Editor’s note: In this interview, Michael R. Cohen, RPH, MS, ScD (hon.), DPS (hon.), FASHP, founder and president of the Institute for Safe Medication Practices (ISMP), discusses the current state of medication safety. The interview was conducted by J.W. Beard, MD, MBA, medical director for medical affairs at ICU Medical.
Q: In the June 2019 ISMP newsletter, there was a list of notable medication errors reported in 2019 that continue to occur despite advances in practice and technology. What are your views of medication errors and safety today compared to 10 years ago?
Cohen: I have to say we are a lot safer in hospitals than we were 10 years ago. I believe it speaks to increased efforts that providers (hospitals, practitioners) have placed on safety, plus the availability of technologies such as smart infusion pumps, bar coding, IT, and EHRs [electronic health records], and we continue to improve. We have truly made progress. I know we certainly still hear about medication errors. Most of the ones that we were getting 10–15 years ago that were fatal events don’t happen anymore; either the drug is taken off the market or other actions have been taken, such as repackaging. An example is vincristine, a drug that is used for cancer therapy in children. It is only used intravenously, but we have over 130 cases where it was accidentally mixed up with an intrathecal drug such as methotrexate and was given intraspinally, which is uniformly fatal. We made a strong recommendation to the United States Pharmacopeia (USP) and manufacturers to change the product labeling and for providers to only put it in a mini-bag, which has reduced the incidence of the route of administration error.
Another one of the successes that ISMP had was a national meeting that made a recommendation to remove concentrated potassium chloride vials from nursing units. At the time, these vials looked like saline vials and could be mixed up. A bolus of concentrated potassium chloride can lead to cardiac arrest and death. Hospitals are now committed to eliminating the medication on nursing units. Joint Commission surveyors will cite a hospital if this practice is discovered. This recommendation has also been adopted in Canada.
Q: You said that we are safer now than we have been in the past, in part due to technology. Have you visited hospitals using smart pump–EHR interoperability? What are your thoughts about this technology and its contribution to infusion safety?
Cohen: There is no question interoperability adds significant safety. At a site I recently visited, the smart pumps and EHR were fully integrated with auto-programming and auto-documentation. Bedside clinicians love it; there is no way they would go back to the way it was with freestanding smart pumps. There is no question that interoperability improves safety, improves drug library compliance, and makes infusion management easier for everyone, including the pharmacist. Years ago, we used to joke about having a system where the pump administration sets would be so long, they would allow us to have the pump in the pharmacy but still reach the patient in order to use this awareness to better manage the IV compounding workload. Now with smart pump integration, it is possible to have the awareness of ongoing infusions. As a matter of fact, in the upcoming publication of the ISMP Guidelines for Safe Implementation on the Use of Smart Pumps, we have a section focused on recommendations for smart pump integration.
Q: ISMP has taken an infusion safety leadership role in the development of these guidelines, which are in the process of being updated from those published in 2009. What has been the process for ISMP to update these guidelines, and what do you think are the most important outcomes?
Cohen: As an organization that is about preventing medication errors, we felt the time was right for an update. The best way to do it is to get people involved and learn from them and listen to their ideas, which we did at the summit in May 2018. We sought out leading clinicians and went to every major infusion pump vendor to ask for support and participation. In several cases, we have worked with frontline clinicians plus representatives from industry. We have used this approach with great success in the past. Vendor participation is important, as the companies that are responsible for those products know the best way to use them. We extended invitations to 60 individuals and met for two days. A professional moderator guided the group as we discussed every aspect of infusion safety and focused on specific topics.
Among the most important issues is safety with the delivery of secondary infusions. Clinicians may forget to raise the bag or open the roller clamp, and you can’t have that type of error. Two pumps on the market have addressed this type of error, but unfortunately there are other pump manufacturers out there who haven’t, so we think it is important to continue to raise awareness of this issue.
Obviously, integration with the EHR is critical, and we are really pushing for IV meds and fluids to be administered with smart infusion pumps. We have a whole section, too, on use of data and analytics. With data and analytics, we focused on the importance of drug library compliance. Smart pumps and interoperability add safety by delivering medications and fluids within safety limits established by clinicians to reduce errors associated with incorrect programming. We believe the goal for drug library compliance should be 95% or greater.
I think the other thing that is important is the risk associated with the inadvertent programming of low concentrations at the pump when pharmacists provide the right amount of drug in the bag. When the pump is programmed at a low concentration, it then must speed up to deliver the dose, and the patient gets this tremendous overdose of drug—sometimes an entire bag. Tim Vanderveen pointed this out to us, and, at the time, we had not had it reported and published in the newsletter. Tim worked with us to make the recommendation to include hard limits in the drug library on abnormally low concentrations, and that has saved lives. The revised smart pump guidelines will include a recommendation to only have hard concentration limits.
Q: The guidelines reference clinical practices and technology that are currently available but may be slow to reach full adoption. What steps are needed to drive the adoption of clinical practice or technology that improves safety?
Cohen: This is true and includes smart pumps and other safety technologies. For example, there are still pockets within hospitals with bar-coded medication administration (BCMA) that don’t use it. Also, in procedural areas including anesthesia, BCMA technology may not be available or used, and smart pump safety features are used less consistently. In the OR, we have had some errors, including bupivacaine given intravenously instead of an antibiotic. This is a cardiotoxic local anesthetic when given intravenously and can lead to fatal arrhythmias. We had another error reported where instead of bupivacaine being used for regional anesthesia, it was given intravenously. This all relates to not scanning the drug in certain areas of the hospital that are not using the technology. We would certainly like to see anesthesiologists and clinicians in procedural care areas pick up on electronic bar-code scanning and incorporate the recommendations in the upcoming guidance as fully as possible into their practices.
The guidelines were just closed after public comment. With all the inputs, we pull everyone together, in our ISMP conference room, to brainstorm how to best use the information that has been given to us. The revised guidelines will be released before the end of the year, and we hope that publication of this updated information will help drive adoption of clinical practice and technology to increase patient safety.