A tech-enabled solution to drug diversion: What to do when a provider steals a patient’s medication
Unfortunately, too many patients go without prescribed medication because healthcare professionals – a distinct minority but no small number – “divert” the drugs for other uses, often abusing the medication themselves.
Imagine you’re injured in a car accident and you go to the hospital. While there, you receive pain medication a doctor has prescribed … unless you don’t.
Unfortunately, too many patients go without prescribed medication because healthcare professionals – a distinct minority but no small number – “divert” the drugs for other uses, often abusing the medication themselves. An estimated 10% of healthcare workers abuse drugs, according to the U.S. Substance Abuse and Mental Health Services Administration and the American Nurses Association.
Drug diversion on that scale represents a lot of pain for affected patients, who may illicitly be given other drugs to obscure the deed. The harm goes far beyond one patient, though. Any patient can find themselves at the mercy of an impaired clinician. There’s also the risk to third-party recipients of stolen drugs, some of which can be fatal.
Providers’ diversion of prescribed medication can cost hospital systems millions in fines for flawed drug management processes. And there’s the reputational damage if the public discovers doctors or nurses are stealing patients’ medications.
At risk are hospitals, healthcare systems, pharmacies, clinics, nursing homes and emergency centers. Each one has multiple medication-lifecycle processes that are vulnerable to diversion, including wholesaler procurement and delivery, central pharmacy inventory management, pharmacy distribution, dispensing to doctors and nurses, and, finally, administration to the patient.
The problem of drug diversion has been with us since Hippocrates, but it’s especially serious now. Overdose deaths in the U.S. hit an all-time high in 2020 and are still on the rise, and the Covid-19 pandemic has layered stress on an already strained healthcare workforce. Packed ICUs couldn’t be more demanding settings, making it easier than ever to divert drugs.
Healthcare administrators need support
Hospitals and healthcare systems have worked mightily to address the problem, creating busy oversight committees and laborious processes. On paper and in rudimentary computing systems, they’ve attempted to follow every prescription from the pharmacy to the patient (and — if there’s excess medicine that needs to be disposed of — back to the pharmacy).
The job is so complex and the offenders so elusive that detection is more than humans alone can handle. Combating drug diversion is a job better suited to advanced analytics and machine learning, which many healthcare systems are adopting. Well-designed systems are equipped to understand normal medication flow throughout a sprawling organization and detect anomalies while minimizing false positives.
Of course, the technology challenge is significant. Anomalies in prescription drug flow are meaningful only when contrasted against a valid picture of normal use. If a hospital has only three years of historical medication data and two of those years have been dramatically affected by a pandemic, what’s the true baseline? The answer is no one knows.
Constantly changing pharmacy information systems, automated dispensing systems and electronic health record systems add to the difficulty of synthesizing useful data. So do the different labels that different technology vendors employ for different data, such as branded drugs and their generic equivalents.
Another challenge is that machine learning systems need regular human attention to do their work: Someone needs to tell the system whether its latest alert is accurate – that the identified employee actually is diverting prescription drugs – or a false positive. That’s one way machine learning gets smarter; neglect this step, and you’re wasting your time and money.
Finally, there’s the challenge of keeping data current. Few drug diversion detection systems bring real-world, as-it-happens data streams on drug flow back into the system in near-real-time. But that’s what’s needed to refine the baseline.
Done right, drug diversion prevention powered by advanced computing can save lives, improve patient care, reduce healthcare costs and eliminate wasted effort.
It’s important work, and it’s satisfying. The payoff for me is not in shaming individuals with substance abuse disorders. It’s in protecting patients, healthcare workers, hospital systems and public health – and making life easier for everyone involved.