Interoperability is the ultimate goal of healthcare information systems. Software and cloud-based services need to be able to talk to one another, to exchange clinical and administrative data to enable complete access to a patient’s record and help clinicians deliver the best possible care.
But health IT vendors and healthcare provider organizations still have a long way to go when it comes to attaining interoperability. In the years ahead, though, progress will be made and there will be various next-generation tactics and techniques that help advance this goal.
For one, artificial intelligence will assist interaction with data to push interoperability forward, said Jitin Asnanni, executive director of the CommonWell Health Alliance, a trade association of health IT companies working to create nationwide access to data.
“As exchange enters the mainstream, broad sets of data become more liquid, and an increasing array of end-users start depending on that data, for example, apps specializing in the sub-components of population health, public health, precision medicine, etc.,” Asnanni said. “End users are going to struggle with the signal-to-noise ratio – even when the data itself is intelligently indexed.”
Intelligent assistants will become requisite features in the tools that to engage with the data, helping to sort through the data and unearth the critical pieces of information, analogous to voice assistants like Siri, Alexa or OK Google, he added.
Don Woodlock, vice president of HealthShare at InterSystems, which specializes in health information interchange, said creating a unified health record is the next big focus.
"Healthcare leaders need to make it a priority to implement an interoperability solution that brings different data elements together to create a unified health record that can be shared across the care continuum."
Don Woodlock, HealthShare at InterSystems
“We live in a multi-health world – multiple data sources, multiple providers, multiple patients in a population, and multi-time as you support a patient through their lifetime,” he explained. “The best way to survive and thrive in this very complex world is to empower everyone with a unified health record, be it caregivers, patients, families, physicians, data scientists, case managers and even AI algorithms.”
If a provider organization wants to succeed in value-based care, for example, it needs real-time access to all patient information, Woodlock said. This includes everything from patient data to population and insurance claims data.
“To make unified health records a reality, interoperability technology must be able to combine multiple data types, from structured data such as weight and height to unstructured data in the form of handwritten clinical notes, into a single snapshot of the patient,” he said. “Healthcare leaders need to make it a priority to implement an interoperability solution that brings different data elements together to create a unified health record that can be shared across the care continuum.”
Moving forward, one of the biggest drivers of interoperability technology will be the FHIR standard. Over the past few decades, the number of data sources and the volume of patient data have increased exponentially. Interoperability used to mean just connecting two systems together, but today’s healthcare ecosystem is much more complex.
“In order to consistently deliver high-quality care, provider organizations need real-time access to multiple health information systems to obtain a comprehensive, longitudinal view of their patients’ health history,” Woodlock said. “FHIR is the only standard that enables seamless and real-time data sharing, making it critical for CIOs to ensure that every health information source adheres to the standard.”
Raychelle Fernandez, vice president of Dynamic Health IT, whose products focus on the quality and interoperability of 2015 Edition Health IT Certification Criteria and integrate with EHRs, agreed with Woodlock that the FHIR standard will play a big role in next-generation interoperability.
"Patients also are taking a measured interest in personal health trackers. As a consequence, we’ll see more devices that ‘phone home’ and integrate with portals and personal health record applications."
Raychelle Fernandez, Dynamic Health IT
“In the near term, the involvement of heavy hitters like Amazon, Apple, Google and Microsoft will spur innovation, but so too will smaller startups that are able to bring fresh ideas to areas with relatively low barriers of entry; think FHIR, blockchain or open source data analytics,” Fernandez said. “We foresee growth in patient-centered health applications. This will involve FHIR integrations through API hooks already existing in 2015 Edition-certified Health IT.”
At first, this likely will happen one health system at a time – as with Apple’s foray into bringing PHI to the iPhone – by getting major provider networks on board gradually, she added. API integrations will allow patients to begin consolidating access to their data across disparate providers, she said.
“Patients also are taking a measured interest in personal health trackers, while providers are finding ways to manage population health remotely and health insurers are mandating feedback from medical devices such as CPAP machines,” Fernandez added. “As a consequence, we’ll see more devices that ‘phone home’ and integrate with portals and personal health record applications.”
Another next-generation move for interoperability will involve natural language processing technology, said Tim Kowalski, president and CEO of Halfpenny Technologies, a vendor of healthcare interoperability systems for clinical data exchange.
“The need for both structured and unstructured medical records data is driven by several factors,” he said. “Key among them is using the information in unstructured data to fill gaps in the structured data. Natural language processing is an important next-generation feature. NLP exposes data needed by analytics platforms to identify patients at risk and can efficiently direct care managers to observations and conditions that may otherwise be masked in the narrative of unstructured data.”
The ability to analyze and extract meaning from unstructured data sources such as progress notes and history will contribute greatly to advancements in care coordination.
“NLP systems that can learn from review feedback offer the most promise to impact care coordination,” Kowalski said. “Systems that involve significant manual attention will not be able to keep pace with the demands of risk-sharing models.”
Email the writer: [email protected]
Source: Read Full Article