On this last episode, we look back on 23 years of joyous, difficult and always informative conversation.
Nearly two years after the U.S. government began incentive payments for medical offices to adopt electronic health records, more than 80 percent have gone digital. But health care workers say downsides remain, including disparate systems that can’t communicate together, confusing screen commands and inflexible software that can lead to medical errors. We explore the challenges that electronic health records systems present to doctors, patients and to technologists building systems that must maintain privacy, yet meet the needs of medical professionals and patients.
- Dan Morhaim Physician; Member, Maryland House of Delegates (D-Baltimore County)
- Ben Shneiderman Professor of Computer Science and Founder of the Human Computer Interaction Lab, University of Maryland
- Jacob Reider Acting National Coordinator for Health Information Technology and Director of the Office of the Chief Medical Officer, U.S. Department of Health and Human Services
TwinList Demo Video
TwinList is a user interface prototype developed at the University of Maryland for doing medication reconciliation. In layman terms it helps a physician review two lists of medications coming from different sources and prepare a new list. In this scenario the “intake” list on the left is what the patient said he was taking when he arrived at the hospital, and the “hospital” list is what was given during the hospital stay. Now the physician who discharges the patient from the hospital has to review both lists and decide what to keep and what to discontinue.
MR. KOJO NNAMDIFrom WAMU 88.5 at American University in Washington, welcome to "The Kojo Nnamdi Show," connecting your neighborhood with the world. Ten years ago, visiting the doctor was a decidedly offline experience. Doctors used color-coded files to view a patient's history. They wrote down new information on charts, and they scrawled prescriptions on paper you handed to the pharmacist yourself.
MR. KOJO NNAMDIWhat a difference a decade makes. Today, most of us see doctors who carry laptops. Our prescriptions are sent straight to the pharmacy. And some of us can even view our health records online. These statistics show there's no going back. Nearly half a million medical providers have digitized their records since 2011 alone.
MR. KOJO NNAMDIBut as the recent rollout of the government's health insurance website shows, the massive task of bringing healthcare online can be riddled with problems. From the technologists and vendors who create and build the record systems to the medical staff who must use them accurately, there are 10,000 decisions that must be made and remade to ensure that patients get the best care possible.
MR. KOJO NNAMDISo how far have electronic medical records come? And what are the next hurdles? Here to have that conversation with us is Dan Morhaim. He is a physician. He's also a member of the Maryland House of Delegates representing the 11th district of Baltimore County. He's a Democrat. Dan Morhaim, thank you for joining us.
DR. DAN MORHAIMGreat to be with you.
NNAMDIAlso in studio with us is Jacob Reider, acting national director for Health Information Technology and director of the Office of the Chief Medical Officer of the U.S. Department of Health and Human Services. He's also a family physician. Jacob Reider, thank you for joining us.
DR. JACOB REIDERThanks, Kojo. Happy to be here.
NNAMDIAnd Ben Shneiderman is a professor of computer science at the University of Maryland. He's also founder of the Human Computer Interaction Lab, with which we have some interaction from time to time. Ben Shneiderman, thank you for joining us.
DR. BEN SHNEIDERMANHappy to be back.
NNAMDIGood to see you again. You, too, can join this conversation by calling 800-433-8850. Have your doctors' offices converted to electronic records? Has the change had an effect on your care? Give us a call, 800-433-8850. Jacob Reider, one of the goals behind converting our nation's medical records to electronic format and a main goal of the Affordable Care Act is to reduce the cost of healthcare. Are we seeing evidence yet that electronic health records are helping to bring costs down?
REIDERSo I'd flip the question a little bit and talk about the primary goals because I think reduced cost is a secondary or perhaps even tertiary goal. The primary goal of incorporating electronic health records into our care delivery system is actually to improve quality and to improve efficiency. And a byproduct of improved quality, improved efficiency is actually reduced cost.
REIDERIf we take better care of patients, they don't need to go to the hospital as much. They have fewer complications. We physicians commit fewer errors. And there is good research. There was a study just published a month or so ago about diabetics, comparing -- Kaiser Permanente compared a couple of hundred thousand patients for whom electronic medical records were used with the same number of patients for whom electronic medical records were not used.
REIDERAnd the patients for whom electronic medical records were used had about 5 percent fewer admissions to the emergency department than those who did. And so I think this is one of a number of studies that demonstrates that electronic health records improve the quality because obviously if they didn't have to go to the emergency department, they were getting better care. They were healthier. And as a byproduct of better quality, cost was reduced.
NNAMDIAnd I'm glad you flipped the question because the fundamental question is, where are we? Nearly 10 years ago, in his State of the Union address, President George W. Bush outlined the plan to convert the U.S. medical system to electronic health records by the year 2014. Soon after that speech, your office was established. I think a lot of us now see laptops and tablets creeping into our doctor's visits. But where are we officially in meeting the goal that was set nearly a decade ago?
REIDERSo I happen to have those numbers for you, Kojo.
NNAMDIJust by chance.
REIDERSo the most recent numbers that we have that are a byproduct of a program that was initiated in 2009, which is called the Meaningful Use Incentive Program, and this program is motivating healthcare providers. So one of the reasons that we see more healthcare providers and more hospitals using information technology is that one of the primary barriers to implementation -- and not the only one, as my colleagues will discuss, I'm sure -- but one of the primary barriers was a financial barrier.
REIDERIt costs money to buy these systems and to implement it in your practice. And so an idea that was incorporated into the HITECH Act, which is part of the stimulus act from 2009, was to motivate healthcare providers to purchase these systems by giving them a financial incentive in order to meaningfully use health information technology.
REIDERAnd so the most recent numbers from that program suggests that about 40 percent of eligible providers -- and that's doctors, dentists, optometrists, chiropractors, and podiatrists, so primarily physicians who are participating in it. But there are some other healthcare providers who are participating in this program.
REIDERBut 40 percent of the healthcare providers in the United States are now participating in the program and are meaningfully using health information technology. So we're not at 100 percent as was predicted 10 years ago, but I think we're well on the way.
NNAMDIDan Morhaim, as a doctor, you are on the frontlines of what we're talking about. You recently wrote a piece in The Washington Post pointing out some of the challenges that doctors are facing, even as an increasing number of medical facilities go digital. How are electronic medical records changing the day-to-day work of seeing and of treating patients?
MORHAIMI think we all share the goals that Jacob outlined. But as a frontline clinician, it's simply not working out that way for most of us who actually practice medicine. And it's not just the doctors. It's the nurses, the respiratory therapists -- and just want to respond to something that my friend here said that Kaiser had reduction.
MORHAIMKaiser's a closed-loop system with one consistent set of electronic medical records. In fact, in the American Journal of Medicine just recently, they concluded that there was no reduction in cost. But the -- that's an unfair comparison because most -- what's happened is that the health IT companies were handed 28 to $50 billion.
MORHAIMAnd they have come up with hundreds of different programs that clinicians -- and many of us work in more than one location. That's just a fact of life as a clinician today -- so that each of these systems has a completely different set of programs and alerts. And this is not just a glitch like what we've read about or even when you update the operating system on your home computer.
MORHAIMIt's a basic design flaw, frankly, that could have been avoided. And I just want to give you a couple of points on that. Completing an electronic medical record is not like renting a car or buying an airline ticket. It's vastly more complicated. And here's just two examples. When I write a prescription -- and most people have gotten a prescription and looked at the package insert. It lists all the side effects and drug-drug interactions.
MORHAIMThe computer programmers have programmed all of them into the computer, so it's virtually impossible to write a prescription without some alert coming on. What sets in is alert fatigue because the clinician has to stop and look all those side effects up, finds out numerous times that it's not necessarily important. And then one day, they'll be a problem.
MORHAIMThe programs were not designed with clinicians in mind who actually have to make real-life decisions. And as an emergency physician, sometimes I just have a few seconds to make a decision or a few minutes. They should have tiered them, done other kinds of stratification. But that's not how it works. And one other example, I saw a patient recently who had depression as a medical problem.
MORHAIMAnd I typed in depression. The pull-down screen came. The first thing that came up was acquired deformity of nose, meaning a depressed nasal fracture. That had nothing to do with the mental illness. But that's the kind of unwieldy and bulky systems we have. Last point, this all could have been avoided. The Veterans administration system, VistA, has been around for decades.
MORHAIMIt works well. Most physicians have had experience with it. And they offered it free, free, to every institution in the United States. And it's used in 13 countries. We should have -- instead of throwing this all out to the many IT companies lured by this trough of money, we should have said VistA is the national electronic health record system, so everybody -- clinician would know how to use it and then married -- put the IT companies to work to marry the EMR system, Electronic Medical Record system, to billing.
MORHAIMBilling's been computerized for years. It's money that drives it. But we didn't do that. I still think we can make this to the national electronic medical record system. It's been tried. It's proven. Privacy, confidentiality, security's all done, and then we can go from there.
NNAMDIIn case you're just joining us, it's a conversation on improving electronic health records. You've been listening to Dan Morhaim. He's a physician. He's also a member of the Maryland House of Delegates, representing the 11th district of Baltimore County. In studio with us also is Jacob Reider. He is acting national coordinator for Health Information Technology and director of the Office of the Chief Medical Officer at the U.S. Department of Health and Human Services.
NNAMDIHe's also a family physician. We're taking your calls at 800-433-8850. You can send email to email@example.com. Have your doctors' offices converted to electronic records? Has the change had an effect on your care? Also joining us in studio is Ben Shneiderman. He is a professor of computer science at the University of Maryland and the founder of the Human Computer Interaction Lab.
NNAMDIBen, as researchers and technologists, you and your team at the University of Maryland are part of the back end of this equation. Can you give us an overview of how you're tackling the technology behind these electronic record systems? And then we'll talk about some specifics, even though I know Jacob Reider probably wants to respond to what Dan Morhaim said earlier. But, starting with (unintelligible) Ben.
SHNEIDERMANThank you, Kojo. Yes, Dan Morhaim, like many physicians, tell us the stories of their struggles with using electronic health record systems. It's not just that it's new, but it's difficult. It varies from clinic to clinic. It takes longer. It takes them away from direct contact with their patients. And so there is a problem, and these can be avoided.
SHNEIDERMANNow, our work is supported by Jacob Reider's office of national coordinator. And so, admirably, there are four major national groups of research universities that are working on different aspects of the problem. And we were given the task about working on patient-centered but physician-oriented tools to help physicians do their job more effectively.
SHNEIDERMANAnd so we began to find the kind of problems that Dan Morhaim outlined, that systems were difficult to use. It was difficult to find things. It was difficult to go through many menus, many tabs, click, click, click, click, click, click until maybe you found what you wanted. If not, you had to go back, back, back, click, click, click until you explored.
SHNEIDERMANSo these systems could have been better, and they still can be better. And they are getting better. So there's some positive news in that there is progress in this area, and we work with the companies that are providing. There are some 1,700 companies, maybe more, that provide different aspects of these electronic health systems. And getting them to work together is the central national challenge.
SHNEIDERMANDeveloping common styles of user interface so that simple things like blood pressure that we know is 120-slash-80 appears in the same way, in the same colors, maybe in the same position on the screen, so a physician can easily grasp that information and then take a look, for example, of the last five blood pressure readings. They should be able to find them in seconds, not minutes.
SHNEIDERMANAnd so the systems are inconsistent, which is one problem, and sometimes poorly designed. So the movement towards usability and user-centered design and the use of the science of human computer interaction is steadily growing. And the companies are hiring one, two, three, four and up to fifty professionals in the area of user-experienced design. I don't promise -- as Jacob said, this is enormously complicated, this is enormously complicated. So I won't promise it'll be easy as your iPhone, but it can be a lot better than it is now.
NNAMDIOnto the telephones, here is Michael in Baltimore, Md. Michael, you're on the air. Go ahead, please.
MICHAELThank you. It will take me about 20 seconds to ask this question and then I'm going to hang up and listen. It's largely for Dr. Reider, but I would like the gentleman who just spoke to answer, too. And there's no disrespect here. I'm a clinical professor of medicine at Georgetown University, and I have a busy private practice in Baltimore. You seem to be speaking to us as if we are the spectators at the parade of the emperor's new clothes.
MICHAELThe preponderance of published evidence suggests that there is no interoperability of the 911 systems in private practice, no credible usability and a recent article in New York Times reported that new doctors are spending 12 percent of their time on these devices and nearly 50 percent documenting -- I'm sorry, 12 percent of their time with patients and nearly 50 percent of their time documenting it on these obtuse devices to meet the draconian subterfuge regulations of meaningful use.
MICHAELMy Hopkins in Maryland colleagues refer to their EHR as the post-epic -- which is the name of their EHR -- chaos. They're cutting back their academic responsibilities, resigning editorships and decreasing conferences. At my institution the best word I've heard about MediConnect, it's a disaster. So here's my question and then I'll hang up.
NNAMDILongest 20 seconds I ever heard, but go ahead, please. (laugh)
MICHAELI'm sorry. I'm sorry, sir. When is the government going to stop the carnage and stop pandering to the unconscionable profits of the HIT industry and put a moratorium on new HIT and meaningful use until one that works is well available?
NNAMDIThank you very much for your call. Here now is Jacob Reider.
REIDERI'm collecting my thoughts for a moment.
NNAMDIThe part about you pandering is what I remember most.
REIDERAnd draconian subterfuge perhaps. (laugh) So let me -- I will offer an answer in the context of my own personal experience, which in fact is not dissimilar from this physician's or even Dan's experience, and maybe offer a little bit of historical perspective. I'm a family physician. I practice and used electronic health records until roughly 10 months ago. I still have a medical license and if I get a little bit of spare time away from this job I'll go back to practicing a little bit in upstate New York, which is where I live on the weekends, as my wife will tell you.
REIDERI went into this position and I joined the federal government because I saw an opportunity to improve the quality of the products that were there that were essential to helping us take good care of our patients. The path that I took toward this place was actually at something called a blog post, which nobody knew about in 2004. So I wrote a blog post about how abysmal my electronic health record was in 2004. And was a vocal critic, just like my colleague here, Dan, and our friend on the phone. So I am as passionate about the need to improve the quality of these products as both of you and, in fact, many of your friends.
REIDERThe question is what's government's role in doing what we are doing here in order to improve the quality of the products, and as a byproduct of that, improve the quality of the care that are patients are given. Is the usability, user experience of these products where it needs to be? As Ben described, not yet. And we, the government, have taken a significant position in requiring that the companies that make these products incorporate the things that Ben was talking about, that are standard practice in many other industries. So your iPhone is your iPhone and it anticipates your needs in a way because there was research done around the needs of the users who used those tools.
REIDERAnd the research was initiated by people like Ben, in that field. So we are now requiring that the vendors use those research methods in creating the products that they're creating.
NNAMDIGot to take a short break. Hold that thought for a second, Dan Morhaim. We'll get right back with you. You, too, can join the conversation. Call us at 800-433-8850. You can send email to firstname.lastname@example.org. We're talking about improving electronic health records. What do you think a trip to the doctor's office will be like in 10 years? All of these problems you think will be solved? 800-433-8850. I'm Kojo Nnamdi.
NNAMDIWelcome back to our conversation on improving electronic health records. We're talking with Jacob Reider. He is acting national coordinator for Health Information Technology and director of the Office of the Chief Medical Officer of the U.S. Department of Health and Human Services. He's also a family physician. Ben Shneiderman is a professor of computer science at the University of Maryland and a founder of the Human Computer Interaction Lab. Ben Morhaim is a physician. He's also a member of the Maryland House of Delegates, representing the 11th District of Baltimore County. He's a Democrat.
NNAMDIDan Morhaim, a couple of things. Our physician on the phone referred to the fact that the statistic indicates that the physicians now spend 12 percent of their time seeing patients and 50 percent of their time entering records. How has lugging a laptop or tablet into the exam room changed the way you interact with parents?
MORHAIMWell, there's no lugging it around. Where I work they're stationary desktop computers. Now, there may be individual practices that can do other things, but where most clinicians work are not so fortunate. Yes. There's not only a great deal of more time spent entering data, as a recent article in the New York Times demonstrates, but there's also, to some degree, an erosion of the doctor/patient relationship because doctors are typing. Actually, what I do when I see a patient now is I interview them, make some notes, have the patient exit the room, then I enter all the data because that's such a complex…
NNAMDIWhich is what the doctor was referring to earlier, about how much time it takes to do all of this.
MORHAIMYes. It takes a tremendous amount of time. Also, when people more from different systems -- and I'm in a situation where people work 24/7/365, the emergency room. So if somebody doesn't show up to work, whether it's a respiratory therapist or a nurse you bring in one of the pinch-hitters. They might be much more familiar with another system, so they can't use the system where they work less frequently, but they're needed clinically. So another person does all their data entry. There are all kinds of other problems. Look, I think we all agree that electronic medical records would be a great thing, properly implemented. There's no doubt.
MORHAIMAnd physicians and all my colleagues are very text savvy. We're not anti technology. We love the stuff and we use it in our regular lives. The problem is that the way this system has evolved has created a vast hodge-podge of different, complicated systems that require continuous training. And the VistA system, you know, is available free of charge. Approximately 50 to 60 percent of physicians in the United States have had some experience with it in their training or in their careers. The problem was the VA doesn't bill.
MORHAIMIs it a perfect system? No. But it's been around a long time and to have this, you know, complex group of IT companies pitching products to hospital administrators and clinic administrators simply isn't working and it's time to rethink it.
NNAMDIWell, before we get one of our panelists, Adam in Washington, D.C. seems to offer a slightly different view of his experience with electronic health records. Adam, you're on the air. Go ahead, please.
ADAMHi, Kojo. Well, I've been a diabetic for 47 years and for the last 25 I've been going to the same doctor's office. And about 9 months ago it was the first time that they had an electronic system. And I did feel like the doctor was spending much more time staring at the computer. The second visit went better. And I was there last week for the third time and I was very surprised and pleased with how smoothly everything went. And I left there thinking, well, the doctor clearly was asking me more questions. And as a diabetic I had walked in there with a list of some questions I had, a couple about prescriptions.
ADAMAnd my doctor raised those points before I did. And I don't know whether it was the doctor or maybe some reminders that were coming up on the computer system, but -- and then when I left they told me -- they gave me a password. And they said I could look at my records whenever I wanted to using the password. So I was pleased.
NNAMDIThank you very much for your call. Ben Shneiderman you wanted to make a point?
SHNEIDERMANGreat. I mean, it's wonderful to hear happy stories like that and I hope we'll hear more of them over time. I think the difficulty though is how do we get there? And the U.S. experience is very different from others. You have small countries, Denmark, Israel, Netherlands, Iceland, which as a national effort were able to make a common-user interface and build one system. So that was easy.
SHNEIDERMANThe British government, under the National Health Service, has tried to do that and struggled with 60 million people. The U.S., with 330 million and maybe more patients, is a much more complicated situation and our economic history is more of a competitive development, rather than centralized government management.
SHNEIDERMANSo we have those political problems. And installing a solution from above is a dangerous thing, as well, I think. So I'm not a supporter of the idea that there be a moratorium or that there be a government imposed solution. But I do think the Office of National Coordinator and government offices, others like NIST, the National Institute for Standards and Technology, can take a more forceful role in promoting a more rapid development of higher quality systems.
SHNEIDERMANIn fact, the American Medical Association's recent summer report does stress very strongly that the additional use of usability testing requiring vendors to apply user-center design and the development of more open strategies will likely lead to an accelerated improvement.
SHNEIDERMANAlso, there are just many legal interferences that limit our knowledge. We don't know how good or bad it is, as we heard in the early discussion. We get one story from one promoter, from one company, from one insurance, from one hospital. We get another story from somewhere else. We don’t have an open reporting system as we have in aviation. We know about near misses. We know about even, you know, which planes are on time. We know when baggage is delayed, but we don't know how many patients are suffering from mistakes made by physicians.
SHNEIDERMANWe don't know how many problems physicians run into, how much wasted time there is. We need a more open mechanism and we need more independent oversight. And that is a role that government can provide and accelerate the commercial development and the consolidation of these hundreds of different companies that are vying, competitively and they believe that by being different they distinguish themselves. But they can do better by sharing their efforts.
NNAMDIPut your headphones back on, please. We're about to hear from Stacy, in Alexandria, Va. Stacy, you're on the air. Go ahead, please.
STACYHi, Kojo. Thanks for taking my call. I just wanted to let you guys know that I do use the V.A. Medical Center and they are wonderful. And the reason that I continue to receive care there is because of the way they maintain their medical records. They can look up any test that I've had. I see a lot of specialists and they can see any blood work, any test and it really gives me a piece of mind that they can see all of my records at once.
NNAMDIWhat, Dan Morhaim, do you see as the problem with implementing that system nationally?
MORHAIMWell, the problem is that we've already gone so far down the path. I mean, Ben just acknowledged a whole list of problems. Look, I'm happy when electronic health record systems work. And I’m happy when I can give information to patients and they can access it. That's not the issue. The issue is that there's somebody at the end of the line, the clinician, that has to actually enter all this stuff.
MORHAIMAnd that is a real challenge. And that's when a lot of mistakes occur. In fact, there's even a whole new issue of potential -- I'll read the title of an article from a law journal, "Too Many Alerts, Too Much Liability: Sorting Through The Malpractice Implications of Drug-Drug Interaction, Clinical Decisions, Support."
MORHAIMThere's also, frankly -- unfortunately -- a tendency for people to -- clinicians to click the box so that the chart gets done, not really having fully evaluated the patient. Look, the promise of electronic health records is wonderful and I support it, but the problems are out there and I think it's time for that more aggressive approach. But we have been going down this path for 10 years. We should have been doing this two years ago, six years ago, eight years ago fixing these things...
NNAMDIWell, allow me to be devil's advocate here…
NNAMDI…with some help with an email from Beth and question the extent to which clinicians themselves may be responsible. Beth writes, "In my ENT's office I saw a visiting French family paying the father's bill and speaking in French about the paper files and cabinets behind the reception desk. They were snickering and asking each other what on Earth those antiquated records were doing there. The receptionist did not speak French and so thought that they were very charming. Having experience health care in France, I can only blush. American doctors have been so slow to adapt to new technology. It's mind-boggling. They resist everything."
NNAMDITo play devil's advocate for a second, Jacob, is part of ironing out the rough patches in these systems also convincing the medical staff that maybe need to change the way they run their offices?
REIDERPart of what we're doing is transforming the way we do business. And as we transform the way we do business, using new tools always requires a shift in culture. As we look at how technology has changed the way that we do business in many other parts of our lives, from banking to renting cars, as Dan described, to buying airplane tickets.
REIDERTechnology changes what we do. So I actually would not take the position that we blame the victim here. Healthcare providers are, I think, to some degree the recipients of the change in technology. And this, as I think both of my colleagues here and also folks on the phone have expressed, sometimes the technology is evolving more slowly than our expectations.
REIDERBut I think your callers have made great points, that there are so many advantages. You know, personally I have experienced this both as a clinician -- and I'll tell a little anecdote. I was unfortunately in the emergency room with my mom over the weekend and not only did the emergency room physician use the electronic health record with her, so she was part of the conversation, I actually was able to look over his shoulder and help him select one of the drugs that she was on.
REIDERBut I could also, with her, log in to the website, review with her primary care physician the events from the past weekend and really understand in detail, in a way that never would have been possible using paper, what happened in that emergency room 300 miles from her home, and getting that information from that emergency room to her primary care doctor. This would never have been possible with paper and faxes.
NNAMDIBen Shneiderman, the University of -- you were going to say?
MORHAIMI was going to say there was an article in the -- look, it's all good when it works.
MORHAIMWe just have to recognize there are lot of times when it doesn't. There's an article in the Annals of Internal Medicine. I'll just quote the last sentence, "Results support the growing evidence that using the basic data input capabilities on the electronic health record does not translate in the greater opportunity that these technologies promise." I support Jacob. I hope we get there, but we've got a long way to go.
MORHAIMAnd yes, there's some good anecdotes, but this was a group of physicians who are tech savvy. Physicians are not anti -- and nurses are very technologically astute. And no one -- that group would like to have electronic records, but ones that work, that are simple to use and that meet Ben Shneiderman's eight criteria of good computer programming.
NNAMDISpecifically, Ben, the University of Maryland is participating in a program called SHARP that's working to make these electronic health record systems more usable, more secure. Your team has come up with at least one innovation that's already being put to use. Tell us about twinlist…
NNAMDI…and what it does.
SHNEIDERMANI appreciate Jacob's description of how he could get access to his records. I can't. I can't, you know, from my medical provider I cannot get online access to my health records. And so it's still a very hodge-podge system out there. We have good stories. We have bad stories. What I'm asking for is openness, openness, openness, so we know what's happening.
SHNEIDERMANAnd yet, most of the -- Dan is unique in going public about his concern, but most physicians are prohibited from speaking about the failures of their system because the contracts that their organization signed prohibit them from saying anything, the so-called gag order. And so when I get a story from a physician and I say, would you send me an email and tell me that story so I could use it. They say, I'm sorry I can't do that. So we need to know more of what's happening.
SHNEIDERMANYour question about Twinlist is much appreciated. One of the tasks that the ONC office had steered us to was this question of medication reconciliation. It's part of the meaningful use effort that is transforming medication and requires that when you enter your physician's office or you're admitted to a hospital or you're discharged, a transfer point, you would bring your list of the medications you think you're taking.
SHNEIDERMANAnd your physician would have a list or the pharmacy would have a list. And the process that has to happen is to reconcile these two lists to make sure that when you go home from the hospital, you're getting the right medications. The current designs of these systems are incredibly complicated. And so we took our skill in design of user interfaces and it was actually a student project initially in my class.
SHNEIDERMANAnd we worked on it for more than two years to develop a visually organized, spatial layout with animation and a highlighting that simplifies, speeds and most importantly, as Jacob said in the beginning, increases the quality of decision making. And that design, when I presented it first at a medical conference now some two years ago, I not only got applause and cheers, but many people remember I got a standing ovation.
SHNEIDERMANAnd the enthusiasm of professionals was powerful. One nurse who said, I do this every day and she says, you know, I breathe a sigh of relief when I saw what you did. And so, the question was, she said, how can you get this into every medical system as soon as you can? And that's been a struggle. I must say, we're gaining and we're making progress. Earlier this week, I met with three of the designers from one of the leading systems and they showed me their version of their system.
SHNEIDERMANAnd I was very pleased to see that that idea is catching on. And it works not just for medications. It works for vaccinations. It works for problem lists. And so, the deep, deep world of research on human-computer interaction and the science that we have can be applied for many of these problems and improve their quality and, we hope, reduce cost ultimately.
NNAMDIWhat the Twinlist system have helped Nancy. Nancy says, "I keep an up-to-date printed list of medications for doctor visits. Despite providing this each time to doctor offices in a major hospital have entered incorrect information. Examples include having me use a vaginal cream on my face. Inject -- injecting -- don't laugh. Injecting a face topical steroid and listing a medication I quit using two years ago."
NNAMDI"These systems seem to have pre-ordained entries that are almost impossible to correct. Patients beware." Dan Morhaim, managing medication seems hugely important as we gradually do away with the prescription scrawled on a piece of paper. But last month, the Pennsylvania Patient Safety authority released a study that showed that electronic systems that use pre-programmed, quote-unquote, "default values" for medication doses can lead to error.
NNAMDIWhile most of the 324 errors that were reported to the Pennsylvania system didn't do any harm, two patients had to be hospitalized. Have you see these kinds of default settings for medications at work? Do they concern you?
MORHAIMYes, they do. Generally, I think they're helpful because they catch some things that might go wrong. But I want to go back to something that Ben has outlined here.
NNAMDIAnd then we have to take a break, but go ahead.
MORHAIMThese are all things that should have -- we're so far down the road. Why weren't these things that Ben just described, you know, initiated in the basic design of electronic health records at the get go? This is late in the day to start trying to figure out that we're going to reconcile medication list. That's not a clinical surprise. I mean, that's something we've all known for a long time.
MORHAIMThe problem has been that the direction to the companies has led to what he's -- Ben has described as, you know, complex, cumbersome, unwieldy system. It could have been done and we still...
SHNEIDERMANBecause we know how bad the problems are. We just (unintelligible).
NNAMDIBecause technology don't have access to the medical staff.
MORHAIMThe other reason is because the process of all this has not included the clinicians. The people who actually have to enter the data -- look, I'm a state legislator. I've been for 19 years and all those other stuff and I've been to hundreds of meetings about all sorts of aspects of health care. I'm usually the only physician in the room. And if there's another physician, I'm usually the only one that's actually ever practiced medicine for any length of time.
MORHAIMYou know, we elect the clinicians at all levels -- nurses, physician's assistants, paramedics -- out of the process. And when you don't listen to the folks who actually have to do the work, you end up with a lot of very savvy technical people who actually have no understanding of what the end user has to do. That's the problem.
NNAMDIGot to take a short break. When we come back, we'll continue this conversation. If you've called, stay on the line. We'll try to get to your calls. If the lines are busy, shoot us an email to email@example.com. Do you wish you had more control over your medical records? Should hospitals be allowed to share patient records? 800-433-8850. I'm Kojo Nnamdi.
NNAMDIIt's a conversation about improving electronic health records with Ben Shneiderman, he is a professor of computer science at the University of Maryland and founder of the Human Computer Interaction Lab. Jacob Reider is acting national coordinator for health information technology and director of the Office of the Chief Medical Officer, U.S. Department of Health and Human Services. He's also a family physician.
NNAMDIAnd Dan Morhaim is a physician. He's a member of the Maryland House of Delegates. He represents the 11th district in Baltimore County. He is a Democrat. Let's hear now from Carl in Washington, D.C. Carl, you are on the air. Go ahead please.
CARLA few minutes ago, Ben Shneiderman mentioned the National Institutes of Standards and Technology. And they've adopted protocols for user interface in using EHRs. And what I wonder is what the...
NNAMDIEHRs being electronic health records, go ahead please.
CARL(unintelligible) Right. And what I was wondering was whether or not what you would think of mandating that the vendor simply report on how they stand against those protocols. In other words, if the protocol says you have to show the full prescription in a line, whether or not they do. Secondly, the lack of a standard ID. I believe Congress has forbidden us to adopt a standard medical ID for patients, which is really not too bright.
SHNEIDERMANThe (unintelligible) reports which I'm holding in my hands here early on advocated approaches to testing that were developed with collaboration on a voluntary basis with industry. And I think that's the right direction to go. And I should say, those people have been working with Jacob. And they helped right the 174 pages of the federal register that mandate the meaningful use directions for 2014.
SHNEIDERMANSo we are making some progress. But as I make -- my two points are, we need to see more rapid progress and also we need more openness that all vendor reports need to be posted online and we need some forms of independent oversight with teeth in them. Not just a review that reports on it, but some way in which there are penalties for non-performance and where the effective performing companies are recognized and celebrated. And so that we can accelerate the process towards high quality systems.
NNAMDIJacob Reider, we talked about Twinlist. Any other innovations that have come out of the shop program that have been incorporated into electronic health records?
REIDERTwinlist is one example of a number of projects that both Ben and others have been working on to improve, as we've been discussing, both the quality of the user experience, what the users are doing. And as a byproduct the quality of the care that's provided. So what we've done in the most recent stage of our regulations is to use these programs to inform and guide the vendor community.
REIDERSo instead of telling the vendor community, make it look exactly like this or do exactly this and being prescriptive or, as Dan suggests, choosing one product. Can you imagine what would happen if the government said everybody has to buy a Ford Bronco, right? But the people who live in this city wouldn't want a Ford Bronco and the farmers might want something different, too.
REIDERAnd so we know that there's a broad variability in the needs of health care delivery organizations from tiny critical access hospitals to large clinics and integrated delivery network. So we don't think that dictating exactly how something looks or exactly which product to be used is the right way to go. We think that giving them a set of guiding principles, and Twinlist is an example of one.
REIDERThere's been research done along with Ben at the University of Texas Houston where they've looked into how is it that we can quantify the quality of a product, how frequently does a user make errors. How can we reduce that frequency. What are the design guidance? What's the design guidance that can be used by the providers? And so what we've done is we required them to use a set of guidelines published by (word?) to assess the quality of their products.
REIDERAnd we've also required them to publish the results of those tests. So very soon now we'll be able to see it, Ben's requested, and now we delivered in a very open way. Anybody can look and see their vendor's test quality reports.
SHNEIDERMANLet me give another example about the kind of problems that come up. It's so-called wrong patient errors where maybe one in a thousand times a physician clicks on the wrong patient name on the list of their patients and inadvertently assigns them the wrong medication or lab test. And often these are caught in time, but sometimes not and damage is done. The truth is, though, we don't know how often this happens.
SHNEIDERMANWe just don't have the kind of oversight and the record keeping that would let us know how this happens. Yet in the aviation field, we have the Aviation Safety Reporting System by which pilots and air traffic controllers can report these kind of problems. And so we need a reporting system that would allow not only clinicians, nurses, physicians, but also patients to report the problems that they see. And we need to make those open, visible and transparent.
NNAMDIWhich brings to mind those people who would say, why isn't any of these stuff regulated? The FDA has to approve all medical devices, but to my understanding doesn't have any role in approving electronic medical systems. Why not? Should it have more of a role? And whatever it is anyway, Dan Morhaim?
MORHAIMThank you, Kojo. Look, whether mandated or encouraged, the veteran system, I'm just going to -- six million patient visits, large places, small places, overseas, you know, we could have greatly encouraged. It'd already been bought and paid for by all these taxpayers. We could have encouraged that system more. Instead it was left on to the side, so that all the IT companies and all those entrepreneurs can go out and pitch a lot of product and make a lot of money.
MORHAIMAnd we've gone too far overboard into that direction. So I think it's time to bring it back to the middle or a little bit more, so that we an easy to use, consistent system that's already proven itself and just tie it to billing.
NNAMDIAnd Ryan in Washington, D.C. has a question about an issue that is sure to come up more and more. Ryan, you're on the air. Go ahead please.
RYANHi, thanks for taking my call. Love your show. Yeah, I just have a question, to change the conversation a little bit. But with all the talk of ObamaCare and ACA, it seems to me that the necessity of robust analytics and tracking and care coordination and the implementation of these value-based contracts and to be able to give meaningful population health management is essential on these IT systems that we're implementing are key in actually realizing the value that ACA that promises. I wonder if your guests have any comments on that or the long-term value that is promised through these systems.
NNAMDIThe effect of electronic medical records on the Affordable Care Act. Jacob Reider?
REIDERSo, of course, and thank you for the question. Health information technology is the foundation of care transformation. As we look at population health and the things that we can and in fact must do as a population, we can't ask questions of our paper records that we can ask of health information technology. So if I were to say, how many of my patients got flu shots? Five years ago when one tried to do that in paper systems, it was simply impossible.
REIDERToday, in a -- even in a very large practice, I can ask my system how many of my patients didn't have flu shots. It can tell me and then it can tell me how to interact with those patients to find them and get their flu shots administered. And that's just a fraction of the opportunity that we have. So analytics looking at all of our data, giving us good insight into the quality of care that we, as health care providers, are offering to our patients. Is the care that we are rendering aligned with evidence-based practice, we couldn't do that in the paper world.
SHNEIDERMANI agree. I mean, you're right. And that's where we should be and we're not there.
NNAMDIAnd you have (unintelligible)
SHNEIDERMANWe've already brought up example. We've already brought up examples and we're going way down this path. We're throwing billions of dollars at it. Somebody...
NNAMDIWell, let me talk about your recommendation for a second because we're running out of time. You've recommended that Jacob's office declare a moratorium on implementing electronic health records, to review where things stands. You've also reviewed that many more medical offices installed an updated VistA system. Do you think that can happen?
MORHAIMWell, I don't know that it can happen, that's up to them. But I would hope that this conversation (word?) will require them to take a fresh look at this. We've gone down that path. I mean, it was said earlier, what if blood pressure was in the same place and the same color? That would be really nice. That simply isn't out there in the real world. It would be nice if all practices could look up such things as the flu vaccine or whether somebody had it or not.
MORHAIMBut there's a lot of garbage in and garbage out, because clinicians, a whole slew of clinicians have to enter data on screens that are wildly different from place to place, where you click boxes, you enter something that doesn't work, yet it's impossible to get it out. Real problems.
NNAMDIGot only about a minute left. Moratorium and review or just review, Jacob?
REIDERWe're reviewing every single day.
MORHAIMI appreciate that.
REIDERSo the interactions that I have with my colleagues across the nation, including this week in Houston, TX, hearing from a group of clinicians how they are doing. The bright spots of success which I hear about every single day and the challenges that I hear about every single day. So I think it is not so rosy as one might suspect, but it's not so dismal as some are (word?).
NNAMDIBen Shneiderman gets the final word.
SHNEIDERMANJacob's got a tough job. I have to say, appreciate what he's doing in service to the country and to the profession. It's very important. There are not easy answers here. I don't support the idea of a moratorium, but I do support the idea of accelerated development of improved quality. And also a greater openness on the part of the companies and the health providers as to the status of their patient care.
NNAMDIAnd I'm afraid that's all the time we have. Ben Shneiderman is professor of computer science at the University of Maryland. He's also founder of the Human Computer Interaction Lab. Ben, good to see you again. Dan Morhaim is a physician. He's also a member of the Maryland House of Delegates, representing the 11th district of Baltimore County. He's a Democrat. Dan, thank you for joining us.
NNAMDIAnd Jacob Reider is acting national coordinator for health information technology and director of the Office of the Chief Medical Officer of the U.S. Department of Health and Human Services. He's also a family physician. Jacob Reider, thank you for joining us. And thank you all for listening, I'm Kojo Nnamdi.
Most Recent Shows
Kojo talks with author Briana Thomas about her book “Black Broadway In Washington D.C.,” and the District’s rich Black history.
Poet, essayist and editor Kevin Young is the second director of the Smithsonian's National Museum of African American History and Culture. He joins Kojo to talk about his vision for the museum and how it can help us make sense of this moment in history.
Ms. Woodruff joins us to talk about her successful career in broadcasting, how the field of journalism has changed over the decades and why she chose to make D.C. home.