Medical Economics put questions about EHRs to experts and consultants in the field of health information technology. Their answers fall into two broad, interrelated categories: The products themselves, and the ways they are purchased and used. Read the results of our EHR survey.
If you use electronic health records (EHRs), chances are you’ve asked yourself some version of the question, “Why can’t I find an EHR that helps me provide better care for my patients, lets me run my practice more efficiently, and is easy to use?”
It’s certainly a reasonable question to ask. After all, EHRs have been on the market for about 20 years now, and in the last four years their manufacturers have received more than $30 billion in indirect government subsidies through the meaningful use program. Yet in the 2015 Medical Economics Physician EHR Survey, doctors’ mean ratings of their EHRs was only 5.3 out of a possible 10.
And even that result was more positive than some others. In a report published earlier this year by the AmericanEHR consortium, for example, only 33% of respondents said they were “satisfied” or “very satisfied” with their EHR, and fewer than a third said they would purchase the same software again.
Similarly, a landmark 2013 Rand Corporation study of factors affecting physician professional satisfaction found that EHRs lead to lower satisfaction due to the technology’s poor usability, the time required for data entry, and interference with face-to-face patient care, among other reasons. If consumers had similar opinions of technology giants like Apple or Amazon those companies would have been out of business long ago.
So why aren’t EHRs evoking more enthusiasm from their customers? And going forward, what will it take for vendors to create products that are more user-friendly? The latter issue becomes especially urgent as practices increasingly are paid based on patient outcomes, rather than the volume of patients they treat.
Medical Economics put those questions to experts and consultants in the field of health information technology. Their answers fall into two broad, interrelated categories: The products themselves, and the ways they are purchased and used.
Capturing the exam room interaction
While most EHR systems are pretty good at functions such as e-prescribing or providing a portal for patient communication, they fall short when it comes to doing what primary care doctors generally value most: capturing the exam room interactions that lie at the heart of the physician-patient encounter. That’s because such interactions are usually free-flowing and not easily recorded-if at all-by the templates on which most EHR systems rely.
“Doctors don’t walk into the exam room and say, ‘I’m going to ask you a series of questions the way it’s designed on this form,’” says Mark Anderson, FHIMSS, chief executive officer of AC Consultants in Montgomery, Texas, and a former hospital chief information officer. “They want the patient to tell them what the problem is. And the patient will answer, but usually not in the exact order of what’s in the EHR.”
Anderson acknowledges that in the last decade or so EHRs have improved in their ability to capture and summarize information such as patient histories, medications, and lab results. “Where the problem comes now is the need to document things like the review of systems, the history of present illness and the assessment. But the products just aren’t well designed for that,” he says.
The reliance on templates, and the lack of user-friendliness found in many EHRs, in part reflects the outlook of the people who design and build them, says Titus Schleyer, DMD, PhD, director of the Center for Biomedical Informatics at the Regenstrief Institute, an informatics and healthcare research organization in Indianapolis, Indiana.
“Software programs are the domain of engineering, of information scientists,” Schleyer says. And those people are not known for focusing a lot on the warm and fuzzies of how the systems interact with consumers.”
Schleyer notes that other industries, such as banking, have been using computers for longer, and thus have had more time to improve the usability of their consumer technologies.
“It takes a long time for concepts like user-centered design to make their way into various industries, and I don’t think those concepts have been sufficiently ingrained in the healthcare industry,” he says. “In many EHRs, ‘usability’ is kind of a final checkbox, ticked off after the software has been written.”
Compounding the problem, says Schleyer, is the difficulty in changing vendors, whether due to “data lock” or working in a health system or large practice where everyone has to use the same EHR. That slows the winnowing of products with poor usability evident in other industries.
“You see this Darwinian selection process at work a lot on the Internet, where stores with bad usability atrophy because it’s so easy for consumers to switch to another store with better usability,” he says.
Jacob Reider, MD, former deputy national coordinator for health IT in the Office of the National Coordinator for Health IT (ONC), says concerns for patient well-being also factor into EHR design changes. “They’re slow to evolve changes in functionality because the safety of the patients relies on the physicians really understanding what they’re doing,” he says. “You don’t completely change the controls in an airplane and then drop in the pilot and say, ‘go fly.’ It’s how people adjust to technology.”
Robert Rowley, MD, a family physician in Hayward, California, and former chief medical officer for EHR vendor Practice Fusion, Inc., believes the problem lies in vendors’ efforts to appeal to diverse segments of the healthcare market. The result is that “EHRs so far have been big, monolithic things. They try to be all things to everyone, and do a poor job for any given workflow,” he says.
Doctors don’t select the product
That problem can be especially acute in hospital systems or large, multispecialty practices that house a wide variety of specialties, each with its unique requirements. Thus an EHR that works well for one specialty may cause headaches for practitioners in another specialty. Or it may not work well for clinicians at all, but be very good at functions that are important to the people making the purchasing decisions.
“If you’re a doctor in a large organization, you don’t decide which system you’re going to use,” says Mark Friedberg, MD, MPP, senior natural scientist at the RAND Corporation and lead author of the 2013 study on physician satisfaction. “The decision as to which EHR that organization is going to use is not necessarily made by people who do patient care. Their primary interest is not necessarily facilitating patient care but meeting the other goals EHRs have, which are basically to fulfill meaningful use criteria and improve billing,”
“The person making the sales pitch usually is talking to someone like the chief financial officer or the chief information officer, not the front line docs,” notes Steven Waldren, MD, director of the alliance for e-health innovation for the American Academy of Family Physicians. “They want to know what is the cost structure, what’s the throughput of patients, and those generally aren’t the first concerns of the clinicians using the system.”
Adding to the frustration of many employed physicians, Friedberg notes, is the fact that hospitals and health systems often customize the systems they purchase, in ways that the physicians may or may not like, but can do little to alter. Vendors sometimes will cite this as a reason for user unhappiness, an explanation that Friedberg dismisses.
“It’s a little disingenuous for the vendor to say, ‘it’s not our fault, the hospital put in all these customizations that are making you unhappy,’” Friedberg says. “But the truth is, the hospital probably did that for a reason, which is that the core EHR product wasn’t functional. They needed to change it to make it work for their physicians.”
RAND’s 2013 study showed that while EHRs improved physician satisfaction in some ways and detracted from it in others, the key determinant for almost all doctors was the technology’s impact on patient care. “If they thought they were doing the right thing by their patient, they went home happy,” he says. “But if they felt like something was limiting their ability to do well by their patients they were frustrated and miserable.”
Smaller, physician-owned practices face a different challenge in purchasing an EHR. Unlike in a large system, the physician is making the buying decision. The problem there, Friedberg says, is “you’re taking people whose expertise is in clinical care and asking them to make a big technology purchasing decision. And it’s hard for them to know exactly what impact each product will have on their clinical life.”
Compounding their challenge is the difficulty of finding reliable information about many EHR products due to “gag clauses” that many of the biggest vendors include in their user contracts. “If you’re looking for an underlying problem causing a lot of these surface issues, that’s a biggie,” says Friedberg. “How is the industry supposed to evolve when you can’t talk about your EHR without violating the contract you signed with them? It would be like going to a restaurant and having to sign a document saying you can’t rate them on Yelp before you can get a meal.”
HITECH, MU impact
Underlying much of the discussion over EHR usability-or lack of it-is the impact of the Health Information Technology for Economic and Clinical Health (HITECH) Act of 2009, and the Meaningful Use (MU) program and EHR certification requirements that it spawned, on the EHR industry and on doctors.
Related: More trouble for EHRs?
Few people on either side would dispute that the HITECH Act has achieved its goal of dramatically increasing EHR use. The question is whether that increased usage has come at the cost of stifling innovation in the industry and creating greater unhappiness among EHR users.
“Because of the timelines in place [for implementing MU] the vendors had to move very quickly because they knew they had to get certified or become irrelevant in the market,” says the AAFP’s Waldren. “And what we found [in the AmericanEHR consortium study, in which the AAFP participated] was a massive decrease in user satisfaction in the 2011-2014 MU time frame. And I think it’s because vendors were required to get the products out ASAP, and so the functionalities weren’t aligned with what physicians actually needed to take care of the patient in front of them.”
“Physicians, like most consumers, will go where the money is,” notes Rowley. And if you’re being rewarded for incorporating EHRs, you’ll get one, even though it may be the wrong one. It’s like saying ‘we’ll subsidize you for buying a suit,’ and it turns out it fits you all wrong.”
Data presented at a September meeting of ONC’s Health IT policy committee support Waldren’s and Rowley’s theory. It shows the percentage of MU participants, known as “eligible professionals” who changed their EHR vendor quadrupled from 2% to 8% between the 2013 and 2014 program years.
Schleyer says EHR vendor representatives have told him that “innovation came to a halt because they had to focus on all the stipulations of complying with Meaningful Use.”
“That wasn’t a bad thing per se. The adoption curve of EHRs speaks for itself,” he adds. “But usability was somewhat of an unintended casualty on the road to getting more people to use these systems.”
For his part, Reider scoffs at the notion of a tradeoff between speed of adoption and usability. “The automobile industry must meet certain regulatory requirements, but does that mean they can’t make heated seats and other cool things? It’s a question of resource allocation.
“There are many ways to skin a cat, and many HIT companies only looked at the surface of what the regulations required and didn’t think of creative ways of meeting them,” he adds. Reider cites the example of the requirement to capture smoking status. Many pre-MU versions of EHRs had templates that merely asked how long the patient had been smoking and how many packs per day. For MU, however, doctors had to classify patients according to one of eight Centers for Disease Control and Prevention-designated categories ranging from “never smoked” to “heavy tobacco smoker.”
“Companies that weren’t very thoughtful about this created a new workflow, pointed the finger at ONC and said ‘those bastards made us do it this way. Don’t blame us, blame them,’’’ Reider says. “Smart companies maintained their existing workflows, then mapped the concepts those captured to the concepts required for MU. The workflow wasn’t changed but the MU requirement was satisfied. Others said, ‘we’re just going to shovel this crap onto our customers and blame the government.’ Then the customers called us and said ‘you guys made my user experience miserable.’”
Reasons for hope?
So are doctors doomed to eternal unhappiness with their EHRs? Not necessarily. Experts say there is reason to believe that the technology’s usability will improve, and customer satisfaction along with it. But the process will be slow.
“The more that advocates for usability publish studies and get engaged in the industry, the more that awareness [of the importance of usability] takes hold,” says Schleyer of the Regenstrief Institute. “Of course, you also need a certain armamentarium, to know the methods. For a company just to say ‘starting tomorrow we care about usability’ is not going to fix the problem. You have to hire people who are trained in human-computer interaction and build user-centered design groups, and so on. And all that takes time.”
The user experience committee of the Healthcare Information and Management Systems Society (HIMSS) includes “very passionate clinicians, who want systems they can trust and that make it easy for them to provide safe care,” says Rod Piechowski, MA, senior director for health information systems at HIMSS. “I think you’ll see a lot more voices being raised about that in the coming years.”
Reider says that before leaving ONC last year he had begun to see EHR vendors focusing more on user-centered design, but that the improvements haven’t yet shown up in the products themselves. “Their release cycles for upgrades to the products can be quite slow,” he says. “So there may have been work done a year ago that hasn’t made it into the production version of a product. And even if it has gotten into the production version, it’s possible customers haven’t made that upgrade.”
Usability is actually among the criteria for EHRs to attain 2014 Meaningful Use certification, Reider notes, albeit in an indirect fashion. For a subset of certification criteria, he says, vendors are required to demonstrate to ONC that they incorporated a user-centered design (UCD) process (as defined by the National Institute for Standards and Technology). ONC then would publish the results on its website. “Our hope was that by requiring the publication of usability testing that usability would improve,” he explains.
A research letter in the Journal of the American Medical Association examined EHR adherence to the certification requirements for UCD among the 50 vendors with the largest number of providers attesting to MU certification with their products from April 2013 through November 2014. Of the 41 reports available for review, 34% had not stated their UCD process, 63% used fewer than the standard of 15 test participants, and only 22% used at least 15 participants with clinical backgrounds.
Asked to comment on the study, Reider replied by e-mail that “the key policy objective here was in fact reached: it’s very transparent which authorized certification bodies enforced the certification requirements, and it’s very transparent which health IT developers met those requirements. Anyone can go to the CHPL [Certified Health IT Product List] and review this information…. I would certainly avoid purchasing any health IT product that sidestepped this certification requirement, or responded to the requirement with a vacuous ‘check the box’ answer. We’ll have to wait and see if physicians and hospitals will make the same decision as I would. “
The usability of EHRs will only grow in importance as healthcare moves away from fee-for-service and towards value-based payment models-a trend that is bound to accelerate as a result of Congress’s approval earlier this year of the Merit-Based Incentive Payment System. Earlier this year the Rand Corporation published a study looking at the effects of healthcare payment models on physician practices.
“One of the main takeaways was that as payment models shift from fee-for-service to other kinds of payments, the importance of having a data infrastructure and performance data that can be used in actionable ways increases dramatically,” says RAND’s Mark Friedberg, the study’s lead author. “You have to know who your patients are, who are your high utilizers, and the reasons that make them high users so you can target appropriate interventions. And you just don’t do that without an EHR coupled with some pretty hefty in-house analytics.”
The study also revealed a surprising level of optimism among clinicians regarding EHRs’ potential. “I don’t think we’ve yet gotten to the point where the reality of EHRs has stemmed the general enthusiasm for the technology,” Friedberg says. “They love the technology when it works and there’s great optimism that it’s going to work better. But there’s also impatience.”