A Conversation With the Nation’s Health IT Chiefs
Attestation for stage 1 meaningful use is underway in radiology, and expectations are rising about the ability of IT to reduce cost and increase quality in health care. In separate interviews with Radiology Business Journal, Farzad Mostashari, MD, ScM, national coordinator for health IT, and Todd Park, CTO for the US DHHS, clarify their perspectives and positions as the nation’s most influential health IT appointees. Framing National Health IT Policy Farzad Mostashari, MD, ScM, joined the DHHS Office of the National Coordinator (ONC) for Health IT as national coordinator in July 2009. RBJ: What are the obstacles to having a national health-care identifier issued at birth to help in aggregating each person’s health information? Mostashari: The one obstacle to that is that Congress has been quite clear that it’s not something it wishes to see any money spent on, and that’s been consistent over several years. RBJ: What’s your opinion about how valuable it would be? Mostashari: I think we should probably leave it at that. I’ll say this: A lot of people assume that if we had a single patient national identifier, it would solve all problems having to do with patient identifiers, patient matching, and records being misattributed. I think there’s a little bit of magical thinking around that. I’ve suggested we move on and think about how we can find better ways within the world we live in, where we don’t have a national health identifier at birth, where we can do a better job at having identity validation as a service (whether it’s commercial based or state based) and work on improving our workflow around data collection and data quality. That may end up doing a lot more for the issue than I think people assume a national patient identifier would. RBJ: What about interoperability? What can/should government do to encourage vendors to make it easier for health information systems to communicate? Mostashari: We see the government playing a critical and limited role. There are three areas where we are active: first, simply convening people to come up with shared solutions to common problems—bringing people together with a sense of urgency. It’s about providing a place where competitors can come together and work on common solutions around the standards. The second area is curating the collection of the standards and implementation specifications that can be used and reused to solve certain problems, so we don’t end up reinventing the wheel for every problem people have. If you dealt with medication codes for medication ordering in one system, you should reuse that, if we’re talking about quality measures and it involves a medication. The third area is enforcement through certification of electronic medical records (EMRs) and testing tools as part of that process. It really starts first with convening industry, academics, provider groups, and others around the standards and interoperability. RBJ: Are we unlikely to see a top-down approach to setting a national standard? Mostashari: It’s both bottom up and top down. The development of the standards is done from the bottom up in the sense that we’re not going to pay a contract to someone to come up with what the standards are going to be. It really has to come from the people who are going to have to live with it. The top-down approaches are the convening and the enforcement. I actually think the bottom-up approach is more important than the top-down approach. RBJ: Many specialties complain that the generic set of meaningful-use criteria forces them to invest in meaningless IT capabilities or in duplicative work. How are you addressing these complaints? Mostashari: It is a challenge. It’s a national program. Eligible professionals come in many different specialties. To have a common platform and a common set of measures that all specialties are going to feel are equally relevant to them is a real challenge. It becomes particularly difficult when we’re talking about some specialists, like radiologists, who often don’t have direct patient contact and opportunities to do some of the things you would want to have as meaningful use for the large majority of physicians whose job (in large part) is patient contact. It is a challenge, and we and our CMS colleagues have tried to provide accommodations and exclusions where something is not part of the scope of practice. The ACR® has been quite helpful and constructive in its comments and suggestions about how we can make meaningful use even more relevant to radiologists. We thank the ACR for that, and we are going to strive to make it ever more relevant. Journals like yours can help people understand what the accommodations and exclusions really are, and whether they are something that makes sense for their practice. If virtually all of their work is done in the hospital setting, they really are not eligible professionals. They are really hospital-based physicians, and the hospital would be embarking on meaningful use as an entity. RBJ: The free exchange of images among disparate health-care providers could help reduce health-care costs by preventing duplicate studies. How do you think health-care policy and regulations could promote this? Mostashari: It’s a cost issue. It’s also a patient-safety issue. Unnecessary radiation doses have a cost to human health, as well as a financial cost. There certainly is a rationale for including interoperability of images. Our Health IT Policy Committee and Health IT Standards Committee did not include these as part of their latest recommendations for stage 2 of meaningful use. RBJ: Why was that? Mostashari: I’m not sure why they didn’t. In the past, there was some concern about the distinction between EMR systems and RIS/PACS (and the appropriate line to draw there), whether we are talking about being able to view images or being able to include radiology systems as a whole as part of certified electronic health record technologies. There were still some questions about the standards, in terms of the different flavors of DICOM that are currently implemented among different proprietary systems. That being said, we have heard from the ACR and other specialty groups about imaging being a low-hanging fruit—maybe more low-hanging than our federal advisory committee had considered—and we’re looking at those comments very carefully. Cracking Open the National Health Data Vault Before joining the DHHS as CTO in August 2009, Todd Park had already cofounded athenahealth® (Watertown, Massachusetts) in 1997; over the following decade, he helped lead its development into one of the most innovative, socially oriented, and successful health IT companies in the industry. His charge is to help DHHS leaders harness the power of data, technology, and innovation to improve the health and welfare of the nation. RBJ: You’ve compared the DHHS, with its mountains of data that are not being used, to the US National Oceanic and Atmospheric Administration (NOAA) in its past state. What is the practical potential of all those data in achieving the triple aim: improved access and quality at reduced cost? Park: The DHHS has been engaged, for almost two years, in the Health Data Initiative. It’s an effort to open up access to (and improve the usability of) vast amounts of data and information that are sitting in the vaults of the DHHS, including CMS, the National Institutes of Health, the FDA, and the Centers for Disease Control and Prevention. It includes everything from hospital quality data to the latest and greatest medical knowledge in the National Library of Medicine and FDA recall data. We’ve created a new website, healthdata.gov, that’s a one-stop shop for all the data we’re making freely available and downloadable—by anybody, without intellectual-property constraint. We are promoting the existence of these data to innovators across the country, who are using them as fuel for applications and services to help consumers take control of their health care, to help physicians provide better care, to help employers promote health and wellness, and to help local policymakers make better decisions. The whole effort is modeled on what the government previously did with NOAA weather data and GPS data. The government, for many years, has made weather data collected by NOAA’s National Weather Service openly available and machine readable, downloadable by anybody for free, without intellectual-property constraint. That has powered a whole host of innovations in the private sector: weather newscasts, weather news sites, weather mobile apps, and other services that have created huge value for the people of the United States. The government did something similar in the 1980s, when it liberated GPS data, which now fuel everything from foursquare™ to your iPhone to supertanker navigation systems and everything in between. Health Data Initiative is running the same open-data and open-innovation play, but this time, with health-related and health-care–related information that’s been sitting in the vaults of the DHHS and sister agencies. RBJ: You’re basically putting the data into a format that others can take and use; are you doing much analysis yourselves? Park: There’s a famous law we like to quote that’s really an underlying principle behind the Health Data Initiative. It’s called Joy’s Law, attributed to William (Bill) Joy, the cofounder of Sun Microsystems. He is believed to have said that no matter who you are, most of the smartest people work for someone else. Our corollary to that law is that if you want to maximize national social return on DHHS data, don’t just have the smart DHHS people turn the data into tools that can help people. Have all the other smart people in the world—who vastly outnumber us—access and use the data and turn them into tools that can help people. RBJ: What interesting solutions to problems have come out of that program? Park: There’s an extraordinary array of new or upgraded tools and services that people have built. In the category of tools that help consumers, there are applications that help you find the best health-care provider for you and your family. There are applications that help you find the right clinical trial that might be of interest to you, as a patient. There are search engines that ingest large amounts of data from the government and other sources to help you search for information, especially on the Internet, where it’s much more reliable, accurate, and targeted than it is normally. There are tools that help you, as a consumer, get the latest and greatest patient-education information from the National Library of Medicine—at your fingertips, exactly when you need it. There are tools that help you find affordable, healthy food, and other resources that may not be health-care services, strictly speaking, but that definitely can help you take control of (and improve) your health. For health-care providers, there are tools and services that help you deliver better and safer care; that help you actually make the transition into becoming a medical home, an accountable-care organization, or a bundled-care team; and that help you coordinate care and practically manage the health of your patients by leveraging the power of data and technology. There are tools that help you make a referral to the best possible provider and tools that help you deliver the latest and greatest educational material to your patients. What we’re seeing really does prove Joy’s Law. We’re taking data that we have and pushing them into the public domain. Innovators across the country (with all of their talent, capital, developers, existing platforms, and audiences) are leveraging the data and their platforms, capabilities, and capital to do amazing things at incredible speeds. RBJ: As health-care superusers, older people tend to be more interested in health-care issues than the young and invincible are; how are you engaging more tech-savvy young people in the effort to improve the quality and cost of health care? Park: As part of the work to educate developers and innovators across the country about the availability of these data, we have been in partnership with other organizations, such as Health 2.0 (San Francisco, California), on a whole series of challenges and open public competitions that anyone can enter to build the best application or product that, for example, helps consumers find the best health-care providers. We’re seeing a lot of new innovators join the health-care–innovation ecosystem through these challenges and Code-a-Thons. RBJ: What have you learned, through your work at athenahealth and your other endeavors, that has been useful in your current position? What IT solutions from business has health care been slow to adopt? Park: One of the most interesting lessons that has been transferable from my experience as a private entrepreneur to my experience here at the DHHS has been the use of lean startup techniques to drive challenge. Lean startup is a whole philosophy that has been articulated by Eric Ries.¹ It’s a set of principles (see box) that entrepreneurs can follow to maximize the probability their startups will work. The whole notion of rapid interdevelopment is that you don’t have development cycles that are nine months long. Have them be nine days long. It’s about working closely with customers to maximize how quickly and how effectively you learn about what your customers’ real issues are and how you can really help them. That actually describes pretty nicely what I learned the hard way to be key factors in building a successful project at startup. As it turns out, those are the same principles that apply if you really want to drive maximum success and maximum speed for major-change problems in the government. RBJ: Information exchange and general transparency efforts have been generally slow to be adopted. What is the rate of recent progress? Park: I think we’re in a moment of enormous positive change on both of those fronts. I think the genie is out of the bottle, and there is a lot of progress that’s being made that’s very exciting.