The poor state of interoperability between EHRs–target of fulminations and curses from health care activists over the years–is starting to grind its way forward. Dr. Kenneth Mandl, a leader of the SMART Platform and professor at the Boston Children’s Hospital Informatics Program, found that out when his team, including lead architect Josh Mandel, went to HIMSS this year to support Cerner’s implementation of his standard, and discovered three other vendors running it.
That’s the beauty of open source and standards. Put them out there and anyone can use them without a by-your-leave. Standards can diffuse in ways the original developers never anticipated.
A bit of background. The SMART platform, which I covered a few years ago, was developed by Mandl’s team at Harvard Medical School and Children’s Hospital to solve the festering problem of inaccessibility in EHRs and other health care software. SMART fulfilled the long-time vision of open source advocates to provide a common platform for every vendor that chose to support it, and that would allow third-party developers to create useful applications.
Without a standard, third-party developers were in limbo. They had to write special code to support each EHR they want to run on. Worse still, they may have to ask the EHR vendor for permission to connect. This has been stunting the market for apps expanding the use of patient data by clinicians as well as the patients themselves.
SMART’s prospects have been energized by the creation of a modern interoperability resource called FHIR. It breaks with the traditional health care standards by being lean, extendible in controllable ways, and in tune with modern development standards such as REST and JSON.
It helps that SMART was supported by funds from the ONC, and that FHIR was adopted by the leading health care standards group, HL7. HL7’s backing of FHIR in particular lent these standards authority among the vendor and health care provider community. Now the chocolate and peanut butter favored by health IT advocates have come together in the SMART on FHIR project, which I wrote about earlier this year.
Mandl explains that SMART allows innovators to get access to the point of care. As more organizations and products adopt the SMART on FHIR, API, a SMART app written once will run anywhere.
Vendors have been coming to FHIR meetings and expressing approval in the abstract for these standards. But it was still a pleasant surprise for Mandl to hear of SMART implementations demo’d at HIMSS by Intermountain, Hewlett-Packard, and Harris as well as Cerner.
The SMART project has just released guidlines for health care providers who want to issue RFPs soliciting vendors for SMART implementations. This will help ensure that providers get what they ask and pay for: an API that reliably runs any app written for SMART.
It’s wise to be cautious and very specific when soliciting products based on standards. The notion of “openness” is often misunderstood and taken to places it wasn’t meant to go. In health care, one major vendor can trumpet its “openness” while picking and choosing which vendors to allow use of its API, and charging money for every document transferred.
The slipperiness of the “open” concept is not limited to health IT. For years, Microsoft promulgated an “open source” initiative while keeping to the old proprietary practices of exerting patent rights and restricting who had access to code. Currently they have made great progress and are a major contributor to Linux and other projects, including tools used with their HealthVault PHR.
Google, too, although a major supporter of open source projects, plays games with its Android platform. The code is nominally under an open license–and is being exploited by numerous embedded systems developers that way–but is developed in anything but an open manner at Google, and is hedged by so many requirements that it’s hard to release a product with the Android moniker unless one partners closely with Google.
After talking to Mandl, I had a phone interview with Stan Huff, Chief Informatics Officer for Intermountain. Huff is an expert in interoperability and active in HL7. About a year ago he led an effort at Intermountain to improve interoperability. The motivation was not some ethereal vision of openness but the realization that Intermountain couldn’t do everything it needed to be competitive on its own–it would have to seek out the contributions of outsiders.
When Intermountain partnered with Cerner, senior management had by that time received a good education in the value of a standard API. Cerner was also committed to it, luckily, and the two companies collaborated on FHIR and SMART. Cerner’s task was to wrap their services in a FHIR-compliant API and to make sure to use standard technology, such as in codes for lab data.
Intermountain also participated in launching a not-for-profit corporation, the Healthcare Services Platform Consortium, that promotes SMART-on-FHIR and other standards. A lot of vendors have joined up, and Huff encourages other vendors to give up their fears that standardization is a catheter siphoning away business and to try the consortium out.
Intermountain currently is offering several applications that run in web browsers (and therefore should be widely usable on different platforms). Although currently in the prototype stage, the applications should be available later this year. Besides an application developed by Intermountain to monitor hemolytic disease among neonates and suggest paths for doctors to take, they support several demonstration apps produced by the SMART project, including a growth chart app, a blood pressure management app, and a cardiovascular app.
Huff reports that apps are easy to build on SMART. In at least one case, it took just two weeks for the coding.
Attendees at HIMSS were very excited about Intermountain’s support for SMART. The health care providers want more flexible and innovative software with good user interfaces, and see SMART providing that. Many vendors look to replicate what Intermountain has done (although some hold back). Understanding that progress is possible can empower doctors and advocates to call for more.