CDC information shows the increase in the frequency of disease outbreaks stemming from drug diversion.
This article was originally published June 10, 2020 on PSQH by John Palmer
Editor’s note: The following Q&A resulted from a conversation PSQH had with Tom Knight, CEO of Georgia-based Invistics, a provider of cloud-based software solutions that improve healthcare inventory visibility and analytics across complex healthcare systems and global supply chains.
Invistics commissioned a study by Porter Research revealing that while nearly all healthcare professionals agree that drug diversion is occurring across the country, two out of five hospitals still do not have effective drug diversion prevention programs in place.
PSQH: In your opinion, why do so few hospitals have drug diversion programs? As it is such a large problem, one would think it would be more of a priority.
Tom Knight: While many hospitals have drug diversion programs, sadly, many hospitals do not have formal programs. In our experience, there are several common reasons.
One common reason is that many healthcare leaders believe that informal drug diversion programs will be enough, or that drug diversion is not an issue at their facility. Unfortunately, drug diversion is an ongoing issue across the country, causing patient safety problems all too often.
CDC information shows the increase in the frequency of disease outbreaks stemming from drug diversion:
- In the 14-year span between 1985 and 1999, 48 outbreaks from drug diversion were documented.
- Between the 14-year span of 2004 and 2018, though, 148 outbreaks were reported.
- There are other examples, too. In 2017, reports surfaced that incidents of missing or stolen prescription medications at Veterans Administration hospitals increased more than 800% in seven years, from 272 in 2009 to 2,457 in 2016.
A second reason is that healthcare organizations sometimes cite the lack of adequate financial resources—but the real issue may be that they may think other spending should take priority. This is understandable, as hospitals have had to invest millions in infrastructure, electronic health records (EHR), and security to bring themselves into compliance with regulations. Many are afraid of spending too much on solutions for drug diversion unless they’ve been burned or seen a competitor impacted by the problem.
The Porter Research survey commissioned by Invistics found that nine out of 10 surveyed believe their facility’s drug diversion program is the same or even better than other organizations, and two out of three are confident or very confident that their drug diversion program successfully identifies employees who divert drugs. But there is definitely a disconnect, because 70% of participants said they believe most diversion incidents in the U.S. go undetected. It’s important that we continue to increase education to help mitigate these numbers.
PSQH: What advice do you have for hospital officials who would like to start a program, but don’t know where to start?
Knight: I would first advise hospital officials to review best practices for drug diversion programs from organizations who have studied this, like the American Society of Health-System Pharmacists (ASHP) or Council of State and Territorial Epidemiologists (CSTE), or our nonprofit public-private partnership, HealthcareDiversion.org. There are some great materials that can be quickly adopted, such as forming a Drug Diversion Oversight Council or training rapid response teams to quickly investigate diversion when it is suspected.
If needed, hospital officials can also consult with experts or other organizations that have already started drug diversion programs and ask specific questions such as the following:
- How did they launch a program?
- What kind of technology solutions did they use?
- Are there other areas that could bear improvement?
- How did they measure the success of their actions, such as through an increase in the number of reported incidents?
Consider how their best practices can be applied to your own healthcare setting. Through research and planning, it will be easier to see where the gaps exist. Some organizations have had success from putting together an opioid task force with leaders from every department to develop solutions to address drug diversion.
It’s worth mentioning that one way to address drug diversion is to ramp up in-house training. National nursing boards, pharmaceutical associations, and other groups frequently offer free or low-cost web-based programs to help employees understand the scope of diversion, recognize the signs of diversion, and learn what to do when they suspect a colleague is diverting medications. An organization can make this training mandatory for new staff or even host a “lunch and learn” event where staff can watch and learn from these programs.
PSQH: What are some of the most important parts of a successful drug diversion prevention program?
Knight: The foundation of a successful drug diversion prevention program is awareness. A healthcare organization must first recognize drug diversion is an issue and take proactive steps to address it through employee training programs.
For example, go beyond handing a new employee a training manual that includes a section on drug diversion. Take the extra steps to look into a potential employee’s work history (looking closely at job changes or geographical moves), and make sure that individual participates in mandatory drug diversion training before they come into contact with patients. The most successful organizations are committed to preventing drug diversion at every level, from the job interview to the patient’s bedside to the pharmacy.
Resources are also needed. We’ve seen that the equivalent of at least one full-time employee assigned to drug diversion detection can help. There are also impressive advancements in machine-learning technologies, which can help to isolate patterns associated with drug diversion and reduce false positives.
PSQH: Are there parts of the country where the problem is more prevalent? Is the lack of programs due to finances, lack of interest, or something else?
Knight: At this time, we don’t have enough data to speak to geographical hotspots. But we can say that drug diversion is happening in every healthcare setting, from hospice and hospitals to nursing homes and community health centers. It’s not just limited to rural pockets of America or major urban centers. As noted earlier, the lack of formal drug diversion programs could mean that hospital leaders are prioritizing other initiatives or don’t believe diversion will impact their facility.
The lack of drug diversion programs also stems from the fact that drug diversion can be difficult to detect. If you can’t see the problem right in front of you, that makes it harder to rationalize allocating resources to addressing it: 70% of healthcare workers who responded to the 2019 Porter Research survey on drug diversion said they strongly agree that “most” drug diversion goes undetected, a number that has increased since 2017.
PSQH: Your survey shows machine learning is a good way to uncover drug diversion. How does this work, and is it effective?
Knight: Machine-learning technologies are trained to recognize known patterns of behavior consistent with diversion. We have a large data set containing many cases of diversion from hospitals across the country and have used that data set to teach the machine-learning algorithms to detect those same patterns.
These algorithms look at data from multiple computer systems—such as EHRs, employee timecards, and medication dispensing cabinets—to find patterns. For example, when a nurse who is diverting medications at a patient’s bedside enters falsified data into the EHR, the machine-learning algorithm is able to detect patterns in that data that provide clues to detect the diversion. Some of those patterns are fairly easy to detect, like a nurse who accidentally charts the medication administration incorrectly in the EHR, while other patterns are more sophisticated, like a nurse who enters accurate but falsified records for medication administration, pain scores, or other clinical data.
The great thing about the technology is that it grows smarter as our training data set grows. The more health systems contribute data, the larger our data set grows and the more accurate the machine-learning algorithms become. We are seeing two exciting developments with this machine-learning approach.
First, the software is growing more accurate because it learns from its mistakes. For example, if the software flagged diversion incorrectly, and the hospital concludes diversion did not occur, then the “false positive” is logged in the training set to avoid making this mistake in the future.
Second, whenever some new diversion incident occurs, and this new pattern is added to the data set, then the software will be able to detect that pattern elsewhere, avoiding a “false negative.” This is particularly helpful because if one hospital finds a new pattern of diversion, that new pattern will then be recognized if it happens at any hospital in the country using the technology. In short, every time a hospital in the network confirms an incident of drug diversion happened, all hospitals become more effective at detecting diversion.
PSQH: About 70% of your respondents say most drug diversion goes undetected. Is this to say that the problem is much bigger than we know even now? What needs to be done to fix this?
Knight: Based on the survey findings, it is clear that more diversion is happening than what is being reported. In 2018 alone, the Department of Justice made only 50 enforcements and about 280 administrative actions related to drug diversion, and this is clearly just a small fraction of the diversion that is happening.
Moreover, the problem could be worsening. For example, the surge of patients with COVID-19 into hospitals could lead to clinicians missing some of the signs of diversion or clinical staff having an easier time circumventing detection.
While no single approach to fixing the problem is guaranteed, improving training, hiring one or more dedicated drug diversion professionals, and investing in machine-learning technology are three big ways we can reduce diversion in healthcare settings.
We also need to engage in broader conversations throughout the healthcare system. The current healthcare culture is often one where workers are sometimes reluctant to speak up due to the stigma attached to drug diversion. We need to change the culture so workers know that speaking up helps everyone. It helps protect our patients and it helps protect our coworkers. There’s too much at stake to ignore the problem. Drug diversion has clinical, legal, and financial consequences.
John Palmer is a freelance writer who has covered healthcare safety for numerous publications. Palmer can be reached at email@example.com.
Patient Safety & Quality Healthcare’s mission is to provide news, science, research, and a forum for opinion for clinicians, healthcare professionals, and everyone interested in improving quality in healthcare. Learn more.