how-innovative-infostructure-can-power-the-purpose-of-integrated-care-systems

How innovative infostructure can power the purpose of integrated care systems

Sharing organized and complete data to generate insights for better health outcomes is the driving force behind joined-up care in integrated care systems (ICSs).


In brief

  • Three essential building blocks form a future health data platform that will lead to better exchange and actionable insights from the underlying data layer.
  • As health care systems integrate, they have the opportunity to reimagine health information infrastructure to allow for innovation and data fluidity.

Integrated Care Systems (ICSs) tick all the right boxes—ideally providing equitable, well-connected and coordinated care across all aspects of a person’s health and social needs resulting in healthier populations and more sustainable systems. Supported by a contemporary and innovative information infostructure, ICSs could differentiate by enabling and sharing insights through data at scale. Read the full report (pdf).

In England, 42 ICSs now cover all areas and are the result of a steady shift in health policy over the past decade toward joining up of services to better meet the needs of local populations. Globally, similar policy interest can be seen in the desire for better integrating care and adapting care delivery through such things as patient-centered care, shared decision-making, medical homes and social prescribing.1,2,3 To unlock the power of ICSs as coordinated and collaborative ecosystems requires a new approach to health information architecture.

 

As data becomes the core asset of health and care, health systems cannot achieve their stated goals of improved wellness, sustainability and equity without an agile information infrastructure. Care systems that integrate health, social, community and local government services mean that thousands of people may need to access data for different purposes to coordinate care to meet a person’s health and social needs. These systems have the opportunity to remake their health information architectures into open platform environments that enable shared care records across the health journey, and allow emerging technologies to be applied to that infostructure to extract insights out of the data layer.

 

Such infrastructures also need to act as platforms of exchange. This means allowing the easy flow of data within, and among, systems and analytics capabilities that support the use of data for local needs, safely and securely.

 

Build for the future

 

Getting the information backbone of the integrated care ecosystem right is mission critical and infrastructure must be designed for the future, with the entire system in mind. This means sufficient flexibility to benefit from emerging technologies (e.g., artificial intelligence, augmented and virtual reality, hyper automation, decentralized identities, Web 4.0, and digital twins), digital-first consumer and workforce experiences, and the capability to integrate within a national and or multinational system(s).

 

The future vision should be of systemically and semantically architected open systems, built from an ecosystem mindset, that accommodate a plurality of approaches and information needs. This should be a dynamic infrastructure, where data is captured in a smart way, adheres to a shared semantic standard tailored to the specific use case and moves data through modern interfaces like RESTful web application programming interfaces (APIs). In this infrastructure, health data captured in electronic patient records (EPRs) will need to be able to flow through this standards-based information backbone, either directly integrating with the backbone API or by publishing their vendor-controlled data objects in a format that can be mapped against it, like HL7 FHIR that is designed for extracting data.

 

All three largest Nordic vendors of Electronic Health Record (EHR) systems in Norway (DIPS), Sweden (Cambio) and Finland (TietoEvry) and the largest vendor of EHRs for elderly care providers in the Netherlands (Nedap) use a vendor neutral data repository based on openEHR.4 Dedalus, a supplier of clinical and diagnostic solutions in many European countries has recently committed to openEHR.5

 

The system should be federated, where the architecture is one of multiple interconnected nodes and shared principles, governance and open standards. Such an architecture allows shared infrastructure services to scale incrementally over time, links different domains together and provides the means to share sensitive health data safely and securely. If built on an open hybrid cloud platform that mixes on-site and third-party cloud computing infrastructure, standard functions such as patient identity management, and verification can be provided nationally while allowing for locally-led workstreams and applications.

 

Build on the installed base

 

Few opportunities exist for greenfield sites, therefore new uses and users must be interwoven with the pre-existing built environment, the installed base. The primary strategic question facing integrated delivery system decision makers now is how to best build scale with sustainable technology choices that suit both national and local purposes. Considerations include the installed base, required enabling infrastructure, as well as future care model and infrastructure needs. Recently reported, is the decision of a Florida-based health care system to spend US$65m to switch from one monolithic EHR system to another to allow access to patient records irrespective of location.6 The cost of switching from one siloed system to another is something that few can afford. It begs the question as to whether health systems should continuously switch or change as needs evolve or if they might choose to work with a vendor-neutral data layer instead?

 

Three building blocks to create the right environment


Creating the right environment for integrated care systems to thrive will be built upon the information architecture, core features, and shared common terminology and standards.

 

1. Infostructure information architecture
 

For a truly integrated system, a data environment with no connection restrictions other than permissions and security is needed. This, in turn, necessitates an open platform architecture that allows for the storage and linking of structured and unstructured data, and that determines how data flows. A decentralized and networked infrastructure will unify disparate information from multiple sources and make sense of it. This means capturing and linking all relevant data regardless of where it is created and stored.

 

The optimal platform separates content and technology and will be vendor-neutral, distributed and modular — incorporating third-party as well as legacy systems. It provides a stabilizing framework for maintaining governance mechanisms which include standards, interfaces and rules. The architecture should be separated into different layers that organize transactions and interactions: the data layer, the application layer and the logic layer:

  • The data layer is standardized in terms of format, nomenclature, terminologies and definitions, which allows it to flow into other systems.
  • The application layer requires a fully systemic design of workflow based upon triggered events of care or intervention, e.g., clinical workflows.
  • The logic layer contains sets of rules that define boundaries and exceptions, and can form workflows.

As Figure 1 shows, the information architecture of the future will shift from many fragmented systems with limited interoperability to a more harmonzied arrangement.

The health information architecture of tomorrow
2. Core features

To architect for the future requires a reference framework around technology, data and end-user primacy. 

Core features

User-centric

  • Built around consumer trust, preferences and control of their own health, lifestyle, behavior and social data for engagement and better health outcomes. From a business and operational perspective, customizable and adaptable systems of clinical and operational data.

Governed

  • Shared principles and rules that guide and safeguard health data use, with sharing practices to protect privacy, enable efficiencies, promote quality and foster research

Interoperable

  • Common rules govern access and content, referenced to internationally accepted open standards. A set of community-sourced common data models is used for storing and sharing data.

Portable

  • Applications or logic developed on top of the data layer should be able to run without change to any independently developed implementation of the data layer.

Federated

  • Where data is created at multiple points; data provenance is fully documented; and data are sufficiently liquid to move within and across systems.

Vendor neutral

  • The data layer is based on vendor neutral standards. Anyone implementing a node to store and share care information may elect to use any technology from a vendor of their choice supporting these standards.

Flexible

  • An architecture that is modular, built on microservices with no need for reconfiguration. Allows for plug-and-play integration of devices and equipment, and for extensibility. Structural separation of data and application layers. Low-code development functions allow flexible creation of applications for specific use cases.

Open APIs

  • Open RESTful APIs integrate data between participants. Accommodates a variety of legacy systems as well as allowing third-party innovations. These APIs allow for easy integration in web applications. The specifications of these APIs should be freely available.

Secure and Safe

  • Where technical, governance and cybersecurity elements meet accepted data security frameworks, assurance schemes and safety and reliability standards for handling personal health and social care information.
3. Data and Terminology Standards

Health information systems should share a common language (standards, semantics and structure) thus avoiding translational interoperability friction and the need for bridging between systems. Separating the data layer from applications is achieved by putting a semantic rich common data model that references a common set of ontologies at the center of design for every use case, and then selecting and constraining the relevant predefined data elements required for a specific use case across the health and care system.

Data standards fall into three categories: interoperability standards, clinical data models and clinical terminology standards.

Interoperability and workflow

Clinical model and pervasive data

Supporting clinical terminology

  • HL7 provides interoperability standards for exchanging data across systems regardless of how the data is stored. HL7 also includes the FHIR standard offering modern RESTful APIs.
  • openEHR provides semantic rich community sourced health data models, published under a creative commons licence. These common data models are supported by a robust reference model and state-of-the-art software specifications, including a RESTful API and a query language. This allows for rapid, data standards driven, application development where data is safely stored for longevity.
  • Clinical terminologies provide clinical codes for tests, procedures, diagnosis, and other technical clinical terms.
  • The Integrating Healthcare Enterprise (IHE) profiles are used to implement clinical processes and workflows for a particular clinical need or pathway. They are used globally to build together meaningful clinical process from other standards such as HL7 messages and imaging standards.
  • The OMOP common data model allows analysis of data from different observational data bases. Administrative and health data are transformed into a common format with common terminologies, vocabularies and coding schemes to support systematic analyses.7

The following allow for the fine-grained recording of the bulk of health data:

  • SNOMED CT: Systematized Nomenclature of Human Medicine Clinical Terms.
  • LOINC: Logical Observation Identifiers Names and Codes
  • ISO/IEEE 11073: Set of standards for medical and personal health devices 

With these building blocks in mind, open platform systems will produce significant value as a result of seamless data flow across the entire health and care value chain—from primary and community care through to hospital systems.


A special thank you to Professor Rachel Dunscombe, Visiting Professor Imperial College London and UK government strategic advisor; Dr. Sheryl Coughlin, EY Global Health Senior Analyst; and Aishwarya Benjwal, EY Health Sciences and Wellness Analyst, for their contribution to this report.


Related content

How can an intelligent health ecosystem create a smarter health experience?

The most important conversation around the future of smart health care is not about the shift to digital. It is about people. Learn more.

14 Jun 2021 Aloha McBride

Five trends shaping health care’s data-driven future

Health care can become more resilient, agile and innovative by shifting to digitally enabled business models with data at the core.

03 Dec 2020 Aloha McBride

How do you create the care models of tomorrow that are barely imaginable today?

Smart health is about putting the patient at the heart of health services design, and reimagining models of care to realize personalized health at scale.

03 Dec 2020 Aloha McBride

    Summary

    Integrated shareable care records replete with good-quality data are the backbone for any future health and care system. In this early stage of transition to integrated care systems (ICSs), opportunity exists to weigh alternatives carefully. If information infrastructure is viewed through a different lens, that of open platforms and systemic design, the building blocks of a truly transformational change become clear.