Intermountain Healthcare has devoted an enormous amount of resources for ground-up development of decision support algorithms and protocols to fire against its patient population. That effort has paid off in decreased mortality and marked improvements in clinical outcomes for segments of that population.

But it’s not enough, not nearly enough, according to Stan Huff. M.D., the health system’s chief medical informatics officer. “We have about 200 clinical decision support programs running every day against patients, but without exaggerating, there are 5,000 algorithms that we could be doing, and should be doing,” Huff says. “The costs of analyzing problems, working with clinicians to develop solutions, and testing the solution via clinical trials and IT betas is absolutely enormous. And we, like everyone else, are targeting a small set of high-profile, high-cost disease states. For the rest of our patients, we’re not doing much.

“If someone gave us a lot of money, I can see how we could get from 200 protocols to 300, but we’ll never get to where we need to be under the current paradigm. Intermountain should be doing what Mayo Clinic or Regenstrief are doing right now, and vice versa. I’m not talking about ideas, I’m talking about executable programs.”

Huff will give a keynote address at Health Data Management's 3rd annual Healthcare Analytics Symposium & Expo, July 14-16 in Chicago. For more information on the event, click here.

What’s particularly frustrating to Huff is that the industry now has a foundation to share such applications and have them embedded widely in clinical practice. Nearly everyone has an EHR, he notes, and when he asks peers about their systems and often gets the same answer: they do pretty much what users expected. They can do order entry, and they can do documentation, but when they try to build out an EHR’s intelligence, they hit a road block.

“The standard approach is that they go to the annual user group conference, and they get in a room with other users and they list 10 things they’d like to have in terms of decision support and functionality,” he says. “Then the vendor takes 10, 20 or 25 ideas and works on them, and in 18 months they are ready to provide those. And it happens year after year and the unfilled gap between what users want and what they get never really closes.”

Huff doesn’t necessarily blame vendors; he notes they are in the same paradigm as users, and many are doing the best they can to respond to user needs. But it just isn’t working, for the industry or society as a whole: “All the vendors are building vertical stacks, and you can argue from a market point of view we don’t need more than three CPOE systems to serve the entire health care industry. But we’ve paying for 50, or 100 systems to be built; our economy truly is paying for that.”

Huff isn’t throwing bombs from the bleachers. He’s leading an effort called the Healthcare Services Platform Consortium that’s putting the pieces in place to build a services architecture foundation that incorporates standard clinical-data models and terminologies to achieve semantic interoperability, at which point, Huff says, “a good idea, be it from a large corporation or a real smart guy working in his garage, can get into the market.” Central to the effort is the creation of a digital marketplace or app store that would deliver applications and software suites to consortium members and the wider industry.

While Huff reiterates vendors are not entirely at fault for the current state of affairs, he does note that “vendor interest in efforts like the consortium is traditionally inversely proportional to their market share. But we are now seeing general interest from even large vendors.”

“There are many companies that have innovative ideas and well-built applications that have no entry point into the healthcare market, and the situation is becoming more acute with the shift toward enterprise vendors increasing vendor lock. There has to be a way in for these companies, many of which are going to retreat from our market if we can’t create that point of entry.”

Huff and Vishal Agrawal, M.D., president at Harris Healthcare Solutions and a key advocate of the Consortium have been traveling across the industry to get the message out, develop the application development platforms and APIs for the consortium, drum up support from the provider and vendor communities and try to find the funds to hire a staff for the effort.

Semantic interoperability has been a Quixotic effort for plenty of HIT leaders and stakeholders. CMS has talked it up; HIMSS has been running its IHE showcase for years at its annual confab. The industry’s annual Connectathon meetings in Chicago, as well as efforts such as the vendor-driven CommonWell Health Alliance, are moving the interoperability chains, but still haven’t addressed the underlying flaws in the paradigm.

Huff is under no illusions that the Healthcare Services Consortium will have an easy go of it. But he has Intermountain putting its considerable muscle behind the effort. One way Intermountain has already committed is working out an agreement with Cerner Corp, its EHR vendor, to support future applications for Intermountain to be built on the services architecture foundation so that those applications are available to Cerner’s user base, most of which have their EHRs remotely hosted by the vendor. While Huff declined to provide the financial commitment by Intermountain and Cerner, he said there are “large, mandatory investments” involved.

In addition, Huff has an established integrator—Harris Corp.-- in his corner as well as a high-profile peer in Agrawal for the consortium effort. And he also knows from experience how providers can drive a market.

“I remember when HL7 standards were being developed and all the competing interests during that process,” Huff  says. “They didn’t get much traction until providers understood the value proposition—how HL7 messages could cut down interfacing, get your lab data out of a LIS, etc. It wasn’t a matter of selling it to vendors, it was a matter of convincing their customers of the value. Once that happened, HL7 support started showing up in RFP checklists, and vendors started getting on board. We can use that same type of momentum for the consortium’s effort.”

Register or login for access to this item and much more

All Health Data Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access