© Geriatric Times. All rights reserved.

Can the Electronic Medical Record Improve Geriatric Care?

by John F. Hurdle, M.D., Ph.D.

Geriatric Times March/April 2004 Vol. V Issue 2

The proposition that electronic medical records (EMRs) could influence the quality of health care for elderly patients may seem either inevitable or a pipe dream, depending perhaps on when and how one trained. Most physicians and other health care professionals use computers on a regular basis. The entire infrastructure of basic services on which we all depend, from transportation networks to the food supply, are run largely by computer systems. They are systems where the same basic operations are performed repetitively on a large scale, with emphasis on the regular delivery of some product. The consumer and the producer of these services share the same business model, with well-understood and easily modeled operating economics. Health care is a basic societal service and is no less fundamental than food or water. Yet historically, it has been chronically under-computerized (Hayes, 1996).

The reasons for this underutilization of computers in health care are many, but they reduce to a common theme: There has been a fundamental mismatch between what providers expect from computers and what computer systems can actually deliver.

The History of the EMR in Health Care

The basic properties of computer systems are less applicable to health care and especially to the care of elderly patients. Health care is individualized at the patient level--rarely simple repetitions of previous visits. Care is delivered as needed, often not on a regular basis, with varied intensity and locale of service. The underlying economics of health care do not follow supply and demand; in health economy, the value to the consumer of the goods and services is essentially inestimable (i.e., quality of life and life itself), while the costs to the producers bear little relationship to the cost of production.

When industrial giants such as IBM, Hewlett-Packard and 3M tried to apply industrial-style computing techniques to health care in the 1970s, 1980s and 1990s, the results were mixed at best and failures at worst. After more than 20 years of spirited competition, the commercial health care computing market remains fragmented and confused. The most successful, broadly deployed EMR systems have been those that grew from the bottom-up--internal systems that were adapted to the specific needs of the parent organizations. The U.S. Department of Veterans Affairs deploys the largest standardized computerized health care network in the country, which was developed internally. Other successful EMR systems are the Regenstrief Institute in Indianapolis, Brigham and Women's Hospital in Boston, LDS Hospital in Salt Lake City, and Kaiser-Permanente.

Why should we expect EMRs now to be a significant agent for change in health care quality across all settings? What has made an electronic record for all providers more feasible today than it was 20 years ago? One factor has been the maturation of a branch of biomedicine called medical informatics. This discipline of research and engineering works at the interface of computer science and health care practice. Its express focus is to use technology to improve health care delivery. Drawing on computer science as well as cognitive science, decision theory, management science and related subdisciplines, medical informatics has created an excellent foundation of what can and should be done with computers to improve health care. For years, however, medical informatics has had notable problems:

Internet protocol has helped solve both problems; within the past 10 years, the ability to securely move information between any two points on the globe has become not only feasible, but commonplace. However, the Internet does not impose standards on content, just on the transmission itself. Although excellent medical communications standards have existed for years (e.g., Health Level Seven [HL7] and Logical Observation Identifiers Names and Codes [LOINC]), it took the Internet to provide an appealing and affordable communication conduit. Users from solo practitioners to patients to corporations chose voluntarily to adopt the Internet as their communication medium of choice, using the impressive medical communications and security standards like HL7 and LOINC that were made available.

The enormous appeal of the Internet and particularly Web browsers is in part due to the fact that no one program (and thus no one vendor) is required to use the medium. The standards and portability problems were almost solved, but they were made less relevant by a technological advance that initially had nothing to do with medical informatics. That advance would help to make the electronic medical record an indispensable tool in geriatric care.

The EMR in Geriatric Care

Geriatric patients today receive care provided by several medical professionals in a multitude of settings. In a single year, an older patient might receive care in several clinics, a hospital, a skilled nursing facility and at home (Medina-Walpole et al., 2002). This care might be provided by numerous people, including a primary care physician; a hospital team; any of several surgical, medical or other specialist physicians; a physical, occupational or nutritional therapist; an advance practice or other specialized nurse; and a social worker, as well as others.

The medical record for such a patient is a scattered and disjointed repository, which may be stored in paper charts in one locale and electronically at another. There are obvious barriers to any one provider retrieving information recorded by another. Such barriers are far more complex than a simple inconvenience. As the patient moves between providers and locations, hazardous lapses can occur in the care plan due to missing or outdated information (Forster et al., 2003).

The development of the EMR was motivated in part by the desire of health care professionals to overcome the limitations of paper. The EMR would allow multiple providers from different locations to access a patient's information simultaneously. Most EMRs are designed based on the paper chart (Lovis et al., 2000). The progress note remains the basic element of the record; therefore, EMRs facilitate entering and reading computerized progress notes. The goal of multiple-provider access can be achieved in this type of system, but access alone has turned out to be a source of confusion.

For example, in the VA's Computerized Patient Record System (CPRS), one of the largest EMRs in the world, all notes by all providers are displayed on a scrolling selection list. Every note title on that list is no more or less important than any other--from a vital admission history and physical to a phone call from a patient for a pharmacy refill. This system makes it difficult to find essential or relevant information in haste. Inevitably, providers tend to scan note titles to find their own last note, ignoring potentially pertinent notes from other providers (Weir et al., 2003).

The VA's EMR is illustrative of both successes and failures of this type of system. Technology like the Internet and computer software such as Java made possible a national system of standards-based medical records. As workable as the CPRS may be, the VA's system is fundamentally an electronic version of the paper chart. Accessibility is important, but the real advantages of computerized medical records will be realized when the EMR evolves from a simple record to an active agent of care.

The next phase in medical informatics will be facilitated by abandoning the patient chart format in favor of a structure based on processes. Outcomes could be organized along lines of patient problems and health maintenance. However, explicit goals would be assigned for each patient's care, in accordance with available, pertinent evidence from the literature and the patient's own history. Evidence-based tasks would be mapped out automatically, with responsibilities assigned to individual providers. Physicians would assume the ultimate responsibility for any care plan, but the active medical record would integrate the patient's data with the current treatments in order to facilitate routine matters such as suggested blood work or optimal scheduling of procedures.

Of course, the EMR will not automatically resolve all problems; there is still ample room for providers to individualize the patient's care. They would be trained to recognize inappropriate treatment suggestions--even those made by the EMR. The difference is that the integrated EMR, as the sole repository of the most complete patient information and pattern of care, becomes a team member, engaged as much as a provider. Integrating information from multiple sites for the elderly patient, the EMR is the first and best place to start processing the patient's complete medical needs. Patients will always require personal contact with providers, as human interaction is vital to patient care. Thus, the experience and training of physicians will be required for complicated patient problems that do not fit the patterns recognized by the EMR.

EMRs and Quality of Care

As early as 1991, the Institute of Medicine (IOM) called for the retirement of the paper medical record (Steen and Detmer, 1991). It was a sound recommendation, but the Internet had not matured enough to solve the problems of standards and portability. Since then, the IOM has revised their call, most recently in their 2003 report to the U.S. Department of Health of Health and Human Services titled Key Capabilities of an Electronic Health Record System (IOM, 2003). The primary and secondary uses of an EMR system outlined in the IOM report are shown in the Table.

The primary improvements are clearly related to patient care, but the secondary improvements are themselves a reflection of the changing nature of the medical record. Data could be extracted from EMRs to discern infection patterns across populations and still be timely. Whether or not that information could be used to investigate a potential bioterrorist event or an outbreak of severe influenza, the enabling technology is the same: a smart, proactive EMR that participates integrally in the modern health care process.

Dr. Hurdle is the senior medical informaticist at the VA Salt Lake City Health Care System and is co-director of the IDEAS research center.

References

Forster AJ, Murff HJ, Peterson JF et al. (2003), The incidence and severity of adverse events affecting patients after discharge from the hospital. Ann Intern Med 138(3):161-167 [see comments].

Hayes GM (1996), Medical records: past, present, and future. Proc AMIA Annu Fall Symp:454-458.

IOM (2003), Key Capabilities of an Electronic Health Record System. Washington, D.C: The National Academies Press. Available at: www.nap.edu/books/NI000427/html/index.html. Accessed Feb. 20, 2004.

Lovis C, Baud RH, Planche P (2000), Power of expression in the electronic patient record: structured data or narrative text? Int J Med Inf 58-59:101-110.

Medina-Walpole A, Barker WH, Katz PR et al. (2002), The current state of geriatric medicine: a national survey of fellowship-trained geriatricians, 1990 to 1998. J Am Geriatr Soc 50(5):949-955.

Steen EB, Detmer DE (1991), A vision for future patient records. J Am Med Rec Assoc 62(8):48-54.

Weir CR, Hurdle JF, Felgar MA et al. (2003), Direct text entry in electronic progress notes. An evaluation of input errors. Methods Inf Med 42(1):61-67.