Great! Your office has selected the electronic medical records system, and now, in the fifth article in this series, we need to address the process of getting hardware installed, networks set up, people trained, and the software working for you.
To get a system installed, there will clearly be requirements for hardware (the PCs, servers (if a Client-Server system), networking equipment (the connections both for internal wired and/or wireless networking, and external access to the internet), and installation of the actual software program selected (locally, if a Client-Server, or possibly on each PC if an applet or application is required for a web-based (ASP) product). Indentifying and fulfilling these requirements may be managed in whole or in part by combinations of the vendor, the practice, or consultants.
It is extremely important to have qualified people look at the physical placement of servers, understanding the need for physical security as well as environmental requirements for dust, cooling, power supply, back-up power, etc. Security breaches to electronic medical record systems are as often from internal sources as from external (hacking). Being able to lock the server(s), secure PCs with appropriate passwords for log on, etc. are not only good business sense, but specifically addressed in HIPAA and ARRA, with BIG penalties for breaches. Being smart about policies, procedures, and design at the outset will be the most cost-effective solution long term.
Part of the hardware process will need to be assessing any existing technology for its ability to perform appropriately for the new system. Technology that is more than about 18 months old, running on older operating systems, or both should be evaluated for the cost-effectiveness of upgrades versus purchase/lease of new equipment. As mentioned in an earlier article, hardware for a practice really should be of commercial quality, not home office level. The robustness of switches, hubs, routers, wireless devices, etc. is substantially better for business-level, and the cost of down-time, repairs, lost productivity will ultimately greatly exceed the greater expense for the business quality products.
Selection of the specific devices (TabletPCs, notebooks, fixed PCs) should be reflective of the use and role required. Flexibility may suggest that notebooks and tablets be considered even for front office staff (allowing mobility) and nursing staff. The physician staff will also benefit from mobility, particularly if a docking station is used, allowing connection of real keyboards (faster typing) plus recharging of batteries (extending useful battery life and run times). Not having to log in and out when moving from room to room is also a time saver. Wireless access is obviously a requirement for this type of mobility, and proper design of the access points to eliminate dead zones is critical.
Consider placing several PCs in the waiting room for patient check-in, education, and, assuming your vendor has a patient portal, updating of their records.
Be aware of the locations to install printers, and where possible, use fast network printers rather than local printers. The seemingly lower cost of personal printers is in fact much more expensive in the long run: purchasing three or four cheap printers will cost almost as much as a decent network printer and the cost of toner (please don’t use ink!) will certainly be much greater with the small printers than the business class network printer. In other words, the cost per page will be many times greater, and the time it takes to print also many times longer. Critical metrics to look for in a printer are the time to first page out, and pages per minute. Printing of instructions for patients, educational materials (e.g., ACOG brochures), and (hopefully only a relative few) printed prescriptions [electronic prescribing is a requirement for ARRA payment] will be a major source of frustration if it takes 5 minutes to print at the physician’s desk instead of 20 seconds at the central network printer.
Applicable in offices with more than one physician, the need to standardize as much as possible will make implementation easier and cheaper for everyone. While personalizing some things is easy (favorite medications, most frequently used diagnoses), having completely different lists for each provider just makes things harder, more confusing, and more expensive to implement. Unless the practices have vastly different patient mixes (e.g., one physician practicing general OB/GYN and another only doing uro-gynecology), having common data tables and structure is definitely preferred. Taking all of the existing policies and procedures, protocols, etc. and discussing the desired way to standardize can be done well in advance of the implementation process, and will again save time and money. This concept applies equally to multi-site practices, where standardization will aid in maintaining the product, ease of staff moving between offices, etc.
Despite having seen the product in demonstrations, your knowledge of the best way to use the new EMR purchased will be incomplete. Training is essential for you and your staff to understand the impact on workflows, and how to take advantage of the new tool. Part of this process MUST be changing YOUR workflows to suit the EMR rather than the other way around. In other words: cooking in a microwave oven requires a different technique than cooking on the stove top, and using a pneumatic nail gun to drive a nail does not mean that you whack the nail with the gun like you do with a hammer!
Training methods can include what is called “train-the-trainer”, where only a select few people are trained by the vendor, and the newly trained staff member(s) is/are responsible for training the rest of the staff. The advantage of this can be a reduced cost of training. By only have a few people travel to the training location, or requiring fewer of the vendor staff to travel to your office, expenses can be decreased. The disadvantage is that the vendor’s training staff will usually be more knowledgeable, and better able to answer those edge-case questions. Additionally, the risks of having a fewer number of fully trained personnel include re-training or lost knowledge should that staff leave (illness, relocation, quitting, etc.).
Thus, for larger offices, train-the-trainer is a viable solution, but for small offices, having all the staff trained by the vendor probably makes more sense. Super-users will need to be identified in either scenario, for on-going local support that does not require a call to the vendor is much quicker and more cost effective.
On-line training (web-based) or computer-based training (CBT) may be offered by some vendors, and can also be a cost-effective way of learning the system. This avoids the expense of travel, and, providing the web training is recorded, or CBT courses can be re-run at will, allows refresher classes to be held at the office’s convenience.
Depending again on the size and number of offices and staff to be trained, the sequence of training can vary with regards to staff versus physician training. “Just in Time” training is suggested, meaning the training occurs very shortly before go-live. In cases where the implementation takes many months, if training occurs at the beginning, most of what was learned has been forgotten before it ever gets put to use. It may be wise to consider having the staff go live (and be trained) for some time prior to the physician training and use. Using that method, when the physician first starts using the system, there will be sufficient data already in the EMR to make it more valuable. As you may suspect, it will generally take many months for the existing patient data to be entered in the system (as back-filling is problematic at many levels: expensive, questionable data and accuracy, etc.). One should probably look for at least 9 to 12 months of data entry exceeding the value for most OB/GYNs, except, of course, for the obstetrical patients. (If you see a new or return gynecological patient for her annual exam, unless she returns for a problem, your next encounter with value of the just-entered data is one year later at her next annual exam.)
Recognize that for at least a short while, the office will have a reduced ability to move patients through. Depending on the technical abilities of the staff (typing skills, computer literacy, familiarity with similar programs, etc.), anywhere from a 10% to 25% decreased productivity can be expected. Ask your vendor what their experience is with this, and the anticipated duration of the decrease. For some practices, just a few weeks of this lost productivity is all that would be expected, while for others it may be months.
To avoid increasing frustration levels beyond the breaking point, schedule the patient load accordingly. Particularly during the first few days (or weeks, even), having enough slack in the schedule to not only overcome questions but also work out kinks in workflows will keep the experience positive. Blowing up any good will and enthusiasm for the project for the sake of seeing a few more patients for increased revenue is another penny-wise pound-foolish choice.
If you have more than one office and choose to implement the product sequentially, make sure that any issues identified and lessons learned are captured to the benefit of the next office in line. Debriefs with the vendor, trainers, and consultants will help keep the process on track for other offices, and for new modules or functionality introduced by the vendor. Keep track of training issues, and ensure that new employees are brought up to speed incorporating the knowledge learned.
Refresher courses as part of credentialing are recommended; just as your credentialing at the hospital requires CME, so should your practice insist on staff and physicians maintaining their abilities to use the office EMR. ARRA will require capturing significant information on meaningful use and quality metrics as a by-product of use of EMRs.
Now that the system is installed, people are trained, and patients are being seen and the visits documented in the EMR, you might think that you are done with this. Wrong!
Just as medicine is not a stagnant field with no new knowledge coming along, neither is this new world of the EMR. There will be updates to the system, new functionalities added, new rules and alerts for the clinical decision support (expert advice) that should be built into the EMR.
All of this requires continued monitoring, continued tweaking, continued adaptation of the office to the dynamics. In our next article in this series, we will look at this process and how to continue refining your practice to best use the new EMR.
--Michael J. McCoy, M.D