The Children’s Hospital of Orange County (CHOC) includes a state-of-the-art 232-bed pediatric hospital in Orange, California, a pediatric hospital within a hospital in Mission Viejo, and five community clinics. CHOC’s centers of excellence include the Cancer, Heart, Neuroscience and Orthopedic Institutes. Together these institutes provide specialized care in pediatric cardiology, cardiothoracic surgery, oncology, neurology, neurosurgery and orthopaedics.

Intensive care beds comprise approximately 40% of our total beds. Intensive care is provided in the pediatric intensive care unit, cardiac intensive care unit, neonatal intensive care unit, and oncology intensive care unit, which is primarily devoted to bone marrow transplants. CHOC is one of only eight children’s hospitals in the United States to be named to the Leapfrog “Top Hospitals 2007” list.

Automating IV Medication Preparation
While the risk posed by preparing intravenous (IV) medications is higher than oral medications in any patient population, the IV medication preparation process is especially risky in the pediatric population. Dilutions of commercially available products are often required to achieve the concentrations appropriate for pediatric patients. In addition, medication concentrations have to accommodate a wide range of patients, from a 0.5 kilogram neonate to an adolescent patient who may weigh over 100 kilograms. Thus, multiple concentrations of an agent are required to accommodate the needs of all our patients. Because dosing is weight-based, the majority of our doses are patient-specific, making the process even more error-prone. These multiple variables result in significant opportunities for error at every step in the process of preparing proper IV medications for pediatric patients.

Choosing a System
Improving the accuracy and safety of our IV medication administration is paramount within our organization, so implementing a robotic IV preparation device was an obvious choice. While this was a pharmacy driven project, one of the keys to our success was involving the hospital’s executive team early on. In December of 2006, the vice president of patient care services and chief nursing officer, together with the vice president and chief information officer, accompanied pharmacy to the ASHP Midyear Clinical Meeting in Anaheim to observe the robotic demonstrations. This began the quest to justify and acquire an IV robot in the pharmacy.

As part of our due diligence, we reviewed the functionality of three IV robotic systems available on the market, and had each manufacturer present their product specifications to the pharmacy team. Our goal was to implement a system that would compound sterile preparations accurately and safely in a USP <797>-compliant environment with as little human intervention as possible while producing 80-95% of our doses. With this goal in mind, we reviewed each system’s capabilities. Key decision criteria included sufficient flexibility to accommodate our varied dosage needs, precision/accuracy, and safety features of the robotic system.

ROI Analysis for Budget Approval
While safety issues drove our decision to implement a robot, the financial benefits of the system expedited the capital budget approval process. We created a return on investment (ROI) analysis based on the system’s ability to reduce technician manpower and decrease wastage through just-in-time production, while accommodating all of our needs.

We collected data on our IV medication wastage under the system we were using at the time, and it revealed that 15% to 30% of the daily-prepared doses were discarded due to change or discontinuation of orders. The ROI assumed a conservative 4% savings in pharmaceutical costs from the reduction in wastage. It was also estimated that the number of technicians needed in the IV room would decrease by half. By decreasing the number of technicians required to prepare IV medications and decreasing our wastage through just-in-time production, we determined that the robot would deliver a positive return on investment within two and a half years.

Hospital-wide, we were all im­pressed by the sophisticated technology of the robot. Once we were confident that the design elements would meet pharmacy needs, and the finance department was confident in the long-term viability of the manufacturer, we chose a five-year, lease-to-buy option. Additional costs associated with the robotic system include the cost of the maintenance contract, a slight increase in the cost of syringe labels and the cost for the cleaning supplies required to maintain the machine. There was also an additional construction cost to ensure adequate mechanical requirements to manage the airflow of the robotic unit.

Preparation for Implementation                                                                                        Our timing in this process proved quite fortuitous, as when we began investigating the possibility of implementing a robot, we were also in the process of designing a new IV room. Thus, we designed the room to accommodate the robot and also designed the chemotherapy preparation room with sufficient space to accommodate a second robot for compounding chemotherapy, which we hope to acquire when budget becomes available. As part of this design process we had to accommodate additional mechanical requirements, such as adequate airflow and new fan coils to ensure efficient heat dissipation.

As the first site ever to implement this robotic system, the pharmacy team and the manufacturer’s engineers spent an entire week in seclusion to perform a failure modes and effect analysis (FMEA). The team began with outlining the new workflow surrounding the robotic system. Risk points were then identified throughout the process and action plans created to offset the major risks identified. At the end of the exercise, it was clear to all that the highest risk points occurred during initial data entry into the system software. For example, if the wrong drug is loaded into the robot, the safeguards in the robotic software will identify the error and place the incorrect drug on the reject rack. On the other hand, the robot will not be able to identify any incorrect data that is entered into the software. Therefore safeguards and triple-checks were introduced in the data gathering and entry process.

Training for our designated super-users included a two-day classroom-training course followed by a half day of hands-on training. The training for certified users was less intensive, with four hours of classroom-training and four hours of hands-on training. Currently, only a few of our staff members are certified to operate the robot.

Building the Drug Database
The interface between the robot and the pharmacy information system is relatively simple: utilizing print stream information from the pharmacy system, the robot receives the information normally sent to the printer to create the IV labels.

While the interface is simple, adding each drug to the system is a time- and labor-intensive process. The robot needs to be “trained” on each drug with precise information including the correct vial size, reconstituted concentration, final concentration, specific gravity, appropriate shaking speed during reconstitution, appropriate syringe, etc. Once every data point is determined and entered into the system, the data points are triple checked before
final testing is performed to ensure accuracy prior to the preparation of patient doses.

Implementation
The finalized contract was signed in February 2008, the IV room completed in July, and we took delivery of the robot in early September. After extensive testing, the first patient-specific dose was prepared by the robot on December 2nd, 2008.

Currently, we are still in the very early stages of implementation. As the first site to implement this system, we decided to progress slowly and cautiously. We began with the simplest process, drawing up patient specific doses from a manufacturer-prepared vial, and have since moved on to using the robot to prepare reconstituted medications. The next step will involve implementing medications that require reconstitution and/or further dilutions. To date, we have implemented seven drugs in the system en route to our initial goal of having the robot prepare 80% of all IV medications by the end of June. At this point, we have not altered our staffing in the IV room. However, we anticipate the reduction of 24 hours of technician time per day at the end of the implementation period, operating with one technician instead of two, per shift in the IV room. We also continue to track wastage data and compare it with the pre-robot data to validate our projected ROI.

Conclusion
Robotic preparation of IV medications will improve medication safety by decreasing compounding errors and increasing dose accuracy. It will improve efficiency within the pharmacy and can decrease repetitive injury risk, such as carpal tunnel syndrome, among pharmacy personnel. A significant amount of time should be dedicated upfront to creating the drug database and adequately training the staff in order to realize the significant benefits from automating IV drug preparation.

 
Rita K. Jew, PharmD, FASHP has served as the executive director of pharmacy services at Children’s Hospital of Orange County for the past three years. Prior to her current position, she spent 14 years as the clinical manager and clinical specialist in neonatology at The Children’s Hospital of Philadelphia. Dr. Jew received her Doctor of Pharmacy degree from the University of California at San Francisco and completed an ASHP-accredited residency in clinical pharmacy at Thomas Jefferson University Hospital. Dr. Jew received the Jonathan Roberts award from DVSHP in 2004 and the Fellow of ASHP in 2007.