Bringing data to DC: Q&A with health data’s biggest evangelist HHS CTO Bryan Sivak

Bryan Sivak has lived the Silicon Valley dream — in the last 15 years, he co-founded two startups, one of which was acquired by Oracle (s ORCL) two years ago. But instead of sticking around to start another company or taking the venture capital route, he wound up across the country in the center of government.

After holding chief technology positions with the District of Columbia and the State of Maryland, last year he was appointed Chief Technology Officer for the Department of Health and Human Services. Ahead of a trip back West, Sivak talked with me about how open data (an increasingly hot topic in government) can drive big changes in health care, where digital health is evolving and why entrepreneurs should give DC a try. Take a look at a (lightly edited) transcript of our chat.

GigaOM: You went from fast-paced Silicon Valley to bureaucratic Washington, DC, what was the biggest adjustment challenge?  

Sivak: The most disconcerting thing, at first, is that because you’re working for an entity that is essentially operating on behalf of the taxpayer, there is this constant spotlight shining on the work that you do. I don’t think that’s a bad thing at all. In fact, I think it’s great — it’s the reason that, I think, I and a lot of other people actually do it. But you very quickly get used to it — you just kind of take it in stride.

capitolGigaOM: At the SXSW Interactive conference you talked about how government can learn “Lean Startup” principles and other Valley-style ways of thinking. What can the Valley learn from DC?

Sivak: I’m a big fan of disruption and bending or breaking the rules to do interesting things. But, at the same time, I think it’s important to realize that there are some rules that are there for a reason and, in many cases, disruption needs a partner called sustainability. I exist in this massive agency right now — 90,000 people work for HHS — and while I’d love to activate the potential of every person there, it’s important to recognize that there are people who are well-suited and who actually should be working on the sustainability aspect … keeping the trains on time and that sort of stuff.  That’s something that’s often overlooked by people who come to it strictly from the Valley mindset.

GigaOM: You’re speaking to a group of entrepreneurs, programmers and designers at the Health Refactored conference [this] week about innovation in health care. HHS has put a lot of effort into opening up health data – which is obviously an important first step – but what else needs to happen to spur innovation?

Sivak: It’s a massively complex ecosystem and environment. [And] one of the things we can do as experts in this area and the government is help by educating people, by doing a better job of describing our data sets, by doing a better job or doing a job or basically explaining the problems that we have and the problems that we want to see solved. There are millions of examples out there of things that can be worked out but people just don’t know there are problems and don’t know it’s something that should be worked on. And that’s a place we can help.

GigaOM: It’s barely a year into your tenure at HHS. But when you look at where we need to go, how far along are we on the progress bar?

Sivak: We’re just at the very tip of the iceberg here. We’ve been working on this for a few years now, long before I got there. My predecessor Todd Park, now the CTO of the United States, kicked off the idea of the data that HHS has as being critical to revolutionizing the system. A few years ago, we started the process of data liberation — changing the default setting from closed to open within the department. That’s been the big focus and I think we’ve been successful with that but there are still pockets of resistance where people just don’t really understand what the value is and there are other complications, such as privacy restrictions and things we have to take very seriously.

GigaOM: What has to happen next?

Sivak: There are two other phases we have to start working on. The first one is around dissemination of that data – we have a website called and it’s the one-stop shop for HHS data. To date, we’ve probably catalogued 40 percent of the data sets that exist at HHS, about 400+ data sets, and a much smaller number, 34 or so, have APIs attached to them. There’s still work to be done of the dissemination side, and that also includes some questions we’re wrestling with now. For example, an important feature should be an area where people can come and collaborate and discuss and ask questions and get answers. And we’re trying to decide whether that kind of forum should happen on our government website or on a third-party website that’s charged with potentially building that community.

The [other] piece is data education – explaining data better, teaching people how to work with the data in a better way and connecting people with experts in a relatively regular fashion so they can get answers to their questions, understand the best ways to use the data, etc.

Fitbit, mobile healthGigaOM: Judging by booming investment, accelerator programs, startup launches, and other activity in the sector, digital health seems to be seeing a lot of innovation, but what kinds of innovation do you think are missing?

Sivak: The beauty of it is that it’s a massive industry and there’s so much room for people to innovate it’s insane. Some areas that I’m personally interested in and I think are interesting to the department, off the top of my head: there are tons of people running around right now with devices generating personal health-related data from Jawbones, Fitbits (see disclosure), mobile phones that capture stuff, you name it. But that data, right now, is very inactionable. There’s no advice, no pro-active suggestions, none of that – that’s one big area.

And integrating that personal health-generated data with clinical data, i.e. the stuff that your doctor generates, is massively interesting. Imagine that you have some kind of chronic condition, like diabetes, wouldn’t it be interesting if you stepped on a scale every morning or took your blood glucose everyday and that got automatically transmitted to your doctor? And the system that your doctor has in his or her office alerts him or her if you have some kind of a problem or if your stats are going in the wrong direction, so they can intervene early? I think that’s a massive area.

Also, nobody has figured out patient engagement yet. And there are some people out there, myself included, who believe that the patient is the single untapped resource in healthcare right now.

GigaOM: How can open data help support health reform and Obamacare?

Sivak: Data is one of the absolutely critical components to reforming our health care system. The big challenge with our system today is we exist in this transactional, fee-for-service environment. When you see your doctor see you in [her] office or he or she performs a procedure or test, they get paid for those things. The problem with that is the incentives are somewhat backward – it incentivizes transactions. It incentivizes people to go in to be treated when they’re sick, as opposed to being kept healthy. In order to fix it, we have to move to a system where we pay for value and outcomes.

In order to do that, we need the data. If we don’t have the data in terms of what happens when somebody gets treated with a certain drug or what kind of drug interactions exist or how effective a specific treatment is, then we’re never going to be able to incentivize the providers to do the right thing, And if you’re a provide or a doctor you’re not going to be able to figure out what the right thing to do is. Liberating this data is incredibly important to fixing the system.

health data visualizationGigaOM: Every though there’s been a lot of buzz about the Affordable Care Act, there’s still a lot of ignorance out there. A recent survey found that 42 percent of people polled didn’t even know it’s a law. How can data help people understand health reform?

Sivak: What that [survey] indicates is that we need to do a much better job talking about it and marketing the value of it. I think the way we can do that is through interesting and creative uses of the data. People don’t remember statistics… but when you tell the stories with the data, that’s what they start to remember. I heard a great quote the other day, which is perfect for this: the singular of data is anecdote. That’s what we have to get to if we want to sell the value of this thing.

GigaOM: What are some of the most interesting or promising uses of the data you’ve seen so far? 

Sivak: For the first time [last week], we released [hospital pricing data] in an easy-to-access public format. [It’s] the actual prices that hospitals charge Medicare for the top 100 procedures across the country and [it shows] that the prices that hospitals charge, even when they’re right next to each other are wildly different. It’s crazy. But here’s a fun little statistic: in the first day that this data set was available online, we had 110,000 downloads.

Another one of my favorite examples is what [healthcare data journalist] Fred Trotter is doing. I love [it], not necessarily because the work that he’s doing to build this social graph of doctors will determine anything interesting, but the fact that he had this idea. That he’s not a medical professional or a subject matter expert and he had this idea to take these two random numbers in a claim and use them for something that could potentially be interesting is what I think is incredibly indicative of the power of the stuff and bringing people in who are not subject matter experts.

GigaOM: You’re not just trying to convince entrepreneurs in the Valley and elsewhere to work on digital health, you’re trying to recruit them to work in government (at least temporarily).  What does DC offer that the Valley can’t?

Sivak: This is actually a very simple answer: because we can give you the opportunity to solve, literally, the most pressing problem in American society today.