3 Ways to Make Electronic Health Records Less Time-Consuming for Physicians

Physicians in the United States are justifiably upset by the amount of time they spend using electronic health records (EHRs). This is true across primary care physicians and specialists, and it contributes to physician burnout. The annual cost of physicians spending half of their time using EHRs is over $365 billion (a billion dollars per day) — more than the United States spends treating any major class of diseases and about equal to what the country spends on public primary and secondary education instruction. This is a problem that can be solved now by taking three steps.

1. Standardize and reduce payer-imposed requirements. A recent study of U.S. health systems using the same major EHR system found that at the median organization the length of the average patient note in an EHR has more than doubled from 2009 to be about 700 words. To put the issue into context, if a physician sees 15 new patients, he or she would have to enter about 20 single space pages into the EHR. The study found that the average note length at the median health system in Canada, the UK, Australia, the Netherlands, and Denmark is less than one third of what it is in the United States.

These astounding findings indicate that a big part of the problem is the documentation requirements that payers impose on providers. Making matters worse, these requirements vary from payer to payer in the United States. Standardizing and rigorously reviewing their utility is essential. The U.S. Centers for Medicare and Medicaid Services (CMS) is beginning to make strides in reducing requirements with its Patients over Paperwork initiative, and we believe that private payers should adopt the same principles and agree on a set of standards, requiring documentation only when it truly adds clinical value.

In a fee-for-service payment system, there will always be a desire by payers to impose documentation requirements to try to limit the claims they pay. But as the United States shifts to other payment models, it has the potential to lessen the need for documentation to justify billing. Christopher Longhurst, the CIO and associate chief medical officer for quality and safety at UC California San Diego Health and a coauthor of the EHR note length study, explained that organizations that operate under risk-based payment models (in which the volume of services less directly influences payment) had shorter average note lengths. The Veterans Health Administration, which operates with fixed budgets and salaried physicians, has comparatively low documentation requirements.

2. Continuously improve EHR workflows. This is something every provider can do right now. There is significant potential to improve user workflows without any regulatory changes or technology innovation. Colleagues who have seen EHR implementations across multiple organizations estimate that there is the potential to improve workflows in the EHR by about 20%, on average, by removing steps that don’t have any value. For instance, Geisinger, which serves communities in Pennsylvania, streamlined the work to get patients to the right musculoskeletal provider from a frustrating multi-click process (or a five- to 10-minute phone call) to one that simply asked two questions: What is the patient’s complaint? and What is the location of the injury? This led to significant increases in provider satisfaction and decreased time for patients to be seen.

Jim Noga, CIO at Boston-based Partners HealthCare, believes that improving EHR workflows should be like painting the Golden Gate Bridge and ought to be done continually. Mark Vrahas, chair of orthopedics at the Cedars-Sinai Medical Center in Los Angeles, has been through EHR implementations at three hospitals and now assiduously reevaluates processes so that he does not add a click in an EHR workflow without also removing another one. Part of the process of optimizing EHR and related clinical workflows should also include examining the clinician or staff that should be involved in performing each activity. For instance, much of the task of entering information can be shifted from physicians to other staff.

3. Unleash innovation. Tech advancements — such as voice recognition, digital scribes, and connected devices — are already beginning to further automate and reduce time spententering information into the EHR. But once all of the information is in the EHR, clinicians still need help with the other half of the problem: the EHR user experience, which is widely viewed as being many years behind that of other industries.

Innovation is needed to enable clinicians to receive contextually relevant insights without having to comb through reams of unstructured information in the EHR. Innovation is also needed around designing better user experiences and reconsidering what form factors (e.g., mobile) will work best. Today’s EHR user experience is centered around a desktop computer even though by the end of 2018 about two-thirds of website visits in the United States were expected to be from mobile devices.

Where will all this innovation come from? Historically, the EHR vendors developed the software themselves and largely did not open their platforms to others. This is beginning to change: They have recently introduced app stores for third-party developers akin to what smartphone companies did a decade ago. Examples include Epic’s App Orchard, Cerner’s App Gallery, and athenahealth’s More Disruption Please program.

We believe that third-party innovators have the most potential to dramatically improve both the user experience of clinicians and the health of patients. For example, using an open source framework called CareKit, Johns Hopkins developed the Corrie health app, which has helped reduce 30-day readmission rates from 19% to 3% for heart attack patients. Key to enabling similar experiences and fostering further innovation will be the creation of stronger standards so developers do not need to rebuild their apps for each EHR. One hopeful sign is the Argonaut Project, which has been steadily expanding the set of data elements in the Fast Healthcare Interoperability Resources (FHIR) API over the last several years. Another is CMS, as part of its Promoting Interoperability Program, is requiring health care providers as of January 2019 to make their data available through the FHIR API for patient-facing apps to query. For instance, a new feature from Apple, Health Records on iPhone, allows people to securely store and aggregate their own health records across their medical providers.

Another benefit of further standardizing APIs across EHRs is that it will allow data from one EHR to be more seamlessly integrated into another so that a physician can easily see a patient’s full medical history. The groundwork is already being laid to enable EHR-to-EHR data exchange. In November 2018, the two leading networks of EHRs, Commonwell Health Alliance and Carequality, both nonprofits, announced that providers on their networks would now be able to bilaterally exchange information with one another. This is akin to the Eastern and Pacific railroads coming together to form the first U.S. transcontinental railroad in 1869.

We are optimistic that better days are on the horizon for clinicians and that we are past the nadir of the EHR usability problem. Improvements will not occur automatically though, and there needs to be widespread recognition that physicians spending half their time using EHRs is a health care crisis that must be fixed. There should be a mandate for payers to standardize and reduce their documentation requirements. Equally as important, we should strengthen APIs and secure data-sharing standards to unleash transformative innovation. These efforts, combined with providers diligently reviewing and optimizing their EHR workflows, will transform EHRs from being a time sink to a time saver and joy to use.