SummaryPharmacists and pharmacy technicians at the University of Maryland Medical Center use autonomous mobile robots to track and deliver certain medications to nursing units. For these orders, which are primarily nonstat and cannot be delivered by pneumatic tube, pharmacy staff receive an electronic medication order from a physician, print the label, fill and check the order, scan and place the medication in one of the robot’s medication drawers, and then select a destination on computer software that communicates with the robot. The robot navigates its way to the appropriate unit using wireless and other technology. Once it arrives, the nurse enters a personal passcode and uses a built-in fingerprint scanner, which then causes the designated drawer to automatically open, allowing the nurse to remove the medication(s). Special steps allow the system to track and document the chain of custody for controlled substances, as required by the U.S. Drug Enforcement Agency. During its first year in operation, the system expedited and improved the reliability of medication delivery, reduced delivery costs and lost medications, and increased nurse efficiency.Moderate: The evidence consists of pre- and post-implementation comparisons of key outcomes measures, including the timeliness, reliability, predictability, and costs of medication delivery, along with nurse efficiency.
Developing OrganizationsAethon; University of Maryland Medical Center
Date First Implemented2004
In 2004, the University of Maryland Medical Center implemented the first mobile robot in the trauma unit, and then put in place seven additional robots in the same and other areas between 2004 and 2007. In 2010, the hospital added chain-of-custody software that allows the robots to deliver controlled substances. In January 2013, the medical center upgraded the last three robots to complete controlled substance delivery and tracking in all critical care areas.
Problem AddressedHaving pharmacy technicians hand deliver medications to nursing units can create backlogs for order preparation and delay medication delivery, reducing the efficiency of pharmacists and nurses,1 negatively affecting patient care, and driving up labor costs. Automated medication delivery systems such as pneumatic tube systems can help address these problems, but they cannot be used for all medications.
- Backlogs driven by hand delivery: Having pharmacy technicians hand deliver medications to nursing units can be a time-consuming task that reduces time spent helping pharmacists fill medication orders, often leading to backlogs. At the University of Maryland Medical Center (UMMC), technicians are responsible for compounding intravenous (IV) medications and helping to fill medication orders. Each time a technician left to deliver a medication to a nursing unit, the workload on pharmacists increased and order fill times were extended. Technicians also help with telephone calls and other operational duties that minimize distractions and are designed to reduce errors, allowing the pharmacist to concentrate on checking medication orders.
- Negative impact on patient care: Delays in medication administration can result in patient harm or discomfort. In addition, nurses often have less time to spend on direct patient care, as they have to track down and retrieve medications that patients need right away.2 Before implementation of this program, UMMC nurses often complained about excessively long medication delivery times, which averaged 74 minutes from receipt of the order to delivery of the medication to the unit. As a result, nurses frequently spent time calling the pharmacy to track down the medications or made trips to retrieve them, activities that took them away from direct patient care.1
- High labor costs: In the absence of automated delivery systems, hospitals generally must employ enough pharmacy technicians to allow prompt delivery of medications to the units. At UMMC, the costs of having enough technicians available around the clock to ensure reasonable delivery times was difficult to sustain given that the pharmacy technician labor was getting increasingly difficult to find and UMMC would have to find coverage for technicians who called out sick or medications would not be delivered in a timely manner.
- Automated systems only a partial solution: Many hospitals (including UMMC) use automated medication delivery systems, such as pneumatic tubes, to speed delivery of medications to the nursing units and relieve pharmacy technicians of distribution tasks.3 However, such systems may not be feasible for the delivery of certain medications, such as liquid medications that cannot be shaken, hazardous medications, and controlled substances that must follow chain-of-custody procedures required by the U.S. Drug Enforcement Administration.4 Consequently, many hospitals still rely on hand delivery of these medications by pharmacy technicians or courier services.
Description of the Innovative ActivityPharmacists and pharmacy technicians at UMMC use autonomous mobile robots to track and deliver certain medications to nursing units. For these orders, which are primarily nonstat and cannot be delivered by pneumatic tube, pharmacy staff receive an electronic medication order from a physician, print the label, fill and check the order, scan and place the medication in one of the robot’s medication drawers, and then select a destination on computer software that communicates with the robot. The robot navigates its way to the appropriate unit using wireless technology. Once it arrives, the nurse enters a personal passcode and uses a built-in fingerprint scanner, which then causes the designated drawer to automatically open, allowing the nurse to remove the medication(s). Special steps also allow the system to track and document the chain of custody for controlled substances. Key elements of the process include the following:
- Electronic medication order: A physician enters an order for a patient’s medication using a computerized medication order entry application.
- Pharmacist review: The pharmacist uses a medication management application to review an electronic queue of patient orders. Linked to the electronic medical record, the application automatically flags orders that include an inappropriate dose range or are contraindicated based on the patient’s medical history (e.g., due to potential drug-drug interactions, allergies, or current laboratory values). The pharmacist contacts the ordering physician to discuss any alerts and revise medication orders accordingly.
- Pharmacist and technician filling of orders: The pharmacist prints a barcoded label listing the patient name, medication, dose, nursing unit, and prescribing physician. The technician fills the order and places the label on the medication bag (for pills) or IV bag. The pharmacist checks the filled order and electronically or manually documents that it is ready for delivery to the nursing unit.
- Pharmacist decision on appropriate delivery method: The pharmacy uses the robot, TUG® by Aethon, for the delivery of most medications to the nursing units, with pneumatic tube and hand delivery by a technician also being options. For controlled substances, hazardous medications, and liquid medications that cannot be sent via pneumatic tube, the pharmacist decides on the appropriate method—robot or hand delivery by a technician (generally used only for “stat” orders). On the typical day, pharmacists use the robot system—which is available at all times—for 110 medication deliveries to units.
- Loading completed orders on robot: For medications to be delivered via robot, the technician signs on to the system by entering a personal passcode using a keypad and then placing his or her finger on a built-in fingerprint scanner (both located on top). The technician then scans the label’s barcode and places the order into the appropriate drawer for the destination unit. Technicians typically sort the medications to be loaded onto the robot by unit.
- Selecting destination: The pharmacist or technician uses touchscreen software (MedEx™) on a desktop computer to select the robot’s destination, which is communicated to the robot via wireless technology. Multiple destinations can be selected, enabling the robot to make deliveries to several units in one trip.
- Navigating to unit: The robot navigates the hospital corridors using a preprogrammed hallway map, with its computer confirming the distance traveled by calculating wheel rotations. The robot uses wireless technology to call the elevator and select the proper floor. A laser-guidance system, sonar, and infrared sensors allow the robot to avoid or navigate around walls, people, or other obstacles.
- Nurse removal of medications: When the robot arrives on the unit, the nurse enters a personal passcode and then scans his or her fingerprint that opens the appropriate drawer. The nurse removes the medications and either delivers them to patients or stores them on the unit. The nurse manually closes the drawer and presses a “go” button on top of the robot to send it on its way to other units or back to the pharmacy (as determined by the pharmacist or technician).
- Special systems to track controlled substances: The robot system uses special steps to closely track controlled substances. When filling orders, pharmacists attach a radiofrequency identification (RFID) tag to these medications to allow constant tracking. The pharmacist places the medication in a drawer specially designated for controlled substances. The robot has an RFID transceiver that notes the medication, patient name, medication location, routing time, and all hospital personnel involved in the delivery process. The nurse either delivers the medications directly to patients or stores them in an RFID-tracked lockbox on the unit (which also must be accessed using a fingerprint and passcode). Through this process, the system tracks and documents the chain of custody for controlled substances in accordance with U.S. Drug Enforcement Agency requirements. Documentation occurs electronically, eliminating the need to complete time-consuming paperwork. All chain-of-custody information is transmitted back to the pharmacy. If a controlled substance is left in the lockbox for more than 24 hours, the pharmacy department receives an alert and staff remove it and then return it to the pharmacy.
References/Related ArticlesPatient safety and quality: pharmacy services on a roll. University of Maryland Medical Center. 2010 Dec. Available at: http://health.kernan.org/quality/tugs.htm.
Case study: University of Maryland Medical Center. Aethon, Inc. 2012 July. Available at: http://www.aethon.com/ummc-case-study-2011/.
Contact the InnovatorJeremy W. Daniel
Department of Pharmacy Services
University of Maryland Medical Center
621 South Charles Street
Baltimore, MD 21201
Innovator DisclosuresMr. Daniel reported having no financial interests or business/professional affiliations relevant to the work described in this profile.
ResultsDuring its first year in operation, the system expedited and improved the reliability of medication delivery, reduced delivery costs and cases of lost medications, and increased nurse efficiency.
Moderate: The evidence consists of pre- and post-implementation comparisons of key outcomes measures, including the timeliness, reliability, predictability, and costs of medication delivery, along with nurse efficiency.
- Faster and more reliable medication delivery, leading to better patient care: Medication delivery cycle time decreased from 74 to 30 minutes for all medications delivered through the robot system in the areas where the robots deliver. As a result, medications can be administered to patients more quickly, thereby improving patient care. Delivery reliability (i.e., how often the medications actually arrive at the unit as promised) has increased by 23 percent, and delivery predictability (i.e., how often the medications arrive within the time promised) has risen by 50 percent.
- Lower delivery costs and fewer lost medications: The per-trip cost with a robot averages $2.40, less than one-half of the $5.50 cost of the typical hand delivery using a pharmacy technician. The hospital has also experienced a decline in lost medications since implementation of the system (no hard data available).
- More efficient nurses: The system freed up 6,123 hours for nurses by reducing time spent tracking or retrieving medications.
Context of the InnovationUMMC is a 654-bed academic medical center serving patients in Maryland and the mid-Atlantic region. The medical center treats roughly 38,000 inpatients and 300,000 outpatients annually. Its pharmacy includes 225 employees who dispense 25,000 medication doses daily. The impetus for the robot system came from nurses who believed that the pharmacy’s medication delivery cycle time was too slow for medications that could not be delivered by pneumatic tube, and that they spent too much time tracking down these medications, thus reducing time available for direct patient care. These issues prompted pharmacy department leaders to consider use of automated systems to enhance efficiency of delivery for these medications.
Planning and Development ProcessSelected steps included the following:
- Discussing solutions: During meetings of senior leaders from nursing and pharmacy, nurses discussed the implications of long delivery cycle times, and pharmacy leaders emphasized that the department could not staff a delivery position with pharmacy technicians who were needed in other areas and could not backfill the position in case one or two technicians were to call in sick. The group discussed various technologies to help address the issue.
- Working with vendor: The robot vendor, Aethon, Inc., approached the UMMC pharmacy director with a possible automated solution, the TUG® robot.
- Pilot testing: UMMC pilot tested one robot on the trauma unit. After the test, pharmacy and nursing representatives met to share feedback; both groups believed that the system worked well.
- Expanding program: Based on the success of the pilot test, hospital leaders gave budgetary approval for expansion of the program.
- Registering nurse users: Nursing staff on the relevant units registered their fingerprints and set up unique passwords during an organized registration effort led by the vendor, which provided a computer able to read fingerprints. The vendor scheduled registration times between shifts to make it as convenient as possible for nurses. Pharmacy staff handle ongoing registration of new users according to a set schedule.
- Training: A vendor representative remained onsite during the first week of each unit’s implementation of the system. This representative followed the robot around the hospital, training relevant staff on how it worked. The pharmacy night staff helped train nursing staff during the night and weekend shifts.
Resources Used and Skills Needed
- Staffing: The use of the robot requires no new staff, as existing personnel incorporate it into their daily routines. Hospitals that previously used technicians to deliver medications may redeploy or require fewer such staff after adopting the system.
- Costs: Data on program-related costs are unavailable; the primary expense consists of monthly service fees for the robots and chain-of-custody software. Some overtime pay may be required for pharmacy staff during initial registration efforts.
Funding SourcesUniversity of Maryland Medical Center
Tools and Other ResourcesMore information about the TUG® system and the controlled substance software is available from the vendor (Aethon, Inc.) at: http://www.aethon.com/.
Getting Started with This Innovation
- Incorporate nurses in decisionmaking and implementation: Including nurse managers and frontline staff in the decisionmaking process will help them understand and accept the robots. They can also offer important feedback on how best to incorporate the system into their unit's workflow.
- Decide on implementation strategy: Each hospital must decide whether to implement the robot system in phases or all at once throughout the facility. The right choice depends on various factors, including hospital finances and the institution's ability to effectively project manage multiple installations simultaneously. There may also be a shortage of resource needs, such as pharmacy technicians, that require a shorter implementation period.
- Set up systems for data collection: Data systems should be created to track the program's impact on key outcomes, such as medication delivery cycle time and costs, staffing resource savings, and nurse efficiency.
- Expect initial resistance from nurses, which should fade over time: Some nurses may initially resist adopting the new system due to an aversion to change. Resistance typically disappears once nurses see how the system speeds up medication delivery and frees up time for direct patient care. Over time, they often become advocates for the system.
Sustaining This Innovation
- Track and share data on program impact: Tracking and sharing data on the program's impact help to sustain support among hospital leaders and staff who use the system.
- Consider expanding program: Hospitals using a phased implementation approach should consider expanding the program over time, once the benefits become clear to key stakeholders, including hospital administrators who have control over the budget.
- Educate new staff members: An annual continuing education program detailing the robot's benefits and savings helps future staff members understand and adopt the technology faster.
Use By Other OrganizationsUMMC was the first hospital in the country to use TUGs®; more than 100 hospitals do so today.
1 Case study: University of Maryland Medical Center. Aethon, Inc. 2012 July. Available at: http://www.aethon.com/ummc-case-study-2011/.
Braswell A, Duggar S. The new look of bedside technology: the point-of-care evolution drives providers to rethink nursing workflow and medication management. Nurs Manage. 2006;37(10):14-18. [PubMed]
Garrelts JC, Koehn L, Snyder V, et al. Automated medication distribution systems and compliance with Joint Commission standards. Am J Health Syst Pharm. 2001;58(23):2267-72. [PubMed]
|Disclaimer: The inclusion of an innovation in the Innovations Exchange does not constitute or imply an endorsement by the U.S. Department of Health and Human Services, the Agency for Healthcare Research and Quality, or Westat of the innovation or of the submitter or developer of the innovation. Read more.|
Service Delivery Innovation Profile
Original publication: October 09, 2013.
Original publication indicates the date the profile was first posted to the Innovations Exchange.
Last updated: October 09, 2013.
Last updated indicates the date the most recent changes to the profile were posted to the Innovations Exchange.