EMR Technology Experiences Growing Pains: Resistant Doctors, Computer Glitches, and Unrealized Benefits

Last week I had my first visit with my new primary care doctor. I picked him based on recommendations (plus he’s one of the few that accepts my insurance), and also because he seemed to be an eager adopter of electronic medical records (EMR). On his website, there was a portal for making appointments on-line, asking questions of the doctor and staff by e-mail and once a registered patient, I could also use a secure system to access my medical records. With EMRs being portrayed as key drivers of quality and savings in health reform, I felt encouraged by my new doctor’s embrace of the technology.

But when he greeted me in the examining room, I was surprised to see the medical assistant hand my doctor a pad of paper with my height, weight and blood pressure written on it. As we talked and he examined me, he wrote notes down on the same pad—even though there was a computer in the room. When I asked how he felt about his EMR system, he said it was a great advance for his practice—but unfortunately it had crashed  that morning and the “tech guy” said it might take a while to get it back on track. “By next week or so we should have you in our system,” he sheepishly explained.

The federal government is committing some $27 billion over the next 10 years to support the adoption and use of electronic health records by doctors and hospitals. On July 13, they released “meaningful use” criteria for hospitals and doctors to meet in order to qualify for financial incentives—as much as $44,000 (through Medicare) and $63,750 (through Medicaid) per clinician and millions for hospitals, depending on their size. These criteria—which are described in detail here are meant to promote the use of EMRs for improving patient care and quality, not just to aid in billing. Maggie recently wrote this post about how electronic records can lead to truly “patient-centered” care.

The incentives are meant to spur more practitioners and hospitals to go the way of my new doctor. But although he saw the promise of technology early, in reality, many American hospitals and physicians have been quite slow in embracing the idea of electronic medical records—with large innovative medical centers like the Geisinger system and younger practitioners being among those quickest to adopt the technology.

In a study published in January 2009 that compared physician use of electronic medical records in seven countries, the Commonwealth Fund found that only 28% of US doctors used EMRs, vs. 89% in the UK and a whopping 98% in the Netherlands.

Another finding from this study: “In the United Kingdom, 88 percent of primary care physicians can easily generate lists of medications that patients take, including prescriptions from other doctors, compared with only 37 percent of doctors in the U.S. The ability to generate medication lists promotes coordination of care among doctors and can help prevent medication errors.”

The Center for Disease Control's National Center for Health Statistics found that in 2009 an estimated 43.9% of doctors reported using full or partial EMRs in their office-based practices. Yet this increased use of the technology is deceptive; just 20% of  the physicians were using “basic systems,” and only 6.3% were employing "fully functional" systems in 2009. A basic system can perform the following functions: provide patient demographic information, patient problem lists, clinical notes, orders for prescriptions, and allow doctors to view laboratory and imaging results on-line. Systems defined as fully functional include all that and more: they can store a patient’s medical history, execute electronic prescribing and test ordering, provide warnings of drug interactions and reminders for guideline-based interventions, among other features.

The rest of the doctors (about 18%) were only using individual elements of the EMR, like electronic prescribing or computerized ordering.

With the push on to get all physicians and hospitals using EMRs by 2015 (or face financial penalties in the form of lowered Medicare reimbursement) it’s important to understand what has kept doctors from fully embracing the technology. It’s not just the financial investment needed to set up EMR systems, says Bruce Carlson, publisher of Kalorama Information, a market research firm in New York that focuses on health information technology. Carlson tells American Medical News (published by the AMA) that “barriers to adoption for physician practices, including a disruption to work flow and a preliminary loss of productivity, are so strong that ‘the stimulus alone is not going to push the issue.’”

On the blogs and in conversations with doctors, I’ve found this same resistance—especially from older practitioners. In a post from Internet Evolution, Mary Shacklett, president of Transworld Data relates some reactions to EMRs that she’s heard from doctors:

“‘I have been practicing medicine nearly thirty years,’ said one New York City physician. ‘Using EMR is time-consuming, and it gets in the way of my relationships with my patients.’ A Midwest general practitioner echoed the sentiment: ‘I am in a situation where the clinic where I work has mandated using EMR, and it takes me six to eight hours a week to learn how to use it and to input information. Instead, I could be using this time to see more patients.’”

Some of the resistance to and difficulty working with EMRs is, frankly, age-related. Shacklett cites a recent CDC study that found that only 7.6 percent of physicians older than 65 used some form of EMR, compared with 47 percent of physicians younger than 35. But, she notes, “the slow movement to EMR has more story lines to it than age.”

Colin Harrington, an associate professor of psychiatry at Brown University’s Alpert Medical School, falls in the middle of these age ranges and also has mixed feelings about the EMR. As a neuropsychiatrist, he sees patients in the hospital as well as in an office or clinic setting. The first application of health IT at Harrington’s hospital was the electronic conversion of lab data and it was a positive experience: “Being able to immediately pull up a patient’s MRI image is unbelievably powerful,” he says, “and being able to pull up lab data proper is out of this world great.” But Harrington does have complaints about the EMR—mainly with how the technology effects his interactions with patients.

“In an outpatient setting, I either don’t type when I’m talking to a patient because it’s rude and I can’t watch patients articulate and observe their movements. For a dinosaur like me, even if I were to type when talking with patients, it would be inefficient.” Instead, Harrington takes notes by hand and then inputs them into the computer later in the day. “The result is that I can see fewer patients and I don’t get paid for the extra time.”

Douglas Perednia, an internist and dermatologist, agrees with this assessment in his recent blog post for KevinMD. Perednia writes about a colleague who hired a “scribe” to input written notes into her practice’s EMR system. Although she’s paying the scribe $10-$15/hour for this service, it could be worth the expense, according to Perednia. “It’s well known that, in the vast majority of cases, EMRs make doctors far less efficient when seeing patients.”

“It’s relatively easy to write with pen and paper while listening and explaining. It’s far harder to keep two hands on a keyboard, a third hand on the mouse, one eye on the screen and another eye on the patient and his family. And the fact that most EMRs and user interfaces are designed by computer geeks with no knowledge of clinical care or workflow certainly doesn’t help matters. As soon as EMRs are deployed, physician productivity typically goes down by about 50%. It rarely gets back to where it was prior to installation.”

A study in the Canadian Family Physician also finds that family physicians feel that they don’t have the time (or rather would have to spend too much uncompensated time) to really learn how to implement health IT systems in their practices; for a fair number, their computer skills were lacking. “For each hour of seeing patients, physicians said they needed 20 to 30 minutes of e
xtra time to process paperwork (eg, complete billing, write consultation letters, review laboratory results).” This helps explain why many of them report dissatisfaction with electronic medical records.

Perednia also mentions another barrier to EMR acceptance—some systems can be unreliable or ill-suited for a particular practice. The colleague who hired a scribe is on her second EMR system: “The first one – purchased for tens of thousands of dollars – was a total disaster and had to be scrapped entirely.” Poor tech support (something I experienced firsthand at my recent doctor visit) is another reason doctors cite for being slow to adopt these systems.

For many doctors—as well as smaller hospitals—choosing an EMR can be overwhelming. The Health and Human Services “meaningful use” criteria provide a basic framework of functions for EMRs that are designed to improve quality of care and safety, among other benefits. But there are more than 300 companies selling EMR software, and many of them are promoting their products as qualified for “ARRA” funding—i.e. they meet the meaningful use requirements to receive incentive payments through the American Recovery and Reinvestment Act of 2009. The truth is that EMR systems must also meet the specific needs of a particular clinical practice, otherwise they will be balky and hard to implement. For example, pediatricians need EMRs that keep track of vaccination schedules and can generate school and camp forms; internists who see a lot of diabetic patients need EMRs that notify staff when test results are unusual or blood pressure should be taken, among other features.

With the government taking a “hands-off” approach to regulating these systems—or even providing standards that will allow interoperability between individual doctors, government payers and hospitals—it can be hard for practitioners to make an informed choice about a system.

As it stands now, there is not even a monitoring system in place for recording adverse events caused by EMR glitches. The reason, according The Huffington Post Investigative Fund, an independent, nonprofit journalism venture affiliated with the Huffington Post, is a “clash of priorities” between the FDA (the agency charged with making sure medical devices are safe and effective) and the Office of National Coordinator for Health Information, “whose central task is to promote the technology’s swift adoption.”

According to the Huffington investigation, “The clash of priorities became public in late February when FDA official Jeffrey Shuren tied 6 deaths and more than 200 injuries to health information technology and said these were likely the ‘tip of the iceberg.’ The data review, based on mostly voluntary reports to the FDA, suggested ‘significant clinical implications and public safety issues surrounding health information technology,’ according to an agency report.

“Shuren laid out three possible options for regulations, none of which have happened. They range from mandatory reporting of ‘adverse events’ to a full blown regulatory structure that would require all digital records system to be approved by the agency prior to marketing.”

The authors conclude; “That no process exists to report and track errors, pinpoint their causes and prevent them from recurring is largely the result of two decades of resistance by the technology industry, a review of government records and interviews by the Huffington Post Investigative Fund shows. The industry argues that even with flaws, digital systems are an improvement over current paper records.”

That may be so. But the improvements may not be immediate. Researchers at Arizona State University’s W.P. Carey School of Business set out to measure the benefits of health IT in a group of more than 300 community hospitals in California that adopted some system of electronic medical records.They were inspired by a 2005 RAND Corporation study that found that America’s health care system could save more than $81 billion annually and improve the quality of care if electronic medical records were widely adopted.  Instead, the researchers, whose work is described in The Fiscal Times “found that the use of IT in a group of California hospitals actually has resulted in higher costs, higher levels of nurse staffing and higher patient complications.”

The problem, note the authors, is that of execution. “Aside from the large, well-funded hospitals, most institutions don't have adequate IT staffs to implement and operate such systems efficiently.” Some of these systems might be only partially implemented—with some medical records still in paper form. Finally, computerized records are only as good as the processes they are computerizing—if the hospital’s own methods for keeping track of in-hospital patient data are sub-par, then you end up “computerizing inefficient processes,” according to the authors.

The Huffington investigation revealed two major glitches with EMRs that occurred in the Trinity Health System, a large chain that operates 46 hospitals, most in Michigan, Iowa and Ohio. In late June, the computer system (sold by the Cerner Corp.) posted some doctors’ orders to the wrong medical charts in a few cases; then two weeks later, Trinity had to shut down its $400 million system for four hours at 10 hospitals because electronic pharmacy orders weren’t being delivered to nurses for dispensing to patients, according to the article. Neither of these system problems led to patient harm, but the authors point out that they do highlight fallibility in EMR—despite their great promise in improving quality and reducing health care costs.

Still, the W.P. Carey researchers did find that the California hospitals with EMRs in place had reduced mortality rates—presumably because the systems helped cut down on medication errors. And they remain optimistic about the long-term benefits of computerized patient records—with improved patient safety and better quality of care being realized sooner than cost savings. The problems being experienced now are “bumps” in the road as hospitals wrestle with new technology; “nobody questions IT's value in enhancing productivity.”

Next year marks the start of the five-year push to get doctors and hospitals to adopt EMRs. The release of the “meaningful use” criteria is the first phase of this process. More government action—in terms of creating standards to drive interoperability between doctors, hospitals, ambulatory care centers and other parts of our vast health care system—will be necessary to help achieve the goals of health reform. And so will better oversight from the FDA or whichever agency takes responsibility for monitoring adverse events and other problems with these systems. Right now we are experiencing the growing pains of an emerging technology—a technology that can feel onerous, balky and time-consuming to many practitioners. But as Colin Harrington told me of his hospital’s EMR system, “I imagine this is going to be 90% good and 10% bad in the long-run. It is still an evolving process.”

24 thoughts on “EMR Technology Experiences Growing Pains: Resistant Doctors, Computer Glitches, and Unrealized Benefits

  1. There’s also some variability in the usefulness of EMR between specialties, I think. I do critical care, and the EMR is a wonderful fit for what I do and how I practice. The EMR replaces large, cumbersome paper flowsheets of the past. I can see other specialties, like psychiatry, being less of a fit for what the EMR does best.
    Your post mentions Cerner — I don’t like their product. I like Epic much better. But, as you point out, adequate IT backup is crucial. When our hospital went live they had 24 hour onsite IT folks in droves available to help. It cost a lot, but without that support I don’t see how we could have done it.

  2. It’s too damn bad that arrogant America will not copy the French system. EMR is all contained on chip on a credit-card sized medical card. No fuss, no paper, records are always current. This process being pushed by Medicare is going to cost medical practice billions of dollars as they replace systems that don’t work. Until the providers are convinced that their productivity will increase not much will happen.

  3. Thank you for the post Naomi. I’d suggest reading HealthCareRenewal for some sobering information on HIT.
    It seems clear to me that HIT is just another case of money driven medicine with a twist — the money goes to the HIT vendors not the doctors.

  4. I want to second what Joe Says wrote. Health Care Renewal blog has sobering and little known information about the numerous problems with EMR systems.
    If only by personal experience I ought to know: I worked for an EMR vendor before getting into medical residency.

  5. Oh for crying out loud!!! Doctors aren’t adopting EMRs because they DON’T HELP US DO WHAT WE NEED TO DO. True they cost more money than they bring in (money driven to HIT vendors indeed!), but that’s strictly secondary.
    Other countries have adopted NATIONALLY INTERACTIVE EMRs! Of course they work better than the non-system we have here.
    If every single credit card company used different sized cards and required different merchant systems to accept them, you’d better believe there would be far more cash than plastic utilized in this country. Get the health IT folks to do for medicine what telecom did for banking and it would be a whole new ball game.

  6. I’m neither a doctor nor a tech guy. I’m just a patient, and I don’t want to hear excuses. Saying it’s rude to type on a keyboard while taking patient information is an excuse. If it means he or she has to take a hammer and chisel and pound my information into granite, I will wait patiently, thank you very much, because ultimately, my interest is in better care for me, and I fully believe that better information will lead to better care.
    So docs, get over yourselves and your excuses, and come up with a not-rude way to say, “Excuse me for a moment while I code that information into your record, because I know it’s important to both of us.”
    Why is that so hard for a guy or gal who was the smartest kid in high school?

  7. The point made by the ASU report is the most germane: If you automate poorly-designed processes, or if you try to automate non-existant processes, you are doomed to failure.
    If the comments of the doctors to this post are indicative of the majority of clinicians, then the appalling truth is that providers still don’t “get” it. The purpose of automating what is a risky (for the patient) and information-driven (for the clinician) business without the necessary information at hand (the way it has been done for thousands of years) is to provide safer, better, more cost-efficient, patient-centered medical care. Whether EMR or CPOE or POS or bedside automation or DSS or data warehouse “suits” of “fits the workflows” of one specialty or sub-sub-specialty, or not, is besides the point. Their patients are still whole people, and their medical care providers need ALL of their information. The way medical care delivery is organized is a historical, political, and economic artifact that can bear changing; their patients’ medical and healthcare needs are not. These responses show, painfully, that providers have not changed from their medieval way of viewing their patients as specific parts or disease manifestations, rather than as people with integrated anatomic and physiologic systems who are asking to be healed.
    A recent McKinsey report describes some of the factors that hospitals should pay attention to to have successful outcomes and provide better care by integrating IT into their delivery processes: http://www.mckinseyquarterly.com/Health_Care/Strategy_Analysis/Reforming_hospitals_with_IT_investment_2653
    The VA consistently out-performs other providers and plans on HEDIS and NQF quality indicators because of their automated, integrated EMR system (among other reasons), in spite of their bigger patient base of older, sicker patients with multiple chronic diseases. http://www.informationweek.com/news/healthcare/clinical-systems/showArticle.jhtml?articleID=224202277
    VA Health IT Generates $3 Billion Savings
    Here is the study: http://content.healthaffairs.org/cgi/reprint/29/4/629?ijkey=zm91rS/ZFWdqE&keytype=ref&siteid=healthaff
    The Value From Investments In Health Information Technology At The U.S. Department Of Veterans Affairs
    The truth is that hospitals don’t want to spend money on patients who might use their competitors and who might never return; why should they make it easier for their competitors? The truth is that doctors are fixated on the tasks of their trades and their back-office needs and not on their responsibilities to their patients to “do no harm.”

  8. #1 Dinosaur has it right. I think it would behoove the HIT industry to show some humility and call a spade a spade.
    Doctors need to start using EHRs, in their current state, so that interoperability is made possible in the future. We have to start somewhere and it would be nice to start with a National EHR (http://www.thehealthcareblog.com/the_health_care_blog/2010/08/the-government-ehr.html ), but just like everything else in health care, we are limited by what is already in place.
    We also have to acknowledge that safety issues exist and make some sort of effort to address them.
    The VA, and Kaiser, have good outcomes with EHRs precisely because they are able to communicate across, a basically, closed system. This is the holy grail, not mindless clicking on boxes to justify higher E&M codes.

  9. Thanks to all for your comments and links. This is obviously an enormous subject (whole blogs are dedicated to EMR)and continually evolving. Margalit, I found your post very interesting and it definitely got me thinking about the concept of a government-driven EMR system. Some states, like North Carolina, for example, are already developing health information exchanges http://www.fiercehealthit.com/story/statewide-hie-efforts-move-forward-north-carolina-kansas/2010-08-23 that will eventually connect all hospitals, doctors and ambulatory centers via an emergency surveillance network that already exists.
    Joe says, thanks for the reference to Health Care Renewal–I think you may be right about HIT having the potential for being another form of money-driven medicine.

  10. Thank you Naomi! I 100% second #1 Dinosaur on this. My EMR does NOT help me provide better care. My resistance is not based on not wanting to use new technology (I’m one of the under-35s and am happy to use my ipod and other devices to help me). But our EMR is useless. The first few weeks I dedicated an extra 2-3 hours per evening filling in fields like vaccination history, preventive care, problem lists. Then when those patients came back, I found it virtually impossible to find those entries in the “flowsheet” because there’s no way to separate out just USEFUL preventive care stuff. And while our EMR is bright enough to realize that men don’t need mammograms or pap smears, it is not bright enough to realize that my 5-year-old patients do not benefit from me having to look at a check box about when the last colonoscopy does, nor to realize that a 16 year old female may be due for shots, but Hib is not one of them (as that is only for young children). The entries for pap smears are antequated with no built-in decision rules and no way to note if someone is HPV positive or not.
    Adding to my frustration is the need to put in a new “problem” for every visit. So some patients have dozens of problems, most of which are “upper respiratory infection”, “lower back pain”, and “visit for form completion”. Of course buried in there are the few actually important problems, like heart disease. So half the time I have to devote my time to deleting problems from the problem list.
    Of course, our inpatient EMR is made up of 3 different systems, none of which talk to each other or to our outpatient EMR. One of the safety issues with the inpatient EMR is the quirks of ordering medications; for instance, if I want a medication given every day, but I order it at noon, the medicine won’t be given until tomorrow because “every day” for the computer means “at 10 AM.” So if 10 AM has already passed, the patient gets the medicine for the first time the next day, a full 22 hours after I wanted him to get it. It’s really easy to make this mistake, especially if you don’t use the system all the time.
    These systems are meant to streamline billing and medication administration procedures. They are not meant to improve patient care or workflow.
    For now, I’ve stopped putting preventive care information in the EMR and have gone back to my prior personal registry: an excel sheet on a jump drive. I have a password-protected file with patient’s initials, med record number, and pertinent tests/results. It’s much much easier and it actually helps me.
    ACarroll, I know you think it’s the providers who don’t get it, but honestly it’s the IT companies that don’t get it. The EMR software available to us does not help us get the whole picture. The VA software does, and all providers I know who use it like it, but that is a closed system and a software that has had lots of time to be improved. These new programs that are thrown at us don’t help us get any information from other hospitals or clinics, and don’t help us create disease registries that would help our patients. They’re all about billing (ie making more money), and the stimulus plan that threw money into the private sector to make more and more different EMRs did not help. We need ONE national EMR. I nominate the VA software; we’ve already paid for it and it has proven results.

  11. Sharon MD, Whew! Sounds like you have some real issues with your EMR—your frustration really jumps off the page! I think the IT people need to have more feedback from users like you who can’t recognize problems until they start using the system in their practices. I have a friend who founded a company that develops and sells EMR systems specifically for pediatricians. His company’s EMR is continually adding new functions to help pediatric practices and they are well-received. Oh, and his wife has been a practicing pediatrician for 30 years and was an early adopter of the software.

  12. One national EMR, particularly Vista is a pipedream. There is simply too much money and too many big time companies with their hands all over this to make that happen. GE and Google wield way too much power to allow Vista to be a national EMR.

  13. Hey Jenga, I think you are on to something — putting more money into HIT doesn’t end up making it better!! Could it be that filling up a hog trough draws more pigs?
    Hmmm, well let’s see, that is the natural result.

  14. If you look at some of the top twenty companies in terms of market capitilization in the US and all of the tech firms have their hands elbow deep in EMRs- Google, Microsoft, GE, Cisco, Oracle, Apple. HIT grew 13% last year in a crap economy and is expected to be a 35 billion dollar industry next year with a potential to go much, much higher with mandates. These players will fight tooth and nail for everything they get. A universal EMR is as likely as a universal car, phone, etc.

  15. I have no doubt that commercial systems don’t meet the needs of doctors in various settings. As an IS consultant for many years, I saw too many incidences of clients who were “snowed” by the sales pitches of hardware and software vendors without investigating whether those systems met their needs.
    It’s a mystery why providers in the US haven’t adopted (with some exceptions) the VistA system (including the EHR application), which is open source and free to anyone (and was created by doctors for doctors). Providers and governments in some foreign countries have done so to their great advantage. In the US, that would be one way to short-circuit the feeding frenzy of the HIT industry which others have complained about in their responses. I also foresee trouble because of the difficulty in getting companies with proprietary systems and “standards” to sit at the same table and come up with universal standards. (Remember the VHS vs. Beta format wars?) I think the feds should create the whole thing, from the ground up. It would also be a way to ensure that the data collected in the system was standardized, manipulable, and shareable, and could be integrated with other applications.
    There is a non-profit development project called WorldVistA which was created to develop VistA for other settings. You might want to follow this up: http://worldvista.org/World_VistA_EHR
    However, I would caution Dr. Sharon about deleting data from fields and other work-arounds (assuming that the system lets you have null fields in the first place). You might want that information some day. What if you want a report on the scope of problems that bring patients to your office; or how many times a patient came to you with the same problem; or an audit of your billing codes; or many other types of information that are important for managing your practice and keeping your patients safe and healthy?

  16. I agree with Mike. There will be enormous cost and then we can wait for enormous amounts of medical records theft!!

  17. Good article. There is no excuse for poor IT systems, but there are many of them. We are rolling out a clinic system (not EMR) and noticed two things. The first is talk to the people who are using the system every single week. And visit them in person once a month. That way you can adjust the system to adapt to them.
    Secondly, having a physician typing on a computer in the corner doesn’t work well. Eye contact with the patient is critical. If the physician can touch type, use a very small netbook and keep your eye contact. After the patient leaves, correct the notes. Or use an iPad. The other method is have a nurse in the room with you and he can input the information.
    There are many ways to make this work. It just takes some time.

  18. Here is a useful new website you might want to consider:
    Biomedical Device Integration Tech Corner
    “An archive of technical documents, protocols, standards and procedures useful for clinical engineers and IT professionals involved in biomedical device integration and connectivity to electronic medical records (EMR)”

  19. There is a lot of poor information online today about using social media for small business – I’m glad to hear from someone who actually has gotten some results! Well done!

  20. The intended effect of EMR systems on the healthcare landscape is wide-ranged depending on the type of system and the environment in which it is being implemented.

  21. Can I just say what a relief to search out somebody who actually is aware of what theyre talking about on the internet. You definitely know the right way to deliver an issue to gentle and make it important. Extra people need to read this and understand this facet of the story. I cant imagine youre not more in style because you definitely have the gift.