| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Milieu in Dental School and Practice |
Key words: electronic patient record, dental, implementation, evaluation
Submitted for publication 10/20/08; accepted 02/11/09
 |
Abstract |
|---|
 TopAuthor information Abstract MethodsResultsDiscussionConclusionsReferences |
|---|
Dental schools have recently begun an ambitious goal of converting undergraduate, graduate, and faculty clinics from paper to electronic patient records (EPRs). The functional requirements of EPRs in dental schools are different from those in medicine and nursing.8 Patient clinics are typically housed in and owned by dental schools, and clinical experiences and productivity of students are tracked meticulously as part of an academically focused system. The users of these systems include students, staff, and faculty. In more advanced systems, even patients have access to parts of the record or the ability to make requests such as appointments. While students are typically the heaviest users, faculty members may be less technically savvy than their student counterparts and pose additional challenges to adoption and user training.
The implementation of an EPR requires a large capital expenditure and human resource effort that includes infrastructure, software, hardware, system configuration, documentation, user training and support, report writing, and maintenance of the new system.9 Other secondary costs include retraining and reassignment of existing staff, ongoing hardware and software costs, compliance, and addition of support staff. The costs of failure or suboptimal implementations are high and can result in less effective patient care, frustration among faculty, students, staff, and patients, and wasted resources.
There are significant opportunities to reduce the rate of failed implementations, costs, and time frame of implementation if experiences are shared among institutions.10 In this article, we seek to systematically review the implementation history of an EPR at our institution, the University of Texas Health Science Center at Houston Dental Branch (UTDB). At implementation, the EPR consisted of two primary applications: clinic management/patient record and digital imaging. We will describe the implementation process, identify characteristics of the implementation that were deemed favorable and those that posed challenges, report the results from end user surveys, and generate recommendations based on lessons we learned that will be useful for other institutions.
|
Methods |
|---|
|
TopAuthor informationAbstractMethodsResultsDiscussionConclusionsReferences |
|---|
Approximately three to four months after the EPR was fully implemented, we conducted in-depth interviews with four key personnel involved in the implementation. Interviews were conducted with each individual separately, and each lasted approximately sixty to ninety minutes. These individuals, who served on the implementation team and were selected based on their involvement with the project from its inception, were the executive associate dean (who served as project leader), the associate dean for patient care, the clinical IT manager, and the director of patient services.
A semistructured interview instrument was developed for this project using a systematic process of defining the goals and objectives of the research and developing open-ended questions to address them (see Figure 1
). Three interviewers (MW, DT, JL) participated in the sessions. The questions were used more as a guide than as a strict protocol during the interview process, so the interviewers and interviewees could fully explore issues that arose. All interviews were audiotaped after receiving consent from the interviewees. A commercial transcription company independently transcribed the audiotape. The transcribed interviews were checked for accuracy by the investigators. Two of the interviewers evaluated each transcript and provided a summary of the account with key points and emerging themes based on the research questions. A grounded theory approach was followed, in which the interviewers independently read the transcripts and identified key phrases, descriptions, and themes.11–13 The findings were then shared between the two interviewers, and any discrepancies or differences were resolved by consensus after review of the transcripts.
|
Two surveys were conducted with faculty, staff, and students. The first (pre-implementation) survey was conducted approximately four months before implementation of the EPR, and the second (post-implementation) was conducted approximately six months after full implementation. The surveys were administered online, using the Zoomerang zPro survey software system. Responses were anonymous, and no identifying information was collected. The pre-implementation survey contained twelve questions designed to gather users’ demographic data and attitudes regarding electronic patient records. An email announcement was sent in May 2006 to approximately 700 individuals (faculty, residents, staff, and students) of UTDB, and seventy-eight responses were received over a period of four weeks for a response rate of approximately 11 percent.
The post-implementation survey included the same twelve questions from the pre-implementation survey, plus four questions to determine the users’ experience with the new EPR; it also added free-response (open-ended) questions so users could describe their experience with and recommendations about the EPR. The post-implementation survey also included a question asking respondents if they had used the EPR. This survey was launched in early December 2007 and closed in late January; there were 138 responders, for a response rate of approximately 20 percent. Eight of these 138 responders indicated that they had not previously used the EPR and were subsequently excluded from the analysis. At least half of the 130 responders that were included in the final analysis provided at least one text comment.
|
Results |
|---|
|
TopAuthor informationAbstractMethodsResultsDiscussionConclusionsReferences |
|---|
shows the overall sequence of implementation with key milestones documented. The project was conducted with a particular focus on creating a participatory and ownership atmosphere in which key stakeholders and users were engaged throughout the project’s life cycle. All clinics, with the exception of the off-site Graduate Pediatric Dentistry Clinic, had the EPR system installed and utilized by 100 percent of the faculty, students, and staff by June 2007, approximately twenty-five months after the project was initiated.
|
, Milestone 10). Following the "go-live" for the pilot program and progress made in resolving technical problems, the digital imaging system was then rolled out in subsequent phases, first to the Faculty Practice Clinic (Table 1, Milestone 14), followed by the Radiology Clinic, and then to an offsite Oral and Maxillofacial Surgery Clinic. As the digital imaging system demonstrated stability and reliability, the implementation team deployed the application on an enterprise-wide basis, i.e., to the remaining undergraduate and graduate clinics (Table 1, Milestone 17). Once the digital imaging system was in full production, the implementation phase of the project was complete because all designated clinics had met the minimum functionality goals originally planned as shown in Table 1. Regarding the second unmet milestone, the implementation team decided to postpone the implementation of the EPR to the off-site Graduate Pediatric Dentistry Clinic until permanent leadership was in place (the department had both an interim chair and program director during the implementation process).
Perspectives of the Implementation Team
The following themes emerged as critical issues from the extensive interviews with the four members of the implementation team relating to implementation of the EPR.
Makeup of implementation team.
The participants noted the importance of having an implementation team with the appropriate mix of technical, clinical, and project management experience and who fully understood both the workings of a university dental clinic and the technical complexities of implementing a mission-critical information system. The makeup of the team was determined by the dean and key personnel from the school’s clinical, IT, and administrative departments, and members were selected to ensure appropriate representation from patient care, clinical education, operations, and technology. The role of a clinician champion, who served as the project leader for the entire implementation, was highlighted as especially important. The clinician champion was a faculty dentist experienced with the workings of the clinical environment and who had a strong grasp of both IT and project management. The clinician champion was trusted by the clinical faculty and staff and also garnered respect among the IT and operations teams. The ability to converse in both the clinical and IT languages proved to be a great facilitator.
Sharing responsibility with the project leader was an experienced clinical IT manager with many years of experience in computing in dentistry and medicine and an additional, newly hired IT staff member with a computer programming background. On the clinical side, the role of the associate dean for patient care was essential, as this person was intimately familiar with the actual workings of the clinic and possessed the administrative authority to effect change. Finally, the director of patient services, who manages patient records, patient accounts, and support staff (many of whose roles would change dramatically), provided an insightful perspective on clinical operations for the implementation team. The team met weekly to manage the project and also utilized email, web-based conference calls with the vendor, and ad hoc meetings. In addition to the core team, several working groups were formed to analyze the number and content of forms and map out clinical workflow in undergraduate and graduate programs.
Understanding clinical workflow.
Members of the implementation team anticipated that moving from a paper-based system to an EPR would dramatically change the clinical workflow of faculty, students, and staff. To address these changes, a workflow analysis task force (WATF) was created for the undergraduate programs, while representatives of each graduate program met individually with the project leader to describe and document its own workflow. All workflow groups were comprised of key clinic faculty members, staff, and students and were tasked to analyze and document the existing pre-EPR workflow in detail. They were also charged to determine and document the changes in workflow that would have to occur in order to best use the EPR—or, vice versa, how components of the EPR might be designed to support an existing process. Meeting weekly over the course of several months, the workflow groups carefully examined the work of the clinic and prepared numerous flowcharts. For example, the undergraduate workflow group produced an overview flowchart of the entire clinic operation and twelve detailed charts of specific suboperations.14 This workflow exercise and the resulting documents had immediate value: examination and documentation of the workflow brought to light at least forty instances of inefficiencies and inequities in the system that could be confronted and corrected before the new technology was installed. There was much discussion about the history and thinking behind many policies, and several were changed on the spot. New workflow documents ultimately served as references for configuration of the EPR application, training session content, "how to" training documents, and revision of job descriptions for appropriate staff. Finally, activities of the workflow groups were also credited with actively engaging key stakeholders, instilling support, and creating understanding and ownership of the implementation.
Despite the efforts of the workflow groups and the implementation team to extensively document work processes and understand the impact of the EPR, some unanticipated workflow anomalies were discovered. For example, the team uncovered a puzzling lack of financial documentation from one graduate clinic. Further investigation revealed that providers sometimes forgot to document a patient visit for cases in which patients had paid in advance; therefore, no charge was generated for subsequent visits. In the old paper-based workflow, an encounter form was generated regardless of whether a charge was needed, and this form served as a physical artifact to remind the provider to document the visit.
Other unexpected workflow issues included the following: 1) bottlenecks in radiology due to the conversion to digital imaging, as students, staff, and faculty members adapted to phosphor storage plates and sensors instead of film; 2) digital images not being available outside the clinic in which they were captured, such as endodontics, unless they had been saved to the main storage server; and 3) "cross-over" of procedure coding and fees between adult and pediatric dentistry clinics. In hindsight, implementation team members reported that more simulation and walk-throughs of the new clinical workflows might have been helpful in minimizing problems associated with transitioning to the EPR.
Support from centralized university IT.
In our environment, the dental school is responsible for the end user support of the functional aspects of its clinical IT applications. The university IT department manages all infrastructure (i.e., network, servers, storage, security, etc.), desktops, and peripherals. As a result, the implementation team included a number of university IT staff members who made major contributions to the project and worked closely with the dental school.
Before implementation, the dental school used approximately 350 leased desktop computers for faculty, staff, student, and administrative use, with comparable local and networked printers. The EPR implementation—essentially a move to an all-electronic environment—required approximately 350 additional workstations, primarily in the clinic arena, for a total of approximately 700 computers. Selection of the workstation type and associated platform was reported as the second most important decision behind selection of the clinical management/patient record and digital imaging software applications. Consideration of the best platform for the end user workstations received extensive discussion, and the university IT department led several fact-finding sessions, including site visits to various hardware and software vendors. Ultimately, a "thin client" solution using "blade" technology was selected for approximately 250 of the 350 added workstations. In this platform, a workstation port (thin client) with a monitor, keyboard, and mouse connects to a centrally located, remote PC (blade) system, which delivers the EPR, Internet, and other applications to the end user. The thin client/blade solution was selected for the following reasons: its small size, making it ideal for operatory use; the ease of making changes to the blades as a group; the absence of a local (hard) drive or CPU in the thin clients; and the ability to manage thin clients and blades remotely. The remote management was a significant feature for the university IT department as it did not need to increase its staffing support to the dental school as a result of the implementation. The implementation team, however, reported the following drawbacks with the thin client/blade platform: difficulty at times with thin clients connecting to blades; slower response than PCs; and inability to capture patient signatures directly on screen (using touch screen monitors).
Pilot-testing.
Implementation team members credited pilot-testing of the EPR as another key to success. Without pilot-testing, the school would have been forced to "live in two worlds"—both paper and electronic—until conversion to the new EPR was completed. The offsite GPR clinic was chosen as the pilot site for two primary reasons: implementing the new EPR there had no impact on the current dental school clinic information system (CIS) since the GPR clinic used a separate, stand-alone system; and the faculty and staff at the GPR were seen as early adopters eager to embrace the EPR and other new technology.
The pilot phase allowed the implementation team to identify problems early and refine the configuration and workflow issues associated with EPR implementation (Table 1, Milestone 10). As a key example, piloting provided firm evidence supporting the decision to postpone rollout of the digital imaging system. During the pilot, serious technical problems surfaced with the digital imaging software, primarily configuration, user-friendliness, and system stability. These problems resulted in delaying enterprise-wide implementation of the imaging system for almost a year. Other technical and application issues, such as thin client/blade configuration, server performance, electronic storage, system configuration, and end user training needs, were resolved during this period, both by the vendor and the implementation team.
Implementation approaches.
From the earliest stages of planning, the implementation team debated whether to take a phased implementation of the EPR versus a "big bang" or all-at-once launch. This question could be further broken down into two related decisions: 1) whether to deploy both the clinic management and patient record features of the EPR or to start with clinic management and then add electronic patient records later; and 2) whether to deploy digital imaging at the same time as the electronic record. While those involved in the discussion recognized the colossal challenge of a "big bang" approach, the majority of the team members felt that implementing in phases—and thus being forced to juggle a combination of electronic and paper-based patient records—was less desirable. With this in mind, the decision was made to implement all phases of the EPR simultaneously.
As implementation proceeded, implementation team members uniformly expressed their growing concerns with the documented technical problems in the digital imaging system in the pilot GPR clinic. Although postponing implementation of the digital imaging system was considered a setback, in hindsight it simplified the process of user adoption and training. During this time, implementation team members perceived a need for more time to test and configure the imaging system than originally planned, and part of the delay was due to waiting on the vendor for bug fixes and enhancements to the software. The team’s consensus was that a more mature digital imaging product would have improved the process.
Data conversion.
The implementation team initially hoped to convert and make available all the existing patient-related data from both the existing paper and electronic systems to the new EPR. However, it soon became clear it was no trivial task to convert ten years of billing and transaction data from the old CIS to a format compatible with the new system. One team member described it as "an absolute chore" that made for difficult days and nights. The implementation team reluctantly decided that much of the existing data (such as appointment history and treatment plans), while potentially valuable in the new EPR, would not be converted. Instead, students would be required to enter existing treatment plans and future appointments into the new EPR after the "go-live" date. This decision enhanced training as it allowed students and faculty to gain experience on the new system. This exercise was also an important prerequisite for simulated patient encounters during the go-live week as part of provider training.
The reality of limited time and resources forced other alterations in the original plan. For example, all paper-based records were to be scanned into the new EPR, but time, cost, additional electronic storage needs, and the cumbersome nature of the scanning process were too great. The implementation team ultimately devised a hybrid system in which existing paper charts would be available in hard copy for the first two years after implementation in order to maintain continuity of care. Also, any paper documents from the paper record would be scanned if requested by patients or providers, such as for medical consultations. Although this hybrid system was not an ideal solution for viewing historical patient data, team members suggested that this resulted in a reasonable compromise, given available resources.
Consensus on electronic forms for data collection.
A forms committee was created and charged with three tasks: reviewing all existing paper patient record forms, defining the required data elements (content), and determining the number of forms to be used with the new EPR. Although the EPR application had some standard pre-built data elements by default, it provided great flexibility and the ability to customize. The implementation team sought consensus from the forms committee on key forms, such as medical and dental health histories. Although consensus was eventually achieved, the end result was that many of the new electronic forms were lengthy and new forms had a learning curve of their own. The consensus medical history form contained approximately 150 questions. There was also debate on the benefits and costs of using "free response" answer fields versus input fields using structured terms. Both the forms committee and implementation team were successful in designing most forms with structured format responses, which greatly facilitates data retrieval and analysis. Many forms were structured to allow providers to make more detailed responses using clinical notes where appropriate.
Communications and user training.
From prior experience with technology innovations, the implementation team felt that building awareness and support for the project among key users was essential. A communications plan evolved with a focus on educating users about the benefits of an EPR and how it would change their workflow and patient interaction. The first step was to create a separate "identity" for the new system, one that would decidedly differentiate it from the old Clinic Information System, or "C-I-S." Therefore, "electronic patient record" or "E-P-R" was selected as the name to replace the CIS, which had been in place since 1996, and establish a new identity for the new system. At the same time, implementation team members were aware of the dangers in overstating or "hyping" benefits of electronic records. They were careful to moderate expectations by explaining that the transition from paper records to the EPR was likely to slow workflow at first and result in some initial inefficiencies. A common statement made by the project leader prior to go-live was "This will not make us faster, at least not at first."
Multiple techniques were used to communicate with stakeholders and end users, including ad hoc meetings and regular hands-on "evaluation" sessions (using the Simulation Clinic) during which faculty members, staff, and students were given time to interact with test versions of the software, ask questions, and offer suggestions regarding configuration and workflow. The implementation team also developed a project website that served as a repository of information about the project and the main source of training material, such as FAQs, training videos, and how-to guides. Finally, the project leader sent out periodic email newsletters, dubbed "EPR Instant Messages," to faculty, students, and staff, which highlighted behavioral issues such as how electronic records might impact the patient-provider relationship. The objective of the pre-implementation phase was that as many users as possible would have had hands-on experience with the new system prior to formal training at go-live.
The newly constructed Simulation Clinic played a critical role in training large groups of users due to its uniqueness as an actual clinic with chairside network computers.15 In this facility, users were provided frequent opportunities to familiarize themselves with the EPR and receive training on the system. The training, conducted by the project leader, was provided to the faculty, students, and staff and tailored to learner type. For example, student training sessions were conducted in large groups with upbeat background music and a training curriculum that required students to work together, while faculty training sessions were more individualized and in a quieter atmosphere. For all groups, the core training was performed on a training version of the EPR and designed to simulate the tasks and security each user group would have in the production system. The training sessions were perceived as effective, and it was noted that students rapidly learned how to use the system. In fact, it appeared that students were so proficient with the system early on that they graciously assisted those faculty members who needed a little help during the early days soon after go-live.
To complete training for undergraduate dental and dental hygiene students, direct patient care was simulated in clinic immediately before go-live with students pairing up as provider/patient and, led by faculty group leaders, performing scripted tasks in the production EPR. These tasks included scheduling appointments, creating and changing treatment plans, generating informed consents, and obtaining faculty approvals.
User Perspectives Before and After Implementation
To assess the level of user acceptance and awareness, two surveys were conducted among faculty, students, and staff (pre- and post-implementation). Demographic results are shown in Table 2, and comparative results for the questions are shown in Table 3. Although there were differences in the sample size between the pre- and post-implementation surveys, there were no significant differences between the types of responders (percentages of faculty, staff, etc.) who replied to the two surveys (chi-square=4.5, df=5, p=0.48); similarly, there were no significant differences in the gender of respondents (chi-square=0.210, df=1, p=0.65). However, direct comparisons between the pre- and post-implementation surveys need to be interpreted with caution as the respondents were not matched and the post-implementation survey had a higher response rate.
|
|
After implementation, users seemed to have confirmed their opinions regarding the impact of the EPR on their work in the clinic. About the same percentage thought that an EPR improved their efficiency, but a significant percentage changed their opinion from neutral to negative regarding efficiency. Similarly, fewer were neutral and more agreed that the EPR required more time to complete than a comparable paper record. Despite these somewhat negative results, an overwhelming percentage of users (91 percent) would recommend an EPR to a dentist starting a new practice, and over two-thirds thought prospective dental students should expect a school to use an EPR.
In addition to verifying if respondents had used the EPR, three questions were added in the post-implementation survey concerning users’ likes and dislikes of the new system and suggestions for improvement. In a departure from the previous survey, they were also offered an opportunity to enter free text, under "other—please specify."
When asked "What do you like about the EPR?," users seem to have been satisfied with the options offered by the survey team, as only eight (6 percent) of the 130 respondents entered comments explaining their "likes" (see Table 4). The eight respondents who positively commented mentioned presentations of the treatment plan, better system security, ease of following past treatment, better tracking of data, improvement in patient interaction (two), less paper, and excellent assignment of procedures to students.
|
) that indicated dislikes for the system and the sixty-eight suggestions for improvement were grouped into the following six categories:
|
Hardware (54 percent) and usability (28 percent) issues were the most frequently mentioned dislikes about the EPR. Users provided suggestions for improving usability (47 percent), hardware (31 percent), and digital imaging (28 percent). Similarities in users’ dislikes and suggestions indicated that improvements were needed in the responsiveness of some of the hardware and difficulty in using the digital imaging system.
|
Discussion |
|---|
|
TopAuthor informationAbstractMethodsResultsDiscussionConclusionsReferences |
|---|
|
Conclusions |
|---|
|
TopAuthor informationAbstractMethodsResultsDiscussionConclusionsReferences |
|---|
In this study, we primarily used interviews and surveys, which provided a wealth of information to help document the implementation process and to explore themes. However, our analysis could have been improved if we had used additional qualitative assessment methods in order to improve validity.25 For example, focus groups and "shadowing," ideally by trained and/or independent facilitators or observers, are excellent sources of additional valuable data. We also had a low response rate for our surveys. Responses may have increased with more frequent reminders or lengthening the period in which the survey was open.
In summary, we identified eight critical issues as important contributors of implementing an EPR. Although there are likely other ways to implement information systems and our findings are local in nature, the general principles and considerations presented may be useful for other dental schools as they too embark on their journey for implementing complex clinical information systems.
|
Acknowledgments |
|---|
|
Author Information |
|---|
|
TopAuthor informationAbstractMethodsResultsDiscussionConclusionsReferences |
|---|
|
REFERENCES |
|---|
|
TopAuthor informationAbstractMethodsResultsDiscussionConclusionsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Glick, D. Taylor, J. A. Valenza, and M. F. Walji Assessing the Content, Presentation, and Readability of Dental Informed Consents J Dent Educ., August 1, 2010; 74(8): 849 - 861. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. K. Hill, D. C. L. Stewart, and J. S. Ash Health Information Technology Systems Profoundly Impact Users: A Case Study in a Dental School J Dent Educ., April 1, 2010; 74(4): 434 - 445. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
