You’ve talked about the over-architected HIE in the past. Do you think that is the cause of some of the state HIE shutterings like Tennessee, California, and Kansas? Is that the main challenge that states are running up against—trying to do too much?
In general, I certainly think there is something to that. I think there are a certain number of local drivers that are very specific to the states. I think that one common thread across them is there was a sense of a big bang, a do-it-all-at-once type of thing, and a presumption that if we build that type of thing, that a) you can build that in short order and b) the market will value that; they’ll want to connect to it; and want to pay for those services to sustain it going forward. I think a number of them have that issue, along with the other local issues. It is partly driven by the way the federal program was designed, the grants very much called for organizations to direct states to create those types of plans. In part they were responding to what the people giving the grants were asking.
Do you think the Office of the National Coordinator (ONC) has been doing enough to support state HIE efforts?
I think the reality of this is that it’s incredibly hard on a number of dimensions. It takes a lot of time to sort through a lot of the issues we are talking about in a country that is as heterogeneous as the U.S. One of the challenges was that these [state HIEs] were funded from the economic stimulus, which by its very nature was trying to get dollars back out into the economy. I think with health information exchange, what we’ve seen is that it’s more complex than what was originally envisioned. [HIEs] tends to differ a lot by state because state regulations start to matter and privacy laws are different. It just takes a while to figure all that out.
The other complexity of health information exchange is that it has to wait for EHRs to be widely adopted in order for it to work, and we didn’t even have EHRs widely adopted. Implementing eClinicalworks in a practice in Idaho is pretty much the same as implementing eClinicalworks in a practice in New Hampshire, whereas implementing a health information exchange in New Hampshire is really, really different than implementing a health information exchange in Idaho.
I think when we look back and reflect on all this, we’re going to see that most state-level HIEs don’t survive. Not one of them is really sustainable right now. Some of the regional ones like NEHEN, Indiana, Cincinnati, Buffalo, those that have been here for awhile. They were sustainable before and will be sustainable after. But a lot of the ones created from HITECH are probably not going to survive.
I would argue that [state HIEs] were very successful in that they brought focus to the policy and business questions that are fundamental in understanding how to get health information exchange to work. And that infrastructure and knowledge in every state will be the foundation for private HIEs to take off with accountable care. From what I’ve seen is a lot of [private HIEs] have benefited a lot from that statewide HIE activity, not because statewide HIE has created technology, but they’ve grappled with [questions like] what is state law on sending data back and forth. Also it may be that [private HIEs] are a measure of success of the public HIE, which short of having actually established a statewide network, have created a huge corpus of knowledge for the private sector to take and use their own investments with much more confidence and more aggressively than they would have before because of all these questions of legality and HIPAA that are vexing.
Do you think that the federal Direct program is the right approach for all state exchanges or one of the many approaches? Some have said that they had already made progress on interoperability and ONC’s push of Direct was forcing them to take their eye off all the other balls they were juggling.
I think it’s a fair point for some of the more mature exchanges that feel that. My perspective is in order to succeed, it’s probably necessary to offer to the market some type of directed exchange capability. Direct, with a capital ‘D,’ is one way to accomplish directed exchange, but it’s not the only way. I think directed exchange is absolutely required. I’m just extrapolating from what you just said, that the organizations that feel that this is a step backward, they probably have some type of repository for interoperability. That’s one way of thinking of how that would work. But that’s a tall lift to get everyone to a place where they’re willing to make all their data available so others can query that data. The idea of directed exchange is that it is another way for the primary care physician to make the information available to the specialist by sending it point to point. The reason that I would say that is absolutely fundamental is that it provides a lower level accessible service for provider organizations to be able to participate in your HIE.
What are the main obstacles that still must be overcome to get health information flowing on a larger scale?
In order for the statewide HIE to really be successful, it needs to think about the lowest common denominator service which it can provide that would be a benefit to all but doesn’t pretend to solve all of the problems that any local provider organization has. If you say from the beginning that a statewide HIE’s way of solving interoperability is to create a big database in the sky that allows everyone to query and have access to everyone’s records, that’s a very tall lift, and it’s going to take a lot of time. There are all sorts of reasons. Some of them are related to the understandable, yet cumbersome nature of state laws and regulations around the way money has to be spent and the accounting of taxpayer dollars that slows down processes in state government.
If you have a very diverse state like Massachusetts or Illinois, it’s very hard for state HIEs to think of a complex set of services that are going to be universally valued. To me that common denominator is just to start with the ability with everyone to have access, at affordable terms, and be able to send and receive clinical information securely with other participants on the network. I would argue that it complements what private HIEs are trying to do. What you don’t want to do is be in the position of trying to get in the middle of what the provider organization is trying to do with its network participants. For example, you might have a large hospital system that is trying to create an ACO-type model. If you have a state-level HIE trying to come in the middle of that, it’s a recipe for disaster, and you’re going to have all sorts of conflicts. Then the private HIE thinks it’s in competition, and that is getting in the way of what it’s trying to do in the market.
In Massachusetts there are organizations like Baystate Health, Partners HealthCare, and Beth Israel Deaconess that have very robust private HIEs, and the statewide HIE is not trying to solve the problems that those private HIEs are trying to solve. What [private HIEs] do value in the statewide HIE is the connection between the ACOs. If I can use the statewide HIE to send stuff to Beth Israel for that 10 percent of patients that go there, then I’m happy to connect to it and have it solve that problem for me. That’s where a lot of these [state HIEs] are starting to see value. If it offers basic services at a relatively low cost and is really just a complement to what you’re doing, so your private HIE can connect to others in a way that will allow the passing of information for patients that go across these private HIEs.
