How blockchain can help improve patient identity matching and consent
BOSTON - "We have to come up with new paradigms if we are going to succeed in this industry," said Amar Gupta, a professor at MIT's Institute for Medical Engineering and Science, as well as its CSAIL Lab, in a keynote address at Converge2Xcelerate on Monday.
At ConV2X – hosted by Blockchain in Healthcare Today and Telehealth and Medicine Today, and presented as part of the pre-conference events of the HIMSS Connected Health Conference – Gupta (who was recently named editor-in-chief of TMT) made the case that healthcare has been too insular and too hidebound for too long.
"You really need to respect the interdisciplinary work that needs to be done in this area," he said. "If you don't do that, you will not succeed in new innovations."
As if to take Gupta up on his challenge, Scott Stuewe, CEO of DirectTrust, and Vincent Albanese, CEO, of blockchain services company Haven Health Solutions – two different stakeholders who might not have been collaborating as recently as a few years ago – took the stage to describe how distributed ledger technology could help address one of healthcare's most perennial challenges: patient identity matching and consent.
"We have a major logistics problem in healthcare," said Albanese.
Indeed, said Steuwe, whose CV includes more than two decades spent with Cerner and close work with interoperability groups such as CommonWell Health Alliance, since the time he first set up a master patient index back in the late '90s, the industry is still in "just about the same place we were with regard to patient matching."
Clearly it's time for some new ideas. And the issue "has to be solved from an industry consortium perspective," said Albanese.
Stuewe laid out the "four big problems that we have yet to solve fundamentally": patient matching, identity, consent and the tension between what he called mass centralization and mass synchronization.
"Patient matching is the problem of trying to get the artifacts in care to match up," ensuring that data produced in two different settings is in sync.
Additionally, "you have this problem of duplication, where at an individual institution, frequently up to 20 percent of the records are duplicates," he said. "But I think of that as an identity problem. The question of identity is still a fundamental issue."
As for consent, that's an even thornier issue. "It's straightforward for treatment between the providers as long as the data is not sensitive in any way," Stuewe explained. "But if it's substance abuse treatment or behavioral health, all that data actually needs to have different consents in different states, which is a complex problem."
And with the fourth challenge – the friction between centralization and synchronization – "if I'm going to centralize everything, if I've got one system and everybody is on one platform, I only have to get that system to work," he said.
"If, on the other hand, I've got lots and lots of systems probing, I'm going to have to get all of those systems to synchronize – and the more of those systems there are, the harder that synchronization problem is. This tension remains in our world, and is actually codified in TEFCA, which basically set up the tension by saying that there will be a relatively small number of highly-centralized organizations that will nonetheless synchronize between each other for the purpose of patient identification."
To get past this, he said, "the fundamental issue is to grapple with a solution that might actually start with the identity problem. So if we actually build a system that was built on the notion that identity was at the source," that could go a long way toward solving the patient matching.
"Let's get it right to start," said Stuewe. "Let's identity proof people when they show up, wherever they show up, but then also issue them a credential that can be utilized in various settings."
That could also help with consent, if "in the process of doing that we actually also involve the patient in the process and could capture their consent in the process of enrolling them a care portal, for example."
Blockchain, he said, "has a potential role to play around undermining this fundamental tension between centralization and synchronization."
Stuewe explained. "So if you start with something like the subscriptions on a patient, you could have the subscriptions contain the who, what, where and why about this particular subscription: What identity-proof individual is it about? What identity-proof individual is interested in receiving information about this? What do I want to know about, and is consent obtained?
"All that could be managed in a distributed ledger," he added. "That's what makes the ledgers most useful, is that they have this ability for different authorities to manage a single asset."
With blockchain, said Alabanese, there's "newness and ge-whiz cool stuff inside of it." But from his point of view, "the truth of the matter is what blockchain has done for us, in the beginning, was it caused people to get together and talk about what that means."
In Haven Health Solutions' case, "out get to market strategy has been the medical doctors and medical societies in various states," he explained. Delaware has been its first use case.
"We're finding that small states are taking a progressive posture," he explained. "They want to be first with blockchain. Delaware was the first state to put their hand up and do something with blockchain and said it's now moving across the country."
There, said Albanese, "we have managed, and I believe this is historic, to get the hospital association, the medical society, the Delaware Health Information Network, the state government, to all sit around one table to talk about one set of problems."
In Delaware's particular case, it's focused on prior authorization. "So with this group sitting together, actually working ands talking as human beings, with the technology – what Scott challenged me was, 'Can you federate the source of truth out there?' – and that's what the blockchain is going to do, it's going to let all the rest of these pieces operate and flow."
At the same time, Albanese recognizes that blockchain and DLT are not perfect technologies, and skepticism about them exists across healthcare. He pointed to four takeaways he's learned from his efforts to apply new technologies to old healthcare challenges.
The first is to choose a suitable use case, one that could use fixing. "Let's acknowledge that there is risk with blockchain, and with the projects you need to start, we have run head on into, 'We're already doing that,' a thousand times." A use case that's already trusted and working well is not the place to be disrupted, he said. "We're looking for new applications out there. That's lesson number one."
The second lesson is that "blockchain will always be slower, less scalable and more complex because you're pulling more nodes." However, "there is an ROI to be determined," he said. More and more industries, healthcare included, are recognizing more and more instances where DLT can drive value
The third is that "this, to us, is really a data play," said Albanese. "And what that means is, it is different than an asset-based transaction on a blockchain. When you're replicating data everybody is getting a pointer to the copy of that. We are not doing any smart contracts, certainly no crypto currency, none of those things. in this case, all we're doing is really making sure that we are replicating data in a reliable safe and usable manner.
And the fourth is that "it isn't really so much about transformation and revolution – we're talking about we're going to be in 2025," he said. "And that, I think is the appropriate time frame. This is slow adoption, by risk-averse institutions, where the privacy of our data as patients comes together."