12 April 2023
A feature on the requirements of a medical device nomenclature written by Edward Glenn, Senior Nomenclature Developer & Business Development Lead at the GMDN Agency
Why is medical device nomenclature harmonisation important?
The subject of medical device nomenclature is currently a hot topic. As we enter a world of medical device big data, the healthcare community is realising the importance of a well-structured and maintained nomenclature to make sense of all this information. The idea of nomenclature harmonisation is a tremendously exciting prospect, facilitating analysis across borders and between different types of stakeholders.
When used as a common language and a data sorting tool, a harmonised nomenclature can only serve to drive improvements in patient safety and efficiency, and support regulation in the developing world. Not only that, harmonisation is also an asset for the regulatory affairs community, reducing time spent assigning their products to different nomenclatures in different regions.
Can a single medical device nomenclature meet the needs of all stakeholders?
Of the medical device nomenclatures available, each was designed in a different way, with some having specific end users in mind. It is of course, possible to create a nomenclature/classification system that meets the needs of a specific user group while being of low value to other users. However, if we want harmonisation, we must choose a nomenclature that meets or exceeds the needs of all users.
The primary use cases we are aware of (both regulatory and healthcare) are listed below:
- Pre-market approval
- Device registration
- Vigilance and signal detection
- Real world evidence for post-market surveillance (medical records, clinical trials, registries)
- Medical equipment management
- Purchasing and tender
Firstly, we can see that the combined needs of these uses cover all medical devices, including implantables, medical equipment, consumables, and diagnostic devices, so a harmonising nomenclature needs to represent all types of medical devices. This not only includes established device types but accurately representing novel devices is highly important. While in an ideal world it would be more convenient if a nomenclature didn’t change often, as this would provide a stable platform for analysis, but the world of medical device innovation is not a stable place. We have what some refer to as the nomenclature dilemma, a nomenclature is either up-to-date or stable, but cannot be both.
The ideal way to capture novel device variations and innovation is through good communication with manufacturers. For the GMDN Agency, our manufacturer membership and enquiry service are crucial in meeting this demand.
How does a dynamic nomenclature support analysis?
One of the most critical issues to users, and fundamental, is the way devices are grouped. Deciding what attributes to draw a distinction on, and to what level of detail, clearly relies on both a clear rational approach and effective communication with stakeholders. For meaningful analysis devices must be assigned to well-defined lowest-level terms, a fundamental starting point for generic device grouping. Mutual exclusivity of these terms is important to avoid comparable devices falling into different groups. To achieve mutual exclusive terms in the face of such massive device variation, term definitions are essential.
On top of the lowest level terms, broader ways of grouping are paramount for dynamic data analysis, so mechanisms for accurate grouping at many levels and based on different attributes is highly sort after. An under-engineered nomenclature or classification system will truly struggle under the weight of such demands. A well-defined polyhierarchy of categories, that group the precisely defined and mutually exclusive terms, is the only way we can imagine such demands can be met. Furthermore, a well-defined category hierarchy provides more stability for longitudinal analysis.
Why are “others” groups a problem?
Some might contest that the use of “others” groups, to group un-specified device attributes, is a pragmatic way of managing a world of huge device variation. The issue of course is that “others” (or “various”) groups are impossible to define and therefore not much use. Analysis of “others” groups shows that they often become identifiers for novel technology, interesting device variations, and device types overlooked by those developing the nomenclature. Not only that, but these groups also make mutual exclusivity an impossibility.
At the GMDN Agency we have a nomenclature structure and processes put in place to meet the needs outlined above. We have processes in place to ensure good data integrity, achievable only with dynamic updating based on new and novel devices. We continually strive to improve transparency so stakeholders understand our processes and so they can help us improve.
What are the requirements for a truly global medical device nomenclature?
With nomenclature harmonisation a desirable outcome, we must address the elephant in the room: which medical device nomenclature should be chosen for use globally. We have discussed some of the key attributes of a quality nomenclature, but for reference and to complete the picture, we at the GMDN Agency have compiled a medical device nomenclature requirements list. This summarises what we feel are the essentials. We welcome all stakeholders to scrutinise this list and give us feedback. Ask yourself, do you use a nomenclature that meets these requirements?
Medical Device Nomenclature Requirements
The Data Structure and Integrity
Transparency and Harmonization
- The nomenclature should represent all types of medical devices.
- Lowest level terms must be mutually exclusive to allow for precise grouping.
- Terms should be created based on real device examples.
- There must be a mechanism whereby manufacturers can inform the organisation responsible for the nomenclature, of devices not well represented, so these devices can be represented in a timely fashion.
- Terms should be created and published in real-time by a nomenclature team such that innovative technology can be included.
- A history of changes must be captured.
- Higher-level categories are required to group the lowest-level terms to facilitate analysis at various levels of granularity.
- All concepts (terms and categories) must include:
- A code (for identification)
- A name (as a prerequisite for a nomenclature)
- A definition (to facilitate precise grouping and mutual exclusivity)
- Higher level categories should be arranged in a methodically structured poly hierarchy allowing for meaningful analysis in a variety of ways and to facilitate searching.
- Horizontal categorisation should be included within the polyhierarchy such that analysis can be done based on specific attributes, and where the attribute is not necessarily reflected in the structure of the naming (e.g., identification of implants)
- Absence of ‘others’ groups/categories which are ill-defined and become a depository for devices overlooked by nomenclature developers such as novel technology.
- Language used should be non-proprietary, accessible to all users, and translated.
- Training should be provided by the organisation responsible for the nomenclature to all stakeholders to ensure the best possible data integrity and utilisation.
- A Medical Device Nomenclature should be maintained full-time by a professional organisation.
- Organisations should promote international harmonisation to allow data sharing across borders for the benefit of healthcare worldwide.
- Funding of the organisation should be independent of any single jurisdiction to avoid bias as a prerequisite for international harmonisation.
- There should be a mechanism where all stakeholders can provide feedback.
- The nomenclature development process should be transparent.
- Regulators and healthcare providers who need to view, export, and analyse the entire data set should be easy to access and with support from the organisation responsible for the nomenclature.
- Manufacturers who require specific codes should be easy to access.