Last May, this column discussed how neurologists are benefiting from electronic medical records (EMR) in their medical practices. But one year later, we have learned that EMR adoption has fallen short of its promise, with fewer than 10 percent of physicians switching over from manual charting.
Why have so many of us been so slow to implement this technology to eliminate manual charting? Interviews with neurologists who had expressed interest in EMR systems revealed that there are still obstacles to overcome: among them, systemic problems, cost, and poor vendor support.
NEED FOR CUSTOMIZATION
Randolph W. Evans, MD, a neurologist in solo practice in Houston, TX, and President of the Harris County Neurological Society, has been looking for an EMR system for several years but has not found one that meets his needs. Like others interviewed here, he said most systems are too generic for his practice.
James N. Domingue, MD, a solo practitioner from Lafayette, LA, agrees. “We need software that emphasizes initial evaluations rather than follow-ups. Family physicians spend their lives doing revisits and most programs are directed towards those concerns.”
Indeed, Dr. Domingue's quest for a better system spurred his interest in taking computer programming courses so he could design his own applications that generate EMG and toxin reports. He eventually obtained a master's degree in computer science.
And even if a suitable program is found, most neurologists report that current products – which run from the tens to even the hundreds of thousands of dollars – are just too expensive.
Bruce Sigsbee, MD, a long-time member of the AAN Medical Economics and Management (MEM) subcommittee, said, “With shrinking revenues, the economic value of EMR is difficult to establish. There are benefits– such as lower transcription costs and chart and paper handling costs – but they don't justify the overall costs.”
Paul Caudill Miller, MD, a neurologist in group practice in Montgomery, AL, looked for an EMR system to reduce his overhead expenses. “My practice of four physicians spends$70,000 per year on transcription alone!” he said. But he found that the notes the system generated did not make sense.“If I tried to send out notes like that, I would lose all my referring physicians.”
PROBLEMS WITH COMPATIBILITY
Many neurologists expressed an interest in a product that was fully compatible with hospital records as well as laboratory and radiology facilities. Dr. Miller said, “Even if a time-neutral product were available, it would need to interface with outside sources to be useful.”
But Laurence Haber, MD, who is part of group practice in Manhasset, NY, said he thinks such a product would be worth the investment.
“I would be willing to spend extra if it meant never having to search for a lost chart or needing to weed through piles and piles of paper reports,” he said. Still, he has not yet found a program that is fully integrated with billing and compatible with external reports. He hopes that his hospital, which is looking into an EMR program, will somehow make it accessible and affordable to community physicians such as himself.
Eric A. Kelts, MD, a neurologist at the University of Rochester, is halfway through both an ophthalmology residency and an MBA program. “The EMR at our university hospital is limited – it includes billing, labs, and tele-radiology – but all clinic charts are paper-based – a source of constant frustration for all involved,” he said. “We live in a time of extraordinary computing power, yet my institution remains hamstrung by a simple paper-based system.”
Clearly, part of the problem is the lack of inter-operability between products. Standards used in developing EMR software are archaic, disparate, or non-existent. “EMR will not work until there are consistent ways of communicating with laboratories, x-ray departments, hospital medical records, other physicians – and then incorporating their data into our electronic charts,” Dr. Domingue said. “This standardization would not be difficult, it just takes time and money.”
Most of the feedback also reflected concern about the reliability of vendors. Dr. Haber observed that even good companies have gone out of business and the smaller ones are not always available for customer support. Dr. Domingue added: “And when they do so there is no guarantee that the EMR in their program will be accessible to other programs. This is inexcusable, there is no reason that why another vendor's interface should not be able to access standardized tables in database format.”
Ron Nath, MD, a neurologist who completed a fellowship in medical informatics at Yale and works in the health care information technology industry, said: “We're stuck with a mess of vendors, each trying to lock you into their proprietary way of doing things and each going out of business, merging with another company, or forcing users to upgrade to their new and improved version. Each user is faced with the daunting prospect of which vendor to choose, only to end up having a bad outcome in the long run.”
Dr. Kelts was also concerned about vendor sustainability in view of the lack of standards. “Almost all of the EMR systems we have looked at use a proprietary file system – if the software company ever goes out of business, there is little recourse but to buy an entirely different system or continue to use an outdated one.”
Vinson J. Hudson, President of the only independent market research and analysis firm in the United States that specializes in the physician's office management and medical information systems industry advised,“It is critical for doctors to do due diligence prior to purchasing an EMR. This is a particularly important time to investigate your vendor; some lack the resources needed to meet HIPAA requirements and will ultimately go out of business or become acquired.”
Last December, the American Academy of Family Physicians (AAFP) received the green light to further develop an Open Source Electronic Health Record (EHR). The product tackles several of the obstacles that AAN members described. It is low cost and free of licensing fees (addressing the issue of expense), allows for customizable input for specialty-specific documentation (addressing content and workflow issues), supports seamless exchange of clinical data between health care providers and institutions, and is consistent with standards and regulations(supporting the data portability and cross-institutional issues).
David C. Kibbe, MD, the project leader and Director of Information Technology for the AAFP, said: “It is entirely within the scope of the Open Source EHR project for neurologists to customize their own interface and workflow version of the application's front end, while taking advantage of and utilizing the standardized core components of the application that are being used by many thousands of their colleagues in other specialties.”
He also welcomes neurologists' input and experience with regard to best practices and the linkage of evidence-based clinical information for use by primary care clinicians, patients, and other non-neurologists.
“The Open Source EHR is truly‘open’ to ideas as to the best means of secure communications between primary doctors and subspecialists, and its leaders would like to see new channels of information flow for referrals and follow-up become easier to use as a result of this technology,” he said.
Dr. Nath said, “We know open source and open standards work – look at Linux, Apache web server, Open Office, as examples.”
Dr. Kelts also finds the idea of the Open Source EHR exciting. He said: “As we have seen with Linux, the benefits of Open Source are many. It would allow for uniformity in EMR file systems and customizability in the form of independently developed modules that could be compiled together. I think it would solve almost all of my institution's needs for a robust EMR without costing an arm and a leg.”
But there are also obstacles with an Open Source EHR. Dr. Haber expressed concern about having to rely on the Internet and did not want his patient data stored remotely. Dr. Evans said, “There is still the hassle factor of conversion, including training my staff who are already harried trying to keep up with Medicare.”
Neil Busis, MD, a neurologist in private practice in Pittsburgh, PA, and a member of the MEM subcommittee, is waiting to see how his hospital's EMR project turns out before he buys one of his own.
“There really are two markets for EMRs out there, and they are very different. One is for the MD's office and the other is for the hospital. A potential conflict arises if the two systems can't talk to one another. The open source AAFP EMR will be greater the more it can integrate with hospital information technology systems and decision support systems and the more it can be adapted to use by specialists.”
Despite obstacles, there is clearly interest in continued development of new technologies to improve practice efficiency. Whether more neurologists will continue to experiment with the next generation of EMR systems remains to be seen.