AHRQ expert discusses trends and challenges in health information technology
Research Activities, April 2012
Following is a Research Activities interview with Jon White, M.D., health information technology (IT) portfolio director at the Agency for Healthcare Research and Quality (AHRQ), about current trends and challenges in health IT.
Research Activities (RA): At the February Health Information Management Systems Society meeting in Las Vegas, you were on a panel to discuss the Federal response to the Institute of Medicine (IOM) report, Health IT and Patient Safety: Building Safer Systems for Better Care. The report notes emerging concerns about safety hazards introduced by complex health IT systems. What are some of these safety hazards?
Jon White, M.D. (JW): There are two sides to the safety issue. First, does health IT improve safety as much as research shows? The answer is when health IT is correctly implemented and used, it has the beneficial effects that we have observed in our research. For example, computerized provider order entry (CPOE) can alert providers to drug-drug interactions, drug-allergy interactions, or drug-condition issues. The alerting system catches the fact that a patient is on drug x and you shouldn't give him drug y with drug x, which helps avoid medication errors. If the provider turns that alert off or gets the alert and ignores it, then you don't stop that problem from happening. That's an example of ineffective or less effective implementation of health IT.
The other side of the coin is, does health IT cause problems or safety issues? This can be a thing like entering orders on the wrong patient. The system says I'm entering them on Mr. Smith, but instead I'm entering them on Mrs. Brown. Or information gets changed as it goes through the system. For example, I order 1 mg of morphine, but due to a software error the patient is administered 100 mg of morphine. There are issues that people like Dr. Ross Koppel of the University of Pennsylvania have described over time, cognitive problems in information that are presented or are presented in a way that is not obvious or useful. For example, you put all your alerts in red because you think that will call them out to the reader. But someone who is colorblind may not notice the alert, because it's in red. So it's not just implementation. Sometimes it's how the systems are put together and subsequently used.
RA: Does the work environment influence the safety impact of health IT?
JW: Absolutely. When you implement a health IT tool in a system that is not working well, it may not improve the situation. A recent example is a pediatric intensive care unit in Pittsburgh that implemented CPOE. Apparently, they were already having communication problems, and after implementation, they documented a short-term rise in their mortality rate. The problem was that people weren't talking to one another, and the order entry tools weren't well integrated into their workflow. As a result, medications were getting to patients slower, as well as other issues. So you don't see improvements in care just by virtuous use of a well-designed product. When you've got problems with a system to begin with, don't expect health IT to fix them, unless you also address underlying workflow and related issues. Health IT is a tool that can help you do a job better, but only if you use the tool correctly.
RA: What actions does the IOM report recommend that AHRQ undertake to improve patient safety in health IT?
JW: The report identifies that health IT can be a safety problem. We don't have a handle on whether it's a big or little problem, medium-sized problem, or if it is happening all the time. We really don't know the scope of the problem. So the IOM report wants AHRQ to use the Patient Safety Organizations (PSOs) and the safe-harbor construct of PSOs as a mechanism for gathering more information on safety issues and medical errors related to health IT. AHRQ has been asked to work with the HHS Office of the National Coordinator for Health IT (ONC) on this issue, and indeed we have been.
RA: How is this reflected in the work of AHRQ's health IT portfolio?
JW: We've always focused on safety as a component of quality. We've funded a lot of projects over the years in the inpatient and outpatient setting, both demonstrating how health IT can improve safety and how health IT may facilitate unintended consequences or medical errors, as well as constructive ways to address good implementation and good use. For example, in the past AHRQ funded a CPOE evaluation tool, which the Leapfrog Group now uses in its annual survey. Leapfrog used to ask, "Do you have CPOE?" as part of the survey. Now systems that have CPOE are assessed on whether or not they are as effective as they are supposed to be—did the CPOE system catch the mistakes it was supposed to catch.
RA: Health IT, when designed and properly implemented, can markedly improve the quality and safety of care. Have health IT tools like CPOE, clinical decision support systems, and e-prescribing become better over time and how do they still need to improve to fulfill their promise to improve care safety and quality?
JW: They've absolutely improved over time. Anyone who's gone from Windows 3.1 to Windows 7 knows that our computing power capabilities and the sophistication of computing tools have increased, and health IT has similarly improved. You no longer have to have a high degree of comfort with computer systems or even a high degree of training in some cases to use them well. Witness the explosive growth of the iPad®. That doesn't mean that we are dumbing down medicine in any way. It's just as complex, but we have better information tools available. We also have a better understanding of different capabilities within the health care system. There have been some providers who have always been early adopters. They like technology and like to use technology tools in the delivery of their care. Then there are providers who have been waiting to use them but aren't going to actively seek them out. Finally, there are people who stubbornly refuse to use them. In our world, the early adopters are using health IT. The middle group is being pulled there by meaningful use incentives provided by the Federal government as well as increased appreciation and understanding within the clinical community of the value of these tools. They're starting to feel they are not delivering care as well as they could without the tools, which is good. That's where we've wanted to be for a while. There may sometimes be some very good reasons why some don't want to make a change to use these tools, so it isn't always stubbornness.
RA: Is cost or lack of resources a barrier for some providers?
JW: Cost is becoming less of a barrier as time goes by. It's not necessarily the up-front cost of the system per se. There are expensive systems, but there are very inexpensive systems. You can actually buy an electronic medical record system at Costco. So there are lots of different ways to get access to the software, and some are even free. The business model for some companies is not to charge you up front, but to charge you a certain percentage of your billings. So there are different ways to afford the information systems. The real cost to practices is the time it takes to train people to use the system, and an initial decreased productivity for those practices that are fee-for-service and driven by volume. Practices are slower when they start using this system as would happen with any new system or tool. Meaningful use incentives have helped provide people with the resources to get over that initial "entropy hump." The majority of clinicians then say, "I'd never go back to paper. It's so clear to me that having the information I need where and when I need it is so useful, I can't imagine practicing without it."
RA: Would you say that health IT underlies patient-centered care?
JW: Underlies and underlines. When I talk about patient-centered care, it means making decisions according to the patients' values for what they want out of their health care and with their engagement and shared participation in the decision-making. You can't do that unless the patient understands the basis on which they are making decisions. And health information systems are going to drive that.
RA: Has health IT helped reduce disparities among disadvantaged populations?
JW: Health IT does not "auto-magically" reduce disparities. If you can identify a specific information-related cause or factor in disparity, health IT can help fix that. For example, rural providers don't have a lot of connection with their peers and patients have to drive long distances for specialty care. In New Mexico we funded a telemedicine intervention called Project Echo that originally focused on hepatitis C and later on a number of other chronic diseases. We created a provider network where providers in Truth or Consequences, New Mexico, can do a teleconsult with the esteemed gastroenterology faculty at the University of New Mexico, Albuquerque. Rural providers can spend time discussing cases with each other, learning about hepatitis C, and become better connected to other providers. As a result, provider satisfaction and confidence has gone up, and more importantly, patient outcomes improved. The cure rate for hepatitis C, which had been much lower than that of medical centers, rose to the higher cure rate seen at the university. An area currently under investigation is pushing information to disadvantaged populations. We can't expect economically disadvantaged people to have the latest and therefore most expensive gadgets, but a common technology such as cell phones can receive texts on how or when to take medicine, or if you are pregnant, information about how to keep you and your baby healthy. That's a promising route to get individuals the information they need that they would not have access to otherwise.
RA: There remain privacy concerns about the exchange of electronic health record and other health information. Is AHRQ working on projects to remedy this situation?
JW: AHRQ has historically worked in close collaboration with ONC. ONC now has a chief privacy officer that works with the Office of Civil Rights very closely. We provide them with input to help them understand the technological capabilities of health IT products, as well as the information needs of clinicians and patients' information needs and desires. AHRQ has a lot of work going on looking at patients' information needs, where and when do they want information, what their attitudes about that information are, and who they need to see it. For example, if I'm the caregiver for my 95-year-old mother, I need access to that information and chances are she's going to want me to have access to that information. There shouldn't be a barrier to my getting access to that information. I'm guessing the clinician will want me to have that information too so I can be fully informed about I need to do. So they are trying to work out those issues to enable better quality. ONC and the Office of Civil Rights convened a March 16 roundtable on mobile apps, privacy, and security, where I moderated one of the panels composed of providers who are using mobile technology to improve the care they deliver.
Editor's Note: More information on AHRQ's Health IT program can be found at http://healthit.ahrq.gov/portal/server.pt/community/about/562.