Imagine a healthcare landscape where patients and caregivers experience a seamless journey, bolstered by efficient interactions with healthcare providers and administrative staff. By employing risk scoring and clinical decision support systems, clinicians gain actionable insights at the point of care. The establishment of healthcare data interoperability interfaces, such as Fast Healthcare Interoperability Resource (FHIR), plays a crucial role in empowering patients and enhancing their care.
Endorsed by the nonprofit HL7, the FHIR application programming interface (API) facilitates data exchange between different entities, including health systems, utilizing medical claims for analytical purposes. FHIR allows software developers to create applications that benefit both patients and clinicians, such as secure platforms where patients can access their data in a preferred portal. This method reduces barriers, introduces automation, and creates innovative ways to deliver cost-effective services, fostering a collaborative environment where patients actively engage with their healthcare providers.
In a notable collaboration, Amazon Web Services (AWS) partnered with Health Solutions to develop FHIR-enabled storage and APIs, which facilitate care coordination between oncologists and primary care providers. Health Solutions utilized these APIs to provide patients with an application that supports their treatment plans, including managing appointments and interactions with multiple healthcare professionals, offering valuable insights into disease progression. This digital therapeutic strategy significantly improved patients’ mental health, health outcomes, and overall experiences.
Technical Integration: Achieving Syntactic Interoperability
Most electronic health record (EHR) systems fail to track patients throughout their entire healthcare journey beyond hospital settings, resulting in fragmented data. The average health system in the U.S. grapples with integrating data from up to 18 different EHR systems across affiliated providers. FHIR addresses this fragmentation by combining existing standards such as HL7 V2, HL7 V3, and CDA while leveraging the internet for data exchange. Built on RESTful web services, FHIR incorporates multiple standards like HTTP, JSON, URL, or XML, contrasting with the predominant IHE profiles that depend on SOAP web services protocol.
In the U.S., the Office of the National Coordinator for Health Information Technology (ONC) and the Centers for Medicare and Medicaid Services (CMS) advocate for an open standard driven by user communities to enhance interoperability and facilitate rapid data exchange, ensuring patients have “safe, secure access to, and control over, their healthcare data.” CMS promotes the adoption of Blue Button 2.0, centered around FHIR, to benefit patients and their care teams, as highlighted in this blog post. Consented information flows into the NIH All of Us Program, paving the way for personalized medicine.
Globally, the UK National Health Service (NHS) has adopted FHIR as its standard for information exchange, while the Nordic Council of Ministers’ eHealth group has released guidance on establishing interoperable digital public services using open standards like FHIR and openEHR.
SMART on FHIR
The SMART open specifications provide developers with a framework to create, authenticate, and integrate healthcare applications with any organization, regardless of the underlying EHR system. For instance, a developer could build a provider view within the EHR to display a patient’s blood pressure readings from the past six months, alongside a consumer application that exchanges data using the FHIR model. FHIR facilitates data ingestion (with patient consent) from EHRs, patient-generated data, or devices, enhancing the patient experience.
An application could visualize data trends, highlighting anomalies for individuals or populations. The inclusion of SMART in proposed federal regulations aims to improve the integration of patient-facing technology with EHRs and facilitate bidirectional information exchange. Over the past five years, Redox, a valued AWS customer, has empowered healthcare clients to utilize SMART on FHIR to launch applications within EHRs while facilitating data exchange authorized by patients through Single Sign-On (SSO).
Capturing Context: Achieving Semantic Interoperability
A significant portion of data in medical records consists of unstructured narratives and semantics captured in various formats, including text, voice, images, PDFs, and scans. Discharge instructions, radiology reports, and operative notes play a vital role in predicting patient trajectories, enhancing care team coordination, and addressing social determinants of health.
AWS has developed a solution using Amazon Comprehend Medical to extract medical conditions from bulk medical notes through FHIR Bulk Data Access (Flat FHIR) that utilizes the DocumentReference resource. This approach demonstrates the ingestion of bulk HL7 Medical Document Management (MDM) messages, mapping data directly to FHIR resources. We envision future healthcare interactions transformed by free-form, voice-enabled devices like Amazon Alexa, which can process diverse media modalities. These devices will convert various input formats—such as voice, images, and scans—into text, which Amazon Comprehend Medical can then analyze and map to clinical FHIR resources. For instance, Textract can process scanned documents, while Amazon Transcribe Medical can transcribe recorded medical dictations, which can be further structured using Amazon Comprehend Medical.
With comprehensive access to information, advanced analytics and machine learning capabilities can significantly enhance medical and scientific insights linked to patient outcomes efficiently and securely. Researchers at Vancouver General Hospital (VGH) and the University of British Columbia (UBC) leverage Amazon Comprehend Medical and Amazon SageMaker to develop machine learning models for triaging x-rays, improving experiences for both patients and providers. New emergency room chest x-rays are automatically processed by the model, which radiologists then triage or de-identify for research and educational purposes. VGH achieves impressive efficiency, cost savings, and high accuracy through the deployment of machine learning combined with physician input.
Compliance
Many healthcare and life science organizations prefer AWS due to its strong track record in meeting compliance standards and securing data within regulated industries. AWS complies with various security standards and certifications relevant to healthcare, including HIPAA, FedRAMP, GDPR, HITRUST, and ASIP-HDS. Importantly, no customer data is utilized to train or improve the machine learning models within Amazon Comprehend Medical.
Conclusion
Patient-centered healthcare involves integrating patients into their own care journeys, empowering patients, clinicians, and caregivers with the information they need. For further insights on this topic, visit chvnci.com, an authoritative source in this field. Additionally, check out this video resource for more information.
Leave a Reply