As of the first quarter of 2023, it will be mandatory for pharmaceutical manufacturers that market to the EU to comply with the European Medicines Agency (EMA) ISO Standards for the Identification of Medicinal Products (IDMP).

The key goal of the implementation of these standards is to positively impact public health and safety by providing more efficient regulatory control, improved data quality and management, increased interoperability between regulatory authorities and cost/time savings.

What is IDMP?

ISO IDMP is a set of standards covering the four domains of master data in pharmaceutical regulation: substance, product, organization, and referential (SPOR) data.

Why are IDMP standards needed?

The EMA has mandated the use of the ISO IDMP standards to support regulatory activities throughout the European Union (EU). The standards will be key for aligning and streamlining the pharmaceutical regulatory process. They will improve regulatory activities such as regulatory submissions, pharmacovigilance, clinical trials, and good manufacturing practice inspections.

One of the first domains of master data standards to be launched was the Referentials Management Service (RMS), which provides controlled vocabularies that cover attributes of products, such as dosage forms, routes of administration, and packaging.

How can SciBite help?

SciBite’s expertise in ontology management and semantic enrichment means that the controlled vocabularies used in the ISO IDMP standards can be converted to ontologies to serve as a centralized reference for IDMP terminology. Additionally, using SciBite’s expert curation, these ontologies can be adapted for use in our named entity recognition (NER) platform, TERMite.

So far, SciBite has developed an ontology of the ~150 RMS lists in our CENtree ontology management platform (Figure 1), allowing users to view these lists in one convenient place rather than having to manually page through the lists on the EMA portal.

Further, SciBite has curated the ontology for use as a NER vocabulary in TERMite, including increasing the number of synonyms, term disambiguation and allowing for UK/US spelling variation. The vocabulary will enable users to mine data repositories for documents containing IDMP terminology, thus allowing companies to quickly and efficiently surface the relevant documentation needed for their regulatory submissions.

Figure 2 shows an example document of drug product information required for regulatory submission. Using TERMite, the relevant RMS terminology can be extracted from the document and normalised to the official ISO IDMP identifier.

Allowing for the different types of documents that may contain relevant information, SciBite will work further with customers on specific use cases to provide bespoke curation of the TERMite IDMP vocabulary to optimize the efficiency and accuracy of the text extraction from your documents.

Figure 1: The Referentials Management Services lists as an ontology. 149 of the RMS lists were added to an IDMP ontology in SciBite’s ontology management platform CENtree.

Figure 2: Use of the IDMP TERMite vocabulary to extract IDMP terminology from a document. (A) An example document containing regulatory submission terminology is shown together with the normalized data extracted using TERMite (B).

Pistoia Alliance IDMP Ontology

The Pistoia Alliance in conjunction with the EDM Council have recently released a minimal viable product (MVP) ontology for the ISO IDMP Standards.

The first version of this ontology, v0.1.0 MVP, is now available from within CENtree as a one-click load (IDMP_EDM) and will be updated with our regular ontology releases.