The weakest link in building a safer medication use model

I’ve just spent four days at the ASHP Summer Meeting in Denver, CO. The meeting offered a nice variety of topics, but seemed to focus on medication safety and informatics more this year than in the past. In fact, this is the first year that ASHP has offered a medication safety tract at one of their meetings.

I avoided the more traditional sessions on therapeutics, choosing instead to focus on the informatics and medication safety sessions. Based on the information presented it was obvious to me that these two disciplines are intimately linked. After all, the idea behind much of the technology we use in healthcare today is to improve patient safety.

In 1999, the Institute of Medicine (IOM) published the now infamous To Err Is Human: Building a Safer Health System. The information presented in that report sent shockwaves through the healthcare industry. More than a decade later we haven’t seen much improvement in the number of mistakes made in hospitals, but To Err Is Human effectively changed the foundation of healthcare forever. While healthcare remains squarely focused on caring for patients, the approach to how we provide that care has changed dramatically. The interest on patient safety has generated an immense body of literature aimed at using automation and technology to improve patient care.

Before diving too deep, it’s important to understand where the errors within the healthcare system occur. Leape’s landmark paper in 1995(1) analyzing ADEs in hospitalized patients found that adverse events occurred as follows: ordering 38%, transcription 12%, dispensing 11%, and administration 39%. Bates found similar results in a study also published in 1995 in the same issue of JAMA (2). Bates found that of ADEs that were considered preventable, 49% occurred during the ordering stage, 11% occurred during the transcription stage, 14% occurred during the dispensing stage and 26% occurred during the administration stage.

Since the publications by Leape and Bates much work has gone into making the medication use process safer. At the forefront of this work has been an advance in automation and technology. Among those technologies being explored include: 1) computerized provider order entry (CPOE) for ordering; 2) pharmacy information systems and clinical decision support for transcription; 3) automated carousels, barcoding and automated dispensing cabinets for dispensing; and 4) barcode medication administration (BCMA) and smart pumps for administration. This isn’t an all-inclusive list, but rather an example to demonstrate the extent to which healthcare has gone to improve patient safety through the use of automation and technology.

With that said, I find it interesting that one of the most error prone steps in the medication distribution phase is often overlooked. I’m speaking specifically about the process of returning/restocking medications in the pharmacy. I have observed the process many times and outside the use of robotics, the system is completely manual, open to selection bias, full of interruption and fraught with error.

Example return/restocking process:

  1. A series of medication are returned to the pharmacy.
  2. The medications are placed in a return bin regardless of medication type, dosage form, storage requirements, etc.
  3. Tablets in the return bin are sorted for restocking.
  4. Someone, most likely a pharmacy technician takes the sorted medications and places them back into pharmacy stock.
  5. The medications are now ready for use.

Notice that step number three above is highlighted in red. This is the step in the process that is most open to error.

Let’s just say that during the sorting process the medications are not sorted properly and a hydrALAZINE tablet finds its way into a hydrOXYzine bin. I’ve seen this happen many, many times. The packaging and names are similar so the single hydrALAZINE tablet goes undetected in the wrong bin. So the next time hydrOXYzine is needed in bulk, i.e. for an ADC replenishment, the hydrALAZINE ends up in the pile of hydrOXYzine tablets. Since the tablets are loose, the pharmacist checking the bag full of hydrOXYzine fails to see the single hydrALAZINE tablet.

The hydrALAZINE is mistakenly sent to an ADC cabinet along with the hydrOXYzine where a nurse pulls the hydrALAZINE from the ADC thinking it is hydrOXYzine. Sometimes the nurse fails to recognize the error and the hydrALAZINE is administered to the patient in place of hydrOXYzine.

Hopefully the facility utilizes BCMA and the error is avoided. However, only about 35% of hospitals in the country were using BCMA as of 2010(3). However, if the facility is not utilizing technology like BCMA, the incorrect medication is administered to the patient where it could potentially cause harm.

Although the example above involves several failures during the medication use process, it all began with a breakdown during the restocking phase. I’ve seen this exact error many times during my career, as well as many others caused by sound-alike-look-alike medications.

It’s clear to me that the return/restocking phase of the medication distribution process is the weakest link, and is rarely acknowledged when thoughts of improving the process come to mind. So what’s the answer? Does the process need to be automated or is a better manual process the answer? I don’t know what the solution is, but I think it’s time we gave it some thought.

References

  1. Leape L.L., D.W. Bates, D.J. Cullen, J.W. Cooper, H.J. Demonaco and T. Gallivan et al. 1995. “Systems Analysis of Adverse Drug Events.” ADE Prevention Study Group. JAMA 274: 35-43.
  2. Bates D.W., D.J. Cullen, N. Laird, L.A. Petersen, S.D. Small and D. Servi et al. 1995. “Incidence of Adverse Drug Events and Potential Adverse Drug Events: Implications for Prevention. ADE Prevention Study Group.”JAMA 274: 29-34.
  3. Pedersen C.A., Schneider P.J., Scheckelhoff D.J. 2011. “ASHP National Survey of Pharmacy Practice in Hospital Setting: Prescribing and Transcribing – 2010” Am J Health-Syst Pharm 68: 669-88.

2 thoughts on “The weakest link in building a safer medication use model”

  1. Well over a decade ago, at my previous place of employment (large metropolitan hospital), I installed the McKesson RobotRx. Returned medications were simply hung on the return rack, in any order, and the robot put them away flawlessly every time. If the bar-code was damaged, the item was rejected for repackaging. This took care of 90% of our oral and injectable dosage forms. Unfortunately, while technology has improved, the cost of this product is still out of reach for most hospitals (including my current, small, community hospital). In another 8 weeks, we are hoping that BCMA will help intercept the these misplaced meds, but people are still human.

  2. I agree, Bruce. Robotics is a decent method for accurately storing and retrieving oral solid dosage forms. The problem is as you describe, i.e. injectables, topicals, oral liquids, and so on. The other thing is the speed, or rather the lack there of with robotics. I think robotics is the future, but it’s still maturing in my opinion. Thanks for stopping by.

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.