Q&A: IT's role in care revolution

Beacon leaders discuss the quality revolution in healthcare IT
By Kate Spies
10:26 AM

Flour, sugar, the last pinch of salt… in crafting the perfect confection, there's often some singular component that holds the rest together, and makes the cake stand up, or the frosting stick. 

[See also: Beacons lead quality 'revolution']

In the realm of public health, IT is proving to be the vital ingredient behind the revolution of patient care. 

At the recent HIMSS 2012 Virtual Conference, leaders from two Beacon communities spoke with Jason Kunzman, project officer for the Office of the National Coordinator for Health Information Technology. The session was titled, "Beacon Comunities: Leveraging Health IT to Fuel the Quality Revolution."

Chris Chute, MD, a principal investigator for the Southeastern Minnesota Beacon Community and Lacey Hart, SE MN Beacon's program manager, outlined the CDR and HIE technology that's driving enhancements in their community. Keystone Beacon Community HIE Director Jim Younkin and Doreen Salek, case manager at Geisinger Health System, highlighted the technology behind the central-PA health system. 

To conclude the event, Kunzman asked the officials questions that were posed by the audience. In this Q&A session, the Beacon leaders discussed the future of HIEs, the importance of trust models, data normalization in relationship to CDRs and HIEs, and the measures they've taken to meet security requirements as IT plays an increasingly prominent role in public health. 

[See also: Beacons put community healthcare strategies to the test]

What will it take to make HIEs national? Will we see that in our lifetime?

Kunzman: I will just say that the office of the national coordinator wants to see substantive and robust exchange really take root in 2012 – for those of you that have seen the proposed ruling for Stage 2 meaningful use, I'm sure you've picked up on the fact that we're driving hard towards interoperability and exchange. ONC has sponsored two sort of open-source solutions toward that end, one being the Direct Project, and the other being the Nation-wide Health Information Network. 

Chute: Clearly it's not a technology challenge or problem. There are a number of technology solutions with respect to push or pull models that are sustainable. What is a major challenge, of course, is the trust model – how do we understand to whom we're sending, and to whom we're requesting? Think of the scenario; without a very well-formed governance model and trust network, I could conceivably set myself up as a spam healthcare organization and solicit health information exchange records - this is the nightmare scenario - on celebrities or politicians from their care providers. And who's to know that I am who I say I am? And who's to know that these requests are legitimate and validated? That's when we get into notions of a trust network and a governance model. Now, this is well-recognized by everybody, and I believe a request for information on a governance model was released fairly recently. It is a major topic of ONC, and the standards and interoperability framework. It's something that is very well recognized. But at the end of the day, we will have to have a trust network that works, and more pertinently, one that the public will have confidence in. 

The two Beacon communities that are on the line right now (SE MN and Keystone) are led by two large delivery networks. Some may walk away from this with the impression 'Well of course they can do this, they're large IDNs; they have the resources and the tools to be able to pull something like this off.' What do you say in response to this 'IDN-or-bust' type of perspective?

Younkin: Well I think that's a good point, and a good question, and you're right; we do get asked that a lot. We've certainly been able to learn from a lot of the work that Geisinger has done in its medical home model in terms of how to create these kinds of initiatives. But what we have done through the Beacon program is we have developed a community–based model that really uses some of the same infrastructure, but it pushes the tools out to community members that are not part of the Geisinger Health System, and in fact in many cases are competitors of Geisinger. We believe through the right kind of governance, and by having a care coordination team that is working on behalf of all of those community members, that this model is entirely possible to be replicated nationwide, regardless of what the organization is that's leading it. And of course this is the predecessor to the accountable care model; we'll be working on setting up an accountable care company to use these tools exactly in that community model, as described. 

Salek: I would just say that, as Jim said, some of our participating providers could even be considered competitors, and all but one of our sites are non-Geisinger physician sites. So we feel that the care model, that Jim had spoken too, didn't have to really change or be tweaked going out to these sites. How we connect with those sites, how the nurses have technology at their sites to document, and how they stay connected between all those facilities, is really not about Geisinger per se, but really about our HIE, and how we also leverage the already-existing EHRs in those practices and how we connect… What we do is bring the common care model and, as Jim said, push out to them some of the ideas, but it's not our Geisinger technical model, because these are non-Geisinger sites. 

In blending the varied environments, what has proven to be the biggest challenge?

Chute: It depends upon the use case. Quite frankly for health information exchange where you're going to present the information to a human being, humans are very adaptable in terms of the consistency and comparability of the data they get. They sort of 'get-it' if they read it. If on the other hand we're tying to do quality metrics, or best practice discovery, then the data-normalization thing starts to become really important. We make very limited efforts to do data normalization in our health information exchange. We make extraordinary efforts to do data normalization in our clinical data repository because that's when we're really posing questions. If the data's not comparable, we can't really ask it questions. 

How does your model take care of complying by Faith Harbor HIPAA requirements when releasing data for healthcare clinical research related activities?

Chute: Simplistically, if it's used for discovery research that is intended to be published, then the usual patient privacy and protection rules of institutional review boards, patient consents, and corresponding types of issues, pertain. In the state of Minnesota we have what's called Minnesota Research Authorization, which is required by Minnesota law before we can inquire a patient record, even if it's an anonymous inquiry. So we're extremely respectful of the protection and constraints on use for research. On a quality-metric perspective, that's considered part of doing business, and our ability to establish whether metrics are being met, or what those metrics actually are, is allowed under Minnesota law, and I believe under HIPAA. Of course then we are not able to publish or disseminate that, because what we achieve under quality by IRB convention and use is not equivalent to what we would disseminate as a shared learning resource and public knowledge.