Known Pitfalls and Proven Methods for a Successful EMR Implementation

Deciding to adopt an EMR is one of the most important decisions made by any practice. The transition to an EMR from a paper system can be challenging due to the fact that it will change the way everyone works. EMR’s can change current documentation method(s), workflows, billing practices, scheduling, patient follow-up methods, communication/messaging, etc.. EMR adoption usually requires reengineering current systems and can dramatically change the way practices run. Considering the vast changes that have to occur to adopt an EMR, extensive planning must occur for a successful implementation.

Below are common pitfalls that have been identified by experts in the field. Use this information to help you plan your implementation and to not fall prey to common errors that may be avoided.

A. Planning Phase:
As the saying goes “Fail to Plan; Plan to Fail” and isn’t that the truth. The planning phase is the most extensive and time consuming phase of the implementation process. The planning phase provides a great opportunity to map out the entire process which may include planning the following: conversion of data from the paper charts and what information to convert, current workflow analysis, redesigning new workflows for the EMR, deciding on methods of documentation (template creation, voice recognition, voice capture, partial dictation), staff training strategies, software testing, hardware testing (whether to consider using mobile devices and wireless technology), security rights and authorized access and system piloting. EMR adoption should be an evolution not a revolution and with proper planning you can get your EMR up and running smoothly with a minimal amount of staff frustration and loss of productivity.

• Identify goals and base your planning strategies around these goals.
  First identify broad goals for the EMR and then develop more refined goals. Examples of broad goals may be: to identify and follow-up all patients who are not meeting the preventive health maintenance guidelines; analyze patient profiles based on demographics; create a referral tracking system; create tight security controls to reduce the risk of compromising the integrity of the chart; ensure that the hardware configuration will allow the provider to maintain eye contact with the patient etc.. Identify specific areas within the EMR to reach goals successfully. Share all goals with the staff as well.

• Decide what data needs to be retrievable: It is common for practices to begin entering data into an EMR only to discover that the data is in a non-reportable format, not been consistently entered or not entered in any standardized manner by all providers. Therefore, this data is not reportable or incomplete, rendering it useless for queries. Identify what data will be useful for reporting purposes such as certain diagnoses and medications prescribed per physician; graph of BMI in a pediatric population after a pediatric exercise program was introduced; incidence of tobacco use within the patient population; diabetic patients who have not received a HbgA1c in a specified period of time etc…

Your pre-determined goals and data that you want captured for reporting purposes should drive the decisions made during the planning phase. Utilize this information to create customized libraries, pick-lists, standardized and/or required data fields that everyone will use consistently for desired reportable information. Ask the vendor how data in certain areas of the system is stored and ask if this data is reportable in that format.

• Be aware that “Free Text” may not be reportable. For many EMR programs, if the data is not in discrete data fields, the information cannot be captured by an internal report writing program or a third-party report writing program. Utilizing a fully-integrated speech recognition software programs within the EMR, which captures voice dictated text, is in a free-text format as well and therefore may be non-reportable. There is a growing trend in the industry at utilizing artificial intelligence to attempt to capture free text as discrete data usable by the EMR for reporting. This functionality may be available in the not too distant future.

• Phased implementation is highly recommended. Most EMR’s lend themselves well for phased implementation because many of their functions are in discrete modules such as lab order entry, messaging, E&M coding, preventive health maintenance, patient tracking, e-prescribing etc. If a phased implementation is chosen, map out the phasing and rationale for the order of implementation. The staff will appreciate adding additional modules after they have adequately digested previous modules.

• Create timelines but be flexible. Time Lines are great tools for project planning but be aware that they must constantly be re-evaluated especially if you are designing time lines for phased implementation. Keep assessing progress as the implementation process proceeds and ensure staff that time lines are adaptable to current situations to help reduce their stress level. Entire implementations including training can span a couple of weeks for small practices (1-2 physicians) to several months for larger practices.

• Perform a workflow analysis: analyze existing work processes while looking for opportunities for improved productivity and efficiency. Design new work flows that could be accomplished with the tools available in the EMR and develop a transition plan.

Staff Considerations and Planning:

• Appoint a Physician Champion. A physician champion can be instrumental in the success of the EMR adoption. This person should be motivating, enthusiastic, have a good working knowledge of the EMR and be able to articulate the specific benefits that the EMR will provide.

• Appoint an in-house Project Manager. Most vendors will supply a project manager for large group installations but in addition, have a key person on staff to oversee the entire project. This person should have extensive knowledge of all areas of the EMR as well as how the EMR will interact with each type of provider and support staff. This person is crucial for the “Big Picture” viewpoint and to know the rationale for decisions that are made.

• Communicate to the staff the practice’s desire to acquire an EMR before the purchase. Better yet, have them be included in the decision of which EMR vendor to choose. It is common for a physician to choose an EMR with no input from the support staff. This can create a feeling of resentment among staff and a feeling that their input is not useful or necessary. The staff will more likely embrace a system that they have had input in choosing and will be more acceptable to the adoption.

• Be aware that support staff may feel that they could be replaced by an EMR. In certain cases this may be accurate particularly with file clerks or other types of staff but be sensitive to this possible concern.

• Have end-user staff be involved in the system set-up. Many times practices rely on only one person to set-up system files, pick-lists, defaults, templates or libraries, customizable options etc... This presents a problem in that only one person has an understanding of the rationale for the decisions that were made at that time and that knowledge will be lost if that person leaves the practice. It is best to utilize the end-users for system-set-up decisions because they are the ones who will be performing the tasks that the system parameters will affect. They have the detailed knowledge of present procedures and workflows and therefore may know ramifications of such system set-up parameters on other functionality.

• Map out Workflows utilizing current staff members: Map out current workflows on paper and bring in the end-users who perform the current workflows to help design new workflows for the EMR. No one knows their job better than the person who does it everyday but more often practices do not go to the source for their crucial input.

• Learning curves are usually underestimated. The learning curve for complete and successful adoption of the EMR is usually vastly underestimated. Even if productivity is not affected initially during the go-live phase, most providers do report an increase in the length of time necessary for documentation, especially if templates are used and the provider’s are not familiar with them. Most providers will spend additional time at the end of the day documenting notes after a go-live. Usually within 6 months to one year, most providers are leaving the office at their normal times. It is difficult to predict length of learning curves and the impact of learning curves on productivity. Utilize the vendor’s knowledge for benchmark learning curve estimates.

B. Testing phase:

Software/Hardware Testing:

• Test software extensively before implementation. Never assume that the software functions in the way you think it should. Set-up a test database for software testing and for staff training. Thoroughly and completely test all areas of the software and utilize the end-users to test their specific functions.

• Perform Volume testing, if possible. Take a typical day and do a dry run in a test database. This step is often overlooked but can provide important information regarding the time it takes to enter data with typical volume or increased volume.

• Ask for a list of known bugs from the vendor for the version you are about to install. If bugs exist, ask the vendor to create work-arounds and identify dates for patch fixes. You do not want to identify a major system flaw or bug during the go-live phase when this could be prevented.

Hardware Testing:

• Prepare Infrastructure:
  A crucial part of the success of implementation will rely on the success of the hardware infrastructure readiness. Note: the hardware testing will be much more extensive if a client/server environment exists or is chosen as opposed to a web based or ASP environment where the software and server is hosted by a vendor off-site.
  
  

For a client/server environment, the project should be planned in advance to define locations of workstations, printers, kiosks, servers, and/or wireless device access points etc. Existing hardware systems may need to be upgraded and/or reviewed to determine the stability of the system prior to any software installation. In addition, cabling may need to be run to new locations to accommodate access to the network. New systems need to be purchased and delivered well in advance of implementation to allow for testing. Once the infrastructure is in place the testing phase should begin to ensure all aspects of the network and hardware are functioning properly. Phase 2 of testing begins once the EMR software has been installed complete with a dummy database to enable appropriate testing of the applications in the new environment. All testing should be complete before staff training dates are scheduled. A test environment should be established for future updates, this will allow the IT Director to install future software updates/upgrades in a non-production environment for testing prior to updating live units.

Staff and Testing:

• Pilot systems before implementation. Pilot workflows, procedures, modules, templates, documentation time etc. in a live environment utilizing a small group of staff long before go-live. This is critical to identifying issues that are unforeseen during the planning phase.

C. Training Phase:

• Not enough time is allocated for training. This is a very common error made by most practices. Keep in mind that not only is staff required to learn the EMR but also new workflow and procedures. Training sessions are best if kept short and scheduled in increments. Small groups are more beneficial for more personalized training. Allow staff to practice what they have learned using a hands-on approach before introducing new information. Utilize the vendor’s experience with training time but be willing to alter for your individual practice.

• Training should be performed outside of clinical work sessions.
  Practice administrators, in their concern to not adversely affect productivity, will attempt to train staff as they try to perform their clinical duties. This leads to poor understanding of the software and frustration. Train users right the first time. There are several methods practices can utilize to effectively train staff such as reducing or blocking schedules, hiring temporary employees, training outside of clinical time etc...
  Staff should also be paid if they are being trained outside of their usual work schedule.

• Set-up a training room for staff to practice. Giving staff time and a quiet location to practice. This can lead to a comfort level with the software and lessen the apprehension of go-live.

• Appoint Superusers. Designate certain users to be “Superusers”. Their role is to provide immediate, first line response to staff with questions and issues during go-live. Designate a superuser for each type of clinical role (MA, nurse, receptionist, provider). Superusers should have a more extensive knowledge of the software and workflows. Being able to provide immediate support to staff during a go-live situation will more likely ensure that productivity is not interrupted.

• Miscommunication risk with Train the Trainer method. One concern with Train the Trainer method is the potential miscommunication and/or misunderstanding of information from one person to another. Trainers supplied by the vendor usually train large groups of users simultaneously and are more experienced with training the software. Train the trainer methods can provide a cost savings to the practice however.

• Evaluate staff’s readiness for go-live. Assess staff’s knowledge of the software and workflows. Create mock live situations and walk-through the workflows considering all possible scenarios. Be prepared to delay go-live if staff is not sufficiently prepared.

D. Go-Live Phase:

• Schedule the go-live in close proximity to the end of the training sessions. Try to avoid a long delay between the training sessions and the go-live. No more than a week should be allowed between the end of training and the go-live. This will ensure better retention of the information.

• Reduce provider schedules: Reduce the number of patients a provider is required to see during the go-live phase. Learning an EMR can be a difficult process, especially for providers. By reducing schedules for some period of time this can take the pressure off significantly. Many practices reduce schedules by 50% for one to two weeks after the go-live and then 25% for several additional weeks. Another method that has been used is to add 15 minutes onto comprehensive examinations and 5 minutes onto follow-up visits. Note: this method may involve some planning ahead to accommodate the scheduling templates.

• Provider Adequate Resources. Be certain to supply the staff with well trained individuals such as vendor trainers, superusers, in-house project manager etc. during the go-live phase. Create a Help Desk Hotline in case trained personnel are not immediately available. Communicate the chain of support method to all users before go-live. Put a sticky label on each PC with the help desk hotline phone number. Have systems in place if bugs or issues are discovered.

E. Post Go-Live:

• Post Go-Live Assessment is necessary: Now that the EMR has been implemented, many practices feel as though the installation is complete. However, nothing could be further from the truth. Practice administrators must continue to assess the staff’s level of frustration, monitor productivity, measure patient cycle times, re-evaluate workflows, learning curve assessment, is the EMR meeting the established goals etc...

• Evaluate the Go-Live with Staff: Query the staff regarding the go-live process. Get their feedback as to what was helpful and what was lacking. This information can help with future implementations especially if new modules are to be introduced in the near future.

• Provider on-Going training and support: Practice administrators should continue to offer training sessions well after the go-live for reinforcement and refreshment. Staff usually cannot absorb all the information given during the initial training sessions and therefore follow-up training sessions should be offered.