The Electronic Health Record Problem
It’s no secret that many physicians are unhappy with their electronic health records (EHRs). They say they spend too much time keying in data and too little making eye contact with patients. They say their electronic records are clunky, poorly designed, hard to navigate, and cluttered with useless detail that colleagues have cut and pasted to meet documentation requirements. Meanwhile, the data they really need are buried almost beyond retrieval.
Not all physicians feel this way. Two-thirds of primary care physicians say there are satisfied with their current EHRs, according to a 2018 survey by The Harris Poll. But the critics have a point. Current EHRs are not well-designed to meet the needs of users. And they don’t do enough to make clinicians smarter and more efficient. This doesn’t mean we would be better off in the paper world of 10 years ago. But it does mean that EHRs need improvement.
As we think about improving them, we need to broaden the discussion of EHRs and their role. We need to reckon with the underlying causes of EHRs’ problems, how to correct them, and how to ensure that their enormous potential benefits are understood and realized.
EHRs are a technology. Like most technologies, they can be used in a variety of ways for a variety of purposes. Their human masters decide.
In our current health system, EHRs have one critical performance requirement: generating clinical revenues. In the fee-for-service world, this means supporting providers’ billing and documentation to generate as much revenue as possible for each clinical service. EHRs also must help providers meet regulatory requirements that may have financial or accreditation implications.
This means that current EHRs were not created to support many of the things that physicians, patients, and policymakers value: better care experiences, reduced costs, or improved care quality and population health management. They were not created to make physicians better diagnosticians or more cost-effective prescribers. The reason: our health care system has mostly not rewarded these activities. They have not been mission-critical for providers or, therefore, EHR designers.
For that reason, EHRs have only the most minimal capabilities related to clinical decision support, which has been proven to increase the quality of care, or to the collection of information on duplicate and unnecessary testing, or on the aggregate health of providers’ patient populations.
To put it simply, improving EHRs will require changing the priorities governing their design. That means moving away from fee-for-service payment toward risk-sharing by providers and, ultimately, some form of prospective compensation. Until then, optimizing the usability and value of EHRs will be an uphill struggle.
EHRs’ Undervalued Benefits: Empowering Patients and Advancing Human Health
Because the benefits of EHRs may be less visible than their burdens, some of their contributions are overlooked and undervalued.
One of these benefits is giving patients access to their medical information. Meaningful-use requirements spurred the adoption of patient portals, which, though sometimes clunky, have enabled patients for the first time to routinely see their test and procedure results. Patients can also now download their entire digital record and share it with third parties that can analyze its contents and educate them on their significance. Apple, for example, has agreements with over 100 health systems and practices to perform this function, which is likely to spawn a deluge of consumer-friendly health care applications based on patients’ own information.
Another underrated EHR benefit is that, by capturing billions of patient encounters worldwide, electronic records are generating a vast store of digital health data that are available for novel uses, including research into the causes and cures of disease and the detection and prevention of threats to public health.
Think of these data as the equivalent of a new natural resource, like water or minerals; they sit in the cloud, ready for extraction, refinement, and application. Their value is increasingly understood by technology companies, new startups as well as old stalwarts, that are pouring billions into exploiting them. There are obvious privacy and security issues raised by this development. But never before in human history have we had access to this novel (un)natural resource.
In entering all that data at the point of clinical care, health professionals and patients are creating a public good. But they get little tangible in return — at least in the short run. This maldistribution of benefit and cost lies at the heart of the current EHR controversy.
To make health professionals’ work easier, and to exploit the vast potential of EHRs, a number of interventions make sense.
The most important is unrelated to the technology. Clinicians unhappy with EHRs have a huge stake in moving from fee-for-service to value-based payment, so that providers and their EHR vendors start to prioritize the production of health and the reduction of waste in health systems. This will reduce documentation requirements, spur the creation of decision support and information exchange that make clinicians’ lives better, and focus attention on getting value from the information so laboriously recorded by doctors and other health professionals.
A second requirement will be to lower the burden of data entry. Many providers have started using scribes to take notes during visits. While many physicians love scribes, they are expensive. A better long-term solution would be to use natural language processing and artificial intelligence to enable clinicians’ conversations with patients and their subsequent assessment and treatment plan to be recorded in real time. Given the increasing power of these technologies, such applications will soon be available.
Another approach to assisting data entry is to systematically redesign records for ease of use and to prune away unnecessary recording requirements. A recent New England Journal of Medicine commentary provided an excellent example of the benefits of this intervention.
A third requirement for EHR improvement falls to health professionals. When I was a medical student, I spent hundreds of hours learning how to take notes in the paper world. More experienced clinicians reviewed and graded these write-ups. Later, as a young physician, I observed the notes of clinicians I admired, and emulated them. This process of professional education in record-keeping unfolded over years and forever shaped my note-writing habits. If physicians are unhappy with how their colleagues use EHRs, they should start educating young physicians — and their peers — on how to properly keep records in the electronic world. What and how data get recorded are ultimately a professional responsibility.
Lastly, we need to find a way to correct the maldistribution of costs and benefits that now plagues the use of EHRs. By creating vast troves of electronic data and enabling patient empowerment, clinicians and their patients perform a valuable public service that has thus far been unrecognized and unrewarded. Reducing the cost of data entry will help, but as the benefits of EHRs and their data become monetized — as they will — some way to share those gains with clinicians and patients at the frontlines should be considered. This could be accomplished in a variety of ways such as voluntary contributions from businesses that rely on EHR data to an EHR innovation fund and/or directing a share of the taxes paid by these businesses to EHR improvement. But at least until EHRs become much more user friendly, this problem of unfair allocation of benefit and cost needs attention.
We are not going back to the paper world, but EHRs need to work better. As they pursue this goal, clinicians, policymakers, managers, and vendors need to understand and address the root causes of the problem they are trying to solve, and the full array of options for addressing it.