Electronic Health Records Their Time Has Come
After almost two decades of advocacy, the health care system might finally be ready to take full advantage of information technology to improve quality and efficiency.
In 1991, when portable computers were the size of sewing machines and the World Wide Web was aborning, the Institute of Medicine proposed a plan for how emerging technologies could be used to improve medical recordkeeping. The plan highlighted the potential of health information systems in general, and computer-based patient records specifically, to support health care professionals as they make decisions at the point of care. It also called for developing a national health information infrastructure. The goal was to achieve ubiquitous use of such patient records by all U.S. health care delivery organizations by 2001.
The goal was overly ambitious. But the proposed plan proved to be an important milestone in the evolution of thinking about patient data and the health information infrastructure needed within organizations and the nation. And such thought is now turning into action. In early 2009, with passage of the American Recovery and Reinvestment Act, the government committed its first serious investment in electronic health records (EHRs) and in developing a national health information infrastructure. The act calls for achieving widespread use of EHRs by 2014, and it provides $36 billion to support the use of EHRs in clinical settings and another $2 billion to coordinate their implementation.
EHRs are much more than computer-based versions of paper medical records or stand-alone data repositories, and their successful implementation is not without challenges. Indeed, the federal government’s newly appointed national coordinator for health information technology, David Blumenthal, said in his first public statement that technical assistance is a “critical factor” in advancing EHRs to reduce health risks.
As an illustration of how EHRs and EHR systems may bring about multiple benefits in medicine, consider how two other industries have used similar technologies to provide convenient, efficient, and customer-centered services. In the banking industry, automatic teller machines and online Web sites provide customers with ways to conduct their banking when and where they choose and with confidence that their personal information is protected. Banks also provide alerts to customers about sensitive activity in their accounts and reminders about payment deadlines. These easy-to-use tools depend on a secure, seamless information infrastructure that enables data to cross organizational and national lines. In the online retail industry, companies such as Amazon.com not only offer convenience in shopping but also provide personalized shopping recommendations based on past purchases or selections made by other customers who have shown similar interests. This feature depends on the ability to capture and analyze data on individual and population levels. Amazon also provides a mechanism for used-book sellers to offer their products via its Web site—a process that is possible, in part, because there is a shared format (technically, interoperability standards) for the information presented to customers.
Now consider how data, information, and knowledge could securely and seamlessly flow through health care organizations. As a case in point, begin with a patient who has a chronic condition and is tracked by an electronic record of her health history, including any unusual symptoms, an accurate list of numerous medications, and reminders of when lab work is needed to ensure that the medications are not causing kidney damage. Lab results are forwarded directly to her electronic health record, which is maintained by her primary care clinician. If lab results are outside of the normal range, the physician receives an alert and, in turn, sends the patient an e-mail requesting that she repeat the lab work and schedule an appointment.
During the appointment, the physician has a comprehensive view of the patient’s health history and receives a reminder that the recommended protocol for treating her condition has been changed. After reviewing options with the patient, the physician prescribes a new medication, and the prescription is sent directly to the patient’s preferred pharmacy and to the patient’s EHR. The physician also recommends increased physical activity, and so the patient elects to receive weekly exercise programs and commits to recording her daily exercise in her health record.
After the visit, selected data elements without personal identifiers are automatically forwarded to a larger population data set maintained by the health organization in which the physician works. The organization can use the data to compare outcomes for its patients to regional or national benchmarks. For example, a hospital may learn that its post-surgery infection rate is higher than the national trend and then compare its practices to those used by other organizations with lower infection rates. Or a physician practice group may learn that its outcomes for a particular diagnosis meet national norms, but that there are less expensive alternatives that yield comparable results.
On a broader scale, outside authorized users—say, university researchers—can access the population data in conducting clinical research. Pooling data from an entire region or state, or even nationwide, will enable more comprehensive and efficient research on the effectiveness of treatments and clinical processes. For example, bioinformaticians might benefit from using large data sets as they seek to advance the intellectual foundation of medicine from its current focus on organs and systems to one based on molecules. Public health professionals can use the data to monitor health trends across various populations. Further, selected EHR data elements may flow into biosurveillance systems so that analysts can detect new outbreaks of disease, whether due to natural infections or bioterrorism.
For the full impact of EHRs and EHR systems to be realized, the results from these studies, when fully verified, must flow back to clinical professionals and patients so that they can base their decisions on the most current knowledge available. This cycle of using knowledge to support decisions, capturing data on the outcomes of those decisions, analyzing the data, and using insights gained to refine the knowledge base is the essence of how to develop a “learning” health care system that is safe, timely, efficient, effective, equitable, and patient-centered. EHRs are the beginning and end of that cycle. Each time that an EHR is used, there is an opportunity to enhance current and future decisions. But EHRs are only a part of the complex infrastructure that is needed to enable learning cycles within health care organizations across the country.
Because information and communications technologies continue to evolve at a rapid pace, technology itself is not a current constraint. There is, however, a need to improve the design of EHR systems to take advantage of technological capabilities. For instance, user interfaces (how the system presents information to the user and how the user interacts with the system) must support rather than interfere with clinician-patient encounters. An EHR system that requires a clinician to focus on a screen rather than a patient may adversely affect the quality of care and will probably result in user resistance to adopting the system. One cure for this is for the clinician and patient to share the screen and work as a team to ensure that data are both accurate and understood. It may also result in clinicians recording observations on paper and then entering the data into the EHR, adding extra time to the care process. EHR interfaces must be easy to learn and use, capture data with minimal intrusion during a patient visit, and provide information in ways that are intuitive to the user. Further, EHR functions such as clinical decision support tools (such as alerts or patient-specific recommendations) must be a byproduct of the user’s professional routine rather than multiple add-on tasks. Thus, system designers must work with clinicians to understand the information and work needs of caregivers at various points in the care process and align the processing power of computers with the cognitive abilities of people.
The medical community also needs to establish a set of accepted terminologies that will underpin consistent understanding of their meanings across settings, among countries, and eventually over time. The terminology standards will need to be continually updated to reflect advancements in the practice of medicine. Although there has been genuine progress nationally and internationally in this area, the efforts have been chronically underfunded.
In addition, new standards are needed to improve the flow of data within health systems. A variety of data exchange (or interchange) standards now enable EHRs to receive data from or send data to certain segments of the health care system, such as physicians’ offices, pharmacies, or laboratories. But considerable work remains to develop standards that go beyond the routine exchange of data to enable users to query distributed databases and process responses. Advancing this area of research will require additional public funding.
The ability to access and transfer health data also depends on consistency in state laws governing such activities. A long and sometimes heated debate on the appropriate approach to the privacy of patients and confidentiality of their data has stalled the development of a framework that protects confidentiality while supporting the legitimate use of data for improving quality, research, and public health. Unquestionably, privacy is a valuable and valued social good. But so too are altruism, health, and freedom. Currently, health information policy seems to be giving too much weight to privacy at the expense of freedom and health.
For example, it is peculiar that national public policy does not help citizens create and maintain individualized personal health identifiers to support their health care and their privacy, while at the same time citizens, at public expense, can call a telephone number to keep phone calls from disturbing the privacy of their evening meal. Public policy must find a way to allow citizens who are so inclined to opt to share their data with researchers for legitimate biomedical and health research. Similarly, citizens should be allowed to choose to let researchers use their genetic information for research purposes.
To secure a foundation for progress, the nation needs to expand efforts to develop a health care workforce that is prepared to develop robust EHRs and EHR systems and to use the technologies to their full potential. To capitalize on the clinical transformations promised by EHRs and EHR systems, clinical teams will need to undergo a basic culture shift in their beliefs and working habits. Achieving this shift will require rigorously training some clinicians to become “champions” who will lead their organizations into the new way of working together, networked via EHRs. The American Medical Informatics Association has demonstrated the potential of this approach, as its 10×10 program has trained over 1,000 health care professionals in applied health and biomedical and health informatics. (The program takes its name from its stated goal: to train 10,000 health care professionals in these skills by the year 2010.) In another effort to train more clinical informaticians, the American Board of Preventive Medicine is working with various other groups to develop a new medical subspecialty of clinical informatics. Other training and certification efforts will be needed to prepare workers across the health care disciplines, including nursing, dentistry, pharmacy, public health, and psychology, among others.
The current cadre of public health informaticians, joined by coming additions to their ranks, will need to help states move into the future world of medical information technology. States will need to be encouraged to develop and maintain public health reporting requirements linked to EHRs as well as to develop regional networks for information exchange. Health care workers and patients need help in learning how to use Web portals and how to protect patients’ personal health data. All health professionals need to know how to use information and communications technology to work in teams, to practice evidence-based medicine, to continuously improve quality, and to keep the patient at the center of their care.
Underlying all of these challenges is a critical need to develop a robust clinical research capacity to take advantage of the expected deluge of data. Thus, research and evaluation must be part of the national agenda. The aim should be not only to improve how EHRs and other related technologies are used, but also to develop new products and processes that will provide even greater health benefits.
The path forward
The challenges are clear, and fortunately the newly enacted American Recovery and Reinvestment Act creates a formal structure for addressing them. The act codifies the Office of the National Coordinator for Health Information Technology, housed in the Department of Health and Human Services (HHS), and assigns a focus on “development of a nationwide health information technology infrastructure that allows for electronic use and exchange of information.” In addition, the Health Information Technology for Economic and Clinical Health (HITECH) Act, enacted as part of the larger stimulus package, establishes within the national coordinator’s office two committees to directly oversee policies and standards development and creates a position of chief privacy officer. Further, the HITECH Act stipulates that the HHS develop an initial set of standards covering various aspects of health information technology, as well as implementation specifications and certification criteria, by December 31, 2009. The standards will be voluntary for the private sector.
Another important provision of the HITECH Act provides assistance to educational institutions “to establish or expand health informatics education programs, including certification, undergraduate, and masters degree programs for both health care and information technology students.” This funding is essential to address workforce needs and particularly to foster implementation of EHR systems and the competent use of EHRs.
Perhaps most important, the HITECH Act provides rewards for those who make meaningful use of EHRs and penalties for those who fail to act. Beginning in 2011, approximately $17.2 billion will be distributed through Medicare and Medicaid to physicians and hospitals who use EHRs at least for electronic prescribing and information exchange to improve health care quality. Beginning in 2015, Medicare fees will be progressively reduced for physicians who do not use EHRs. This policy could provide the impetus for those wary of EHRs to take the plunge. Expanding demand for EHRs will also probably increase pressure to resolve the other barriers to EHR use.
So what might the next five years bring? It is my hope that the nation will be in the midst of a true transformation in health care and that the benefits of EHRs will spread to organizations across the land. I do not expect that every physician group or practice will be using EHRs, but the majority of them should be. And all hospitals should be moving toward adopting a learning culture through the implementation of robust EHRs; this learning curve should include expanding technical capabilities to support the use of EHRs and doing more to encourage patients to participate.
Advocates of expanded use of EHRs and EHR systems are emerging in all quarters. Growing numbers of health care professionals and organizations are expressing support. Students are showing increasing interest in careers in informatics and information technology. Patients are voicing their demands that EHRs be added to their arsenals for managing their own health. In government, officials at many levels are recognizing the growing support for developing a framework that protects patient confidentiality while enabling authorized use of patient data for public health and research.
For those of us who have committed our professional lives to the use of information and communications technologies to advance health and health care, the next five years are certain to be exciting. Some observers compare the current and future challenges to the Project Apollo Moon-shot program of the 1960s. Indeed, the tasks in aggregate are comparable in complexity. Changing the health care system will involve change at the most basic level in how health professionals do their daily work, while also empowering the public and patients to take a more active role in protecting their health. Both national efforts represent major investments in people and technology to achieve transformative visions. But as history shows, the United States did successfully reach for the Moon.
Jacob Goldstein, “Big Challenges Await Health-Records Transition,” New York Times, April 21, 2009.
Institute of Medicine, The Computer-Based Patient Record: An Essential Technology for Health Care (Washington, DC: National Academy Press, 1991).
Don E. Detmer (firstname.lastname@example.org) is a professor of medical education in the Department of Public Health Sciences at the University of Virginia School of Medicine, in Charlottesville, Virginia, and a senior advisor at the American Medical Informatics Association, in Bethesda, Maryland.