on the most critical concerns. Additionally, providing unit dose level visibility assists teams in easily maintaining thorough documentation of the medication handling process. This not only aids in investigations but also ensures compliance with regulatory standards and simplifies audits and reviews.

By pinpointing specific areas of concern, hospitals and health systems can allocate their investigative resources more effectively. Instead of broad, unfocused investigations, staff can concentrate on high-risk areas identified through data analytics. These systems support quick, collaborative investigations by compiling a complete evidence package in a central location for teams to work with. Multidisciplinary drug diversion response teams can decrease the time to confirm diversion as well, while more effectively working together and fostering a mindset of continuous program improvement.

Seamless flow of information from various systems, such as EMRs and ADCs, improves the overall investigative process by ensuring the integrity and security of controlled substance accountability, ultimately contributing to improved patient safety. Overall, second-generation technologies support hospitals in focusing their investigative efforts by providing automated, AI, and data-driven insights that streamline the identification and resolution of medication and practice-related issues.

Utilizing ASHP Guidelines

In addition to leveraging second-generation technologies, hospitals should adhere to ASHP's guidelines for drug diversion monitoring.¹ These guidelines provide a comprehensive framework for establishing effective diversion prevention programs, including:

• Implementing robust policies and procedures for medication management,
• Conducting regular audits and assessments to identify vulnerabilities,
• Providing ongoing education and training for staff on diversion prevention,
• And establishing a multidisciplinary diversion response team to oversee investigations and response efforts.

Data Insights

My company has worked to refine the approach of compiling anonymized data from hospitals to identify diversion trends. In the first quarter of 2024, data from 865 sites revealed anomalies in 7% of 24 million medication transactions tracked.² These variances are discrepancies between the order of a medication and its dispensing, administration, waste, and return. Variances typically highlight either a gap in documentation, a potential instance of diversion, or an opportunity to reimagine workflows for improving patient safety and documentation practices, prevent additional waste, and avoid regulatory fines. The top 5 drugs with variances were opioids and benzodiazepines, such as fentanyl, midazolam, hydromorphone (Exalgo; Mallinckrodt Inc), morphine, and lorazepam.

The diversion monitoring technology allowed facilities to confirm diversion in 275 cases in quarter one of 2024. The detailed scoring and supporting data virtually eliminated "false positives." Additionally, at the beginning of 2022, investigations took an average of 84 days to close. They are now being resolved 46 days sooner, averaging just 38 days to confirm diversion. This is a testament to the work of diversion teams within health care organizations, and the impact enhancements in technologies have had on diversion monitoring.

Hospitals possess a wealth of data crucial for effective drug diversion detection. By leveraging second-generation technologies and adhering to ASHP guidelines, they can enhance detection accuracy, improve investigative efforts, benchmark their programs, and advance overall outcomes. Advanced tools and data analytics provide the opportunity to transform these insights into actionable strategies, ensuring patient safety and maintaining the integrity of health care systems. It's imperative for health care organizations to fully harness the power of these technologies, taking a proactive stance in the fight against drug diversion and safeguarding the trust placed in them by their communities.

REFERENCES

1. Grissinger M. Safeguards for Using and designing automated dispensing cabinets. P T. 2012;37(9):490-530.
2. Diversion Quarterly Report, Q1 2024. BLUESIGHT. 2024. Accessed August 15, 2024. https://bluesight.com/resource/diversion-quarterly-report-q1-2024/?utm_campaign=ControlCheck%20Q2%202024%20Campaign&utm_medium=email&_hsenc=p2ANqtz-_kQVGg-0IQXPcK7SG6wCoRiGxeSnXuuUbDD-d1JLqKNA-MEFc8jtfoyLKL3wU4H4nIdEJ-7tEd7EUimbdhxdCY3S2CBQ&_hsmi=308096492&utm_content=308096492&utm_source=hs_email