How to select and implement an electronic health record in a neurology practice
Summary
The purchase, implementation, and maintenance of an electronic health record (EHR) are among the most significant financial investments a practice will make. A practice's choice of EHR will have long-term and wide-ranging implications for how that practice operates. A successful EHR implementation may potentially result in increased efficiency, improved quality of patient care, and a possibly more successful practice. Extensive research and thoughtful planning, done with the involvement of all stakeholders, the commitment of adequate time, staff, and financial resources to the process, and sufficient training will increase the chances for a successful EHR implementation.
Health care remains the only major industry that has not successfully transitioned to electronic records. The complexity of even routine health care processes, along with financial, technical, and legal factors, are barriers to implementation of electronic health records (EHRs).1,2 Recent government incentive programs, most notably the Health Information Technology for Economic and Clinical Health provision of the American Reinvestment and Recovery Act, have attempted to reduce these barriers by providing payments for the meaningful use of an EHR to eligible Medicare and Medicaid providers.3 Payment adjustments, or penalties, will begin in 2015 for eligible providers who have not yet attested to meaningful use.
As of 2008, few neurologists were using EHRs to their fullest potential.4 The purchase, implementation, and maintenance of an EHR are among the most significant financial investments a practice will make, and the choice will have long-term and wide-ranging implications for how that practice operates. An intensive vetting process by multiple practice stakeholders is required before any purchase is made. A successful EHR implementation may result in increased efficiency, financial gains, return on investment, improved quality of patient care, and overall a potentially more successful practice.5 An unsuccessful implementation will result in the opposite, and will come with a high price.6 This article is intended to guide neurologists toward an effective, robust implementation of an EHR that is well-suited to their practice type.
Selecting an EHRInitial steps
The first step in choosing and implementing an EHR is to complete a Health Information Technology readiness assessment; free tools are available online.7,8 This step will assess financial and operational readiness and practice willingness to implement an EHR. If both clinicians and staff are not committed to the process, implementation will fail.6 Clinicians and staff may be resistant for a variety of reasons, including fears over the EHR adversely impacting the patient–physician relationship and concern over their ability to master new technology. This resistance may be a considerable impediment to a successful EHR implementation. In addition to ensuring practice willingness upfront, a strong physician champion, thorough validation, and adequate training are necessary to overcome it.
Financially, both the initial cost of implementation and the total cost of ownership of an EHR should be considered. The average initial cost of adoption of a complete EHR system is $25,000 to $54,000 per clinician.9 Maintenance costs may range from $3,000 to over $20,000 per clinician per year.9 There is significant variability depending on the hardware and software options, implementation model, and service plan selected. The cost of training and what exactly is covered by this cost (number of trainers for classroom sessions, online modules) may vary greatly between vendors. Other “hidden” costs of EHR adoption include the loss of productivity and resultant decreased revenue that often occurs during the implementation period.10 There may be costs associated with the long-term storage of existing medical records. Costs may be offset by incentive payments and by eventual improved efficiency and decreased operating costs (such as need for less staff) that will result from a successful EHR implementation; the latter should be vetted through an online EHR financial assessment tool.11 If practice transitions to employed or affiliated models are imminent, consider adopting the same EHR; approach the group to determine if there is a monetary subsidy for choosing the same EHR. If retirement looms, not converting to an EHR is a legitimate option.
EHRs may be purchased as complete systems or as modules. Complete systems automate all aspects of the practice, from patient scheduling and registration, to clinical documentation and order entry, through charge capture and billing. If the practice already possesses electronic systems, buying a complete EHR may not be needed; a modular approach can be adopted, which can reduce costs. However, back-end integration must be fully assessed before proceeding with modular solutions. A modular approach may require a longer implementation schedule. This may increase the period of reduced productivity and efficiency. A complete and official list of certified complete and modular EHR products is available online.12
There are 3 major EHR hardware options: the self-hosted model, the application service provider model, and Software as a Service model (SaaS). Each has significant advantages and disadvantages, including considerable differences in cost. The first 2 options are server-based, and are often utilized by larger practices with dedicated information technology (IT) staff. The SaaS model, in which the practice subscribes to Internet-based software, is the least expensive and is a practical option for small practices.
Choosing an EHR
The authors cannot recommend a specific EHR for the readers' practices. One size does not fit all. Web-based resources,8,13 including those available on the American Academy of Neurology (AAN) Web site,14 and other AAN resources, such as courses on this subject at the annual meeting and EHR-specific user groups, are helpful in guiding a practice in this choice. A vendor questionnaire or request for proposal, which asks the vendor to explain how their product will address the practice's specific needs and goals of EHR adoption, should be issued to a practice's list of top potential vendors.15
The figure describes the checklist a practice should employ in choosing an EHR. Many vendors provide access to a model system so practice members may try out the system and offer their assessment.10 Representation from across the practice should try the model system and provide feedback. This will ensure that the EHR under evaluation meets the needs of all roles in the practice and will improve buy-in once a final decision is made.
While many EHRs allow for customization, the ease with which this may be accomplished by the end user varies greatly. The ability to easily create custom content and tools (“the build”), including documentation templates or order preference lists, offers a significant advantage over a vendor-dependent process. Certain components such as patient portals or sophisticated reporting tools may require extra cost; however, meaningful use eligibility necessitates these functionalities. The integration of laboratory and radiology data is a requirement to achieve stage 2 of the Meaningful Use program. Neurologists must deal with the additional challenges of integrating other ancillary data, such as EMG and EEG results. Interfaces to these systems are not standardized and therefore will likely add considerable additional costs. Extensive planning must be undertaken with the vendor in order to address the complexities of the possible interfaces. More information on the Meaningful Use program and its requirements can be found online.16
The physician champion role is critical to the success of implementation. The physician champion provides high-level direction and oversight, ensuring the ultimate goals of the project stay in focus. It is the responsibility of this physician to communicate clinical priorities to the implementation team and realistic expectations to the clinicians, facilitate compromise when necessary, and to accomplish this in a nonconfrontational way that engenders mutual respect and willingness to cooperate. The physician in this role should be clinically respected and liked, have some understanding of technology, be passionate about the cause, and be able to resolve conflict. This will greatly increase buy-in from members of the practice.17
The project manager is responsible for assigning roles and responsibilities to other team members, developing and maintaining timelines, tracking the progress of each aspect of the implementation, and managing routine issues that arise. The physician champion may also be the project manager, depending on practice size, though this time commitment will likely reduce clinical productivity.
Workflows are the interaction of processes through which a clinic or hospital provides care to patients.18 A typical patient visit is comprised of numerous workflows, from how the patient schedules an appointment to how he or she checks out when this appointment is completed. These must be mapped out step by step, including all possible variations, and analyzed for efficiency. Vendors have tools to assist this process, but they cannot complete these steps. All paper forms should be critically evaluated. Only those that cannot be reproduced electronically (external testing forms, patient benefit programs) should be kept for routine use.
Construction, hardware, and infrastructure needs should be simultaneously assessed while workflows analysis is occurring. Laptops may provide more flexibility in examination rooms than desktop computers. If vendor-supported, tablets and smartphones should be considered. A formal walk-through of the practice space should be conducted to ensure that it is optimally set up for use with an EHR, specifically that eye-to-eye patient contact can be maintained. If a practice is not large enough to support its own IT department, it may be worthwhile to contract with an IT services company that will supply, install, and maintain all hardware and network equipment.
Once software and hardware requirements are well-established, a timeline should be created and the go-live date set and communicated; no absences from the practice should be allowed during this time. This date will dictate the entire roadmap of implementation, including training, reduction of patient schedules during the implementation period, and many other steps. The vendor's project manager should provide guidance on setting these timelines.
Validation sessions should be held to review the new and revised workflows and any custom build that was created. These sessions should be open to all clinicians and staff of the practice, and the EHR should be demonstrated exactly as it will appear at implementation. End users should be satisfied with the proposed workflows and build. Extensive testing of the system should then be performed with the vendor.
Decisions around data conversion, specifically what, how much, and in what form old data should be brought into the new EHR, will have significant postimplementation consequences. Existing data can be placed into the EHR in a number of ways, including electronically through an interface with preexisting electronic systems, by manual entry into the new system, and through scanning of the paper records, collectively referred to as chart abstraction. The best solution is often a hybrid of all 3 approaches.19 Table 1 discusses the 2 main implementation strategies: a big-bang approach vs a staggered approach.
Training
Adequate, role-specific, and appropriately timed training of the entire practice is essential to a successful implementation.6 The vendor should have specific recommendations as to the length and format of the training, and it may provide trainers and other training tools, such as online modules. The training resources provided are highly vendor-specific; the implementation team should be aware of what was contractually agreed upon regarding this critical aspect of the implementation and plan accordingly. Clinicians and staff must be trained not only on the technical aspects of the EHR, but also on workflows and any new EHR-related policies and procedures. In addition to formal training, a “dress rehearsal” of the new EHR in the clinic setting utilizing mock patients should be considered to identify issues prior to go-live.
Members of the practice who will require training in basic computer or typing skills prior to EHR training should be identified early enough to accommodate these needs. Clinicians and staff who are already facile with computers and will likely adapt easily to the EHR should also be identified early. These “super users” should then be provided with earlier and more robust training, enabling them to serve as additional resources during the implementation period.
Final details
It is strongly advised that clinic volume be reduced during the initial weeks of implementation. A big-bang implementation often requires a 50% reduction the first 2 weeks, followed by a 25% reduction for the third and possibly fourth week. While this will result in a loss of revenue during the implementation period, the possible negative long-term consequences of having inadequate time for clinicians and staff to adjust to the new system are potentially much greater.
A frequent concern raised by clinicians about EHR adoption is the impact on the patient–physician relationship, especially during the implementation period. However, as long as patients are well-informed upfront, they are almost universally tolerant of the process and supportive of their neurologist as a new learner. They are often enthusiastic about being part of this transition. A variety of methods, including verbal communication, signs, brochures, and letters, can be utilized to explain the conversion and ask for understanding.
At-the-elbow support is a necessity at the time of go-live. This may be partly supplied by the vendor, but supplementation by members of the practice's implementation team is often necessary. Preidentified super users will bridge the gap between the resources needed and what is provided by the vendor. Super users will still require dedicated time to function in their usual roles seeing patients or doing clinical support work. The length of at-the-elbow support needed will vary between providers; 5–10 days is usually sufficient. Providers who work part time may require an extended support period.
Troubleshooting and optimization
Despite thorough planning, there will be problems at implementation. Potential issues include unaddressed workflows, hardware, software, and network problems, and build errors not detected in testing. Each issue should be tracked and prioritized as it arises and communicated to the end-users. End-users will have a much greater tolerance for dealing with implementation snags if they know the implementation team is aware of the problem and working diligently on a resolution. There should be a clearly defined vendor reporting process. Interruptions of EHR service, known as downtime, are required for routine software upgrades, and though hopefully rare, unexpected outages will occur. Downtime procedures should be well-defined, so that the clinical operations continue with the least possible amount of disruption (table 2).
Optimization requests, such as documentation template modifications and additions to order preference lists, are expected once clinicians and staff become familiar with the EHR. These requests should also be tracked and prioritized, and any changes made communicated back to the users. Many EHRs provide reporting functionality that can handle routine queries and at least basic data mining, which in turn may be utilized for quality improvement initiatives, registry reporting needs, and clinical research. However, more extensive and sophisticated analysis of the data generated by the usage of an EHR often requires the expertise of an informaticist, usually an external consultant for small practices.
DISCUSSION
Health Information Technology projects fail at an exceedingly high rate (up to 70%) for a number of reasons.6 Extensive research and thoughtful planning, done with the involvement of all stakeholders, the commitment of adequate time, staff, and financial resources to the process, and sufficient training will increase the chances for a successful EHR implementation. The adoption of an EHR can be broken down into countless decisions, with each choice bringing distinct advantages and disadvantages. As long as the implications of all options are carefully considered, the decisions made will be the correct ones for the practice. Continued success will depend on having as robust a strategy for optimization as was in place for implementation. Staying up to date with vendor software upgrades and taking advantage of resources such as vendor and AAN EHR-specific user groups and webinars will ensure that neurologists get the most out of their EHR.
STUDY FUNDING
No targeted funding reported.
DISCLOSURES
The authors report no disclosures relevant to the manuscript. Go to Neurology.org/cp for full disclosures.
Notes
Correspondence to: ude.hsur@srehtaeW;f5000x#&L;f5000x#&nosillA
Footnotes
Correspondence to: ude.hsur@srehtaeW;f5000x#&L;f5000x#&nosillA