- Email: [email protected]
Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations
Health Policy and Technology (]]]]) ], ]]]–]]]
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/hlpt
Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations Zachary Davis, Lara Khansan Department of Business Information Technology, Pamplin College of Business, Virginia Tech, 1007 Pamplin Hall (0235), Blacksburg, VA 24061, USA
KEYWORDS Electronic medical record (EMR); Epic; UVa; Usability; Patient safety; Cost of healthcare
Abstract Objectives: Electronic Medical Record (EMR) systems have become an integral part of patient care, in both inpatient and outpatient settings. The objective of this paper is to propose a set of recommendations on how the Epic EMR system can be used to improve patient care. To this end, we present findings on the use of the Epic EMR system in the University of Virginia (UVa)'s Health System. Target audience: Healthcare organizations implementing electronic medical record systems and health technology managers. Methods: Face-to-face interviews with 30 of UVa's hospital personnel and others in the Epic department at UVa. Results and conclusions: Three key areas are discussed to determine the feasibility of improvement including a decrease in medical errors and the resulting parallel improvement in patient safety, inter-disciplinary collaboration, and a decrease in the overall cost of healthcare. We identified many discrepancies between the Epic EMR system’s intended use, and the workaround system that clinicians have used to document patient care. In addition, we discuss a dichotomy in perspectives amongst the Health System and Technology Services department at UVa, and healthcare staff end users, with regard to the intended functionality and the usability of the Epic EMR system. In light of our findings, we provide a set of recommendations on how to decrease the gap between the intended and actual use of EMR systems, in general & 2015 Fellowship of Postgraduate Medicine. Published by Elsevier Ltd. All rights reserved.
Introduction n
Corresponding author. Tel.: +1 540 231 5003; fax: +1 540 231 3752. E-mail addresses: [email protected] (Z. Davis), [email protected] (L. Khansa).
With the passing into law of the Health Information Technology for Economic and Clinical Health (HITECH) Act, every
http://dx.doi.org/10.1016/j.hlpt.2015.10.004 2211-8837/& 2015 Fellowship of Postgraduate Medicine. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: Davis Z, Khansa L. Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations. Health Policy and Technology (2015), http://dx.doi.org/10.1016/j.hlpt.2015.10.004
2
Z. Davis, L. Khansa
private and public healthcare institution in the United States (US) is now required to use an Electronic Medical Record (EMR) system to store, integrate, and consolidate patients' protected health information, or PHI [1]. Specifically, Section 4101(b) of the HITECH Act states that beginning January 1, 2015, healthcare organizations that do not have an EMR system will be subject to a “negative adjustment” [1, p. 2]. The National Alliance for Health Information Technology (NAHIT) defines an EMR as “an electronic record of health-related information on an individual that can be created, gathered, managed, and consulted by authorized clinicians and staff within one healthcare organization”[2]. As a result of the recent legislation, nearly every hospital in the US now uses an EMR system for storing and processing patients' health information. Much of this effort has been accomplished through incentives provided by the government, but also through the movement of healthcare organizations toward more technologicallybased care [3]. The EMR system should support and enhance the work of healthcare providers working with these advancing medical technologies but such is not always the case. Among the challenges that hospitals face when implementing EMR systems is the need to improve usability, ensure interoperability, and keep costs affordable. This is often not feasible as healthcare institutions, already striving to afford the care they provide, often struggle to budget for the hefty costs of purchasing, implementing, and maintaining an EMR system [4]. Maintenance and education are often overlooked when considering the initial and ongoing costs of EMR implementation. If the ongoing costs are not budgeted for then follow up education and support is often limited, which, in turn, constrains the overall success of the EMR system in practice [5, 6]. As the government continues to push for the “Meaningful Use” of EMR systems nationwide, the need to train healthcare professionals and the costs required to do that will continue to grow [6]. “Meaningful Use” criteria are milestones that were put forth by the US government to incentivize healthcare institutions to adopt EMR systems and, more broadly, health information technology (HIT) [7]. In an effort to better explore the opportunities and challenges of EMR implementation as they relate to patient care, we look in this paper into the workings of the Epic EMR system in the University of Virginia (UVa)'s Health System. Specifically, we conducted interviews with employees of various roles who have used the Epic EMR system at UVa, and with employees at UVa's Epic department, with the goal of collecting information that would shed light into the actual practice, use, and implementation of EMR systems, particularly the Epic EMR system. Epic is known for its excellence in customer service and its ability to be used across healthcare settings (http://www.epic.com/softwareindex.php); however in practice some of this is not as apparent to the Epic's end users. In the remainder of the paper, we first review the extant literature and identify salient research streams. We then discuss UVa's setting and their Epic implementation, and present our data collection methods. Next, we present our results, followed by a discussion of our findings. In the last section of the paper, we provide a summary and a set of recommendations, and we subsequently discuss some of our study’s limitations and finally conclude.
Literature review A comprehensive literature review was performed using an EBSCO database for recent peer reviewed articles that were published after 2010. The following search qualifiers were used: “Electronic Medical Records,” “Implementation,” “Hospital,” and “United States.” This search originally returned 97 articles, but after removing duplicates, a total of 56 articles remained. The articles were reviewed and 35 were omitted because they were not relevant to the research topic. After carefully reading the articles further, we excluded another 8 papers for their limited scope and/or lack of applicability, leaving a total of 13 relevant articles. Review of these 13 articles revealed that providers are generally cautious about adopting an EMR system and often find that productivity decreases after implementation, at least in the short term; however financial incentives and potential penalties from the government have caused most healthcare institutions to acquire an EMR system despite potential setbacks [6–9]. We found that prior research has focused on three major themes: (1) the effect of an EMR system implementation on mitigating/exacerbating medical errors/patient safety [4–6,8,9,11–14]; (2) the effect an EMR system has on interdisciplinary collaboration and increased communication, particularly amongst, but also within, disciplines [4–6, 9,11,12,15,16]; and (3) the effect of implementing an EMR system in relation to overall healthcare costs [3–8,12,16]. Our findings from reviewing the literature are summarized in Table 1. While some articles reported that the implementation of an EMR system could result in an increase in medical errors if clinicians choose to circumvent system controls so as not to adjust their workflow [14], the majority of the reviewed articles found that, by decreasing medical errors, EMR systems have had a positive impact on patient safety and quality of care. In particular, Estrada and Dunn [11] reported that the use of an EMR system allowed nurses to better individualize patient treatment plans. Other articles found that the use of an EMR system improved clinicians' workflow, which in turn benefited patient safety and reduced the cost of providing care [4,6,9,15,16]. Similarly, some other articles reported that the use of an EMR system improved clinical outcomes, resulting in improved quality of care [4,6,12]. Further, many of the reviewed articles found that improved documentation, faster and more informed decision making, and improved medication safety are especially beneficial [4,5–8,11–13]. In particular, Hsieh [8] and Palvia et al. [6] found that EMR systems improved patient safety by safeguarding patients’ health information through more streamlined documentation. Similarly, Estrada and Dunn [11] and Shen et al. [4] found that the use of an EMR system improved the accuracy of documentation. Palvia et al [6] and Cook et al. [5] found that quality could be improved with the use of technology-driven clinical decision support systems and computerized physician order entry (CPOE). A similar intervention that has been found to improve patient safety is an automated electronic provider alert, which scans the medical record for specific identifiers and brings these to the attention of healthcare providers [13]. EMR systems were also found to improve medication safety in some way [4,5,12,13]. Cook et al. [5], for example, identified the EMR system as a tool that allowed
Please cite this article as: Davis Z, Khansa L. Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations. Health Policy and Technology (2015), http://dx.doi.org/10.1016/j.hlpt.2015.10.004
Advantages of EMR systems in prior literature.
Prior literature
Improvement summary
Adler-Milstein et al. Decreased cost [2] Cook et al. [4] Improved documentation, decreased infections, improved drug safety, increased collaboration/communication, decreased cost Estrada and Dunn [10] Improved collaboration/communication, improved documentation, improved patient education Ford et al. [7] Decreased cost Hsieh [8] Decreased cost, improved documentation Kim et al. [14] Improved workflow, improved collaboration/communication Palvia et al. [5] Decreased cost, improved collaboration/communication, improved clinical outcomes, improved workflow Reed et al. [11] Improved clinical outcomes, improved documentation, improved collaboration/communication, improved drug safety, decreased cost Schweiger et al. [12] Improved patient education, improved documentation, improved drug safety Ser et al. [13] Decreased safety Shen et al. [3] Does not exhibit cost savings, increased collaboration/communication, improved clinical outcomes, improved workflow, improved drug safety, improved documentation Sittig et al. [9] Improved workflow, improved collaboration/communication Smith et al. [15] Decreased cost, improved workflow, improved collaboration/communication
Impact errors
on
safety/quality/ Collaboration Overall costs
Decreased Improved
Improved
Improved
Improved
Decreased
Decreased Decreased
Improved Improved
Improved Improved
Decreased
Improved
Improved
Decreased
Improved Decreased (workarounds) Improved
Improved
Increased
Improved Improved
Decreased
Improved
Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations
Please cite this article as: Davis Z, Khansa L. Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations. Health Policy and Technology (2015), http://dx.doi.org/10.1016/j.hlpt.2015.10.004
Table 1
3
4
Z. Davis, L. Khansa
physicians to better select the appropriate medications for each patient. Overall, these articles reported that increasing the quality of care and patient safety through improved clinical decision making is a primary benefit of using EMR systems. Another stream of research studied how the use of an EMR system could affect collaboration and communication among healthcare providers, particularly those having different roles within the same healthcare institution [4–6,10– 12,15,16]. One such study found no improvement in communication among nurses as a result of using an EMR system as a care planning tool [11]. On the other hand, Shen et al. [4] and Sittig et al. [10] found that the implementation of an EMR system resulted in an increase in collaboration between the IT department and clinicians through improved workflow. Smith et al. [16] looked at the executive structure of organizations and found that if the CEO and CIO were peers and collaborated on the implementation of the EMR system, the implementation was more successful. Cook et al. [5] also found that the EMR system could be successfully utilized for increased collaboration between pharmacists and physicians in selecting the appropriate medications for patients through the ability of accessibility of charts for review. Similarly, Reed et al. [12] found that using an EMR system in the outpatient setting increased communication between providers and patients. Kim et al. [15] evaluated the use of an EMR system for computerized physician order entry in the ordering of add-on oratory lab tests and found that it improved communication among physicians and laboratory workers, which also improved their workflow [15]. Palvia et al. [6] also discussed the ability of clinical decision support systems within EMR systems to facilitate collaboration and improve workflow. Overall, if used effectively, EMR systems have been found to improve communication and collaboration among healthcare providers who are responsible for different roles within the same healthcare institution. A third important characteristic of EMR systems that has been studied extensively relates to cost savings that accompany EMR system implementations [3,5–7,9,12,16] or cost barriers that hinder successful EMR system implementation [4]. While cost barriers have been found to plague EMR system implementations especially in smaller healthcare facilities [4], cost savings have been widely reported. For example, Smith et al. [16] and Hsich [9] found that the use of a sophisticated EMR system enabled cost savings through increased employee productivity. Reed et al. [12] and Adler-Milstein et al. [3] found that the use of an EMR system assisted clinicians in selecting medical tests for patients, thus eliminating the overhead cost of over-testing. Palvia et al. [6] found that costs are increased when patient discharge is delayed, but cost savings can be realized by streamlining storing and dissemination of information. Reducing unnecessary use of medications through prescribing the appropriate type and only the needed amount of medications with the help of an EMR system was also found to generate cost savings [3,5]. EMR system implementation was also found to generate additional cost savings as a result of fewer rejected insurance claims [3]. There are various ways in which an EMR system implementation relates to added or reduced costs, but overall the extant literature suggests that EMR system use is associated with decreasing healthcare costs. As Ford et al. [8, p. 11] stated,
“As hospitals reengineer their work processes to accommodate the new systems, they may realize greater savings that translate into larger fund balances or profits.”
Research design Research setting UVa is an academic medical center that comprises a School of Medicine, a School of Nursing, a Health Sciences Library, the University Physicians Group, and the Health System. The Health System includes a 604-bed hospital, a Level 1 trauma center, specialty heart and cancer centers, multiple primary and specialty clinics, a transitional care hospital, and a variety of other in-patient and out-patient services (http:// www.virginia.edu/bov/). UVa is well known as a forerunner in the utilization of new technologies and advancements in medical care. They are also involved in various types of medical research and, frequently have both medical and nursing students working in many areas of the health system. Most healthcare departments in the UVa Health System, both in-patient and out-patient, use the Epic system, a widely-used EMR system by hospitals across the US [17]. The community hospital in which interviews were conducted is no exception, and Epic is the only charting system the hospital has used since its inception in 2011.
Methods The interview questions were initially developed with the help of the Director of Clinical Operations and a Nurse Educator at the healthcare institution. The interviews were conducted with healthcare employees of various roles at a Long Term Acute Care Hospital, including managers, hospital administration employees, and licensed and unlicensed staff. The Long Term Acute Care Hospital was determined to be ideal for this research as they have never used another charting system, having been in business for less than three years at the time of the interviews. To conduct the interviews one of the authors, accompanied by the Director of Clinical Operations, announced to all hospital staff at their morning meeting that interviewees are needed for a study on the evaluation of the usability of the EMR system in use at the hospital. Only two hospital staff members accepted to be interviewed at the time of the meeting. The interviewing author left contact information for those wishing to be interviewed subsequently but no additional staff members came forward. After obtaining and analyzing the results of the two initial pilot interviews, the author subsequently interviewed the CTO and the Director of Epic to clarify some points raised by the two staff members. It was at that time that the theme of the dichotomy of views emerged. Initial research and pretest interviews took place in the summer of 2014, followed in the Fall of 2014 by the actual interviews that we analyze and report in this paper. Semi-structured interviews lasted about 30 min each. The interviewees received no incentives. With the help of the Nurse Educator, additional staff members were subsequently recruited for more interviews. The interviewees were selected at random during both the
Please cite this article as: Davis Z, Khansa L. Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations. Health Policy and Technology (2015), http://dx.doi.org/10.1016/j.hlpt.2015.10.004
Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations day and night shifts with the help of a day-shift and nightshift nurse respectively. The author, accompanied by the nurses, conducted the interviews on a one-on-one basis. There were a total of 30 interviews conducted over a period of several days during July 2014. Because of their busy work schedule, the interviewees agreed to answer our questions while working, which resulted in a quasi-informal interview format. The interview method used was that of the General Interview Guide Approach discussed by Patton [18]. The interviews were conducted by one person to minimize interview style biases. The interviewees' responses were hand-written. At the completion of their interview, the interviewees were asked to double check their answers and confirm that they were properly quoted (via read-back). The interview questions and the position of those interviewed are given in Appendix A and Table 2 respectively. Table 2 categorizes the interviewees by their roles within the organization, and shows their distribution across the various roles. We had the opportunity to interview a member of UVa's executive leadership, several department directors and managers, nurse educators and members of the quality department, physicians and user staff including registered nurses (RNs), licensed practical nurses (LPNs), certified nursing assistants (CNAs), and registered respiratory therapists (RRTs). Our sample was diverse in that it included individuals with various tenure lengths and experience levels. Some of those interviewed started their careers using paper charting; other more novice personnel started their jobs using electronic charting. Table 3 shows the demographics of the respondents including age, gender, ethnicity, and educational level. After obtaining the results of the interviews we sorted the responses by question and then analyzed the responses for recurring themes.
Results Unique to UVa is the way they have chosen to tailor the Epic EMR system to suit their needs. However, as a result of modifying the EPIC system by more than 10%, some of the functionalities have changed, creating unintended complications. One such challenge emanates from the
Table 2 at UVa.
Distribution of interviewees and their roles
Interviewees' role at the healthcare organization
Number of interviews
Executive leadership Department director Managers Nurse educator/quality Physicians RN LPN RRT CNA HUC Total
1 3 2 4 2 9 1 2 4 2 30
Table 3
5
Demographics of interviewees.
Demographic Gender Male Female Age Over 50 35–50 20–35 Ethnicity Caucasian African American Education level Doctoral Masters Bachelors Associates Certificate High school
Number of interviewees
7 23 7 12 11 29 1 4 5 3 11 5 2
complexities of the Epic EMR system, and the extent to which it was customized to offer different views for different roles within the organization. While personalization has allowed hospital personnel to focus on the information they need to fulfill their various roles as healthcare professionals, it has created new barriers resulting from data inconsistency. In particular, the physician view shows patients' intake and output in a table that helps doctors uncover trends and make more informed decisions about their patients' plan of care. However, the nurse and nursing assistant views that include boxes to input this information do not all link to the table accessed by physicians. An example of an unsafe situation resulting from these differing views is related to the charting of a patient's intake and output. If a patient has a bowel movement, it is charted on the flowsheets in Epic by the nursing assistant, but this information is not viewable on the screen where physicians’ view intake and output. This can lead the physician to believe that the patient is constipated because it appears they have not had any bowel movement. As a result, the physician might order an enema simply because their view did not show the bowel movement that was charted. One nurse stated, “As the nurse you have to be aware of the problems [with the EMR system] because it is possible to have a patient having diarrhea and then a stool softener is ordered because the physician can’t tell that from the chart and sometimes the patient is not a good historian. The nurse has to be looking for these things to make sure this doesn’t cause harm to the patient but if you were just giving the medications as ordered it doesn’t work out.” Interviews with hospital administrators who were part of the implementation of the Epic EMR system revealed that they believe the tables do interface and that the charted bowel movement should in fact be viewable on the physician’s table [17]. This demonstrates a dichotomy between the understanding of administration and that of healthcare providers regarding the EMR system.
Please cite this article as: Davis Z, Khansa L. Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations. Health Policy and Technology (2015), http://dx.doi.org/10.1016/j.hlpt.2015.10.004
6
Z. Davis, L. Khansa
Additionally, inter-disciplinary care and collaboration is not fostered by the varying views of the different platforms. Similarly, various healthcare roles have different areas to chart the same patient activities leading to confusion about what care has been provided by other co-workers assuming the same role. For example, there is more than one area in which to document tracheal suctioning of a patient. One place to chart this activity is on a flowsheet commonly used by nursing staff. Tracheal suctioning is a shared task between nurses and respiratory therapists. There is a different place to chart tracheal suctioning in an area that is commonly used by respiratory therapists. If each discipline charts this activity on their commonly used flowsheet then the activity is only viewable if you are viewing that specific flowsheet. A respiratory therapist stated, “We have a lot of patients that need a lot of suctioning and I used to get really annoyed with the nursing staff because I thought they were leaving all the suctioning for the RTs. After talking to a nurse one day about her not suctioning a patient we finally found the different views in the record. I was not even able to see the place where she was charting suctioning but she was doing it and she was documenting it.” These contradictions are not only potential safety hazards to the patient, but they also lead to feelings of frustration and lack of collaboration between doctors and staff. Not having the right information as a result of being given access to a stale view of patient activity charts can potentially put everyone involved at risk. In discussing these potentially hazardous situations and staff concerns with the developers of the Epic EMR system, there was a belief that all staff had access to every area of the chart and that this difference in perspectives did not exist [17]. This also demonstrates the disparity between what the Epic department at UVa thinks should happen, and what hospital staff members are really experiencing. It is unclear whether these hazardous situations are the result of the staff’s lack of training in the Epic system, or if a true duplication of charting areas exists without a consolidated user interface. The goal of this paper it to get a better understanding of how a collective and comprehensive EMR system, such as the Epic EMR system, can be better tailored and utilized to generate a reduction in healthcare costs and medical errors, and an improvement in inter-disciplinary collaboration. It is important to note that the concerns discovered in this paper are far from unique to the UVa Health System, but many institutions shy away from such research. This is because so much time and money are invested in creating an EMR system that is intended to be safer for patients and easier to use for hospital staff that healthcare institutions are reticent to divulge the flaws of their EMR implementation to their constituents and the public.
Discussion Many of the interviewees felt that the EMR indeed reduced medical errors, but they were concerned that the automation of some tasks reduced deliberation and rational thinking. Others thought that some parts of the EMR system could be rendered even more automated, by including additional alerts and “hard stops”, i.e. tasks that a staff member cannot proceed before checking, such as taking a blood
pressure reading before giving a blood pressure medication [19]. This specific feature does exist, which is an excellent safeguard, but some staff and administrators would like to see more of this to increase patient safety. UVa's medical laboratory uses a different medical record system which creates a variety of delay and implementation problems. Lab results do populate into the Epic EMR system, but lab labels cannot be created in the Epic EMR system, making the process less than seamless and a potential source of confusion, error, and delay. A similar system exists with the Pyxis medication retrieval software, which does not fully interface with the Epic EMR system either. Steps are being taken by UVa to better integrate Pyxis in the Epic EMR system and to increase interoperability among various systems [17]. During our interviews, a nurse we have spoken to noted that, “Healthcare providers are torn on the use of these systems, often citing that it makes giving care more difficult, takes time away from the patient and introduces patient safety concerns when the system is complex and at other times noting it’s superiority over paper charting.” Additionally, there is a lack of ongoing training related to the Epic EMR system, and the training that clinicians receive at the time of hire is often all the live training they ever get. Online training is required at the time of each update but this does not allow users to ask questions or interact with Epic experts. This also contributes to staff not understanding the finer points or changes made to the EMR system. UVa has personnel who support the various departments that use the Epic EMR system, but clinicians often have little to no interaction with these support staff. One nurse stated, “On night shift we usually feel less supported and it really comes with our role but I don’t think that our need for support, especially with charting, is actually less.” Another finding is that healthcare providers tend to prefer the method they were originally trained on. This indicates that perhaps the group that did not use electronic charting initially may need additional training and support to fully adopt the EMR in their everyday practice. This type of training without ongoing support-style education is common among hospitals. Conducting a thorough review of the literature complemented our face to face interviews by revealing key themes, namely the reduction of medical errors/increased patient safety, improved interdisciplinary communication and collaboration, and a decrease in healthcare cost. These same themes emerged during our interviews and are therefore further drawn upon and examined in relation to their impact on the UVa Health System.
Reduction in medical errors and increased patient safety Medical errors are a widely acknowledged concern, although the depth and breadth of the problem is difficult to assess. UVa has begun a new initiative called ‘Be Safe’ that makes it easier for hospital personnel to report problems, and subsequently includes them in the laborious process of finding long term solutions to these problems. ‘Be Safe’ events are an opportunity for staff to document errors, near-misses, or processes needing improvement to increase patient safety. Hospital personnel are invited on a
Please cite this article as: Davis Z, Khansa L. Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations. Health Policy and Technology (2015), http://dx.doi.org/10.1016/j.hlpt.2015.10.004
Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations regular basis to participate in planning sessions to devise solutions [20]. The EMR has many opportunities to prevent errors by prompting clinicians to look at symptoms and results while also grouping them in meaningful ways. Currently Epic uses this ability to identify the symptoms of sepsis and prompts clinicians to implement the sepsis protocol. However this feature is currently not in wide use for other purposes although it has the potential to considerably improve patient safety by potentially decreasing medical errors. Various other methods such as pop-up notifications for critical lab values or vital signs have also been shown to improve patient safety. Many methods that would visually stimulate the attention of healthcare professionals are promising capabilities that have yet to be utilized. One interviewee stated, “Even with computers in the patient’s rooms you don’t want to stand there for the 30 min it will take you to chart your care so you wait and chart it after you have seen all four patients, which really isn’t a good practice as far as documenting but at least you saw all of your patients.” Far from being unique to UVa, staff noted this to be a charting habit at other institutions they had worked at due to the time it takes to document in an EMR system. Patient safety is every clinician’s first concern; however, EMR systems are not necessarily supportive of this goal and remain, similar to their predecessor, the paper chart, merely flat despite their potentially dynamic capabilities.
Improved inter-disciplinary collaboration Improved collaboration amongst personnel with different roles within the organization is a goal set by many institutions because inadequate communication has by far been identified as the major cause of medical errors [21]. UVa's transitional care hospital recently began using the ‘Care Plan’ component in Epic as a common platform where different personnel can work together on shared goals and collaborate to devise a common care plan for patients. Nurses, respiratory therapists, physical therapists, occupational therapists, and speech therapists can all access the same shared space and are authorized to contribute their ideas and assist one another in ensuring patient safety. Using existing collaborative tools in such a manner has the potential to expand collaboration among disciplines and improving patient outcomes. The notes section of the EMR system also allows for personnel with various roles to write diagnoses, observations, and reflections that are then viewable by personnel in other disciplines to further communication and improve patient care. For example, this allows the night-shift nurse to write a note that might be useful to the daytime on-call physician, with whom she might not get the opportunity to communicate otherwise. One physician noted, “I appreciate the ability to read a note written about the patient's status from the night before but this system existed with paper and we have moved away from it and decreased communication in many ways.” These small details that enhance communication can make great improvements in patient care and EMR systems are poised to enhance such asynchronous encounters. In the Epic EMR system many of these areas already exist but are underutilized by clinicians.
7
Decreased cost of healthcare UVa's status as a large academic medical center comes with an increased cost of providing care. Allowing such a large number of residents, new graduate physicians, fellows, and medical students to order and review tests for more accurate diagnosis comes with an increased cost. If a system was made available to employees that would assist them in finding solutions far more advanced than what their novice experience allows them to, they would able to perform their job much more efficiently with less risk of medical error and at a much lower cost. Achieving return on investment following the implementation of an EMR system requires much time and effort due to the lofty costs of implementation, training, and maintenance, to say the least. Theoretically speaking, EMR systems can improve billing ability and decrease the necessary number of staff, but in practice this is often not the case [16]. Interestingly, EMR systems have not yet been able to actually demonstrate their cost saving abilities [4]. The ability of EMR systems to analyze the data and aid healthcare providers in their diagnosis is not yet a reality at UVa, so the cost savings have yet to be realized. This functionality is marginally noted in the ability of the Epic EMR system to diagnose sepsis from its symptoms and prompt providers to follow a step-by-step protocol for early treatment and faster recovery. Nevertheless, this functionality is by far underutilized. In summary, examination of the Epic EMR system in use in UVa's Health System revealed many aspects of EMR system use that are often overlooked. While usability is often discussed as a necessary area of improvement, the need to improve communication among various healthcare entities within the same organization and ways that an EMR can help are often overlooked. The lack of usability of these additional features could be a trigger for medical errors rather than error prevention. EMR systems have the ability to prevent medical errors and improve patient care provided proper training is offered and additional features are embedded.
Evaluation summary, recommendations, limitations, and conclusions During our thorough evaluation, it became clear that the Epic EMR system is both a tool that can improve care and a barrier to care when its lack of usability consumes healthcare personnel's time and effort and keeps them away from their patients. Many changes are needed to increase the ability of the Epic EMR system to better serve healthcare personnel in their various roles. Some required system features are already in use in Epic's EMR system and simply need improvement; others still need to be added. Changing existing features or installing brand new ones are challenging as they are usually met with employee resistance and require both time and money. The EMR system should be able to be seamlessly integrated into the clinician's day, but in practice there tends to be clustered moments of patient care, followed by extended periods of computer charting time. Many clinicians spend more time feeding information into the system than they do treating their patients and interacting with them to answer their questions and
Please cite this article as: Davis Z, Khansa L. Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations. Health Policy and Technology (2015), http://dx.doi.org/10.1016/j.hlpt.2015.10.004
8
Z. Davis, L. Khansa
concerns. While this problem will not be resolved overnight, the ability exists to make changes to the current system to make it more helpful to healthcare personnel in various disciplines. EMR systems have the potential to dramatically impact patient care in ways that can be positive or negative and the reality and risks associated with that need to be addressed in all healthcare settings. To determine how this could be accomplished would require more research but this area has the ability to impact both healthcare employees and patients in a very real way. The evaluation of the Epic EMR system has highlighted many areas that could be changed in order to meet the goals of increasing patient safety/decreasing medical errors, improving inter-disciplinary collaboration, and decreasing costs. Recommendations for future modification to either the Epic EMR system or to how it is utilized include: complete integration of the Pyxis software to the Medication Administration Record (MAR) in the Epic EMR system, education and mandatory use of the Care Plan component by all disciplines, increased tools that prompt for “Best Practices” and/or for the clinician to examine the symptoms/results, new functionality to group symptoms and results to give suggestions for orders and/or diagnoses, ability to interface with the ‘Be Safe’ practices identified by UVa staff, creation of pop-ups to alert clinicians to new orders, changes in patient conditions, and critical lab results, and the creation of hard stops to force clinicians to address a safety concern before being allowed to progress. Such an improved system would also simplify/ remove differences in Epic EMR platforms, which can in turn improve communication among personnel in various healthcare roles, improve the system's ability to chart meaningful education plans in the employees' Care Plan section, and seamlessly integrate with the laboratory ordering, collecting, and result reporting systems. Further, clinicians need to regularly undergo additional training throughout their careers in order for the EMR system capabilities to be fully utilized. Limitations that exist in this research include the inability to interview sufficient staff to gain a statistically significant quantity leading to a lack of quantitative analysis, the breadth of the use of the EMR system does not lend itself to specific findings, duration of the research limited to a five week period, and the ever-changing nature of a large healthcare institution made it difficult to capture exact results across departments at any given moment. In conclusion, the UVa Health System has put great effort into implementing the Epic EMR system with many successes, but various unforeseen challenges. They remain committed to the improvement of patient care and the use of the EMR system in this endeavor. The dichotomy that exists between what healthcare workers experience and what administrators believe healthcare workers experience poses an additional challenge. While this paper was written with the intention of examining the Epic EMR system in use in the UVa Health System, our findings suggest that these challenges are likely being experienced at other healthcare institutions. EMR systems have become a necessary part of healthcare delivery. Clinicians use them to chart the care they give, to communicate with one another, and to simply document what needs to be recorded in case there is future need for such vital information. Due to the importance of
EMR systems, usability is key. To increase EMR system usability and improve the overall EMR functionality, major changes must be implemented. It is a fact that EMR systems will be a part of the future of healthcare, but their potential have yet to be realized.
Author statements The authors played no role in the implementation, execution, maintenance or support of the Epic EMR system. One author did use Epic in a clinical role at UVa in the Medical Laboratory department.
Funding The authors certify that no funding has been received to conduct this research.
Appendix A.
– Interview questions
1. How do healthcare providers view using Epic? 2. Describe Epic's initial creation in the UVa system and the current method for updating the system. 3. How safe do you think the information stored in Epic is? 4. How easy is it to use Epic to chart your care? 5. Explain the of the reports that come out of Epic? 6. Do you think the use of an EMR decreases medical errors? 7. What is the effect of the EMR on patient safety? 8. Do you know someone who has accessed information through the EMR that they should not have? 9. Have you used a paper-based system and if so how does it compare to the EMR system we use now? 10. Do you think EMRs promote a HIPPA violation?
References [1] Department of Health and Human Services – Centers for Medicare & Medicaid Services. Article MM8667. Available from: (〈https://www.cms.gov/Outreach-and-Education/MedicareLearning-Network-MLN/MLNMattersArticles/downloads/ MM8667.pdf〉 ); 2015 [accessed 10.09.15]. [2] National Alliance for Health Information Technology. Defining Key Health Information Technology Terms. Available from: (〈http://www.nacua.org/documents/HealthInfoTechTerms. pdf〉 ); 2008 [accessed 10.09.15]. [3] Adler-Milstein J, Salzberg C, Franz C, Orav J, Newhouse JP, Bates DW. Effect of electronic health records on health care costs: longitudinal comparative evidence from community practices. Ann Intern Med 2013;159(2):97–104. [4] Shen X, Dicker AP, Doyle L, Showalter TN, Harrison AS, DesHarnais SI. Pilot study of meaningful use of electronic health records in radiation oncology. J Oncol Pract 2012;8(4):219–23. [5] Cook PP, Rizzo S, Gooch M, Jordan M, Fang X, Hudson S. Sustained reduction in antimicrobial use and decrease in methicillin-resistant staphylococcus aureus and clostridium difficile infections following implementation of an electronic medical record at a tertiary-care teaching hospital. J Antimicrob Therapy 2010;66:205–90. [6] Palvia P, Lowe K, Nemati H, Jacks T. Information technology issues in healthcare: hospital CEO and CIO perspectives. Commun Assoc Inform Syst 2012;30(19):293–312.
Please cite this article as: Davis Z, Khansa L. Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations. Health Policy and Technology (2015), http://dx.doi.org/10.1016/j.hlpt.2015.10.004
Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations [7] Kruse CS, Bolton K, Freriks G. The effect of patient portals on quality outcomes and its implications to meaningful use: a systematic review. J Med Internet Res 2015;17(2):e44. [8] Ford EW, Menachemi N, Huerta TR, Yu F. Hospital IT adoption strategies associated with implementation stress: Implications for achieving meaningful use. J Healthcare Manag 2010;55(3):175–88. [9] Hsieh C. Electronic medical record system: current status and its use to support universal healthcare system. Commun IIMA 2010;10(3):79–97. [10] Sittig DF, Wright A, Meltzer S, Simonaitis L, Evans RS, Nichol WP, Ash JS, Middleton B. Comparison of clinical knowledge management capabilities of commercially-available and leading internally-developed electronic health records. Med Inform Decis Mak 2011;11(13):1–9. [11] Estrada NA, Dunn CR. Standardized nursing diagnoses in an electronic health record: nursing survey results. Int J Nurs Knowl 2012;23(2):86–95. [12] Reed M, Huang J, Graetz I, Brand R, Hsu J, Fireman B. Outpatient electronic health records and the clinical care and outcomes of patients with diabetes mellitus. Ann Intern Med 2012;157(7):482–9. [13] Schweiger B, Zeitler P, Eppley S, Swietlik M, Barker J. Using the electronic medical record to improve education in patients at risk for adrenal insufficiency. Int J Pediatr Endocrinol 2010:964525.
9
[14] Ser G, Robertson A, Sheikh A. A qualitative exploration of workarounds related to the implementation of national electronic health records in early adopter mental health hospitals. PLOS One 2014;9(1):e77669. Public Library of Science ONE. [15] Kim JY, Kamis IK, Singh B, Shalini B, Dixon RH, Dighe AS. Implementation of computerized add-on testing for hospitalized patients in a large academic medical center. Clin Chem Lab Med 2011;49(5):845–50. [16] Smith AL, Bradley RV, Bichescu BC, Tremblay MC. IT governance characteristics, electronic medical records sophistication, and financial performance in U.S. hospitals: an empirical investigation. Decis Sci 2013;44(3):483–516. [17] Hughes J. Personal interview. Stacey Hall, Charlottesville, VA: University of Virginia; 2014. [18] Patton MQ. Qualitative research and evaluation methods.3rd ed. California: Sage; 2002. [19] McDaniel M. Personal interview. Transitional care hospital. Charlottesville, VA: University of Virginia; 2014. [20] Webmaster, S. Who are UVA’s leaders, anyway? Available from: (〈http://med.virginia.edu/asp/2014/05/19/who-are-uvas-lea ders-anyway/〉); 2014 [accessed 10.09.15]. [21] Agency for Healthcare Research and Quality. AHRQ’s Patient Safety Initiative: Building Foundations, Reducing Risk. Available from: (〈http://www.ahrq.gov/research/findings/final-re ports/pscongrpt/index.html〉); 2003 [accessed 10.09.15].
Please cite this article as: Davis Z, Khansa L. Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations. Health Policy and Technology (2015), http://dx.doi.org/10.1016/j.hlpt.2015.10.004
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/hlpt
Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations Zachary Davis, Lara Khansan Department of Business Information Technology, Pamplin College of Business, Virginia Tech, 1007 Pamplin Hall (0235), Blacksburg, VA 24061, USA
KEYWORDS Electronic medical record (EMR); Epic; UVa; Usability; Patient safety; Cost of healthcare
Abstract Objectives: Electronic Medical Record (EMR) systems have become an integral part of patient care, in both inpatient and outpatient settings. The objective of this paper is to propose a set of recommendations on how the Epic EMR system can be used to improve patient care. To this end, we present findings on the use of the Epic EMR system in the University of Virginia (UVa)'s Health System. Target audience: Healthcare organizations implementing electronic medical record systems and health technology managers. Methods: Face-to-face interviews with 30 of UVa's hospital personnel and others in the Epic department at UVa. Results and conclusions: Three key areas are discussed to determine the feasibility of improvement including a decrease in medical errors and the resulting parallel improvement in patient safety, inter-disciplinary collaboration, and a decrease in the overall cost of healthcare. We identified many discrepancies between the Epic EMR system’s intended use, and the workaround system that clinicians have used to document patient care. In addition, we discuss a dichotomy in perspectives amongst the Health System and Technology Services department at UVa, and healthcare staff end users, with regard to the intended functionality and the usability of the Epic EMR system. In light of our findings, we provide a set of recommendations on how to decrease the gap between the intended and actual use of EMR systems, in general & 2015 Fellowship of Postgraduate Medicine. Published by Elsevier Ltd. All rights reserved.
Introduction n
Corresponding author. Tel.: +1 540 231 5003; fax: +1 540 231 3752. E-mail addresses: [email protected] (Z. Davis), [email protected] (L. Khansa).
With the passing into law of the Health Information Technology for Economic and Clinical Health (HITECH) Act, every
http://dx.doi.org/10.1016/j.hlpt.2015.10.004 2211-8837/& 2015 Fellowship of Postgraduate Medicine. Published by Elsevier Ltd. All rights reserved.
Please cite this article as: Davis Z, Khansa L. Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations. Health Policy and Technology (2015), http://dx.doi.org/10.1016/j.hlpt.2015.10.004
2
Z. Davis, L. Khansa
private and public healthcare institution in the United States (US) is now required to use an Electronic Medical Record (EMR) system to store, integrate, and consolidate patients' protected health information, or PHI [1]. Specifically, Section 4101(b) of the HITECH Act states that beginning January 1, 2015, healthcare organizations that do not have an EMR system will be subject to a “negative adjustment” [1, p. 2]. The National Alliance for Health Information Technology (NAHIT) defines an EMR as “an electronic record of health-related information on an individual that can be created, gathered, managed, and consulted by authorized clinicians and staff within one healthcare organization”[2]. As a result of the recent legislation, nearly every hospital in the US now uses an EMR system for storing and processing patients' health information. Much of this effort has been accomplished through incentives provided by the government, but also through the movement of healthcare organizations toward more technologicallybased care [3]. The EMR system should support and enhance the work of healthcare providers working with these advancing medical technologies but such is not always the case. Among the challenges that hospitals face when implementing EMR systems is the need to improve usability, ensure interoperability, and keep costs affordable. This is often not feasible as healthcare institutions, already striving to afford the care they provide, often struggle to budget for the hefty costs of purchasing, implementing, and maintaining an EMR system [4]. Maintenance and education are often overlooked when considering the initial and ongoing costs of EMR implementation. If the ongoing costs are not budgeted for then follow up education and support is often limited, which, in turn, constrains the overall success of the EMR system in practice [5, 6]. As the government continues to push for the “Meaningful Use” of EMR systems nationwide, the need to train healthcare professionals and the costs required to do that will continue to grow [6]. “Meaningful Use” criteria are milestones that were put forth by the US government to incentivize healthcare institutions to adopt EMR systems and, more broadly, health information technology (HIT) [7]. In an effort to better explore the opportunities and challenges of EMR implementation as they relate to patient care, we look in this paper into the workings of the Epic EMR system in the University of Virginia (UVa)'s Health System. Specifically, we conducted interviews with employees of various roles who have used the Epic EMR system at UVa, and with employees at UVa's Epic department, with the goal of collecting information that would shed light into the actual practice, use, and implementation of EMR systems, particularly the Epic EMR system. Epic is known for its excellence in customer service and its ability to be used across healthcare settings (http://www.epic.com/softwareindex.php); however in practice some of this is not as apparent to the Epic's end users. In the remainder of the paper, we first review the extant literature and identify salient research streams. We then discuss UVa's setting and their Epic implementation, and present our data collection methods. Next, we present our results, followed by a discussion of our findings. In the last section of the paper, we provide a summary and a set of recommendations, and we subsequently discuss some of our study’s limitations and finally conclude.
Literature review A comprehensive literature review was performed using an EBSCO database for recent peer reviewed articles that were published after 2010. The following search qualifiers were used: “Electronic Medical Records,” “Implementation,” “Hospital,” and “United States.” This search originally returned 97 articles, but after removing duplicates, a total of 56 articles remained. The articles were reviewed and 35 were omitted because they were not relevant to the research topic. After carefully reading the articles further, we excluded another 8 papers for their limited scope and/or lack of applicability, leaving a total of 13 relevant articles. Review of these 13 articles revealed that providers are generally cautious about adopting an EMR system and often find that productivity decreases after implementation, at least in the short term; however financial incentives and potential penalties from the government have caused most healthcare institutions to acquire an EMR system despite potential setbacks [6–9]. We found that prior research has focused on three major themes: (1) the effect of an EMR system implementation on mitigating/exacerbating medical errors/patient safety [4–6,8,9,11–14]; (2) the effect an EMR system has on interdisciplinary collaboration and increased communication, particularly amongst, but also within, disciplines [4–6, 9,11,12,15,16]; and (3) the effect of implementing an EMR system in relation to overall healthcare costs [3–8,12,16]. Our findings from reviewing the literature are summarized in Table 1. While some articles reported that the implementation of an EMR system could result in an increase in medical errors if clinicians choose to circumvent system controls so as not to adjust their workflow [14], the majority of the reviewed articles found that, by decreasing medical errors, EMR systems have had a positive impact on patient safety and quality of care. In particular, Estrada and Dunn [11] reported that the use of an EMR system allowed nurses to better individualize patient treatment plans. Other articles found that the use of an EMR system improved clinicians' workflow, which in turn benefited patient safety and reduced the cost of providing care [4,6,9,15,16]. Similarly, some other articles reported that the use of an EMR system improved clinical outcomes, resulting in improved quality of care [4,6,12]. Further, many of the reviewed articles found that improved documentation, faster and more informed decision making, and improved medication safety are especially beneficial [4,5–8,11–13]. In particular, Hsieh [8] and Palvia et al. [6] found that EMR systems improved patient safety by safeguarding patients’ health information through more streamlined documentation. Similarly, Estrada and Dunn [11] and Shen et al. [4] found that the use of an EMR system improved the accuracy of documentation. Palvia et al [6] and Cook et al. [5] found that quality could be improved with the use of technology-driven clinical decision support systems and computerized physician order entry (CPOE). A similar intervention that has been found to improve patient safety is an automated electronic provider alert, which scans the medical record for specific identifiers and brings these to the attention of healthcare providers [13]. EMR systems were also found to improve medication safety in some way [4,5,12,13]. Cook et al. [5], for example, identified the EMR system as a tool that allowed
Please cite this article as: Davis Z, Khansa L. Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations. Health Policy and Technology (2015), http://dx.doi.org/10.1016/j.hlpt.2015.10.004
Advantages of EMR systems in prior literature.
Prior literature
Improvement summary
Adler-Milstein et al. Decreased cost [2] Cook et al. [4] Improved documentation, decreased infections, improved drug safety, increased collaboration/communication, decreased cost Estrada and Dunn [10] Improved collaboration/communication, improved documentation, improved patient education Ford et al. [7] Decreased cost Hsieh [8] Decreased cost, improved documentation Kim et al. [14] Improved workflow, improved collaboration/communication Palvia et al. [5] Decreased cost, improved collaboration/communication, improved clinical outcomes, improved workflow Reed et al. [11] Improved clinical outcomes, improved documentation, improved collaboration/communication, improved drug safety, decreased cost Schweiger et al. [12] Improved patient education, improved documentation, improved drug safety Ser et al. [13] Decreased safety Shen et al. [3] Does not exhibit cost savings, increased collaboration/communication, improved clinical outcomes, improved workflow, improved drug safety, improved documentation Sittig et al. [9] Improved workflow, improved collaboration/communication Smith et al. [15] Decreased cost, improved workflow, improved collaboration/communication
Impact errors
on
safety/quality/ Collaboration Overall costs
Decreased Improved
Improved
Improved
Improved
Decreased
Decreased Decreased
Improved Improved
Improved Improved
Decreased
Improved
Improved
Decreased
Improved Decreased (workarounds) Improved
Improved
Increased
Improved Improved
Decreased
Improved
Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations
Please cite this article as: Davis Z, Khansa L. Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations. Health Policy and Technology (2015), http://dx.doi.org/10.1016/j.hlpt.2015.10.004
Table 1
3
4
Z. Davis, L. Khansa
physicians to better select the appropriate medications for each patient. Overall, these articles reported that increasing the quality of care and patient safety through improved clinical decision making is a primary benefit of using EMR systems. Another stream of research studied how the use of an EMR system could affect collaboration and communication among healthcare providers, particularly those having different roles within the same healthcare institution [4–6,10– 12,15,16]. One such study found no improvement in communication among nurses as a result of using an EMR system as a care planning tool [11]. On the other hand, Shen et al. [4] and Sittig et al. [10] found that the implementation of an EMR system resulted in an increase in collaboration between the IT department and clinicians through improved workflow. Smith et al. [16] looked at the executive structure of organizations and found that if the CEO and CIO were peers and collaborated on the implementation of the EMR system, the implementation was more successful. Cook et al. [5] also found that the EMR system could be successfully utilized for increased collaboration between pharmacists and physicians in selecting the appropriate medications for patients through the ability of accessibility of charts for review. Similarly, Reed et al. [12] found that using an EMR system in the outpatient setting increased communication between providers and patients. Kim et al. [15] evaluated the use of an EMR system for computerized physician order entry in the ordering of add-on oratory lab tests and found that it improved communication among physicians and laboratory workers, which also improved their workflow [15]. Palvia et al. [6] also discussed the ability of clinical decision support systems within EMR systems to facilitate collaboration and improve workflow. Overall, if used effectively, EMR systems have been found to improve communication and collaboration among healthcare providers who are responsible for different roles within the same healthcare institution. A third important characteristic of EMR systems that has been studied extensively relates to cost savings that accompany EMR system implementations [3,5–7,9,12,16] or cost barriers that hinder successful EMR system implementation [4]. While cost barriers have been found to plague EMR system implementations especially in smaller healthcare facilities [4], cost savings have been widely reported. For example, Smith et al. [16] and Hsich [9] found that the use of a sophisticated EMR system enabled cost savings through increased employee productivity. Reed et al. [12] and Adler-Milstein et al. [3] found that the use of an EMR system assisted clinicians in selecting medical tests for patients, thus eliminating the overhead cost of over-testing. Palvia et al. [6] found that costs are increased when patient discharge is delayed, but cost savings can be realized by streamlining storing and dissemination of information. Reducing unnecessary use of medications through prescribing the appropriate type and only the needed amount of medications with the help of an EMR system was also found to generate cost savings [3,5]. EMR system implementation was also found to generate additional cost savings as a result of fewer rejected insurance claims [3]. There are various ways in which an EMR system implementation relates to added or reduced costs, but overall the extant literature suggests that EMR system use is associated with decreasing healthcare costs. As Ford et al. [8, p. 11] stated,
“As hospitals reengineer their work processes to accommodate the new systems, they may realize greater savings that translate into larger fund balances or profits.”
Research design Research setting UVa is an academic medical center that comprises a School of Medicine, a School of Nursing, a Health Sciences Library, the University Physicians Group, and the Health System. The Health System includes a 604-bed hospital, a Level 1 trauma center, specialty heart and cancer centers, multiple primary and specialty clinics, a transitional care hospital, and a variety of other in-patient and out-patient services (http:// www.virginia.edu/bov/). UVa is well known as a forerunner in the utilization of new technologies and advancements in medical care. They are also involved in various types of medical research and, frequently have both medical and nursing students working in many areas of the health system. Most healthcare departments in the UVa Health System, both in-patient and out-patient, use the Epic system, a widely-used EMR system by hospitals across the US [17]. The community hospital in which interviews were conducted is no exception, and Epic is the only charting system the hospital has used since its inception in 2011.
Methods The interview questions were initially developed with the help of the Director of Clinical Operations and a Nurse Educator at the healthcare institution. The interviews were conducted with healthcare employees of various roles at a Long Term Acute Care Hospital, including managers, hospital administration employees, and licensed and unlicensed staff. The Long Term Acute Care Hospital was determined to be ideal for this research as they have never used another charting system, having been in business for less than three years at the time of the interviews. To conduct the interviews one of the authors, accompanied by the Director of Clinical Operations, announced to all hospital staff at their morning meeting that interviewees are needed for a study on the evaluation of the usability of the EMR system in use at the hospital. Only two hospital staff members accepted to be interviewed at the time of the meeting. The interviewing author left contact information for those wishing to be interviewed subsequently but no additional staff members came forward. After obtaining and analyzing the results of the two initial pilot interviews, the author subsequently interviewed the CTO and the Director of Epic to clarify some points raised by the two staff members. It was at that time that the theme of the dichotomy of views emerged. Initial research and pretest interviews took place in the summer of 2014, followed in the Fall of 2014 by the actual interviews that we analyze and report in this paper. Semi-structured interviews lasted about 30 min each. The interviewees received no incentives. With the help of the Nurse Educator, additional staff members were subsequently recruited for more interviews. The interviewees were selected at random during both the
Please cite this article as: Davis Z, Khansa L. Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations. Health Policy and Technology (2015), http://dx.doi.org/10.1016/j.hlpt.2015.10.004
Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations day and night shifts with the help of a day-shift and nightshift nurse respectively. The author, accompanied by the nurses, conducted the interviews on a one-on-one basis. There were a total of 30 interviews conducted over a period of several days during July 2014. Because of their busy work schedule, the interviewees agreed to answer our questions while working, which resulted in a quasi-informal interview format. The interview method used was that of the General Interview Guide Approach discussed by Patton [18]. The interviews were conducted by one person to minimize interview style biases. The interviewees' responses were hand-written. At the completion of their interview, the interviewees were asked to double check their answers and confirm that they were properly quoted (via read-back). The interview questions and the position of those interviewed are given in Appendix A and Table 2 respectively. Table 2 categorizes the interviewees by their roles within the organization, and shows their distribution across the various roles. We had the opportunity to interview a member of UVa's executive leadership, several department directors and managers, nurse educators and members of the quality department, physicians and user staff including registered nurses (RNs), licensed practical nurses (LPNs), certified nursing assistants (CNAs), and registered respiratory therapists (RRTs). Our sample was diverse in that it included individuals with various tenure lengths and experience levels. Some of those interviewed started their careers using paper charting; other more novice personnel started their jobs using electronic charting. Table 3 shows the demographics of the respondents including age, gender, ethnicity, and educational level. After obtaining the results of the interviews we sorted the responses by question and then analyzed the responses for recurring themes.
Results Unique to UVa is the way they have chosen to tailor the Epic EMR system to suit their needs. However, as a result of modifying the EPIC system by more than 10%, some of the functionalities have changed, creating unintended complications. One such challenge emanates from the
Table 2 at UVa.
Distribution of interviewees and their roles
Interviewees' role at the healthcare organization
Number of interviews
Executive leadership Department director Managers Nurse educator/quality Physicians RN LPN RRT CNA HUC Total
1 3 2 4 2 9 1 2 4 2 30
Table 3
5
Demographics of interviewees.
Demographic Gender Male Female Age Over 50 35–50 20–35 Ethnicity Caucasian African American Education level Doctoral Masters Bachelors Associates Certificate High school
Number of interviewees
7 23 7 12 11 29 1 4 5 3 11 5 2
complexities of the Epic EMR system, and the extent to which it was customized to offer different views for different roles within the organization. While personalization has allowed hospital personnel to focus on the information they need to fulfill their various roles as healthcare professionals, it has created new barriers resulting from data inconsistency. In particular, the physician view shows patients' intake and output in a table that helps doctors uncover trends and make more informed decisions about their patients' plan of care. However, the nurse and nursing assistant views that include boxes to input this information do not all link to the table accessed by physicians. An example of an unsafe situation resulting from these differing views is related to the charting of a patient's intake and output. If a patient has a bowel movement, it is charted on the flowsheets in Epic by the nursing assistant, but this information is not viewable on the screen where physicians’ view intake and output. This can lead the physician to believe that the patient is constipated because it appears they have not had any bowel movement. As a result, the physician might order an enema simply because their view did not show the bowel movement that was charted. One nurse stated, “As the nurse you have to be aware of the problems [with the EMR system] because it is possible to have a patient having diarrhea and then a stool softener is ordered because the physician can’t tell that from the chart and sometimes the patient is not a good historian. The nurse has to be looking for these things to make sure this doesn’t cause harm to the patient but if you were just giving the medications as ordered it doesn’t work out.” Interviews with hospital administrators who were part of the implementation of the Epic EMR system revealed that they believe the tables do interface and that the charted bowel movement should in fact be viewable on the physician’s table [17]. This demonstrates a dichotomy between the understanding of administration and that of healthcare providers regarding the EMR system.
Please cite this article as: Davis Z, Khansa L. Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations. Health Policy and Technology (2015), http://dx.doi.org/10.1016/j.hlpt.2015.10.004
6
Z. Davis, L. Khansa
Additionally, inter-disciplinary care and collaboration is not fostered by the varying views of the different platforms. Similarly, various healthcare roles have different areas to chart the same patient activities leading to confusion about what care has been provided by other co-workers assuming the same role. For example, there is more than one area in which to document tracheal suctioning of a patient. One place to chart this activity is on a flowsheet commonly used by nursing staff. Tracheal suctioning is a shared task between nurses and respiratory therapists. There is a different place to chart tracheal suctioning in an area that is commonly used by respiratory therapists. If each discipline charts this activity on their commonly used flowsheet then the activity is only viewable if you are viewing that specific flowsheet. A respiratory therapist stated, “We have a lot of patients that need a lot of suctioning and I used to get really annoyed with the nursing staff because I thought they were leaving all the suctioning for the RTs. After talking to a nurse one day about her not suctioning a patient we finally found the different views in the record. I was not even able to see the place where she was charting suctioning but she was doing it and she was documenting it.” These contradictions are not only potential safety hazards to the patient, but they also lead to feelings of frustration and lack of collaboration between doctors and staff. Not having the right information as a result of being given access to a stale view of patient activity charts can potentially put everyone involved at risk. In discussing these potentially hazardous situations and staff concerns with the developers of the Epic EMR system, there was a belief that all staff had access to every area of the chart and that this difference in perspectives did not exist [17]. This also demonstrates the disparity between what the Epic department at UVa thinks should happen, and what hospital staff members are really experiencing. It is unclear whether these hazardous situations are the result of the staff’s lack of training in the Epic system, or if a true duplication of charting areas exists without a consolidated user interface. The goal of this paper it to get a better understanding of how a collective and comprehensive EMR system, such as the Epic EMR system, can be better tailored and utilized to generate a reduction in healthcare costs and medical errors, and an improvement in inter-disciplinary collaboration. It is important to note that the concerns discovered in this paper are far from unique to the UVa Health System, but many institutions shy away from such research. This is because so much time and money are invested in creating an EMR system that is intended to be safer for patients and easier to use for hospital staff that healthcare institutions are reticent to divulge the flaws of their EMR implementation to their constituents and the public.
Discussion Many of the interviewees felt that the EMR indeed reduced medical errors, but they were concerned that the automation of some tasks reduced deliberation and rational thinking. Others thought that some parts of the EMR system could be rendered even more automated, by including additional alerts and “hard stops”, i.e. tasks that a staff member cannot proceed before checking, such as taking a blood
pressure reading before giving a blood pressure medication [19]. This specific feature does exist, which is an excellent safeguard, but some staff and administrators would like to see more of this to increase patient safety. UVa's medical laboratory uses a different medical record system which creates a variety of delay and implementation problems. Lab results do populate into the Epic EMR system, but lab labels cannot be created in the Epic EMR system, making the process less than seamless and a potential source of confusion, error, and delay. A similar system exists with the Pyxis medication retrieval software, which does not fully interface with the Epic EMR system either. Steps are being taken by UVa to better integrate Pyxis in the Epic EMR system and to increase interoperability among various systems [17]. During our interviews, a nurse we have spoken to noted that, “Healthcare providers are torn on the use of these systems, often citing that it makes giving care more difficult, takes time away from the patient and introduces patient safety concerns when the system is complex and at other times noting it’s superiority over paper charting.” Additionally, there is a lack of ongoing training related to the Epic EMR system, and the training that clinicians receive at the time of hire is often all the live training they ever get. Online training is required at the time of each update but this does not allow users to ask questions or interact with Epic experts. This also contributes to staff not understanding the finer points or changes made to the EMR system. UVa has personnel who support the various departments that use the Epic EMR system, but clinicians often have little to no interaction with these support staff. One nurse stated, “On night shift we usually feel less supported and it really comes with our role but I don’t think that our need for support, especially with charting, is actually less.” Another finding is that healthcare providers tend to prefer the method they were originally trained on. This indicates that perhaps the group that did not use electronic charting initially may need additional training and support to fully adopt the EMR in their everyday practice. This type of training without ongoing support-style education is common among hospitals. Conducting a thorough review of the literature complemented our face to face interviews by revealing key themes, namely the reduction of medical errors/increased patient safety, improved interdisciplinary communication and collaboration, and a decrease in healthcare cost. These same themes emerged during our interviews and are therefore further drawn upon and examined in relation to their impact on the UVa Health System.
Reduction in medical errors and increased patient safety Medical errors are a widely acknowledged concern, although the depth and breadth of the problem is difficult to assess. UVa has begun a new initiative called ‘Be Safe’ that makes it easier for hospital personnel to report problems, and subsequently includes them in the laborious process of finding long term solutions to these problems. ‘Be Safe’ events are an opportunity for staff to document errors, near-misses, or processes needing improvement to increase patient safety. Hospital personnel are invited on a
Please cite this article as: Davis Z, Khansa L. Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations. Health Policy and Technology (2015), http://dx.doi.org/10.1016/j.hlpt.2015.10.004
Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations regular basis to participate in planning sessions to devise solutions [20]. The EMR has many opportunities to prevent errors by prompting clinicians to look at symptoms and results while also grouping them in meaningful ways. Currently Epic uses this ability to identify the symptoms of sepsis and prompts clinicians to implement the sepsis protocol. However this feature is currently not in wide use for other purposes although it has the potential to considerably improve patient safety by potentially decreasing medical errors. Various other methods such as pop-up notifications for critical lab values or vital signs have also been shown to improve patient safety. Many methods that would visually stimulate the attention of healthcare professionals are promising capabilities that have yet to be utilized. One interviewee stated, “Even with computers in the patient’s rooms you don’t want to stand there for the 30 min it will take you to chart your care so you wait and chart it after you have seen all four patients, which really isn’t a good practice as far as documenting but at least you saw all of your patients.” Far from being unique to UVa, staff noted this to be a charting habit at other institutions they had worked at due to the time it takes to document in an EMR system. Patient safety is every clinician’s first concern; however, EMR systems are not necessarily supportive of this goal and remain, similar to their predecessor, the paper chart, merely flat despite their potentially dynamic capabilities.
Improved inter-disciplinary collaboration Improved collaboration amongst personnel with different roles within the organization is a goal set by many institutions because inadequate communication has by far been identified as the major cause of medical errors [21]. UVa's transitional care hospital recently began using the ‘Care Plan’ component in Epic as a common platform where different personnel can work together on shared goals and collaborate to devise a common care plan for patients. Nurses, respiratory therapists, physical therapists, occupational therapists, and speech therapists can all access the same shared space and are authorized to contribute their ideas and assist one another in ensuring patient safety. Using existing collaborative tools in such a manner has the potential to expand collaboration among disciplines and improving patient outcomes. The notes section of the EMR system also allows for personnel with various roles to write diagnoses, observations, and reflections that are then viewable by personnel in other disciplines to further communication and improve patient care. For example, this allows the night-shift nurse to write a note that might be useful to the daytime on-call physician, with whom she might not get the opportunity to communicate otherwise. One physician noted, “I appreciate the ability to read a note written about the patient's status from the night before but this system existed with paper and we have moved away from it and decreased communication in many ways.” These small details that enhance communication can make great improvements in patient care and EMR systems are poised to enhance such asynchronous encounters. In the Epic EMR system many of these areas already exist but are underutilized by clinicians.
7
Decreased cost of healthcare UVa's status as a large academic medical center comes with an increased cost of providing care. Allowing such a large number of residents, new graduate physicians, fellows, and medical students to order and review tests for more accurate diagnosis comes with an increased cost. If a system was made available to employees that would assist them in finding solutions far more advanced than what their novice experience allows them to, they would able to perform their job much more efficiently with less risk of medical error and at a much lower cost. Achieving return on investment following the implementation of an EMR system requires much time and effort due to the lofty costs of implementation, training, and maintenance, to say the least. Theoretically speaking, EMR systems can improve billing ability and decrease the necessary number of staff, but in practice this is often not the case [16]. Interestingly, EMR systems have not yet been able to actually demonstrate their cost saving abilities [4]. The ability of EMR systems to analyze the data and aid healthcare providers in their diagnosis is not yet a reality at UVa, so the cost savings have yet to be realized. This functionality is marginally noted in the ability of the Epic EMR system to diagnose sepsis from its symptoms and prompt providers to follow a step-by-step protocol for early treatment and faster recovery. Nevertheless, this functionality is by far underutilized. In summary, examination of the Epic EMR system in use in UVa's Health System revealed many aspects of EMR system use that are often overlooked. While usability is often discussed as a necessary area of improvement, the need to improve communication among various healthcare entities within the same organization and ways that an EMR can help are often overlooked. The lack of usability of these additional features could be a trigger for medical errors rather than error prevention. EMR systems have the ability to prevent medical errors and improve patient care provided proper training is offered and additional features are embedded.
Evaluation summary, recommendations, limitations, and conclusions During our thorough evaluation, it became clear that the Epic EMR system is both a tool that can improve care and a barrier to care when its lack of usability consumes healthcare personnel's time and effort and keeps them away from their patients. Many changes are needed to increase the ability of the Epic EMR system to better serve healthcare personnel in their various roles. Some required system features are already in use in Epic's EMR system and simply need improvement; others still need to be added. Changing existing features or installing brand new ones are challenging as they are usually met with employee resistance and require both time and money. The EMR system should be able to be seamlessly integrated into the clinician's day, but in practice there tends to be clustered moments of patient care, followed by extended periods of computer charting time. Many clinicians spend more time feeding information into the system than they do treating their patients and interacting with them to answer their questions and
Please cite this article as: Davis Z, Khansa L. Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations. Health Policy and Technology (2015), http://dx.doi.org/10.1016/j.hlpt.2015.10.004
8
Z. Davis, L. Khansa
concerns. While this problem will not be resolved overnight, the ability exists to make changes to the current system to make it more helpful to healthcare personnel in various disciplines. EMR systems have the potential to dramatically impact patient care in ways that can be positive or negative and the reality and risks associated with that need to be addressed in all healthcare settings. To determine how this could be accomplished would require more research but this area has the ability to impact both healthcare employees and patients in a very real way. The evaluation of the Epic EMR system has highlighted many areas that could be changed in order to meet the goals of increasing patient safety/decreasing medical errors, improving inter-disciplinary collaboration, and decreasing costs. Recommendations for future modification to either the Epic EMR system or to how it is utilized include: complete integration of the Pyxis software to the Medication Administration Record (MAR) in the Epic EMR system, education and mandatory use of the Care Plan component by all disciplines, increased tools that prompt for “Best Practices” and/or for the clinician to examine the symptoms/results, new functionality to group symptoms and results to give suggestions for orders and/or diagnoses, ability to interface with the ‘Be Safe’ practices identified by UVa staff, creation of pop-ups to alert clinicians to new orders, changes in patient conditions, and critical lab results, and the creation of hard stops to force clinicians to address a safety concern before being allowed to progress. Such an improved system would also simplify/ remove differences in Epic EMR platforms, which can in turn improve communication among personnel in various healthcare roles, improve the system's ability to chart meaningful education plans in the employees' Care Plan section, and seamlessly integrate with the laboratory ordering, collecting, and result reporting systems. Further, clinicians need to regularly undergo additional training throughout their careers in order for the EMR system capabilities to be fully utilized. Limitations that exist in this research include the inability to interview sufficient staff to gain a statistically significant quantity leading to a lack of quantitative analysis, the breadth of the use of the EMR system does not lend itself to specific findings, duration of the research limited to a five week period, and the ever-changing nature of a large healthcare institution made it difficult to capture exact results across departments at any given moment. In conclusion, the UVa Health System has put great effort into implementing the Epic EMR system with many successes, but various unforeseen challenges. They remain committed to the improvement of patient care and the use of the EMR system in this endeavor. The dichotomy that exists between what healthcare workers experience and what administrators believe healthcare workers experience poses an additional challenge. While this paper was written with the intention of examining the Epic EMR system in use in the UVa Health System, our findings suggest that these challenges are likely being experienced at other healthcare institutions. EMR systems have become a necessary part of healthcare delivery. Clinicians use them to chart the care they give, to communicate with one another, and to simply document what needs to be recorded in case there is future need for such vital information. Due to the importance of
EMR systems, usability is key. To increase EMR system usability and improve the overall EMR functionality, major changes must be implemented. It is a fact that EMR systems will be a part of the future of healthcare, but their potential have yet to be realized.
Author statements The authors played no role in the implementation, execution, maintenance or support of the Epic EMR system. One author did use Epic in a clinical role at UVa in the Medical Laboratory department.
Funding The authors certify that no funding has been received to conduct this research.
Appendix A.
– Interview questions
1. How do healthcare providers view using Epic? 2. Describe Epic's initial creation in the UVa system and the current method for updating the system. 3. How safe do you think the information stored in Epic is? 4. How easy is it to use Epic to chart your care? 5. Explain the of the reports that come out of Epic? 6. Do you think the use of an EMR decreases medical errors? 7. What is the effect of the EMR on patient safety? 8. Do you know someone who has accessed information through the EMR that they should not have? 9. Have you used a paper-based system and if so how does it compare to the EMR system we use now? 10. Do you think EMRs promote a HIPPA violation?
References [1] Department of Health and Human Services – Centers for Medicare & Medicaid Services. Article MM8667. Available from: (〈https://www.cms.gov/Outreach-and-Education/MedicareLearning-Network-MLN/MLNMattersArticles/downloads/ MM8667.pdf〉 ); 2015 [accessed 10.09.15]. [2] National Alliance for Health Information Technology. Defining Key Health Information Technology Terms. Available from: (〈http://www.nacua.org/documents/HealthInfoTechTerms. pdf〉 ); 2008 [accessed 10.09.15]. [3] Adler-Milstein J, Salzberg C, Franz C, Orav J, Newhouse JP, Bates DW. Effect of electronic health records on health care costs: longitudinal comparative evidence from community practices. Ann Intern Med 2013;159(2):97–104. [4] Shen X, Dicker AP, Doyle L, Showalter TN, Harrison AS, DesHarnais SI. Pilot study of meaningful use of electronic health records in radiation oncology. J Oncol Pract 2012;8(4):219–23. [5] Cook PP, Rizzo S, Gooch M, Jordan M, Fang X, Hudson S. Sustained reduction in antimicrobial use and decrease in methicillin-resistant staphylococcus aureus and clostridium difficile infections following implementation of an electronic medical record at a tertiary-care teaching hospital. J Antimicrob Therapy 2010;66:205–90. [6] Palvia P, Lowe K, Nemati H, Jacks T. Information technology issues in healthcare: hospital CEO and CIO perspectives. Commun Assoc Inform Syst 2012;30(19):293–312.
Please cite this article as: Davis Z, Khansa L. Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations. Health Policy and Technology (2015), http://dx.doi.org/10.1016/j.hlpt.2015.10.004
Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations [7] Kruse CS, Bolton K, Freriks G. The effect of patient portals on quality outcomes and its implications to meaningful use: a systematic review. J Med Internet Res 2015;17(2):e44. [8] Ford EW, Menachemi N, Huerta TR, Yu F. Hospital IT adoption strategies associated with implementation stress: Implications for achieving meaningful use. J Healthcare Manag 2010;55(3):175–88. [9] Hsieh C. Electronic medical record system: current status and its use to support universal healthcare system. Commun IIMA 2010;10(3):79–97. [10] Sittig DF, Wright A, Meltzer S, Simonaitis L, Evans RS, Nichol WP, Ash JS, Middleton B. Comparison of clinical knowledge management capabilities of commercially-available and leading internally-developed electronic health records. Med Inform Decis Mak 2011;11(13):1–9. [11] Estrada NA, Dunn CR. Standardized nursing diagnoses in an electronic health record: nursing survey results. Int J Nurs Knowl 2012;23(2):86–95. [12] Reed M, Huang J, Graetz I, Brand R, Hsu J, Fireman B. Outpatient electronic health records and the clinical care and outcomes of patients with diabetes mellitus. Ann Intern Med 2012;157(7):482–9. [13] Schweiger B, Zeitler P, Eppley S, Swietlik M, Barker J. Using the electronic medical record to improve education in patients at risk for adrenal insufficiency. Int J Pediatr Endocrinol 2010:964525.
9
[14] Ser G, Robertson A, Sheikh A. A qualitative exploration of workarounds related to the implementation of national electronic health records in early adopter mental health hospitals. PLOS One 2014;9(1):e77669. Public Library of Science ONE. [15] Kim JY, Kamis IK, Singh B, Shalini B, Dixon RH, Dighe AS. Implementation of computerized add-on testing for hospitalized patients in a large academic medical center. Clin Chem Lab Med 2011;49(5):845–50. [16] Smith AL, Bradley RV, Bichescu BC, Tremblay MC. IT governance characteristics, electronic medical records sophistication, and financial performance in U.S. hospitals: an empirical investigation. Decis Sci 2013;44(3):483–516. [17] Hughes J. Personal interview. Stacey Hall, Charlottesville, VA: University of Virginia; 2014. [18] Patton MQ. Qualitative research and evaluation methods.3rd ed. California: Sage; 2002. [19] McDaniel M. Personal interview. Transitional care hospital. Charlottesville, VA: University of Virginia; 2014. [20] Webmaster, S. Who are UVA’s leaders, anyway? Available from: (〈http://med.virginia.edu/asp/2014/05/19/who-are-uvas-lea ders-anyway/〉); 2014 [accessed 10.09.15]. [21] Agency for Healthcare Research and Quality. AHRQ’s Patient Safety Initiative: Building Foundations, Reducing Risk. Available from: (〈http://www.ahrq.gov/research/findings/final-re ports/pscongrpt/index.html〉); 2003 [accessed 10.09.15].
Please cite this article as: Davis Z, Khansa L. Evaluating the epic electronic medical record system: A dichotomy in perspectives and solution recommendations. Health Policy and Technology (2015), http://dx.doi.org/10.1016/j.hlpt.2015.10.004