What has SHARP accomplished for EHRs?

'I would say in general that the SHARP programs were successful, but some were more successful than others'
By Neil Versel
11:13 AM
confused man at laptop

After four years and $60 million, the ONC's Strategic Health IT Advanced Research Projects program is essentially done, and each project has made strides in improving various elements of EHRs. Those involved in SHARP believe their work has advanced the health IT field, but it remains to be seen if any of their innovations find their way into widespread use.

[See also: Object of beauty, or ungainly nuisance?]

"I would say in general that the SHARP programs were successful, but some were more successful than others," said Douglas Fridsma, MD, former ONC chief scientist who departed in September to become president and CEO of the American Medical Informatics Association. Fridsma was effectively the point person for SHARP while at ONC.

With money appropriated by the 2009 HITECH Act, ONC issued grants in 2010 to four major institutions, to "address well-documented problems that impede the adoption of health IT," according to the program's website.

[See also: EMR usability seen lacking]

"We wanted to bring relatively mature technologies to the marketplace," said Fridsma.

The University of Illinois at Urbana-Champaign studied privacy and security, hoping to increase public trust in EHRs. Harvard Medical School built an architecture called SMART, or Substitutable Medical Apps, Reusable Technologies, to facilitate easy extensions of EHRs. Mayo Clinic received a grant to develop best practices for secondary use of EHR data.

The University of Texas Health Science Center in Houston focused on patient-centered cognitive support, consisting of EHR usability, workflow and clinical decision support. "From our perspective, they are the same thing," said Jiajie Zhang, director of the school's National Center for Cognitive Informatics and Decision Making in Healthcare. "They are three pieces of the same topic."

A fifth project, one funded by the NIH's National Institute of Biomedical Imaging and Bioengineering and later moved under the auspices of ONC, became known as the Medical Device Strategic Health IT Advanced Research Project, or MD SHARP.

That five-year, $10 million grant, led by the Medical Device Plug-and-Play (MD PnP) Interoperability Program based at the Center for Integration of Medicine & Innovative Technology and Massachusetts General Hospital, is developing software, standards and tools to promote interoperability between medical devices and IT systems.

MD SHARP continues through the summer, but the other projects are officially complete since the HITECH money has run out.

"It was never intended to be an ongoing funding stream," Fridsma said.

In Fridsma's view, the most successful grantee was Harvard.

"They were somewhat critical of some of the national standards being promulgated," he said. "Their idea was to develop apps."

At AMIA's annual conference in November, several presenters, including some from the Harvard team, discussed apps and open application programmer interfaces, particularly "SMART on FHIR," an implementation of Health Level Seven International's forthcoming Fast Healthcare Interoperability Resources standard, or FHIR.

"Any system that wants to implement these open specs can run these apps in any way they like," said Josh Mandel, MD, lead architect for the SMART Platforms team at Boston Children Hospital's and Harvard Medical School, speaking at AMIA.

The Texas NCCD project culminated with the publication of an e-book titled, "Better EHR: Usability, Workflow and Cognitive Support in Electronic Health Records," which included methods, tools, a knowledge base and guidelines for vendors and EHR implementers to improve usability.

"The book is a summary of only part of what we have achieved," Zhang said.

Zhang's team developed a framework called Task User Representation Function, or TURF.

"This is a unified framework for EHR usability," he explained. They also developed a software implementation of the system to assess usability, as well as "inspirational" functionality prototypes for future systems, such as a software module for medication reconciliation.

There were some changes during the four-year period. NCCD originally had two sets of goals, one for the short term and one with a longer time horizon, but after the contract began, the provider community started raising their voices about EHR usability issues in meeting meaningful use.

"ONC directed us to put more resources into short-term goals," Zhang said.

The group subcontracted to Arizona State University to create a tool set for authoring clinical decision support rules, then made the resource publicly available. "Other people can pick it up and use it," Zhang said, adding that ONC's open-source Health eDecisions initiative is doing so.

Likewise, as its own contract requires, CIMIT is making its work products free to the public, such as a resource package called Medical Device Free Interoperability Requirements for the Enterprise, or MD FIRE. This includes a white paper on device interoperability, a sample request for proposal and contracting language to "promote the adoption of fully interoperable medical devices and systems in support of patient safety," according to the center.

Another element of MD SHARP is the website openICE.info, which stands for Open-Source Integrated Clinical Environment.

"We've changed the paradigm for how data is shared" between devices, said Julian Goldman, MD, medical director of biomedical engineering at Partners HealthCare and director of interoperability for CIMIT. It's now network-based rather than point-to-point sharing. "We call this OpenICE."

"We take a standards-based approach," Goldman said.

They also have tried to make smarter systems with alerts that noticeable without being intrusive and also able to prevent errors on their own.

"Do what clinicians do in the first place," he said. "They look for (patient) reactions." If the heart rate is dropping on multiple devices, for example, that is a sign that something probably is wrong with the patient. Instead of sending an alarm, OpenICE stops the medication first. This leads to a "more intelligent alarm," Goldman said.

[See also: Seven questions with Jiajie Zhang]