Electronic health records lack tools to improve clinical care

Author
Lloyd B. Minor, M.D., is professor of otolaryngology — head and neck surgery and dean of the Stanford University School of Medicine.

Early in my career as an  eye surgeon, an ear nose and throat surgeon told me about two different patients who came to him with the same set of bizarre symptoms. Certain noises made their eyes move involuntarily, and objects they were looking at appeared to move around in patterns. After months of careful investigation, these patients led me to discover a rare disorder called superior semicircular canal dehiscence and to develop a corrective surgery for it.

Today, anyone with Internet access can search the symptoms of superior semicircular canal dehiscence and get multiple hits for diagnosis and information about treatment. Yet most doctors who use electronic health records are years away from this kind of capability. There is no search engine to support our clinical decision making. That essential part of health care remains a practice of informed guesswork, and we are often unable to access information that could improve decisions in the moment of care and make health care far safer and more effective and efficient than it is today.

A recent article published jointly by Kaiser Health News and Fortune and my own personal experience are stark reminders for me that electronic health records have not yet lived up to their true potential. In addition to highlighting their inability to share information across proprietary platforms, as well as the fact that physicians report spending more time than ever on data entry rather than interacting with patients, the in-depth investigation uncovered thousands of incidents in which errors caused by faulty electronic health records harmed patients.

As a physician, I find this deeply worrisome. A full decade after passage of the Health Information Technology for Economic and Clinical Health (HITECH) Act — legislation that originally set aside $27 billion to incentivize physicians and hospitals to adopt EHRs — we still have a long way to go. I see little if no progress in remediation of this serious deficiency.  

I see this failing to progress as the total lack of incentive for existing vendors to re-write their software. The federal government has overemphasized MACRA, MIPS as measures of meaningful use, a bureaucratic means of totally ignoring this situation. Their measures are a poor measure of quality and outcomes. They have mistakenly or on purpose tied reimbursement to these measures.  If anything it has reduced the quality of care by overloading physicians and professionals with clerical duties, mindless clicking, opening and closing screens, in a search for information and entering new information. This is not a new problem. For the past ten years physicians, and their organizations have been vocal about the loss of patient facetime, longer work days, and physician burnout.  Many doctors have retired well before the end of a useful career life, choosing to do non-clinical work. This is adding to the physician shortage. It is an easy calculation.

In 2018, a Stanford Medicine/Harris Poll found that nearly half of U.S. primary care physicians said that electronic health records actually detract from their effectiveness as clinicians, and 44% said they believed that the primary value of these systems is data storage. Far from being a transformative health care tool to support clinical decision-making, a large portion of physicians feel they have traded physical filing cabinets for digital ones.

Electronic health records still have the potential to make health care more predictive, preventive, and precise — but only if we can achieve sustained collaboration among health care providers, technology companies, and health insurers to address their shortcomings. One step in that direction took place on Stanford’s campus last June, where we convened leaders in patient care, technology, design thinking, and policy to discuss a path forward for electronic health records. In principle, the group agreed on three points:

First, electronic health record systems must become interoperable, meaning that a doctor using an Epic system should be easily able to send patient information to a doctor using a Cerner system, or one from athenahealth. Fewer than one in three hospitals can functionally share and use patient information received from another provider. This is the most important challenge, one that will require a combination of technical and operational solutions. The health care industry, including insurers, must agree to common technical standards to effortlessly exchange data, and providers must enforce these standards through shared contracting requirements with technology vendors.

Second, electronic health records must be redesigned to better respond to physicians’ needs. Doctors complain about the dozens of clicks it can take to order a simple test or submit a prescription. Physicians and developers must work together to build new systems or update existing ones so they better reflect the rhythms of clinical care. This will take time to build and training to implement, but such investments are worthwhile given their potential to eliminate well-documented frustrations that physicians have with their electronic health records.

Third, building a more clinically relevant electronic health record system should incorporate artificial intelligence that can synthesize anonymized patient records; combine them with the medical literature, and provide insights at the point of care. Artificial intelligence, neural networking and deep learning were unheard of at the beginning of the HIT age. Now we have the computing power to perform this kind of analysis, the anonymized records to study, and vast swaths of digitized medical literature. Now it’s time to make better use of them. Though this may sound ambitious, there is encouraging work being done.

For example, a collaboration between Google and three academic medical centers — the University of California, San Francisco, the University of Chicago, and Stanford Medicine — is testing the ability of artificial intelligence to analyze raw electronic health records and generate accurate predictions about patient outcomes based on 46 billion pieces of anonymous patient data. While early in development, this effort has shown promise and could one day help physicians extract greater value from their electronic health records. Even so, this initiative and others like it must contend with fundamental gaps in our health care system’s IT infrastructure.

Issues of interoperability and user interface optimization have been successfully addressed in industries as complex as aviation, telecommunications, and banking. You can easily withdraw money from any ATM in the world because the industry came together to design secure systems that would talk to each other to satisfy consumer demand. These examples provide valuable lessons and hold out hope that the health care sector will achieve its long-awaited transformation.

Electronic health records lack tools to improve clinical care - STAT: To make electronic health record systems clinically relevant, they need artificial intelligence that can help provide insights at the point of care.