The promise of decreased medical errors, increased office efficiency and cost savings is driving interest of physicians and others in the electronic medical record, or EMR. However, very little data exists to support these claims, there are a wide variety of EMR systems available, and the technology is often expensive and time-consuming to implement. This article offers an overview of the EMR, summarizes some available data regarding office cost savings, and details some decisions to be made when purchasing and implementing an EMR.
The EMR often is confused with the EHR, the electronic health record. The terms frequently are used interchangeably, but there are fundamental differences between the two. An EHR is patient-specific, while an EMR is physician-specific. Thus, theoretically an EHR should exist independently for every person, and should be able to be accessed by an EMR.
Another distinction to be made is between a document management system and a true EMR. A document management system is basically a “paper” chart on a computer, while an EMR is distinguished by its functionality: patient visit charting and document management, diagnostic test ordering and results management, electronic prescribing, decision support features for evaluation and management coding, treatment options based on clinical protocols or guidelines, and a patient portal through which patients can access and amend various portions of their medical record. An EMR also is distinguished by interoperability, such as between a hospital and laboratory.
Financial considerations are paramount when considering EMR adoption. Like any other asset, the benefits must outweigh the costs. The learning curve of the software and added time spent charting—the EMR almost always will take longer than the trusty dictaphone—also must be factored into the decision.
Cost Versus Savings
A report by Miller and colleagues in Health Affairs detailed the implementation
of an EMR by several primary care practices. The average cost per physician
to implement the EMR was $44,000. The average break-even time on their investment
in the EMR was 2.5 years, which caused cash flow problems for some smaller
practices. After the break-even period, the financial benefits from EMR conversion
were approximately $33,000 per physician, which, after subtracting the ongoing
expenses, amounted to $23,000 per year. For these primary care practices, about
half of the financial benefit derived from the EMR was realized from decreased
personnel costs for transcription and chart management, and the other half
was attributed to increased revenue from more accurate evaluation and management
coding. However, almost all physicians noted an increase in the amount of time
spent in the office for an average of four months after EMR implementation.
Another study, by Grieger and colleagues in the Journal of the American College of Surgeons, detailed the implementation of an EMR in an academic surgical practice. This study reported annual total cost savings of $14,000 per physician, with ongoing costs of $4,000 per year. In contrast to the previous study, almost all of the cost savings were due to reduced personnel costs, and the effect on revenue from evaluation and management coding was neutral.
Based on these limited studies, it would seem that for a practice that derives most of its revenue from procedures, such as neurosurgery, the cost savings generated by an EMR are attributable to a 0.5 to 1.0 full-time-equivalent expense reduction rather than to increased revenue from more accurate coding. Thus, adding full-time-equivalent expenses for transcription and chart management and subtracting the ongoing expenses from the EMR cost offers a rough idea of the additional income that could be derived from an EMR implementation.
EMR Selection
Once the decision to move ahead with an EMR has been made, the first step
is to choose the EMR functionality most important to one’s practice. Probably
the most important consideration is the method for charting patient visits.
In order to support evaluation and management coding, most EMRs organize the
visit charting based on Medicare coding guidelines. Thus, a patient visit is
opened by picking a chief complaint from an exhaustive list, which usually
can be customized. Based on the chief complaint chosen, the illness history
will be broken down into the required fields such as location, duration, modifying
factors, etc. The two most common ways to fill in these fields are by drop-down
menu (for example, for a patient with left leg pain, select leg, then left,
etc.) or via third-party voice recognition software. Drop-down menus can be
filled in on a laptop or tablet computer during the office visit, which some
patients may find impersonal, or after the patient visit, which is less efficient.
The EMR selected should be compatible with the practice’s current practice management software for billing and appointment scheduling. Also check with area hospitals, imaging centers, and laboratories to see if they support electronic ordering, as many of these organizations have not yet implemented interoperable information technology.
Choosing a Vendor
The next step is to choose a vendor. The Certification Commission for Health
Information Technology, created in 2004 to develop standards and certification
criteria for health information systems, lists all CCHIT-certified EMR software
programs on its Web site, www.cchit.org. Choosing a CCHIT-certified EMR will
ensure maximum interoperability.
The most important element of vendor selection is whether to choose a Web-based model or to house the EMR software on a server in the office. Web-based models typically have less up-front costs for implementation, but the data resides elsewhere and there may be bandwidth issues that limit the connection speed if traffic is high on the company’s server. Also, the ongoing charges for many Web-based models are based on a certain percentage of the practice’s revenue rather than fixed. EMRs housed on an office-based server are more expensive initially—the server must be purchased and set up—but typically have lower, fixed maintenance costs. Especially for larger practices, the fixed costs of maintaining and upgrading the server and software may represent a smaller percentage of revenue than that charged by a Web-based software vendor. However, when working from varying locations it can be more difficult to access the EMR program remotely through an office-based server without losing some functionality such as voice recognition software compatibility, e-prescribing, and test ordering.
Lastly, studies suggest that EMR implementation is more likely to be successful when there is a physician champion who is able to convince the inevitable Luddites in the office to persevere.
Health information technology adoption is one of the fundamental tenets of the American Recovery and Reinvestment Act of 2009. There is a possibility that EMR adoption may be mandated in the near future. Barring a mandate, the decision to implement an EMR should be based on whether the potential cost savings are worth the extra time. If you decide to adopt an EMR, proceed carefully if your practice lacks an enthusiastic champion, as there is a very real chance you will decide in a pique of frustration to scrap the EMR, irretrievably losing your initial investment. Choose the vendor and functionality your practice needs and, in time, the cost savings can make your practice more competitive in today’s healthcare environment.
K. Michael Webb, MD, is a member of the AANS Neurosurgeon Editorial Board. He is a founding partner with NeuroTexas PLLC, Austin, Texas, and an executive medical director of the NeuroTexas Institute at St. David’s HealthCare. The author reported no conflicts for disclosure.
For Further Information
- Grieger DL, Cohen SH, Krusch DA: A pilot study to document the return on investment for implementing an ambulatory electronic health record at an academic medical center. J Am Coll Surg 205(1):89-96, 2007
- Miller RH, West C, Brown TM, Sim I, Ganchoff C: The value of electronic health records in solo or small group practices. Health Aff 24(5):1127-1137, 2005