Robotic antineoplastic preparation and adjuvant medication compounding improved medication preparation accuracy, reduced staff safety events and reduced overall costs but did not reduce the incidence of serious errors, during a recent observational study by researchers from two centers in Boston.

Investigator Jeffrey M. Rothschild, MD, MPH, an associate professor of medicine at Brigham and Women’s Hospital and Harvard Medical School, said the study, the results of which were published online by the Journal of Oncology Practice on Sept. 25, was the first of its kind in the United States. The study was conducted to determine whether robotic antineoplastic and adjuvant medication compounding of bags and syringes could provide incremental safety and efficiency advantages compared with standard manual pharmacy preparation.

The antineoplastic preparation process includes several stages that are vulnerable to opportunities for potentially harmful medication errors, including using the incorrect drug, dose, concentration or storage, said Dr. Rothschild. So these drugs have to be prepared with “great care because of their toxicity and narrow therapeutic window.” Intravenous formulations of antineoplastic agents present additional safety challenges.

Table. Medication Errors, Staff Safety Events With Robotic Versus Manual Antineoplastic Preparations
Study Period Serious Medication Errors, % Staff Safety Events, %
Manual preparations (baseline period) 0.7 5.1
Robotic preparations (intervention period) 0.7 2.9
Based on Seger AC, et al. J Oncol Pract. 2012 [ePub ahead of print]. doi:10.1200/JOP.2012.000600.

Dr. Rothschild and his co-investigators found that although the incidence of medication errors remained constant (0.7%; Table) during the baseline and robotic periods, robotic compounding improved the accuracy of prepared antineoplastic and adjuvant medications. From May 2009 when the study began at Brigham and Women’s and Dana-Farber Cancer Care Center, to April 2011 when it ended, medication accuracy rates improved from a failure rate of 12.5% in the manual preparation baseline period to a rate of 0.9% in the trial period with the robot (P=0002). Using a ±5% variance, Dr. Rothschild said, that represented an accuracy improvement rate of 96%, which he called “impressive.” A secondary analysis using a ±10% variance measured eight failures, a 4.5% failure rate, in the baseline period compared with none in the intervention period.

The percentage of observed staff safety events that had the potential to cause staff harm—such as spills/leaks, closed-system transfer device failures, wrong technique and the failure to wear protective equipment—declined from 5.1% to 2.9% (P=0.007).

In addition to patient safety concerns, manual antineoplastic preparation, dispensing and waste disposal “create significant staff risks and consume significant staff time and costs,” Dr. Rothschild said. The many safety steps required include the correct matching of the order to the patient and the medication, drug transfers to both intermediate and final delivery containers and the careful disposal of waste. Additionally, the costs for the process done manually, including materials, equipment and labor, are quite high.

Commenting on the study, Luci A. Power, MS, RPh, the senior pharmacy consultant for San Francisco, Calif.-based Power Enterprises, described the findings as “very interesting, though preliminary,” adding that this study of robotic technology supporting a cancer treatment center “definitely sets the stage for further research.”

Although the percentage of spills and leaks, which accounted for 34 of the 73 staff safety events observed in the manual phase of the study, declined to 23 of the 28 events observed in the robotic phase, both Dr. Rothschild and Ms. Power noted that the decline could not be considered significant because the number of spill/leak events measured were too low. Both said more studies are required to determine the benefits of robotic preparation in this area of staff safety.

Overall Costs Decreased

During the trial, the mean costs for ancillary materials per preparation declined from $13.36 in the manual baseline period to $6.44 in the robotic intervention period. Dr. Rothschild explained that “by eliminating the pharmacy technician’s handling of open/exposed antineoplastic and adjuvant medications during robotic preparations, we were able to reduce ancillary costs associated with several components of the closed-system transfer device, particularly the vial and injection adapter components.” He further noted that “the savings for those components accounted for 60% of the overall costs of the closed-system components. When annualized for the 16,500 antineoplastic bags/syringes that were prepared for the hospital in 2009, the savings in material costs would have been $115,500.”

Although the pharmacy technician’s mean drug preparation time increased 160%, from four minutes, 12 seconds in the baseline period to 10 minutes, five seconds in the intervention period, the study showed that pharmacist’s mean drug preparation time decreased by 76%, dropping to 46 seconds in the intervention period from three minutes, 13 seconds in the baseline period.

Dr. Rothschild attributed the decrease in the pharmacist’s time during the intervention period to “the quicker scanning process associated with the robotic software package. The pharmacist could scan the final administration container with all the product information displayed on a single screen, saving the time that pharmacists need to manually gather various verifying data, both in computer and written form.”

The total mean drug preparation times for pharmacists and technicians together increased from seven minutes, 24 seconds in the baseline period to 10 minutes, 51 seconds in the intervention periods.

Based on hourly salary and fringe benefit costs of $25.44 for technicians and $64.23 for staff pharmacists, mean total pharmacy labor costs per preparation dropped from $5.22 during the baseline period to $5.10 for the intervention period.

During the intervention stage, the medications were prepared by the Brigham and Women’s Hospital pharmacy using a CytoCare Robot compounder made by Bolzano, Italy-based Health Robotics. After the trial, as a way to further increase accuracy in the future, the pharmacy installed Health Robotics’ i.v. SOFT gravimetric workflow software for manual preparations of antineoplastic drugs and adjuvant medication compounding and also re-educated its staff to improve accurate manual preparations, Dr. Rothschild noted.

Although robotic compounding environments offer the potential for safer and more cost-effective antineoplastic and adjuvant medication preparation, they also could introduce unintended consequences including potential or actual harmful outcomes. Over the course of this study, Dr. Rothschild reported, there were 45 events of unintended consequence specifically attributable to the new technology (4.6%), including 41 mechanical and four software failure events. None of these events resulted in a final product, and they were not judged to be serious medication errors. They included 24 cases with no drug injected into the bag or syringe; four cases in which the internal checking system of the robot rejected the drug due to incorrect weight; three cases of bags, syringes or needles dropped or cracked before they were completed; three reconstitution shaker failures; two malfunctions of medication carousels; five other mechanical failures; and four software failures that prevented the completion of drug preparation.

Future additional studies, Dr. Rothschild and Ms. Power both said, are needed to address the overall costs and benefits of using what Dr. Rothschild called “this expensive but potentially valuable alternative to manual preparations of antineoplastic and adjuvant medications and the clinical impact of the improved accuracy of preparations.”

Ms. Power said that she believes that IV robots “will be major contributors to future IV compounding” and was “pleased that authors of this caliber are investigating robot use in compounding cancer treatments.”