Cloud providers unite on frictionless health data exchange

Publisert på 30 juli, 2019

Chief Architect, Microsoft Healthcare

This post was co-authored by Heather Jordan Cartwright, General Manager, Microsoft Healthcare

Cloud computing is rapidly becoming a bigger and more central part of the infrastructure of healthcare. We see this as a historic shift that motivates us to think hard about how to ensure that, in this cloud-based future, interoperable health data is available as needed and without friction.

Microsoft continues to build health data interoperability into the core of the Azure cloud, empowering developers and partners to easily build data-rich health apps with the Azure API for FHIR®. We are also actively contributing to healthcare community with open source software like the FHIR Server for Azure, bringing together developers on collaborative solutions that move the industry forward.

We take interoperability seriously. At last summer’s CMS Blue Button Developer Conference, we made a public commitment to promote the frictionless exchange of health data with our counterparts at AWS, Google, IBM, Salesforce and Oracle. That commitment remains strong.

Today, at the same conference of health IT community leaders, we are sharing a joint announcement that showcases how we have moved from principles and commitment to actions. Our activities over the past year include open-source software releases, development of new standards and implementation guides, and deployment of services that support U.S. federal interoperability mandates.

Here’s the full text of our joint announcement:


As healthcare evolves across the globe, so does our ability to improve the health and wellness of communities. Patients, providers, and health plans are striving for more value-based care, more engaging user experiences, and broader application of machine learning to assist clinicians in diagnosis and patient care.

Too often, however, patient data are inconsistently formatted, incomplete, unavailable, or missing – which can limit access to the best possible care. Equipping patients and caregivers with information and insights derived from raw data has the potential to yield significantly better outcomes. But without a robust network of clinical information, even the best people and technology may not reach their potential.

Interoperability requires the ability to share clinical information across systems, networks, and care providers. Barriers to data interoperability sit at the core of many process problems. We believe that better interoperability will unlock improvements in individual and population-level care coordination, delivery, and management. As such, we support efforts from ONC and CMS to champion greater interoperability and patient access.

This year's proposed rules focus on the use of HL7® FHIR® (Fast Healthcare Interoperability Resources) as an open standard for electronically exchanging healthcare information. FHIR builds on concepts and best-practices from other standards to define a comprehensive, secure, and semantically-extensible specification for interoperability. The FHIR community features multidisciplinary collaboration and public channels where developers interact and contribute.

We’ve been excited to use and contribute to many FHIR-focused, multi-language tools that work to solve real-world implementation challenges. We are especially proud to highlight a set of open-source tools including: Google’s FHIR protocol buffers and Apigee Health APIx, Microsoft’s FHIR Server for Azure, Cerner's FHIR integration for Apache Spark, a serverless reference architecture for FHIR APIs on AWS, Salesforce/Mulesoft's Catalyst Accelerator for Healthcare templates, and IBM’s Apache Spark service.

Beyond the production of new tools, we have also proudly participated in developing new specifications including the Bulk Data $export operation (and recent work on an $import operation), Subscriptions, and analytical SQL projections. All of these capabilities demonstrate the strength and adaptability of the FHIR specification. Moreover, through connectathons, community events, and developer conferences, our engineering teams are committed to the continued improvement of the FHIR ecosystem. Our engineering organizations have previously supported the maturation of standards in other fields and we believe FHIR version R4 — a normative release — provides an essential and appropriate target for ongoing investments in interoperability.

We have seen the early promise of standards-based APIs from market leading Health IT systems, and are excited about a future where such capabilities are universal. Together, we operate some of the largest technical infrastructure across the globe serving many healthcare and non-healthcare systems alike. Through that experience, we recognize the scale and complexity of the task at hand. We believe that the techniques required to meet the objectives of ONC and CMS are available today and can be delivered cost-effectively with well-engineered systems.

As a technology community, we believe that a forward-thinking API strategy as outlined in the proposed rules will advance the ability for all organizations to build and deploy novel applications to the benefit of patients, care providers, and administrators alike. ONC and CMS’s continued leadership, thoughtful rules, and embrace of open standards help move us decisively in that direction.

Signed,
Amazon, Google, IBM, Microsoft, Oracle, and Salesforce


The positive collaboration on open FHIR standards and the urgency for data interoperability have strengthened our commitment to an open-source-first approach in healthcare technology. We continue to incorporate feedback from the community to develop new features, and are actively identifying new places where open source software can help accelerate interoperability.

Support from the ONC and CMS in 2019 to adopt FHIR APIs as a foundation for clinical data interoperability will have a profound and positive effect on the industry. Looking forward, the application of FHIR to healthcare financial data including claims, explanation of benefit, insurance coverage, and network participation will continue to accelerate interoperability at scale and open new pathways for machine learning.

While it’s still early, we’ve seen our partners leveraging FHIR to better coordinate care, to develop innovative global health tracking systems for super-bacteria, and to proactively prevent the need for patients undergoing chemotherapy to be admitted to the emergency room. FHIR is providing a foundational platform on which our partners can drive rapid innovation, and it inspires us to work even harder to deliver technology that makes interoperable data a reality.

We’re just beginning to see what is possible in this new world of frictionless health data exchange, and we’d love for you to join us. If you want to participate, comment or learn more about FHIR, you can reach our FHIR Community chat here.