Got Class(ification)?

A common complaint about standards in healthcare is that there are simply too many, which includes classification systems for medical devices. From the Global Medical Device Nomenclature (GMDN) to the United Nations Standard Product and Services Code (UNSPSC), there is certainly no shortage of systems, let alone acronyms.

While there have been (and still are) efforts to develop and promote use of a single international system, it’s not an easy proposition, not only because of the number of systems in place today, but also because those systems are used for different purposes, including procurement, spend analysis, inventory control, and regulatory oversight. The question is what kind of taxonomy system (or systems) are needed in a value-based healthcare system where products need to be managed and evaluated to understand the risks and benefits to multiple stakeholders, but most importantly the patient.

Early efforts

In the mid-1990s, the European Commission tried to create an internationally recognized nomenclature to facilitate clearer product-related communications between regulators, manufacturers, hospitals and other parties. That led to the creation of the GMDN, which was based on a number of existing nomenclatures that were primarily in use in different regions of the world. Since then it has received considerable uptake, most notably as the classification system recommended by the International Medical Device Regulatory Forum and chosen by the Food and Drug Administration (FDA) for compliance with unique device identification (UDI) regulations. Since then, some have questioned its future suitability given that access to GMDN codes requires a fee-based license issued by the GMDN Agency. As of press time, the European Commission (EC) had not yet announced if it would also name the GMDN as the required nomenclature for UDI compliance.

The cost issue is one reason why the Office of the National Coordinator for Health IT (ONC) in the U.S. decided to allow hospitals to use either the GMDN or the associated SNOMED Clinical Terminology (SNOMED CT), which is free to participating countries, when creating the required list of a patient’s implantable devices in electronic health records (EHRs). In 2012, GMDN and the organization responsible for SNOMED (The International Health Terminology Standards Development Organization or IHTSDO) entered into a cooperation agreement under which GMDN codes would be used to identify medical devices within the SNOMED system, which was developed to improve how clinically relevant information, including diseases, procedures, pharmaceuticals and devices (referred to as physical objects), is entered in EHRs. With this relationship, specific classes of medical devices can be associated with the procedures in which they are used. One of the challenges with GMDN codes is that they are tied to the approved use of the product, and in the real world, devices are sometimes used for multiple and even off-label purposes.

Classification for value

Unlike pharmaceuticals, there has been little application of classification schemas to study the value — both in terms of cost and performance — for medical devices. That’s because there are some unique challenges around devices compared to drugs. For example, there is relatively little clinical evidence for many devices that are iterations of previously approved products; many devices require additional investments, such as for training; and the performance of a device can be dependent on other factors, such as the skill of the clinician.

In 2015, researchers in Berlin tried to address these issues by creating a matrix1 that classifies products based on the level of risk from a variety of perspectives, including the risk classes assigned by regulators, whether the products are for diagnostic and/or therapeutic use, the primary user (patient or clinician), and if the product is implantable. But beyond implants, the researchers did not distinguish the clinical use, e.g., for cardiovascular care or orthopedics, and some question the clinical relevancy without such distinctions.

Those who study taxonomies suggest their primary value is to help individuals find information they need more quickly and efficiently — especially in this world of seemingly endless data. Betsy Walli, PhD, writes in her article on Taxonomy 101 that multiple taxonomies can be used simultaneously to provide multiple views into the same data. For example, one taxonomy could be used to understand spend on a particular class of product, while others could be used to understand usage, risk and reimbursement. That begs the question, what factors must be in place to ensure different taxonomies can work together? Does there need to be some kind of underlying logic or validation to ensure we are comparing like to like? And what will it take to foster international efforts to develop and adopt such a system or group of interoperable systems?

Last month, the World Health Organization (WHO) held its fourth global forum on medical devices. Recognizing the increasing role of medical devices in healthcare delivery, WHO explored a range of topics around how best to regulate, evaluate, procure and use devices on a global level. The use of classification system(s) was on the agenda, and we will continue to explore this highly complex topic in future issues of Standard Practices.

References

1 Henschke C, Panteli D, Perleth M, Busse R. Taxonomy of Medical Devices in the Logic of Health Technology Assessment. International Journal of Technology Assessment in Health Care. 2015; 31:5: 324-330.

2 Walli B. Taxonomy 101: The Basics and Getting Started with Taxonomies. KM World Website. http://www.kmworld.com/Articles/Editorial/What-Is/Taxonomy-101-The-Basics-and-Getting-Started-with-Taxonomies-98787.aspx. Published August 15, 2014. Accessed January 1, 2019.