5. Infostructure and Infrastructure

Site: EUPATI Open Classroom
Course: Digital Health Applications – Infostructure, Infrastructure
Book: 5. Infostructure and Infrastructure
Printed by: Guest user
Date: Sunday, 14 April 2024, 7:15 PM

1. Introduction to Infostructure and Infrastructure

(This section is organised in the form of a book, please follow the blue arrows to navigate through the book or by following the navigation panel on the right side of the page.)

As  utilisation of digital technologies in healthcare systems is expanding, more and more data are being collected. Organisation and integration of the collected data from different sources, called data curation, becomes more obvious in big data. Adequately curated health data can be integrated with other data, such as environmental and geospatial data, which may be beneficial in early trend detection (e.g. forecasting of outbreaks) [1]. The most significant advantage of health data comes from combining data from different sources and creating a big data pool. Researchers can then generate results from diverse data sources by employing big data analytics.

However, proper use of health data is contingent on well-designed info- and infrastructure. This approach also dictates the degree to which health information can be merged with information outside healthcare, e.g., environmental information.

2. Infostructure

Definition 

Infostructure is the basic organisational structure needed for the operation of an organisation1. For example, “eHealth Infostructure is the development and adoption of modern systems of information and communications technologies (ICTs) to define, collect, communicate, manage, disseminate, and use data to enable better access, quality and productivity in health and health care” [2].

Interoperability 

Interoperability is one of the requirements to have a functional infostructure. HIMSS (Healthcare Information and Management Systems Society) defines Interoperability as “The ability of different information systems, devices and applications (systems) to access, exchange, integrate and cooperatively use data in a coordinated manner, within and across organisational, regional and national boundaries, to provide timely and seamless portability of information and optimise the health of individuals and populations globally” [3]Many potential benefits of digital health solutions such as health data integration from different sources (e.g. EHR), real-time feedback and monitoring cannot be achieved efficiently without interoperability.

Interoperability has three dimensions:

  1. Basic Interoperability: when two systems can exchange data. However, the proper interpretation of the data is not guaranteed. Therefore, an individual should read, interpret and act upon the received data.
  2. Structural interoperability: the systems receiving data can interpret it at the data field levels and the clinical and operational value of the shared data will be preserved, resulting in less human interpretation and intervention.
  3. Semantic or advanced interoperability: It is the meaningful exchange of data between two systems where the recipients of data will receive relevant information. For instance, when a patient undergoes surgery, the surgeon will receive information about the patient’s past surgeries from the records of the patient’s family physicians [4].

Building an interoperable system is contingent on a standardised data format, where the information systems, regardless of the language and technologies, can understand the structure and meaning of the data.

Health Data Standards

Standardisation of health data is crucial for designing and building a robust infostructure. Health data standards can be categorised into four groups [5] [6].

  1. Information model standards: Information model standards specify the characteristics and structure of different elements of infostructures. These standards make the data exchange between different systems more reliable and predictable. Information models can be closed (i.e., proprietary and defined by IT vendors) or open source. Open-source information models are defined by communities such as CEN (the European Committee for Standardization) and ISO, which may become information standards. There are no widely adopted information standards for healthcare information; however, ISO 13606, HL7 (Health Level Seven International), and CDISC (Clinical Data Interchange Standards Consortium) are more widely adopted. These standards allow information to be re-used and accessible across various services (e.g., hospital to community service or patient to general practitioner).
  2. Terminology Standards: Terminology standards standardise the medical terminologies by creating standard codes for each term. Sender and receivers of health data can communicate unequivocally by using these codes. For instance, clinicians may use different terms for one disease (e.g., Brain tumour and Brian cancer) or different terms for diseases in different languages. The terminology standards make the exchange of data possible in these circumstances. SNOMED (Systematised Nomenclature of Medicine) and ICD (International Classification of Diseases) codes are two examples of terminology standards.
  3. Content Standards: Content standards determine the structure and content of an exchanged piece of data between two systems. For instance, a standardised prescription template is needed in a pharmacy for dispensing an electronic prescription. The template enables the pharmacy system to accurately recognise and process the necessary information, such as a patient’s name, medication name, billing code, etc. It allows a pharmacist to dispense the right medications to a patient. HL7 (Health Level Seven International)2 is one of the organisations that provide standardised templates.
  4. Communication standards: specify the formats that should be used to transmit health data between two computer systems. FHIR (Fast Healthcare Interoperability Resources) is an example of a communication standard published by HL7 to facilitate the exchange of health care information between organisations.

The main challenge in designing a robust infostructure is the compatibility of patient-generated data with existing health data. The popularity of health monitoring technologies and mobile applications has resulted in the creation of health data by patients. Many digital health applications and tools are not designed based on data standards and the generated data cannot be integrated with other health data appropriately and securely.

Look at this simple example:

Researchers sometimes have to fix problems of how data is represented. In some places, the data might be expressed in one way, and in other areas, the same data can be described in a completely different way. For example, you can classify a disease like diabetes with a certain number (3) in one database and (5) in another. This is one reason for the considerable effort in industries to create standards to share data more easily. This example illustrates the usefulness of the Fast Healthcare Interoperability Resources (FHIR) mentioned above.

Data is the New Oil, HACKERNOON (2019) https://hackernoon.com/data-is-the-new-oil-1227197762b2

A well-designed infostructure system should be able to incorporate existing unstructured data and transform it into a digital format. As discussed above, computer science architects, software engineers, health care professionals, data scientists, and patient advisory groups are engaged in translating existing paper-based systems to support digital-focused systems. This process often involves a translation or mapping process to present the material in a computer-readable format, which requires a different structure and layout. Such a revised structure presents the data in a manner supporting any future data exchange specification requirements for sustainable digital solutions, broader access, and adoption. This process is complex and multifaceted. It involves several different levels of translation layers to ensure that all the devices and applications of the system can connect and coordinate the retrieval of the information correctly [7].


Note 1: An overarching generic definition by Wegner: Wegner[146] stated that "interoperability is the ability of two or more software components to cooperate despite differences in language, interface, and execution platform".  Peter Wegner. 1996. Interoperability. ACM Comput. Surv. 28, 1 (March 1996), 285 - 287. https://doi.org/10.1145/234313.234424

Note 2:  https://www.hl7.org/about/index.cfm?ref=nav Founded in 1987, Health Level Seven International (HL7) is a not-for-profit, ANSI-accredited standards developing organization dedicated to providing a comprehensive framework and related standards for the exchange, integration, sharing, and retrieval of electronic health information that supports clinical practice and the management, delivery and evaluation of health services. HL7 is supported by more than 1,600 members from over 50 countries, including 500+ corporate members representing healthcare providers, government stakeholders, payers, pharmaceutical companies, vendors/suppliers, and consulting firms.

3. Infrastructure

In addition to the infostructure, large-scale information systems like EHRs need a technical infrastructure to allow for secure collection, storage, curation, sharing and analysis of information. IBM defined infrastructure as the “combined components needed for the operation and management of enterprise IT services and environments” [8]. For example, IT infrastructure consists of the following components:

  • Hardware such as computers, servers, switches, and data centres.
  • Software such as Content Management System (CMS), operating system, Customer Relationship Management (CRM), etc.
  • Facilities that provide space for hardware.
  • Networks that enable the transfer of data between computers.
  • Servers

Two main types of IT infrastructure are traditional and cloud infrastructure. The traditional IT infrastructure is usually built on a company's premises with dedicated physical spaces for infrastructure hardware. Cloud infrastructure has the functionality of traditional infrastructure with the advantage that users can access the infrastructure via the internet [8].

While end-users do not need to know the details of all components, understanding the different components provides opportunities for stakeholders to engage with a system without focusing on its entirety. It is also helpful to understand which components are essential to keep and which are replaceable. This knowledge is beneficial to avoid what is called vendor lock-in (or more generally the adoption of a proprietory system). Vendor lock-in occurs when IT vendors keep the structure of the information used within their system private. It means other systems may be unable to access this information conveniently without the involvement of the vendor.

4. References

[1] Pastorino R, De Vito C, Migliara G, et al. Benefits and challenges of Big Data in healthcare: an overview of the European initiatives. 2019;29. doi:10.1093/EURPUB/CKZ168
[2] Health Canada. Guidelines for the FNIHB EHealth Infostructure Program ( EHIP ).2012.
Digital Health Europe. The major categories of digital health standards. Accessed June 22, 2022. https://www.i-hd.eu/health-standards/the-specifics-of-health-data-standards/categories-of-health-data-standards/ 
[6] International Classification of Diseases. 2019. Accessed November 21, 2022. https://www.who.int/standards/classifications/classification-of-diseases
[7] Fennelly O. Factors for Success in Electronic Health Record Implementation: Literature Review and Key Considerations.2019.
[8] IBM. What is IT Infrastructure? Accessed May 28, 2022. https://www.ibm.com/topics/infrastructure      
[3] HIMSS. Interoperability in Healthcare. Accessed June 26, 2022. 
https://www.himss.org/resources/interoperability-healthcare
[4] Digital Health Europe. Interoperability dimensions - Digital Health Standards. Accessed June 26, 2022. https://www.i-hd.eu/health-standards/the-specifics-of-health-data-standards/interoperability-dimensions/