The benefits of electronic medical records (EMRs) have been well established, and the digital tools continue to promote improvement as they are more widely incorporated into the healthcare system. But at the same time, EMRs bring drawbacks and challenges that frustrate clinicians and detract from what could be a more uniform improvement in care.
A recent survey conducted by Stanford Medicine and The Harris Poll found that although most physicians — a clear 63% — think EMRs have improved care, there still is a substantial physician population that doubts the overall value of the technology.
Forty percent of the physicians surveyed said there are more challenges with EMRs than benefits.
The survey also found that “62% of time devoted to each patient is being spent in the EMR.” And 49% of office-based primary care physicians “think using an [EMR] actually detracts from their clinical effectiveness.”
Seventy-one percent of surveyed physicians say EMRs greatly contribute to physician burnout, and 59% say EMRs need “a complete overhaul.” Forty-four percent say the primary value of EMRs is data storage, with only 8% saying the primary value is clinically related.
Almost three-fourths (72%) think improving EMRs’ user interfaces should be the first move to improving their usefulness. Sixty-seven percent say that “solving interoperability deficiencies should be the top priority for EMRs in the next decade, and 43% want improved predictive analytics to support disease diagnosis, prevention, and population health management,” the report says. (The report is available online at: https://stan.md/2ARDL83.)
Key Driver for Burnout
Research by The Doctors Company, a medical liability insurer based in Napa, CA, reveals similar concerns. A recent nationwide survey of the insurer’s members revealed that EMRs are causing high levels of frustration for physicians and are a key contributor to burnout, says chairman and CEO Richard E. Anderson, MD, FACP.
“Additionally, more than half of the 3,400 doctors we surveyed thought EMRs had a negative impact on their relationship with patients,” Anderson says. “Even more said the technology hinders efficiency and productivity.”
The problems manifest in the day-to-day operations of a hospital or physician practice in two main ways, Anderson says. “The first is that they require hours of duplicative and often unnecessary data input, most often by the physician. The second is that the technology itself has failed us; the systems are non-intuitive, vary widely, and don’t connect to one another,” he says. “EMRs increase costs and reduce productivity in the majority of practices that use them.”
A common complaint is that EMRs have become the required interface for anything a physician wishes to do, Anderson says. Instead of a being a tool that the doctor can choose to employ in the most effective way for a particular patient and situation, the EMR demands that the doctor input data and navigate through layers of questions before being allowed to order tests, for instance.
“Mandatory conversations with the computer have devalued the doctor-patient relationship. This is enormously frustrating to physicians seeking to provide quality care,” Anderson says “When a technology impedes, rather than facilitates, patient care, it’s a terrible outcome for all involved — particularly the patient.”
The good news is that there is a growing awareness of this problem, and practitioners are starting to speak out, Anderson says. The president of the American Medical Association (AMA) recently said that EMRs are to blame for physician burnout, calling it “abuse.” EMRs are turning doctors into data entry clerks, AMA president Barbara McAneny, MD, said at the AMA’s Interim Meeting recently in National Harbor, MD.
For physicians, every hour spent directly caring for patients results in two hours of EMR data entry and related tasks, she said, citing a 2016 study. (That study is available online at: https://bit.ly/2MBRXpG.) McAneny called much of the current EMR technology “dysfunctional” because it does not provide doctors with the information needed to effectively care for patients.
Gaps in Organization
About half of practicing physicians are facing burnout, McAneny said, and the primary cause is the EMR.
“Doctors are spending excessive time on data entry, contributing to physician burnout, with implications for quality of care,” McAneny told the meeting attendees. “It grew out of the billing software, so it doesn’t give us the decision support or the information we need. The vendors of these systems like to paint doctors as Luddites who don’t like technology. They need to understand that we love technology — we just want technology that works.”
AMA CEO James Madara, MD, also told the attendees that EMRs suffer from “vast structural gaps (in) achieving true data liquidity and interoperability” and “gaps in how clinical data is organized at the point of care.” He said the lack of reliable and well-organized data leaves physicians “driving a car with a windshield covered in snow.”
AMA is addressing the problem with a new digital platform, the Integrated Health Model Initiative. It is designed to address some of the problems related to EMRs with a common data model. (More information on the initiative is available online at: https://ama-ihmi.org/.)
Hospitals should take active steps to mitigate these negative issues associated with EMRs, Anderson says. The goal should be to ensure EMRs serve their intended purpose: improving patient care, he says.
“A lot of money has been invested in EMRs. There is a growing recognition of the problems, but recognition is not sufficient. We need to see constructive plans for bringing about resolution,” Anderson says. “Hospitals have an important role to play in this conversation. They owe it to their physicians — and their patients — to address the problems electronic health records create in the hospital setting.”
Informatics Can Be Useful
The nearly universal adoption of EMRs in a decade has transformed healthcare, but getting the most benefit from them — and avoiding some of the potential problems — requires dedicated professionals in informatics, says Douglas Fridsma, MD, PhD, FACP, FACMI, president and CEO of the American Medical Informatics Association in Bethesda, MD.
“We see it in the United States and in other countries adopting EMRs. They adopt the technology, but they don’t see the benefit because the workforce isn’t capable of turning all that information into actionable items,” Fridsma says. “It’s like giving everyone a stethoscope and not teaching anyone cardiology. They have this tool, but they don’t know how to interpret the signals and information they’re getting from it to turn it into actionable items.”
The wealth of data available from EMRs means “the future is here, but it’s not evenly distributed,” Fridsma says. The most successful healthcare organizations look at data not as an operational asset but as a strategic asset, he explains.
“It’s not just about running this data operation. It’s about asking the right questions of the data, taking the answers from the analytics, and applying it back into the healthcare process,” he says. “Use that information to guide where the organization is going and how to take care of patients. That requires understanding that the data inherent in the electronic record is a strategic resource that can be used to drive your organization forward, rather than just numbers that tell you where you’ve been.”
Hospital Improves EMR
Provider burnout is a primary concern with EMRs, but so is provider complacency, says Shannon Sims, MD, PhD, chief analytics officer with Vizient, a healthcare performance improvement company based in Irving, TX. He previously worked in similar roles with hospitals.
Burnout is largely related to the amount of time it takes to properly document patient care in the EMR, he says.
“That’s both the number of clicks, the user interface that they don’t like as much as more traditional ways, and also the feeling that the burden of a lot of tasks has shifted from administrative staff to the physicians and other providers,” he says.
Sims worked with a large hospital that wanted to address those concerns. Emergency physicians had complained that they were spending an excessive amount of time documenting care and otherwise interacting with the EMR beyond their normal shift times, so the hospital sent data analysts to determine how much that actually happened and find ways to reduce the EMR time.
“They developed a metric for a proxy of the qualitative sense of how much time physicians spent with the EMR and deployed a host of interventions. Some of it was very workflow-focused, reducing clicks to minimize interruptions to the physicians providing care,” Sims explains.
“Some of it was training-based as well, making sure the physicians and the rest of the team understood how to best use the system,” he adds. “They were able to see substantive reductions in the off-hours documentation, and surveys showed an increase in provider satisfaction.”
Doctors Can Grow Complacent
Provider complacency involves the tendency of some physicians to rely too heavily on the EMR to guide the course of treatment. As much as physicians resist the idea of “cookbook medicine” in which they must adhere to rigid protocols, some physicians let their guard down and put too much trust in the EMR to steer them to the right clinical pathways, Sims says.
“Attending physicians in particular are concerned about how some physicians assume the EMR is paying attention to detail and doing all the heavy thinking, catching potential mistakes. It’s lulling the care team into a sense that ‘the computer is doing it for me, so I don’t have to pay attention to drug interactions,’ for example,” he says.
“They worry that some doctors think the order sets or evidence-based protocols take care of all the contingencies, so they don’t have to worry about it. I hear this concern a lot, particularly about the impact on trainees, from medical students to residents and fellows, [how EMRs are] diminishing the amount of thought that goes into patient care.”
Sims notes that the problems with EMRs are not vendor-specific. Any brand of EMR can introduce these problems to a hospital, with the likelihood and significance of the challenges coming mostly from how the system is designed and implemented, he says.
No Single Solution
Once a hospital recognizes EMR-related problems, there are different ways to address them, he says.
“There is no one-size-fits-all solution. Different organizations tackle it in different ways and with different prioritization,” he says.
“Some organizations make it the top concern and address it accordingly, while others make it a priority but not at the top of the heap. We’ve found that those that put it at the top of the heap are having a higher dissatisfaction with the EMR, and that is causing high turnover, which is very costly for the organization — sometimes hundreds of thousands of dollars for a physician when you count the lost productivity and the cost of finding a replacement.”
EMR efficiency often must be addressed with a mix of strategies to respond to subjective, qualitative concerns by clinicians — how they feel and what impact they think the EMR is having — and with a more data-driven approach that uses metrics to document improvement.
“We’ve found that the highest performing organizations address both of those pieces of the puzzle, typically by installing physician-led governance over the EMR and creating teams to address specific parts of the workflow,” Sims says. “That will mean teams for order sets, clinical decision support, speech recognition, and others.”
Indiana University (IU) Health in Indianapolis is in the process of improving an EMR that had grown organically over 15 years but had given rise to issues that threatened quality of care. IU Health is Indiana’s largest healthcare system, with 18 facilities and almost 33,000 employees.
Seung Park, MD, senior vice president and chief health information officer with IU Health, joined the organization in the spring of 2017 and began improving the IU EMR system. A key issue to address was the average transaction response time (ATRT).
In 2017, that figure was a maximum of 0.72 seconds, and the percentage of transactions greater than five seconds was 1.7%. The national average for transactions greater than five seconds is 0.5%.
“As IU Health acquired new facilities or as our facilities expanded, we threw more and more on to that EMR, which is a common story. By the time I arrived, our EMR was the slowest performing Cerner EMR in the world,” he says.
“We had 10 times the number of alerts as the national average, 10 times the nursing paths, and 10 times the number of required data entry fields in forms like the standard nursing patient intake assessment for inpatient care.”
Park found that part of the problem was how over the years IU Health had completely customized the product’s standard inpatient nursing format. Park says that created an untenable situation with a “slow, customized EMR that couldn’t even bear its own weight. It’s no wonder that our clinicians were dissatisfied with that EMR.”
The first step in improving the EMR was to declare a moratorium on further changes to the system until the issue could be studied and solutions developed.
An analysis revealed how far IU Health’s EMR metrics were out of line with the averages of nationwide users. It also revealed that another part of the issue was that the health system had not properly updated the EMR product.
“We were using the EMR as if Cerner had not updated the system since the 1990s, when we first adopted it,” Park says. “We systematically began broadcasting the message that we were going to move to a single, standard, modernized EMR and that we were not going to allow for individual physician or nurse customization of the EMR anymore. We had to first break down what was wrong before we could build what is right.”
That message was received more positively by nurses and physicians than Park expected. Working with an EMR under extensive reconstruction creates difficulty in the daily care of patients, but for the most part, clinicians accepted the challenges with good cheer, he says.
“Every end user of a system believes that he or she knows what he or she needs, whereas in actuality those things are wants and not needs,” Park says.
“User feedback is very important, but at the same time, you have to do the calculus to determine if this is an actual need or a want. If I do this to the EMR, does this apply to and benefit everyone or only this one individual? If I make this change now it will be give me a short-term gain, but what is the long-term gain or loss?”
Other challenges involved things outside of IU Health’s control. If the EMR’s manufacturer-provided server is down, clinicians blame the EMR itself when the problem actually lies with the network structure supporting the in-house EMR.
IU Health reverted to using the EMR’s standard format for inpatient nursing solutions, which helped decrease the ATRT from 2.39 seconds to 0.4 seconds. The health system also removed 90% of the alerts in the EMR, another aid in reducing the transaction time.
All of the metrics that were so off track in 2017 are now at or near national averages, Park says. Physician burnout measures are down by about 30%, and Park says he is looking forward to further reductions as the EMR improvements take hold. IU Health physicians currently score just under 2.0 in the Physician Well-Being Index, a measure of physician satisfaction and likelihood of burnout. The national average is 1.77, and IU Health physicians scored 1.99 recently, down from 2.27 previously.
“We are continuing to put into place a culture that accepts a standardized EMR because that is the only way we are going to be fast and the only way we are going to get to the streamlined future that our clinicians so desperately need and deserve,” Park says.
“This takes discipline and backbone, the willingness to tell the right story over and over. You have to be willing to partner and be humble if you’re ever going to get people to buy into the idea that this is the right thing to do.”
- Richard E. Anderson, MD, FACP, Chairman and CEO, The Doctors Company, Napa, CA. Phone: (800) 421-2368.
- Douglas Fridsma, MD, PhD, FACP, FACMI, President and CEO, American Medical Informatics Association, Bethesda, MD. Phone: (301) 657-1291.
- Seung Park, MD, Senior Vice President and Chief Health Information Officer, Indiana University Health, Indianapolis. Email: email@example.com.
- Shannon Sims, MD, PhD, Chief Analytics Officer, Vizient, Irving, TX. Phone: (800) 842-5146.