GHTF is gone, not dormant
The Global Harmonization Task Force was permanently replaced by IMDRF. GHTF documents remain useful as history and some are still referenced, but citing GHTF as a live source describes an organisation that no longer exists.
[ HARMONIZER — NOT A MARKET ]
IMDRF
A forum of device regulators producing the vocabulary and the risk framework that device submissions are actually written in.
What this page does not claim
SPEQ synthesis for orientation. It does not determine what applies to a specific product, and an adopted version can move without this page moving with it. Confirm the instrument in force with the authority before relying on it.
Who publishes here (1)
INTERNATIONAL
What makes a requirement binding here
None. The International Medical Device Regulators Forum was established in October 2011 by device regulators from Australia, Brazil, Canada, China, the European Union, Japan and the United States, meeting with the World Health Organization, and it builds on the work of the Global Harmonization Task Force, which it permanently replaced. It is a voluntary group of regulators. It issues no law, grants no approvals and inspects nothing.
Through the vocabulary and the frameworks being taken up in binding instruments elsewhere. IMDRF/SaMD WG/N10, published in December 2013, settled the key definitions for when software is a medical device; IMDRF/SaMD WG/N12, published in September 2014, set out a possible framework for risk categorisation of software as a medical device. Neither is a rule, but both shaped how regulators and manufacturers describe the same product — which is a real effect, because a submission that adopts the shared vocabulary is intelligible across the jurisdictions that use it.
WHAT TRANSFERS
Language and structure, not permissions. A risk categorisation reasoned in N12’s terms is legible to several authorities at once, which shortens the distance between submissions. What does not transfer is any approval: a device authorised in one IMDRF member jurisdiction has no status in another. The Medical Device Single Audit Program is the closest thing to transferable evidence in this space — a single audit that participating authorities accept — and it is worth being precise about what it is: a programme operated by its participating authorities, not an IMDRF product, voluntary in Australia, Brazil, Japan and the United States, and mandatory in Canada for class II, III and IV devices since 1 January 2019.
Operating here
N10 and N12 are definitional, and that is their value
N10 answers when software is a medical device; N12 offers a risk-categorisation framework. Neither imposes a requirement. Their usefulness is that an argument built in their terms can be read the same way by more than one regulator.
MDSAP belongs to its participating authorities
It is run by the authorities that participate in it, not by IMDRF, and its status differs by country — voluntary in Australia, Brazil, Japan and the United States, and mandatory in Canada for class II, III and IV devices since 1 January 2019. Describing it as an IMDRF certification misstates both its owner and its effect.
Membership does not imply a common rulebook
The member jurisdictions run substantially different device regimes — EU MDR, US FDA premarket pathways, Japan’s PMD Act, and others. Shared vocabulary sits on top of those regimes; it does not merge them, and a device still faces each one separately.
What practitioners get wrong
- IMDRF approves nothing. It is a voluntary regulators’ forum that publishes documents; the authorisation you need is granted by a national or regional authority.
- MDSAP is operated by its participating authorities, not by IMDRF, and it is mandatory only in Canada for class II, III and IV devices — voluntary elsewhere.
- GHTF was permanently replaced by IMDRF in 2011. Treating GHTF as a current body is a dated reference rather than a minor imprecision.
- N10 and N12 give you shared language for a SaMD argument. They do not give you a classification any regulator is obliged to accept.
Questions about IMDRF
Does IMDRF approve or certify medical devices?
No. It is a voluntary forum of medical device regulators, established in October 2011, that publishes harmonisation documents. It issues no approvals, no certificates and no inspections. Market authorisation comes from a national or regional authority — FDA, an EU notified body under the MDR, Japan’s authority, and so on.
What are IMDRF N10 and N12?
IMDRF/SaMD WG/N10, published in December 2013, sets out key definitions for software as a medical device, including when software falls into that category. IMDRF/SaMD WG/N12, published in September 2014, proposes a framework for categorising SaMD by risk. Both are harmonisation documents rather than requirements, and their value is a shared vocabulary across jurisdictions.
Is MDSAP an IMDRF programme?
It grew out of an IMDRF working group but is operated by its participating regulatory authorities. Participation is voluntary in Australia, Brazil, Japan and the United States, and it has been mandatory in Canada for class II, III and IV medical devices since 1 January 2019. An MDSAP audit is accepted by participating authorities; it is not an IMDRF certificate.
What happened to the GHTF?
The Global Harmonization Task Force was permanently replaced by IMDRF, which was established in October 2011 to build on its work. GHTF documents are archived and some remain influential, but the organisation no longer exists and should not be cited as a current source.
ADOPTED, AT A PINNED VERSION
What binds here, and which edition
SPEQ has not decoded a pinned adoption edge here yet. That is a gap in this catalog, not a finding about IMDRF: requirements still arrive through the authorities that adopt this body’s output, and the edition in force is theirs to state.
What influences it
SPEQ records no harmonizer participation for IMDRF. Requirements here are set by the authorities above rather than arriving through a harmonized guide, so evidence prepared for another market does not transfer on that basis.