The ramifications of errors made during IV medication preparation can be significant for any patient population, but when treating pediatric patients, the risks posed by errors are even more severe. Given the weight-based dosing and complex dilutions necessary to accommodate patients potentially ranging from a half-kilogram neonate to a 100-kilogram adolescent, producing accurate pharmacy-prepared products is vital for proper patient care in a pediatric hospital.
At the main campus of the Children’s Hospital of Orange County (CHOC Children’s), a 279-bed pediatric hospital located in the City of Orange, California, medication preparation processes are standardized and technology leveraged as much as possible to lower medication error risks. Although most medications are distributed from the central pharmacy, the satellite pharmacies in the OR, ED, and ICU, all perform some IV compounding; as a pediatric hospital, over 95% of the doses dispensed are custom-made and patient-specific. Thus, no matter how structured the processes, the multiple steps and manual preparation tasks involved in pediatric medication compounding invariably increase the chance of errors. Therefore, adopting robotic technology to mitigate or eliminate the number of manual steps and to automate the compounded sterile preparation (CSP) production process has been integral to our operations. CHOC Children’s has benefitted from robotic compounding technology for almost five years now, but the journey has not been without its share of challenges. However, through our experience, we expect to continue reaping patient safety benefits, as well as reducing waste and optimizing the financial impact through robotic compounding.
Focus on Addressing Needs
When we first considered the acquisition of an IV compounding robot in 2007, we reviewed the functionality of the three robotic systems then available and had each manufacturer present their product specifications (see Implementing a Robotic IV Medication Preparation System, in the March 2009 issue of PP&P). Key considerations included precision and accuracy, the flexibility to accommodate varied preparation needs, and the overall safety features of each robot. In order to find the best fit for our operations, we also performed a failure modes and effects analysis (FMEA) on our manual compounding process to identify potential failure points that could lead to errors. Ultimately, the goal was to implement a robotic system that would compound 80% of CSPs, safely and accurately, in a USP <797>-compliant environment with limited human intervention, and eliminate the majority of the failure points identified by the FMEA process.
The robot we chose offered all the necessary safety features we required and its fail-safe mechanisms eliminated most of the FMEA-realized risks. It also has the flexibility of performing various complex compounding processes, from reconstitution to dilution and preparation of patient-specific dosages in most syringe and bag sizes. Among the deciding factors was the robot’s ability to apply a label on the final prepared product, which was a process we wanted to automate given the opportunities for human error during manual labeling. Furthermore, we conducted an additional FMEA on our planned utilization of the robot to prepare CSPs prior to implementation. The majority of the potential error risks identified were addressed by the inherent fail-safe mechanisms that are part of the robot’s design. The remaining error risks were deemed human-based: data entry errors and contamination of the robot’s cell during the cleaning process. To address these issues, we instituted a triple-check process for data entry and a detailed robot cleaning procedure and training program.
The robot was implemented in 2008 and after a four-month testing period, we began producing patient doses. Our facility leased the very first production model of this robot, thus we faced a steep learning curve, especially during the first two years. We first trained the robot to perform straight draws of medications, followed by reconstitutions, and finally, more complicated processes such as dilutions.
Since 2008, there have been multiple software upgrades to enhance the functionality of the robot, but in recent years, these upgrades have become much less frequent as the device has matured. Compared to other large-scale automation and technology software updates, such as those for a CPOE system, the robot’s software upgrades have been relatively painless.
As always, patient safety was the primary motivation for implementing a robotic system, but forecasting an overall projected cost savings helped expedite the budget approval process. We estimated that full adoption of this technology would allow us to reduce the number of necessary IV room technicians by 1.8 FTEs. We also analyzed our past IV medication waste data and found that 15% to 20% of daily CSP doses were discarded due to order changes or discontinuations. After conducting a return on investment (ROI) analysis that factored in reduced technician FTEs and decreased wastage through just-in-time production, we assumed a positive ROI within two and a half years, which was achieved.
In order to meet the projected ROI, we sought to utilize the robot for high-volume drugs, thus reducing the number of technician FTEs needed in the IV room. With the robot sharing the workload of preparing CSPs, we were able to implement just-in-time production, preparing daily doses of medications in four batches instead of two. This helped reduce wastage due to medication changes or discontinuations. After implementation of the four-batch schedule, drug wastage was consistently below 10% of doses prepared (compared to 10-15%, previously), and at times, dropped as low as 6%. This reduction not only mitigates overall pharmaceutical costs, but it also allowed us to absorb additional production volume without increasing the number of FTEs. At the time we implemented the robot, our facility was undergoing rapid growth; however, we did not have to increase the number of technician FTEs working in the IV cleanroom for two years due in large part to the benefits of robotic preparation and the ability to move to just-in-time production.
Continual Device Management
National drug shortages pose a huge challenge to the proper maintenance of hospital technology databases, and their impact extends directly to IV robot automation. Whenever an alternate supplier is brought in due to shortages, the compounding robot must be reprogrammed with new data, and the drug on shortage must be deactivated. In addition, a series of test runs are necessary to confirm that the robot will function properly with the newly programmed data. These scenarios can prove to be time consuming and on occasion, we have to limit the use of the robot in compounding the drug on shortage until thorough programming of the new drug is completed.
Another operational challenge for the IV compounding robot is the need to continually monitor even slight variations in medication container sizes and shapes as received from the manufacturers. While drug manufacturers are required to adhere to strict FDA guidelines when producing a drug to ensure its contents fall within strict limits, there are no requirements for consistency in the production of the container that holds the medication. However, the precision of the robot is such that it has a very low tolerance to any variations in vial height or width, or to the thickness of a vial stopper. Thus, if a drug vial varies in height from another, even for vials of the same drug from the same manufacturer, this could require adjustments to the operational software to ensure the robot can manipulate the vial properly.
Continuous Process Improvement
Today, at CHOC Children’s, the robot prepares dilution bags, patient-specific batches in syringes, and non-patient specific batches. On average, it produces 180 to 250 patient-specific doses and 20 to 30 dilution bags daily. Although we have backed off from the initial goal of utilizing the robot to compound 80% of our CSPs since the implementation of the IV robot, we do continue to prepare high volume medications using the robot, the timing of which depends on the complexity of the order.
In 2012, IV workflow management software was adopted to compliment the robot. The goal for this application was to narrow the safety gap between automated robotic IV preparation and manual production by technicians. This software leverages bar code technology to ensure the correct drug and diluent are being used for preparation. With this software in place, the current initiative is to maximize the utilization of the IV robot to prepare dilution bags needed for manual preparation. The rationale for this strategy is that the preparation of dilution bags is a complex, multi-step process that involves numerous calculations—a process ideal for robotic automation. In addition, since a single dilution bag is often used to draw up multiple patient-specific doses, if that bag is made incorrectly, it will adversely impact a large number of patients. Therefore, it makes sense to rely heavily on the precision and accuracy of the IV robot to perform the high-risk, complex process of preparing dilution bags. The IV workflow management software and its bar code technology can then be utilized to ensure that the correct dilution bag is used to prepare patient-specific doses.
Automating Chemotherapy Preparation
Given the financial impact of acquiring a fully automated, robotic IV preparation device, our initial efforts focused on utilizing the machine to produce products that generated a strong return on investment (primarily through waste reduction), and chemotherapy compounds were not ideal products in this model. Due to its high-risk nature, chemotherapy preparation is tightly controlled, and doses are only prepared after gaining confirmation that the patient’s health status will permit the chemotherapy regimen. Thus, the pharmacy rarely wastes chemotherapy agents. However, given our extensive experience in utilizing the robot for IV preparations, it is now time to realize the safety benefits afforded by automated production of chemotherapy for both the employees preparing the doses and the patients receiving the doses. As such, we hope to acquire a second robot dedicated to chemotherapy compounding to remain in-line with our overall safety goals.
Having acquired the original IV robot in 2008, the end of the five-year lease is approaching. CHOC Children’s originally leased the device, as opposed to purchasing it, because it was such a new technology that we did not know what to expect. Now that we are familiar with it and it has become ingrained in our workflow, we are discussing our options with our lessor for retaining the device. Thus far, the robot has lived up to expectations and we will continue to utilize this technology as it matures to the full potential of its abilities.
Rita K. Jew, PharmD, FASHP, was executive director of pharmacy and clinical nutrition services at Children’s Hospital of Orange County. Since September 2013, she has assumed the position of director of pharmacy at University of California San Francisco Medical Center, Mission Bay campus, where they have implemented three of the same IV compounding robots used at CHOC Children’s in addition to other robotic technologies.
Before Acquiring a Compounding Robot
Below are some key points to keep in mind when considering the acquisition of a compounding robot: