HHS Unlocks Health Data to Empower Public Health

Eli Y. Adashi, MD; Todd Park


March 30, 2011

This feature requires the newest version of Flash. You can download it here.

Eli Y. Adashi, MD: Hello. I'm Eli Adashi, Professor of Medical Science at Brown University, Providence, Rhode Island, and host of Medscape One-on-One. Joining me today is Todd Park, the first-ever Chief Technology Officer of the US Department of Health and Human Services (HHS). Mr. Park is entrusted with harnessing technology to promote openness and foster innovation. Prior to his appointment Mr. Park cofounded athenahealth, Inc., a leading health information technology company. Welcome.

Todd Park: Thanks so much.

Dr. Adashi: Absolutely wonderful to have you.

Mr. Park: It's wonderful to be here.

Dr. Adashi: Openness.

Mr. Park: Yes.

Dr. Adashi: With the Community Health Data Initiative launch you are now in the thick of unlocking the power of governmental databases with consumers, providers, and researchers in mind.

Mr. Park: Yes.

Dr. Adashi: Could you expand on the basic concept and the inspiration this project derives from the weather data?

Mr. Park: Absolutely.

Dr. Adashi: Ecosystem --

Mr. Park: Absolutely.

Dr. Adashi: Seemingly a remote analogy but not really.

Mr. Park: You're absolutely right. The origin of the entire Community Health Data Initiative is an idea that Deputy Secretary Bill Corr had. We were brainstorming one day and he said, "Look, the thing about HHS, Medicare, Medicaid, the NIH [National Institutes of Health], the CDC [US Centers for Disease Control and Prevention], the FDA [US Food and Drug Administration], etc is that we're sitting on tons and tons of data, just extraordinary data that have the potential to help illuminate, inspire, generate insight, and power all kinds of good in the world. It is mostly sitting pretty quietly in our vaults relatively underutilized. What can we do to really get it out there and turn it into massive new social good?"

We studied different models, and the model that came up as the one that we loved the most was the weather, specifically what the National Oceanic and Atmospheric Administration (NOAA) does with weather data. What I actually didn't know before I joined the federal government about a year and a half ago is that virtually all weather data come from NOAA. It's not as if your local newscasters measure the barometric pressure themselves every morning. It's not as if ABC News has its own hurricane sensor network over the Atlantic. NOAA collects all that data and then publishes them online in machine-readable format for free download by anybody without intellectual property constraint.

Then a bunch of innovators outside NOAA turn it into The Weather Channel, weather.com, iPhone weather apps, weather research, the sheet in your hotel that said it's going to rain tomorrow. It's innovators outside government who take data that the government supplies and turns them into products and services and applications that benefit the American people and create a lot of economic value along the way.

So we said, "Look, why can't we do that? Why can't we become as HHS the NOAA of health data?" We tested that idea initially with a group of about 45 innovators who we gathered together -- healthcare experts and tech experts -- and they loved the idea. Over the course of about 6 hours -- this was March 11, 2010 -- they brainstormed about 20 different classes of applications that you could build using specifically community health data: smoking rates, obesity rates, and provider quality data such as our hospital compare data and nursing home compare data, etc.

At the end of that meeting we then challenged this group of innovators and their friends to build the applications that they had brainstormed. We said on June 2, 2010 that Secretary Sebelius will host a giant public meeting with Harvey Fineberg, President of the Institute of Medicine, that if one of you can actually build one of these apps we'll make you famous.

Sure enough 90 days later, less than 90 days later, these innovators debuted over 20 new or upgraded applications that leveraged our community health data and provider data to produce all kinds of useful benefit for consumers, providers, employers, and communities.

Search engine functionality is where health data got integrated in the search so if you typed in the name of a hospital it would actually pull up relevant data on that hospital. Community health dashboards would give a community a sense of where they stood: other communities in terms of health and what other communities are doing to improve health against a given indicator such as smoking or immunizations. Decision-makers could use powerful new analytical software to help make the right health investments: health games that can teach you a lot about health as you are having fun and so forth. It was really inspiring.

The punch line for us was that the government only published data and marketed data to the right innovators who then turned it into incredible stuff that could benefit the public in a very short span of time. We're really excited about this and putting the pedal to the metal on it and publishing greater quantities of data and marketing more broadly to innovators who can turn it into useful products and services, just like what's happened with weather data or in the past with GPS or financial data.

Dr. Adashi: In the interest of our viewers the data can actually be found today under which portal?

Mr. Park: It's funny that you ask that question because it has been scattered in a bunch of different places in the past, but as of now you can go to a site called HealthData.gov, which is a new 1-stop shop where you can find a catalog of every single federal health-related data set that's available for free, downloadable, machine readable, and without intellectual property constraint that anyone can use to power their applications, their research, their insights, their uses, and their services. It's not just HHS data. There are also data from the USDA [US Department of Agriculture], the EPA [US Environmental Protection Agency], etc because a lot of those data sets frankly are not more important to health than healthcare data sites.

Dr. Adashi: It's a government-wide --

Mr. Park: Government-wide, yes --

Dr. Adashi: Data set --

Mr. Park: Yes. In addition we have a growing list of nonfederal government data sets from states, from localities, and from the private sector that we think are useful to the community: data sets that are being provided on the same basis that we're providing our data -- free, machine readable, and downloadable without intellectual property constraint.

On top of that there is a link to a virtual apps expo where you can see a growing cross section of applications that utilize the data to do useful things for consumers, providers, employers, communities, and others. On top of that there is an online community where you can join in on conversations about the data. You can ask questions about the data, complain about the data, suggest improvements to the data, request new data sets, and talk about what other people are doing with the data and what you're doing with the data. It's a 1-stop shop for data to see what other people are doing with the data and to talk about the data.

Dr. Adashi: Talking about apps, perhaps you can share with us one particular application on which your entrepreneurial juices rank high on the list as useful, informative, and fun.

Mr. Park: Absolutely. I think they're all fun. That's just me because I'm such a geek, but I would talk about particular classes of apps emerging that I think are really useful, fun, and powerful. An emerging class of applications aim to help physician groups and health systems make the transition to accountable care organizations such as moving from fee-for-service production to proactive population health management: powerful apps leveraging data that help provide real-time analytics, guidance, and situational awareness for physician groups and health systems that are now going to be practically managing the health of their population. That's very exciting to me as a geek and as a health data and healthcare enthusiast. That's one class of apps.

Another class of apps help consumers and patients research providers: which provider to go to. What kinds of providers might be available? For example, there's an app called iTriage, which is a very popular and rapidly growing Android, iPhone, and Web app that helps you research medical treatments and find providers. They recently integrated the complete directory of HHS community health centers, ie, the community health centers that the HHS helps to fund and make searchable on their app. They've gotten tens of thousands of searches already for community health centers for free and low-cost care, which is just wonderful.

Dr. Adashi: That's terrific.

Mr. Park: So those are 2 classes that I find particularly exciting, and there are others as well. I mean actually one: just a third one. I have so many favorites that it's difficult to narrow it down. Health games. One of the apps that was demoed last year as part of that 90-day challenge to innovators was a game called Community Clash, which you can play at communityclash.com. It's basically healthcare blackjack. Here's what happens: Go to Community Clash; pick a city like Boston, Massachusetts; pick an opposing city like New York, NY. You're dealt 5 cards facedown, each of which corresponds to a health indicator or a driver of health such as obesity, access to fresh fruits and vegetables, smoking, etc.

You're allowed to swap out up to 2 of the cards for other indicators that you think will beat New York. Then you show your cards, and the objective is to, of course, beat New York. It sounds silly, but at the floor show after the main session on this June 2 meeting where people could go and browse any of the 20-something apps that had been built or upgraded the longest line by far was for this Community Clash game because it's fun. I played it for hours that night with my wife. It was very interesting and entertaining. In addition to having fun I actually learned a lot that I didn't know about health, and I've been in healthcare and health for the better part of my years.

For example, I learned that rates of mental illness in the United States are much higher than I had thought. I kept trying to find a city where I didn't think that they were really high, and I couldn't find one. That will influence my decisions as a healthcare leader going forward. Consider Farmville. Are you familiar with Farmville? Farmville went from zero to how many users in 18 months?

Dr. Adashi: No idea.

Mr. Park: Seventy-three million is what I've been told.

Dr. Adashi: That's remarkable.

Mr. Park: About 70 million monthly users in 18 months, and it's just a little game right where you farm and trade sheep with other people on Facebook. I submit to you that the person who invents Healthville and takes Healthville from 0 to 70 million users in 18 months will be one of the most important healthcare education and health education figures of the 21st century because they will do so much to help get the word out about health and healthcare issues while delivering fun where the education is a side effect.

Health gaming is a very interesting additional zone, and employers' health plans and health systems are increasingly interested in health gaming as a way of engaging consumers and helping educate without having to go straight up to people and say, "Eat your spinach because it's good for you." Instead learn more about health because it's fun.

Dr. Adashi: Remarkable. I think what you're describing is really taking a stale, unused, hard-to-access set of information and bringing it to life in ways that probably would not have been possible 10 years ago but is very much in keeping with the times. Let's move on to some other examples of openness in public reporting, and I'm specifically interested in the compare suite of Websites. What can you tell us about the current state of the existing compare Websites, and how do you see us going from here to the next level in terms of amount of information, quality of information, and not least of which user friendliness?

Mr. Park: Wonderful question. There are multiple dimensions to talk about there. First, we're taking our existing compare data sets -- hospital compare, nursing home compare, home health compare, and dialysis compare -- and in the spirit of the Community Health Data Initiative liberating that data so that more than just government Websites can display them. We've not only made that data downloadable. It's been downloadable for some time, but last fall we put APIs on the data. API is an application program interface. Just a fancy name for a gate that you put on your data to enable other applications to easily retrieve on a request basis. It's the way that developers like to access data as opposed to downloading a file and uploading it manually into another Website.

We put APIs on that data to make them vastly easier to obtain. Other folks are turning the data into features and applications that are frankly how the majority of the American public is going to access these data. One of Tim O'Reilly's... I like to think of him as the Thomas Jefferson of the Web. One of Tim O'Reilly's laws about data and how data change people's lives is the following: If you really want data to help people don't make people find the data. Make data find people where they're already making decisions that the data can inform. A classic example of this is that Microsoft Bing has taken our hospital compare data and integrated it into search such that for a lot of hospitals not every hospital, but for a lot of hospitals, if you type in NewYork-Presbyterian Hospital it pulls up in Bing's primary search result the patient satisfaction rating with that hospital vs the average.

I recently used this to help me find the hospital at which my wife will deliver our second child. It's very useful. What's great about that is that you don't have to know that the HHS has a hospital compare data set or a hospital compare Website to benefit because all you're doing is typing in the name of the hospital in the search and boom the data are traveling to you. That's --

Dr. Adashi: So that's a Bing initiative in a sense?

Mr. Park: Yes it's --

Dr. Adashi: It's a Microsoft initiative.

Mr. Park: Yes and other providers of different kinds of applications are also integrating the data such as iTriage into their application to help bring the information to the consumer in a way that's maximally user friendly, maximally accessible, and helping people when they're already thinking about and looking for a provider. So that's one dimension. Free up the data so that private-sector innovators can turn them into all kinds of features and applications that help the public in a wide variety of situations use that data in meaningful ways. Bring the data where people already are and expand the range and power of the ways that you can interface with the data. So that's part one.

Part two is that, as you know, through the Affordable Care Act the compare data sets are going to deepen as well. The hospital compare data set will be expanded, for example. We'll be starting physician compare, which is a big new data set.

Dr. Adashi: And you have started it?

Mr. Park: We have started it, yes.

Dr. Adashi: It is online.

Mr. Park: That's right, as a physician directory and an indicator of who is participating in the PQRI [Physician Quality Reporting Initiative] program, but the real action is coming in 2013 when we put quality metrics online. That's being done through a very open, very transparent process critical to have engagement from all dimensions of the provider community and the physician community engaged in that process to make sure that what we put out there is good stuff, robust stuff, and directionally helpful. That open, transparent dialogue involves providers, researchers, government, consumers, patients, etc: a critical thing to keep going and to intensify as we move down the compare path to make sure that we continually make good iterative progress in making these metrics more and more helpful.

Dr. Adashi: Wave of the future frankly. Public reporting was not all that popular when it got started, but the new spirit of openness and transparency that your office and the administration are fostering will obviously change all of that.

Mr. Park: There has been a tremendous push for the private sector as well in favor of transparency, so I think that it's coming from multiple --

Dr. Adashi: From all directions at this point, yes. Let's move on to the notion of patient health records and their accessibility.

Mr. Park: Yes.

Dr. Adashi: Eventually, of course, the national highway of healthcare would go beyond personal health records (PHRs) but would render electronic health records (EHRs) and hospital-based records accessible anytime, anywhere, and without restraints --

Dr. Adashi: Assuming that it's appropriately protected and whatnot. What's your vision of that in terms of where we are today, where we're likely to be in the next few years, and beyond that your vision of this interconnected healthcare world that seems at some level unstoppable and inevitable?

Mr. Park: We're very about David Blumenthal, our National Coordinator for Health IT at HHS. He has been leading a tremendous team doing tremendous work advancing the ball on this front. We're making great progress toward this vision of unlocking the power of electronic clinical data to help the patient and providers take care of patients. The initial steps have been, first of all, to turn the manila folders in doctors' offices and hospitals into electronic form to liquefy them so that they can actually move. Second is promulgating ways in which the information can move securely wherever the patient needs for it to go.

A major step forward on that front was taken recently through the Direct Project. This is a project that exemplifies the best of how the public and private sectors can work together to make progress. Due to principally input from physicians the government posed a problem that needed to be solved: How do we enable physicians and other providers to be able to move a care summary from a primary care doctor to a cardiologist or a lab result from a lab to a doc's office in a simple secure way that anybody could do? We put that out there as a challenge to the private sector and said that we're going to convene as neutral territory a group of willing private-sector companies or organizations to solve this problem, and so a whole bunch of EHR companies, PHR companies, and big and small companies got together and in 90 days put together a very simple specification: a protocol for how to meet physicians' needs that was a secure healthcare email. This was a simple brilliant solution but one that safeguards privacy.

Just a few months after that in January 2011 the first production transaction went live. Now more and more EHR and PHR companies are announcing that they're enabling all of their clients to be able to transact via direct project secure email. A lot of people say that just in 18 months it's going to represent a huge number of transactions and the bulk of how clinical information is exchanged in this country, which is just phenomenal. Right?

There the government just simply posed a problem and was a neutral convener, and the private sector came together in record time to develop a simple obvious spec and then implement it for the benefit of patients and the providers who care for them, so that's really wonderful. Then deeper forms of exchange are being worked on in parallel, but it's wonderful to take this first step and get helpful exchange happening in a practical way and then move on from there.

Another example of this is something called Blue Button. This is something that I've been working on with Peter Levin, the visionary chief technology officer of the Veterans Administration (VA). We had this very simple idea that was given to us again from a couple of brilliant people in the private sector. The idea is this: The VA and Medicare are 2 of the largest repositories of personal health information in the world. The Medicare program has claims information for Medicare beneficiaries, and the VA has, of course, a lot of personal health information for veterans. Both Medicare and the VA have offered for some time portals -- MyMedicare.gov and My HealtheVet -- where Medicare beneficiaries or veterans can go to this portal, get an account, authenticate themselves, and look at their own data on a Web page.

Blue Button is a very simple idea. We added literally a blue button that you can hit to download an electronic copy of your own data that you can then share with other folks and with your provider, upload into your PHR, upload into applications to help you manage your health, and that simple act turned out to be a big deal because that isn't common across the healthcare system. More and more people are Blue Buttoning their data, taking their patient profiles, and adding a Blue Button and telling the old people to actually get a copy of their old data, which is very exciting. We haven't marketed this, but for Medicare and the VA alone there were 200,000 people in the last handful of months who have actually already downloaded a copy of their own data.

Dr. Adashi: So the capability is there?

Mr. Park: Yeah.

Dr. Adashi: And any Medicare beneficiary who has access to the Internet has the ability to download his or her --

Mr. Park: Own data. That's right. Any veteran can as well.

Dr. Adashi: And the nature of these data would be more transactional or would they be of a different nature?

Mr. Park: It depends on the provider of the data. For Medicare they're self-entered data as well as your claims data. For the VA PHR data are being maintained on My HealtheVet, but it depends on the provider of the data. We're very excited that whatever data they are the principle being established is the patients' data and their ability to access them.

Dr. Adashi: Evolution is inevitable, and more will be added and more meaningful information will be there.

Mr. Park: That's right.

Dr. Adashi: We wanted to take at least a little time to discuss one app that excited the medical world in a big way, and that is text4baby --

Mr. Park: I'm a huge fan of text4baby.

Dr. Adashi: Could you briefly describe the application for our viewers?

Mr. Park: Yes.

Dr. Adashi: And perhaps speculate on the utility of the approach beyond the women's health space. Should you believe that there is such a potential?

Mr. Park: We absolutely do, and we're taking action on that front. Text4baby is a brilliant simple app that was put together by a collection of public and private-sector entrepreneurs. If you're a mother or a father you can text baby or en Espanol bebe to 511 411 along with your estimated delivery date, and it automatically enrolls you in a series of --

Dr. Adashi: Your estimated delivery date.

Mr. Park: Yes, that's right. That automatically enrolls you in a series of text messages that you get on a weekly basis reminding you of things to do or give you tips for how to care for yourself as a pregnant woman and how to care for your baby in the first year of life. It's content that was developed by the HHS, and then private-sector folks turned it into a set of 140-character messages that the HHS then blessed as cool. It's being delivered by this very innovative company Voxiva Inc., which has done a lot of text messaging for health-type work in the developing world. This is one of its first major installations here in the United States. It's free. All of the wireless carriers agreed to do all of the text messages for free, and to date -- I think that its 1-year anniversary just happened -- they have 135,000 mothers who signed up, which is just extraordinary. It's the largest free text messaging program for health in the country and growing like crazy.

Dr. Adashi: These are mothers who logged on, subscribed, and then were the beneficiaries of useful information that was trimester specific to their particular gestation period.

Mr. Park: That's right.

Dr. Adashi: They were given information as they went in keeping with where they were --

Mr. Park: Exactly right. It's just a lot to remember. This is a helpful aid to help you remember the things that you need to remember to make sure that you have the best possible pregnancy and the healthiest, happiest baby. This is a type of service that we think has huge potential. The secretary has created a text for a health work group within the HHS that is now a couple of months into figuring out how to clone text4baby in other areas such as smoking cessation, early childhood health, etc. That group is doing extraordinary work, and we hope to have announcements along those lines very soon.

Dr. Adashi: So the principle is extendable, and we're likely to see other applications that in essence would keep consumers/patients informed.

Mr. Park: That's right.

Dr. Adashi: On a personal note have you given any consideration to what you might be doing next should you be returning to the private sector?

Mr. Park: After the HHS?

Dr. Adashi: Yes.

Mr. Park: I haven't thought about it at all. I have no idea. I'm so busy at the HHS. There are so many busy exciting things to do at the HHS. It's all that I can do just to keep up with the HHS. I've been having an extraordinary time at the HHS, and I'm really excited about what we're working on. I'm really excited about the people with whom we're working. I've had the privilege and benefit of having a lot of entrepreneurial experiences in my life, but my time at the HHS has been the most entrepreneurial of my life. It's such a privilege and joy to be there right now. There is so much good to do. There is so much exciting action happening, and I just can't get enough of it.

Dr. Adashi: I would argue that having you at the HHS has been a privilege and a significant contribution to our healthcare system when it comes to the communication, transparency, and openness arenas -- all of which one cannot help but conclude have gained substantially. Thank you very much.

Mr. Park: Thank you.

Dr. Adashi: On that note sincere thanks to Todd Park and to you, our viewers, for joining Medscape One-on-One. Until next time, I am Eli Adashi.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.