Bar Code Medication Administration

From Wikipedia, the free encyclopedia

Bar Code Medication Administration (BCMA) is a barcode system designed to prevent medication errors in healthcare settings and improve the quality and safety of medication administration. The overall goals of BCMA is to improve accuracy, prevent errors, and generate online records of medication administration. It consists of a barcode reader, a portable or desktop computer with wireless connection, a computer server, and some software. When a nurse gives medicines to a patient in a healthcare setting, the nurse can scan barcode on the wristband on the patient and make sure that the patient is the right patient. The nurse can then scan the barcode on medicine, the nurse and the software can then verify if it is the right medicine at the right dose at the right time by the right route ("Five rights"). [1] Bar Code Medication administration was designed as an additional check to aid the nurse in administering medications, however it can not replace the expertise and the professional judgment of the nurse. BCMA was first implemented in 1995 [2] at the Colmery-O'Neil Veteran Medical Center in Topeka, Kansas, USA. It was conceived by a nurse who was inspired by a car rental service using barcode. From 1999 to 2001, Department of Veterans Affairs promoted the system to 161 facilities.[3]. Cummings and others recommend the BCMA system for its reduction of errors. They suggest healthcare settings to consider the system first while they are waiting for RFID. They also pointed out that adopting the system takes a careful plan and a deep change in work patterns.[4]

Improving Medication Safety with a Wireless, Mobile Barcode System in a Community Hospital
By Mitch Work

Over the past few years, hospital organizations have increasingly looked to new technology solutions to improve patient safety. Barcode technology is an especially
promising approach in the effort to reduce medical errors. While barcode technology has been used for quite some time in many hospital applications, it has only recently been used to address patient safety. The use of barcode technology at the patient's bedside has shown impressive gains in reducing medication administration errors, which may account for as many as 7,000 deaths per year in U.S. hospitals (IOM, 2000). Barcodes provide a valuable verification of medication administration by assuring that the "five rights" are confirmed right patient, right medication, right dose, right time, and right route of administration. While studies conducted in VA hospitals in the early 1990s showed that the use of barcodes reduced medication administration error rates by up to 86% (Meadows, 2003), community hospitals are just beginning to use this technology to improve patient safety. Recent estimates indicate that only 2% to 6% of hospitals are currently using barcodes to reduce medication administration errors (Center for Business Innovation, 2004).

Background
This case study examines the use of a wireless, mobile barcode medication administration system at Beloit Memorial Hospital in Beloit, Wisconsin, a 175-bed community hospital with four off-site clinics, serving a population of approximately 175,000 residents living in southwestern Wisconsin and northwestern Illinois. The hospital has a formal affiliation with the University of Wisconsin Hospital System and an active though informal affiliation with Rockford Hospital in Rockford, Illinois. The hospital's Board of Trustees, composed of 10 medical professionals and community leaders, has committed the hospital to a primary goal of providing highquality healthcare services for all patients. In September 2003, Beloit Memorial installed a wireless, handheld barcode medication administration system in its Family Care Center (FCC) unit, consisting of 35 ob/gyn, pediatric, and labor/delivery beds. This pilot resulted in a 67% decrease in medication administration errors within the first four months of operation. With the pilot successfully completed, the hospital expanded implementation of the barcode system to four additional inpatient units. This case study documents Beloit's experience in these units, which included:

Special Care Center (SCC) Oncology Multi Care Center (MCC) General Medical Surgical Intermediate Care Center (ICC) Telemetry Critical Care Center (CCC) Critical Care

These units presented a more complex environment than the FCC, treating patients with generally more serious medical conditions, who often required more medications. Consequently, these units were more prone to experience medication administration errors.

Goals and Objectives

The main goal of this study was to identify and measure the benefits of a wireless barcode system to reduce medication administration errors. To address this goal the following specific objectives were established: 1. Conduct an onsite visit, both pre- and post-implementation of the wireless barcode system in the four new inpatient units. 2. Determine pre- and post-implementation medication administration error rates in these units.

Methodology
Pre- and post-implementation site visits were conducted by an independent consultant who interviewed senior nursing management, the unit managers, and the nurses charged with medication administration responsibilities; and observed the medication administration rounds. The pre-implementation visit was conducted in late June 2004, and the post-implementation visit was conducted in December 2004, to review the results of the system four months after the "go-live" date. This interval was selected to ensure that the nursing staff was familiar with the barcode system and had completed its learning curve of the new system.

Patient Safety Goals

The hospital Board of Trustees and senior management have focused on establishing a culture of cooperation and communication aimed at improving the quality of patient care. This culture was key to embarking on a successful patient safety program. While the hospital established a Patient Safety Committee in 2001, the barcoding initiative originated from a recommendation by the director of pharmacy, who felt that additional checks and safeguards at the point of medication administration would be valuable in reducing medical errors. As a result, the hospital embarked on its initial barcode pilot in September 2003. The alignment of the board and senior hospital management including the CEO, the vice president of nursing, and department heads created a positive atmosphere and communicated a clear, unified message from senior management that led to the adoption and funding of the new barcode system.

Anticipated Benefits
In expanding the use of barcode technology to the additional inpatient units, the hospital hoped to achieve the following benefits: 1. Improve patient safety through the reduction of medication administration errors, while minimally increasing nurse staffing time requirements. 2. Improve nursing staff job satisfaction. 3. Improve patient satisfaction. 4. Generate positive public relations with the local community.

Barcode Implementation at Beloit

In February 2003, the Board of Trustees unanimously approved funding for a bar-code pilot project in the FCC, which began in July 2003. By January of the following year, the hospital approved the further deployment of barcoding for medication administration to include four additional inpatient units due to the FCC's early success. Implementation of the original pilot and subsequent rollout to the other inpatient units occurred over a 15-month period from July 2003 through October 2004. The SCC went live in March 2004. The MCC, CCC, and ICC all went live in August 2004. This chronology is shown in Figure 1 below.

Figure 1. Timeline of Approvals and Implementation Rollouts of the Wireless, Mobile Barcode System at Beloit Memorial Hospital

Staff Training
The SCC, an oncology unit, was selected as the second unit for barcode utilization implementation due to the fact that it administered many medications yet it was a relatively small unit with seven beds. SCC staff training on the barcode system was conducted over a one-week period immediately prior to go-live. The training was conducted by "super users," FCC nurses experienced and proficient with the system. These super users volunteered to assist in the SCC training. In addition, training staff from the vendor were on site to assist with the 4-hour training sessions, in which all 12 unit nurses were trained. Once the staff training was completed, barcoding went live in March 2004. Daily meetings were held with the nursing staff during the first two weeks following implementation to troubleshoot any problems and to provide rapid response to questions about the system. In addition, a dedicated notebook was placed in the SCC where nurses could write comments, provide feedback, and ask questions about the system. The notebook was reviewed regularly by nursing managers to address issues quickly. In addition, daily management meetings were conducted with the director of pharmacy, director of IT, and the nurse unit manager to ensure that the system was working properly. The implementation proceeded smoothly, and hospital management reported that they were extremely satisfied with the system. Based on the success of the implementation in both the FCC and the SCC units, the decision was made to bring on three additional units simultaneously. A training design similar to that used in the SCC was implemented using super users as well as on-site support from the vendor. The MCC, CCU, and ICC went live in August 2003. A total of approximately 100 nurses were trained for these units. Training was conducted 48 hours prior to go-live. Eight super users were utilized in the training and were present on the units during the implementation. Going forward, new staff will be trained exclusively by the super users, eliminating the need for ongoing training by the vendor and reducing costs. Hospital management was extremely pleased with this training approach. This can be attributed to several factors including careful, advance planning, continual communication with nursing staff both prior to and following implementation, a flexible training schedule to accommodate all shifts, and the fact that the barcode utilization process had been carefully integrated into the medication administration workflow processes. Other factors also had a positive influence on the barcode system adoption in these new units including the initial success of the FCC pilot and nursing staff's generally high level of satisfaction with the new system. These factors were publicized both formally by hospital management and, perhaps of equal importance, were communicated informally by staff nurses throughout the hospital. Components of Wireless Barcode System
Initial implementation of wireless barcode technology at Beloit was with Care Fusion's wCareMed release 3.0 (www.carefusion.com), and it is now being used in all five inpatient units. An upgrade to release 4.0 is expected in the first quarter,

Implementation Challenges and Solutions

Implementing a new system like barcode medication administration can present many challenges as nursing staff adapt to new work processes. Beloit Memorial encountered few objections by the nursing staff, though a few minor concerns were expressed about "having to learn new technology." Implementing the system in the Critical Care Center was somewhat more challenging. Initial concerns were expressed regarding the need to administer some drugs on an emergency basis or administering drugs to newborns within one hour after birth to comply with state law, and the new system changed work processes in a manner that they felt could potentially compromise patient care. The commitment to safe patient care was reinforced by management as the key treatment goal that would take precedence over the use of the barcode system in emergency situations. Once the system was installed and the CCC staff became more familiar with the new system, these concerns did not resurface. Further, several staff were personally involved in situations where the system stopped a medication error from occurring, successfully demonstrating that their patient's safety had been enhanced through the use of the wireless, barcode medication administration system.

2005. An interface to the Meditech HIS was implemented during the pilot by Iatrics Systems. Symbol Technologies PPT 8800 handheld wireless devices with integrated barcode scanners are currently being used. A wireless network (802.11b wireless LAN) with 128 bit encryption for data security was installed with access points and connecting sensors in all patient rooms. A dedicated server (IBM xSeries 345 Server, Pentium 4 Xeon 2.8 GHz, 4 GB RAM) was connected to the wireless network.

Another issue that needed to be addressed during the implementation was the lack of barcodes on some of the unit-dose drugs being used. The barcodes had not been affixed to some unit dose medications, as required, due to a temporary staffing shortage in the pharmacy. This issue was quickly resolved and did not reoccur. However, this does suggest that the hospital pharmacy must be intimately involved in the planning and implementation of any new barcode utilization system for medication administration. The pharmacy department needs to consider the additional time that will be required to affix medications with barcodes.

Despite the implementation of the new barcode system and the changes in work processes that it necessitated, the hospital was not required to hire additional pharmacy or nursing staff.

Benefits Realized: Patient Safety Improvements

A comparison of the medication administration error rates pre- and post-implementation shows that medication administration errors were reduced by an average of 82% for the five units studied (see Figure 2).

Figure 2. Medication Administration Error Rates in Inpatient Units Pre- and Post- Implementation of Wireless, Mobile Barcode System (September 2003 - November 2004)

The post-implementation data was collected from the go-live month in each unit through November 30, 2004, the most recent data available at the time of this study. The reported post-implementation time periods ranged from 4 to 15 months experience, depending on the unit. This data includes the first month that the system was installed, when staff learning curves were steepest and errors were most prone to occur. If the first month of implementation is not included, to allow time for learning the new system, the medication administration error rate is decreased even further to an average of 93%. All participating units reported significant error reduction with the greatest numeric decline in monthly average medication errors occurring in the MCC, which fell from an average of 5.8 to 1.25 medication errors per month. The error rate in this unit was high in comparison to the other units because the MCC is the largest unit with a total of 64 licensed beds. The types of medication errors that occurred in these units varied and included instances of wrong dosage, missed medication, missed drug reaction, or wrong IV bag hung and were documented in the hospital's Med Event Report.

Other Benefits Realized

The benefits of the barcode system extend beyond a decrease in medication administration errors. They include:

Improved nursing staff satisfaction. Increased job satisfaction was identified as an additional benefit of the system. The nursing staff felt that the hospital was truly committed to improve quality of care by investing in state-of-the-art technology to improve patient safety. Recruitment of new staff was also mentioned as an additional potential benefit. Showing the hospital's advanced use of technology was cited as a positive factor in recruiting new staff in a competitive employment environment. Patient satisfaction. During the initial implementation, all patients were briefed about the new medication administration system and why it was being installed. Nurses reported a genuine sense of appreciation from patients and their families that the hospital was committed to patient safety, as demonstrated by use of the new system. Improved community relations. After deciding to implement a new medication administration system, the hospital contacted the local media and sent senior management to speak with local citizen groups (e.g., Rotary International) to educate the community about the new system and why it was being implemented. This public relations exposure helped to build on the existing positive relations between the hospital and the local community and resulted in several stories in the local press.

Future Automation Directions

Currently, the hospital is committed to proceeding with the rollout to make full use of other wireless mobile barcode applications. The rehab unit went live in the fourth quarter, 2004. It is expected that wireless barcoding will be implemented for respiratory therapy by the first quarter, 2005. Other barcode applications will also be installed at the hospital in accordance with the implementation schedule shown in Figure 3.

Figure 3. Proposed Wireless, Mobile Barcode Application Implementation Schedule for Beloit Memorial Hospital

Conclusion
Recent reports indicate that five years after the Institute of Medicine's landmark study, To Err Is Human: Building a Safer Health System (2000), hospitals have made little progress in implementing patient safety solutions to reduce medical errors. This case study has documented the experiences of one community hospital, committed to a culture of patient safety, which has demonstrated the success of a wireless mobile barcode system. This resulted in a reduction of medication administration errors, in the inpatient units observed, by an average of 82% after being installed for a relatively brief period of time (5 to 15 months).

Additional benefits attributed to the barcode system were also identified. These include improved staff and patient satisfaction, and generation of positive press and public relations with the local community. In the final analysis, success of the barcode medication administration system at Beloit Memorial can be attributed to not only the use of state-of-the-art technology, but also to a carefully constructed, fully communicated work plan that included clearly identified new work processes and a thorough staff orientation and training program.

Mitch Work (mitchwork@workgroupinc.net) has over 25 years experience in healthcare as a researcher, educator, and executive for provider and supplier organizations. Currently, he serves as president of The Work Group, Inc., a healthcare consulting company delivering strategic marketing, sales, and public relations services to healthcare IT companies and provider organizations. Over the past 18 years, he has been involved in healthcare information technology consulting, assisting HIT companies and providers to be more successful. Recently, he has focused on new technologies for patient safety including barcoding, RFID, wireless technology, and the Internet. He also assists companies to launch new HIT products and establish vertical healthcare technology businesses. He has conducted ROI and best practices studies to demonstrate financial and clinical benefits from HIT. Over the past year, he has consulted with several national and international healthcare technology conferences and has served as a co-chairman, presenter, and panel moderator.

Nurses Perceptions of a (BCMA) Bar-coded Medication Administration System

A Case-Control Study by Valerie J. Gooder Ph.D., RN This article was made possible by an educational grant from Chamberlain College of Nursing

CITATION
Gooder, V. (June 2011). Nurses Perceptions of a (BCMA) Bar-coded Medication Administration System: A Case-Control Study. Online Journal of Nursing Informatics (OJNI), 15, (2), Available at http://ojni.org/issues/?p=703

ABSTRACT
This case-control study examined the perceived impact of bar-coded medication administration system (BCMA) on nurses ability to give medications, perceptions of medication errors, and nurses satisfaction with the medication administration process. The author developed a questionnaire based on Rogers diffusion of innovation theory and established content validity. The participants (BCMA n= 33; control n= 26) were given the questionnaire 1 month prior and 5 months following the implementation of a pilot unit. Participants in the experimental group indicated difficulty determining which medications had been given (p < .000). There was a decrease in the overall satisfaction with the medication process following implementation of the BCMA

system (p = .001). This study demonstrates that implementation of BCMA systems may have negative impact on nurses attitudes toward the medication administration process and may make the work processes more difficult. Effective implementation of BCMA systems requires an understanding of the impact of the system on nursing work processes. Key Words: BCMA, Medication administration, Barcode, Nurse satisfaction, Medication error

Introduction
In 1999, The Institute of Medicine (IOM) reported that nearly a million patients each year are injured in hospitals in the United States due to error. Medication errors occur more often than other categories of preventable errors (19%), and most medication errors occurred during medication administration (34%) where they were more likely to directly impact the patient and cause harm (Bates, et al., 1995). The IOM recommended systemic changes to hospital processes including medication administration (Kohn, Corrigan, & Donaldson, 1999). Bar coded medication administration systems (BCMAs) are one of the proposed solutions to medication administration errors and may reduce reported medication errors by as much as 86% (Baldwin, 2002; Bates, et al., 2001; Cipriano, 2002; Crane & Crane, 2006; Cummings, Bush, Smith, & Matuszewski, 2005; Johnson, Carlson, Tucker, & Willette, 2002; Paoletti, et al., 2007; Rivish & Modeda, 2010). Success of the BCMA system used in Veterans Administration hospitals in the 1990s prompted a Federal Drug Administration (FDA) mandate to barcode all prescription and most over-the-counter medications by mid 2006 (Traynor, 2004). Due to supportive efforts by the FDA and the Joint Commission (JC), an increase in the number of hospitals purchasing BCMA systems is expected. Bar-coded medication administration will probably be utilized in the majority of hospitals by 2024 (Roark, 2004). Despite optimism about the impact of BCMA systems on medication errors, there is some concern about the safety and effectiveness of these systems (Ash, Berg, & Coiera, 2004; McDonald, 2006; Sakowski, Newman, & Dozier, 2008). Ethnographic and observational studies have documented poor compliance with BCMA systems in several settings (Patterson, Cook, & Render, 2002; Patterson, Rogers, Chapman, & Render, 2006). Years after implementation of the first BCMA systems there is still widespread variation in how the systems are used (Carayon, et al., 2007). One reason for the variations is the fact that implementation of BCMA systems has an impact on the current work processes of nurses, who give the majority of medications in healthcare facilities. Using the BCMA system requires more time than other traditional methods of medication administration documentation such as a paper or computer-based medication administration record (Lawton & Shields, 2005). Implementation of the BCMA prevents workarounds (shortcuts) and other personalized methods used by nurses to administer medications (Englebright & Franklin, 2005). Problems with the BCMA technology may create additional frustration for the already busy bedside nurses. The author of this paper hypothesized that an increasing level of frustration felt by the nurse may lead to a decrease in the level of satisfaction with the medication administration process overall. Bar-coded medication administration systems are implemented to reduce medication administration errors, but it is unclear if the bedside nurses view the systems as effective in error prevention. Evidence of this is the development of system workarounds that bypass the intentional blocks to medication administration presented by the BCMA system. Workarounds are processes that bypass key safety features of the BCMA system, i.e. shortcuts. For

example, a patients armband is removed from the patients wrist and taped to the bed or doorjamb. Instead of scanning an armband on the patients wrist, the nurse scans the armband that is taped to the bed (Koppel, Wetterneck, Telles, & Karsh, 2008). This could result in a patient receiving the wrong medication, one that was intended for the previous occupant of the bed. By using workarounds, nurses bypass safety features of the system and therefore negate the overall purpose of the BCMA. A key to successful implementation and use of a BCMA system is to understand the level of satisfaction the nurses have with the system and how effective they view the system in preventing errors. The widespread use of BCMA systems in healthcare organizations may result in a significant decrease in medication administration errors (Coyle & Heinen, 2005; DeYoung, Vanderkook & Barletta, 2009; Sakowski et al., 2005). Despite this benefit, nurse dissatisfaction with the BCMA systems may impact overall compliance with the systems and decrease the overall effectiveness of the systems. Evaluation of nurse satisfaction with the BCMA system and the nurses perception of these systems to reduce medication errors may be important in the future design and implementation of these systems.

Description of the Problem

Literature: BCMA
A BCMA system utilizes bar-coded medication doses, patient identification bracelets, and nurse staff badges to facilitate the five rights (right patient, right medication, right dose, right time and right route) of medication administration. The BCMA system includes a server and a wireless handheld device (or a tethered device) coupled with software that interfaces with a hospitals information system. The system is often integrated with a patient unit-based automatic dispensing machine (ADM) and a pharmacy packaging and dispensing robot. Prior to administration of the medication, the nurse accesses the software in the handheld bar-coding scanner by scanning her/his badge and using the stylus to type in a secure password. Next, using the stylus, the nurse chooses a patient from a list on the handheld device. In the case of a new medication order, the nurse is prompted to confirm the new medication with the physicians written order. Once the nurse verifies the medication, the nurse scans the medication that is due to be administered. The software on the handheld device will list the patients medication administration record (MAR) and display alerts if the medication is incorrect in any of the following ways: medication, dose of medication, route of medication, or time for administration. Finally, the nurse scans the patients armband at the bedside to confirm the correct patient. The nurse has the option to attend to or override an alert and give the medication.

Literature: Nurse Satisfaction with BCMA

Research has been conducted on the satisfaction of nurses with BCMA technology, but the results have been inconclusive. One preliminary investigation regarding satisfaction of nurses who use BCMA systems have shown that nurses who previously documented medications using a paper-based system were satisfied with BCMA systems (Rough, Ludwig, & Wilson, 2003). The researchers measured satisfaction four months prior to implementation of the BCMA

and six months after and reported a 42% improvement in nurse satisfaction with medication administration and documentation after implementation of the BCMA system. Observations by Coyle and Heinen (2005) indicated nursing satisfaction with a BCMA system, but did not include specific timing of the observations or any attempts to quantify satisfaction. The Medication Administration System-Nurses Assessment of Satisfaction (MAS-NAS) Scale, developed by Hurley, et al. (2006), was used to compare satisfaction level of nursing staff pre- and post-BCMA implementation. This scale has demonstrated early reliability and validity (Hurley, et al., 2006). Hurley et al. (2007) measured the satisfaction of nurses in a large academic medical center 2-4 weeks prior to deployment and 4 months after implementation. The results indicated a statistically significant improvement in satisfaction following deployment. Of note, the results indicated that the nurses viewed the new system as time consuming, but safer. Selected interviews of participants corroborated the results of the study (Hurley, et al., 2007). Not all researchers found such promising results. In a longitudinal, descriptive study, Fowler, Sohler, and Zarillo (2009) used the MAS-NAS Scale to evaluate nursing satisfaction with a BCMA system and Category C medication error rates. Category C errors are medication errors that occur and reach the patient, but do not cause harm. They found no difference in satisfaction for the period prior to implementation compared to 6 and 9 months following implementation of BMCA. The researchers found that Category C errors increased following BCMA implementation, but hypothesized that this was due to increased reporting and surveillance. They did not find a decrease in errors related to medications given to the wrong patient (Fowler, Sohler, & Zarillo, 2009). Marini et al. (2010) developed a 33-question survey based on a model constructed to measure nursing acceptance of a BCMA system based on the image profile of the technology and to determine users attitudes. The image profile of the model included three aspects: system functionality, systems usability, and systems impact on nursing practice. The items on the questionnaire were measured using a five-point Likert scale. The questionnaire was sent by email to members of the CARING e-mail list, an informatics-focused listserve. The instrument demonstrated initial reliability and validity. Their results indicated that when nurses valued the safety features of the system, they viewed the system as more useable (Marini, Hasman, Huijer, & Dimassi, 2010). Despite the potential benefits of BCMA, nurses develop workaround strategies that reduce the overall effectiveness of these systems (Patterson, et al., 2002; Patterson, et al., 2006; Vogelsmeier, Halbesleben, & Scott-Cawiezell, 2008). Workarounds occur as a result of problems with technology, task, organization, patient issues, and the environment (Koppel, et al., 2008). Often workarounds are the result of several of these factors occurring simultaneously. These problems occur because the process has not been reengineered properly (Vogelsmeier, et al., 2008). A case study conducted by Bargren and Lu (2009) described system gaps in the BCMA that created a perceived need for nurses to use workarounds. Use of workarounds may encourage other unsafe practices (Halbesleben, Wakefield, & Wakefield, 2008). The use of workarounds indicates a lack of confidence in the system and may be an indication of decreased satisfaction. A positive correlation between nursing job satisfaction and job performance motivates nurse managers to investigate causes of low satisfaction among nurses. Frustration with the BCMA system may reduce nursing productivity and therefore negatively impact quality of patient care (Marini, et al., 2010). Poor work satisfaction is considered a leading cause of turnover resulting in reduced quality of patient care. The successful implementation of BCMA systems that maintain or improve nursing satisfaction require healthcare leaders to address issues that limit workarounds.

Although BCMA systems provide robust technology designed to reduce medication administration errors, there is concern that the benefits of these systems are not fully realized due to a failure to adequately integrate the systems into the current nursing work processes. The development of workarounds may be an indication that nurses feel that the system is not adequately supporting the medication administration process (Halbesleben, Wakefield, & Wakefield, 2008). The willingness of some nurses to bypass key safety features in the system indicates that they are not fully embracing the ability of the technology to significantly reduce medication administration error rates. Research regarding nursing satisfaction with BCMA systems is limited and conflicting. Understanding the impact of BCMA system implementation on nursing satisfaction with the medication administration process will assist with improving the development and implementation of these systems.

Conceptual Framework
Measuring satisfaction with BCMA technology requires an understanding of how technology is introduced into the healthcare work process. Rogers (2003) Diffusion of Innovation Theory (DoI) discusses the key attributes of innovations as being the result of five characteristics: 1) relative advantage is the innovation is better than what precedes it; 2) compatibility is the innovation is consistent with the goals of the current process; 3) complexity does the user consider the innovation complicated; 4) trialability how much can the new system be tested and experimented with; and 5) observability how easy is it for users to see the results of the innovation, in this case, a reduction in medication errors (Rogers, 2003). For this study the researcher evaluated the success of BCMA implementation based on the nurses views of the relative advantage, compatibility, complexity, and observability. These aspects of the system form a basis of nursing satisfaction with the system.

Purpose
This study investigated nurse satisfaction with the use of a BCMA system as compared to a previous non-barcoded medication administration system. The specific purpose of the research was to: 1) describe the perceived impact of BCMA implementation on nurses ability to give m edications; 2) investigate nurses perceptions of medication errors including near misses (nurse in act of giving medication and then stopped prior to giving it) before and after BCMA implementation; and 3) determine if the implementation of a BCMA system impacted the nurses overall satisfaction with the medication administration process.

Methods

Design, Setting, Participants, and Data Collection

A case-control pilot study was designed to test the nursing satisfaction with a medication administration process prior to and after implementation of a BCMA system. The study was approved by the Institutional Review Board of the participating hospital. Implied informed consent was assumed when subjects completed a questionnaire developed by the researcher. The questionnaire was administered to one unit that was implementing the BCMA system (experimental unit) and one unit that was not (control unit), before and 5 months after the implementation. The time of administration of the questionnaire after the implementation was determined by the date of the end of the pilot study on the BMCA unit.

Setting
The study hospital was a 280-bed acute care facility in the western United States. The hospital employed a comprehensive integrated computer-based documentation system. Prior to implementation of the BCMA, medications were ordered on a paper-based physician order sheet and scheduled on an electronic medication administration record (MAR) by pharmacists. Nurses were required to verify the scheduled medications against the paper-based physician orders at the beginning of each shift. The nurses documented medication administration in the electronic MAR. The policy in the facility was to document the medication prior to administration in order to take advantage of the functions of the electronic MAR including alerting.

Procedures
The nurses on the 28-bed medical unit piloted the BCMA system and were recruited as the experimental group. Nurses working on a 28-bed cardiovascular step-down unit were recruited as a control group. The control unit did not implement the BCMA system and served to demonstrate any impact of extraneous variables on nurse satisfaction with the existing medication process in the facility. The BCMA system implemented on the experimental unit was the AdminRX system (McKesson Automation, Inc. Pittsburgh, PA). A handheld device (Symbol, Motorola, Holtsville, New York) used wireless technology and a bar-coding scanner in the device and microprocessor with a 2.3 x 2.3-inch screen and graphical user interface with a touch pen. The system linked the hospital systems computerized medication ordering and BCMA system. The handheld device displayed the electronic medication administration record and allowed for verification of new orders entered by the pharmacists, double signatures on certain medications, documentation of medication administration, and alerting. After thorough testing of the BCMA-hospital information system interface by staff nurses from the experimental unit, pharmacists, vendor support, and the nurses on the BCMA unit were educated on the use of the BCMA systems using a train-the-trainer approach. Sample policy and procedure documents provided by the vendor were reviewed and modified by a small focus group including the nurse manager, a small group of staff nurses, and the nurse informatician. Workflow process changes were introduced to the nursing staff during the software training activities. During the implementation, vendor and hospital informatics personnel were available 24 hours per day, 7 days per week.

Instrument
The questionnaire consisted of eight questions. The first five questions measured the nurses perception of how easy the medication administration process was. The following is a complete list of questions included on the questionnaire:
1. 2. 3. 4. 5. 6. 7. 8. 9. Finding out which medications are due to be given soon is easy. It takes me too long to give medications. I always document my medications prior to administration. I can easily see what medications my patient has had. The new medications are put in the computer/handheld PC in a timely manner. (This question was designed to detect problems with the delivery and administration of new stat medications and if the nurse could see them on the MAR when ready to give them). I have had a medication error within the last month. I have had a medication error within the last month. I have nearly had a medication error in the last month. Overall, the medication administration process on my unit is:

The first seven questions used a Likert-type scale from 1 (Strongly Disagree) to 5 (Strongly Agree) with a choice of 3 (Dont Know) at the center of the scale. The eighth question on the survey asked the subjects to rate satisfaction with the current overall medication administration process on their unit on a Likert-type scale from 1 (Poor) to 5 (Excellent). The survey questions were developed by the researcher, and content validity was tested by colleagues in informatics. Reliability and validity statistics were not computed on the survey due to the low sample size (Feldt & Ankenmann, 1998). Surveys were coded to protect the identity of the subjects. Codes were consistent throughout the study to provide paired data for statistical analysis. The Statistical Package for Social Sciences (SPSS) (Version 16.0 for Windows, SPSS, Chicago, IL) was used to conduct statistical data analysis. Independent t-tests were used to analyze differences in the individual items on the questionnaire between the experimental and control groups before and after implementation of the BCMA system. Paired t-tests were used to analyze differences the values for individual items on the questionnaire between the experimental group before and after implementation of BCMA and the differences in values for the control group before and after implementation. Following a Bonferroni correction for multiple tests, the a level was set at .016. Chi-square tests were conducted on gender and licensure variables to determine differences between the experimental and control groups. Independent t-tests were used to compare age and years of experience between the experimental and control groups.

Results
Participants
Completion of the questionnaires was voluntary and the return rate of the questionnaires was approximately 42%. A total of 33 staff members returned surveys on the BCMA and 26 returned the surveys on the control unit (Table 1). Twenty-five surveys were collected prior to the implementation on the

experimental unit and 22 were collected on the control unit. Following the implementation of BCMA, 33 surveys were collected on the experimental unit and 14 on the control unit. There were 19 paired surveys on the experimental unit and 10 paired surveys on the control unit. The completed surveys had various amounts of missing data.

There were significant differences in the age and years of experience between the experimental and control groups (Table 1). The control group subjects were older and more experienced. This may have had an impact on differences in satisfaction between the two groups. There were significant amounts of missing data, particularly in the results of the control group surveys. This limits the ability to draw inferences about the demographic similarities or differences in the two groups.

Questionnaire
A comparison of satisfaction between the control and experimental units before BCMA implementation demonstrates differences in two areas. The control group was significantly less satisfied with the overall medication administration process, t(44) = 3.33, p = .002. The control group felt that the process of getting medications scheduled by the pharmacy in the computerized system was less timely on the control unit than on the experimental unit t(44) = 2.95,

p= .005. Both groups indicated that they agreed with the statement that it was easy to determine what medications were due, M = 4.5 (BCMA) and M = 4.09 (Control). Following the implementation, the experimental group had decreases in satisfaction with the medication administration process in three areas (Table 2). First, nurses satisfaction with their ability to determine which medication was due decreased with use of the BCMA. Second, the nurses indicated that it was more difficult to see what medications the patient had already had, t(43.4) = 4.05, p <.000. Finally, there was a decrease in satisfaction with the overall medication administration process for the BCMA group following implementation of the BCMA, t(52) = 3.54, p = .001. The paired t-tests verified these findings.

A comparison of the control group surveys, pre- and post-implementation of the BCMA system on the experimental unit did not yield any statistically significant differences for any of the satisfaction indices. The paired t-test verified these findings.

Discussion
This pilot study indicated that the nurses on the experimental unit perceived that there was a decreased ability to visually see the medications due, as well as medications previously given, on the handheld device following implementation of the BCMA system. Nurses indicated a decrease in the overall satisfaction with the medication process following implementation of the BCMA system. There were no differences in either group related to perceived medication errors or near misses. The use of BCMA systems is viewed as a promising technology to reduce medication errors in hospital settings, but implementation of these systems may be less than optimal if they have unintended outcomes on the medication administration process. To date, this is the first case control study evaluating the satisfaction of nurses following implementation of a new BCMA system. Anecdotal evidence suggests that the inability of nurses to view medications due and medications given previously was due to design of the software rather than the screen size on the handheld devices. This study also demonstrated an overall reduction in nurses satisfaction with the medication administration process when the BCMA system was implemented. The control group had no significant changes in responses following the study, lending confidence that the decrease in the satisfaction with the experimental group was due to the implementation of the BCMA system rather than other factors. These results differ from those of other researchers who indicate that nurses are satisfied with the systems (Hurley, et al., 2007; Coyle & Heinen, 2005; Rough, Ludwig, & Wilson, 2003). Future studies employing larger samples are recommended. Randomization of subjects may be difficult, but the continued use of carefully chosen control groups will provide higher levels of evidence for research in this area. This research provided information that may assist in the future development and implementation of systems that will maximize the benefits rather than introduce new error into an already problematic medication administration system. In this study, nurses were part of the implementation team, but including the nurses at the implementation phase may not be adequate. Research investigating the impact of including nurses in the initial design and development of BCMA systems will provide important answers that may guide future development in ways that maximizes the potential of this new technology. The implementation of new technologies into healthcare systems can be a complicated endeavor. Due to the significant investment of money required to purchase and implement these systems, discussion of negative outcomes is often not desired or encouraged. Implementation of new technologies requires an honest evaluation of the impact these new systems have on current practice in order to maximize the benefits these systems provide to quality and costeffective healthcare.

Caryon et al. (2007) concluded that changes in workflow must be assessed and workflow processes reengineered prior to implementation of these systems. Methods used to provide education and change processes can be enhanced to improve the overall satisfaction with these new technologies. Unless implementation staff and software developers acknowledge the impact these systems have on nurses and make adjustments to improve satisfaction, the intended improvements in care of our patients as a result of these new technological innovations may never be realized.

Limitations
There were several limitations to this study. Due to the nature of the BCMA pilot program, the number of subjects available for study was limited. Although surveys were given to all nurses on the units using their unit mailboxes, completion of the questionnaires was voluntary and the return rate of the questionnaires was low. There were no limitations on communications between the experimental or control groups, so cross contamination of the groups may have occurred. Demographic data on the returned surveys was incomplete. There was no follow up for nursing staff that did not complete their survey, and the sample size was too small to determine statistical reliability of the instrument. The survey was short in order to maximize the response rate by busy clinicians. The MAS-NAS Scale developed by Hurley and colleagues (2006) demonstrates reliability and validity but was unfamiliar to the researcher at the time of this study. Theories of diffusion of innovation set forth by Rogers (2003) indicate that technology is accepted and integrated into work processes in stages and there would therefore be differences in satisfaction depending upon when measurement took place. Waiting 6 months or longer to measure satisfaction post BCMA
may have yielded different results. Measurement of satisfaction and attitudes for this study could not be delayed since the pilot project was ended 5 months after implementation. Ultimately, the healthcare system studied in this research opted to develop a medication bar-coding system rather than to purchase.

Conclusion
Bar-coding medication administration may be a technology that will significantly reduce medication errors in hospitals and therefore greatly improve patient safety. However, this study demonstrates that BCMA systems may have a negative impact on nurses attitudes toward the medication administration process and may make the work processes more difficult. Therefore, introducing BCMA systems into patient care areas may have unintended consequences, such as workarounds, that may reduce the effectiveness of the system. So before any decisions are made regarding the overall effectiveness of BCMA, hospitals first need to determine whether the benefits are negated by nurses resistance to the change and how that resistance can be minimized.

BACKGROUND Serious medication errors are common in hospitals and often occur during order transcription or administration of medication. To help prevent such errors, technology has been developed to verify medications by incorporating bar-code verification technology within an electronic medicationadministration system (bar-code eMAR). Full Text of Background... METHODS We conducted a before-and-after, quasi-experimental study in an academic medical center that was implementing the bar-code eMAR. We assessed rates of errors in order transcription and medication administration on units before and after implementation of the bar-code eMAR. Errors that involved early or late administration of medications were classified as timing errors and all others as nontiming errors. Two clinicians reviewed the errors to determine their potential to harm patients and classified those that could be harmful as potential adverse drug events. Full Text of Methods... RESULTS We observed 14,041 medication administrations and reviewed 3082 order transcriptions. Observers noted 776 nontiming errors in medication administration on units that did not use the bar-code eMAR (an 11.5% error rate) versus 495 such errors on units that did use it (a 6.8% error rate) a 41.4% relative reduction in errors (P<0.001). The rate of potential adverse drug events (other than those associated with timing errors) fell from 3.1% without the use of the bar-code eMAR to 1.6% with its use, representing a 50.8% relative reduction (P<0.001). The rate of timing errors in medication administration fell by 27.3% (P<0.001), but the rate of potential adverse drug events associated with timing errors did not change significantly. Transcription errors occurred at a rate of 6.1% on units that did not use the bar-code eMAR but were completely eliminated on units that did use it. Full Text of Results... CONCLUSIONS Use of the bar-code eMAR substantially reduced the rate of errors in order transcription and in medication administration as well as potential adverse drug events, although it did not eliminate such errors. Our data show that the bar-code eMAR is an important intervention to improve medication safety. (ClinicalTrials.gov number,NCT00243373.)