Medication errors are estimated to cause approximately 7000 deaths annually and cost approximately $2 billion in added health care spending. Improving the safety outcomes of medication-use processes in hospitals has been cited as a means of addressing this issue. Careful and planned use of information technology is a proven approach to solve a majority of medication errors made with paper-based systems (Paoletti et al., 2007). These systems have been found to provide much better compliance with the “five rights” of medication administration: right patient, right dose, right route, right time, and right medication (Hook, 2008). Accordingly, the present paper describes a test plan for the implementation of an electronic medication administration record (eMAR) with a bar-code component in a hypothetical hospital setting. The current technology utilized by in the hypothetical hospital is a manual, paper-based, five day MAR.
Pre-implementation studies in a hospital setting similar to the one selected for this test plan indicated that over a five-day period a total of 188 errors were noted (Paoletti, 2007). In general, the majority of the errors were classified in a first category of wrong time (15 of 188), or late administration. However, a second category of errors included problems such as omission (15), wrong technique (14), wrong dose (6), extra dose (5), wrong medication (3), wrong route (1) and wrong formulation (1) (Paoletti, 2007). The goal of the new eMAR system will be to greatly reduce the errors in the second category, while not impacting workflow that could be reflected in an increase in errors of the first category, that is, missed timing.
An overview of the methods of developing the actual test plan would involve workflow analysis, review of the workflows to make changes for efficiencies, and review of the requirements as perceived by a project committee of information technology staff members of the hospital, physicians, and nurse end users. The foundation of all of these analyses will be evidence-based whenever possible, thus pre-installation testing, like that described above, will likely have to be done in order to have the data needed to make intelligent decisions. After data is gathered and analyzed, the committee will perform a review of the possible commercial products available to meet the identified requirements, and will finally make a selection of a product for use. Once the final system is selected, training and implementation in a pilot unit will occur, followed by changes to the system as needed, and then staged rollout to the entire hospital. Other less labor-intensive test plans could be formulated, but involving evidence-based decision-making and a committee of both experts and end users is commonly used to support buy-in and improve the match between the selected system and practical needs.
The scope of the test plan for this eMAR application is that it will be implemented across all units within the hospital including inpatient, outpatient infusion center, cardiac catheterizations labs, and post-anesthesia care units (Paoletti, 2007). However, a staged implementation approach is recommended, to allow for one nursing unit to become the pilot testers of the system. Based on their experiences and input, issues can be solved in a manageable way before larger-scale implementation is applied in a step-wise fashion. The schedule is anticipated to have the selection process occur and pilot unit system in place within six months, a three-month trial period to allow for adjustments to the system and training, and then a full hospital roll out within the next three months for a total of a one-year schedule for movement from a paper-based to a electronic eMAR system.
Acceptable deliverables in this project would be all components in place as required for a system that provides electronic recording of the medications that are ordered and administered within the nursing unit that the system is monitoring. In essence, this system would replace the current manual based script system with an electronic order and record of administration. This would include a means, based on a bar-code system, for identifying the patient through an ID bracelet, a means for identifying the medication administered as provided by the FDA bar-coding rules, and an electronic record-keeping system for both the medication order by the physician and the execution of that order by the nurse. Some parts of these records would include patient name, medication, dose, route of administration, and timing of order and timing of administration. A projected workflow for the expected deliverables is included as Figure 1.
A key component of the deliverables would be a reporting system that would record discrepancies between orders and administration in two manners. The first would be a prevented medication error report that would show times when warning notices were displayed to the nurse or other provider administering the medication, but changes were made in order to bring the administration into compliance with the order. The second would be a possible medication error report that would show where the system was not being used correctly, that is, a warning error is displayed but medication administration continued (Paoletti, 2007). Both of these reports will provide valuable data concerning the benefits and risks of the newly implemented system.
Many resources would be necessary to carry out this test plan. The initial human need is to staff a project team to oversee the system selection. As mentioned above, such a project team should include both information technology experts from the hospital staff as well as end users. There should also be an executive level member that has been tasked with directing the system selection and purchase. From a human staffing point of view, it would be ideal to have one administrator of the system, preferably with information technology healthcare expertise, hired to oversee the implementation and rollout process. Ideally, this administrator could handle the training component. Experience has shown that these kinds of systems require at least four hours of training and include practical, hands-on activities (Hook, 2008). Another human resource that could be implemented would be “super-users,” or champions of the system, perhaps from the pilot unit, that receive extra training and are stationed in the follow-on units in the first few days of their rollouts. This kind of human support has been shown to be key in getting buy in from nurse users (Hook, 2008).
Some of the hardware required includes computer terminals or tablets that have attached scanners for the bar-coding aspects. If budget allows the ideal situation is one computer/tablet for each patient room, as rolling terminals have proven an issue with workflow and storage space issues. Other hardware needs are for producing labels for both patient bracelets and drug containers that would be located at patient intake and the pharmacy, respectively. Consideration should be given to wireless and/or networked implementation of the system, as this would help relieve the need for wires to connect this system, promote more efficient connections between the components of the system, make the system much more portable, and result in less space impact overall.
Software needs would be the eMAR program itself and any supportive bar-code aspects not included in the eMAR system. Programs available from commercial providers should be reviewed for conformity with the needs of the hospital, as found by the committee appointed to lead up this project. Unless there are highly specialized needs within the hospital, it is anticipated that an off the shelf software system would be the best choice, although customization could be available if needed. Another possibility would be software as a service (SAAS), therefore a cloud-based system, but as discussed more fully below, risks with patient confidentiality would have to be carefully weighed with such a system. Whether a software or cloud based system is selected, consideration should be made for the inclusion of technical support time from the software provider within the purchase contract. An estimated budget for this project would be approximately $500,000 and could be more if specialized software adaptions are required. If an SAAS solution is selected, although there may be less initial up-front installation costs, it should be noted that licensing fees will be an ongoing expense that will need to be budgeted.
Projects of this type have very important risks that need to be addressed. A first risk that needs to be addressed is back-up systems during system failure. As the administration of medication is a required function within the hospital, there will need to be a contingency plan for system failures that will occur. Possible plans include having a redundant system in place that can be accessed if needed, or the selection of a SAAS system that shifts the back-up plan burden to the service provider. Another option would be to utilize the current paper-system in case of a software system failure, although depending on the time lag between eMAR implementation and the first system failure, memory of the paper system may be very dim. Importantly, this system does have the advantage of not requiring electricity, thus if the failure involves a disaster including electrical failure requiring reliance on hospital generated electricity, the paper system would not be an added burden.
A further consideration of risk, discussed briefly above, is patient confidentiality. Certainly it must be verified that any commercial system selected complies with governmental regulations, such as HIPAA, or the applicable equivalent. Further, if SAAS is selected for the eMAR privacy issues will have to be examined even more carefully, as cloud based applications require by definition the movement of highly confidential information outside the control of the hospital. However, other advantages such as cost savings, less need for technology support, and ability to update without additional cost can over-ride the added confidentiality risk.
Thus, the present paper provides a theoretical test case plan for the implementation of a eMAR system with bar-code capabilities within a hospital setting. It provides basic approaches for determining the organizational needs of such a system and a proposed plan for selecting a system that meets those needs. This plan also describes implementing the system including a proposed workflow (Figure 1), and details resource needs such as hardware, software, human resources, space, and estimated budget. Finally, two major risks of the system were addressed in the need for system failure backup and patient confidentiality concerns. In putting this test plan together, it is evident that this process is complex and will require the concerted efforts of many persons to be successful. But the ultimate goal of greatly improved patient safety that is provided by these information technology systems justifies this complex process many times over.
Figure 1 -- Workflow for eMAR (Ola-Weissman, 2013).
References
Hook, J. M., Pearlstein, J., Samarth, A., & Cusack, C. (2008). Using Barcode Medication Administration to Improve quality and safety. Agency for Healthcare Research and Quality. U.S. Department of Health and Human Services. Retrieved from
http://healthit.ahrq.gov/sites/default/files/docs/page/09-0023-EF_bcma_0.pdf
Keohane, C. A., Hurley, A., Bane, A. D., Bates, D. W., Feathersone, E., Gandi, T. K. . . . Woof, S. (2008). Quantifying Nursing Workflow in Medication Adminstration. The Journal of Nursing Administration. 38(1): 19-26.
Ola-Weissman, T. (2013 April 8). Implementation of an electronic medication administration system. Slideshare. Retrieved from
http://www.slideshare.net/Spiderella/implementation-of-an-electronic-charting-system
Paoletti, R. D., Suess, T. M., Lesko, M. G., Feroli, A. A., Kennel, J. A., Mahler, J. M., & Saunders, T. (2007). Using Bar-code Technology and Medication Observation Methodology for Safer Medication Administration. American Journal of Health-System Pharmacy. 65 (5):536-43.
