EBME Forums
Posted By: RoJo GMDN - 15/02/12 9:55 AM
Do any of you who use the GMDN classification system?
If so do you have a decision tree to help with the classification of devices? Or a searchable database? Or any help at all.
Or a copy of ISO 15225:2010
Posted By: Spicer Re: GMDN - 15/02/12 2:56 PM
Hi Robert,

The EU sets the classification of medical devices is based on MDD93/42/EEC (not GMDN) - just see and follow the rules in Article 9 in the link:


GMDN gives a unique number to a device based on the description and its associated definition e.g. "Medical Air Terminal Unit" is defined as "A device that is a component of a medical gas pipeline system........................................", i.e. not a classification. You have to pay for the service to get an actual number and the rest(i.e. become a member). However if you are canny and use the Australian TGA site:


Details of many devices are there including GMDN and supplier classifications (be careful old registrations might give out of date GMDNs - since they get regularly updated or superseded). As to ISO 15225 - it's a copyright publication £136 from BSI (half price if a member).

I hope this helps.

Posted By: RoJo Re: GMDN - 15/02/12 3:54 PM
I think I did not expess myself too well.
I am trying to search the GMDN list for equipment and cannot find certain simple things such as air mattresses - it does have have overlays. (I have paid for access to the official site)
I was wondering if anyone had a simple way of working out what heading things come under. If you start at the top and work down the tree (albeit an up side down one) I end up with not the item I wanted. there seems to be no logic at all in the groupings of equipment and there seems to be multiple paths to the same item. As you can tell I am new to this system and am struggling to understand it.
Any experienced people out there or is it really a mess as I am beginning to suspect?
I thought the collective terms were hierachical but it seems they are not. Am I right?
Posted By: Geoff Hannis Re: GMDN - 15/02/12 6:54 PM

Good luck with that Robert. I fear that your suspicions may be well founded. frown

I took a brief look at the thing a few years ago and, shall we say, failed to be impressed. But to be honest I didn't spend too much time on it once I noticed how much they were charging.

Frankly, my own "home grown" codes have always been good enough for anything that I have ever done, although I appreciate that you must be acting for some third party or other (that is, not free to pick and choose). frown
Posted By: John Stewart Re: GMDN - 16/02/12 10:09 AM
Hi Robert
I have been using the GMDN for some time now and have just recently completed changing all our old medical device equipment codes to GMDN codes in our AIMS database. As the GMDN is being used more and more we felt this was the correct way to go.

There are still quite a few medical devices without GMDN codes but you can request that the manufacturers apply to the GMDN for codes to be created for their devices. I currently use 5 old codes to group devices not on GMDN into a general purpose category until new GMDN codes are set up.

I agree that finding the correct codes can be quite tricky at first but after using the system for some time you soon get a feel for what your search criteria should be. In the case of you’re ‘air mattresses’ you should not use plurals in the search just ‘air mattress’. This will bring up a selection of devices from which you can pick the most appropriate nomenclature and code.

At times the nomenclature will not even contain any of the search words, for your example above it gives… Flotation therapy bed, adult ... as one of the options with the device description as shown below…

A mains electricity (AC-powered) bed designed to minimize pressure points on a patient's body by providing contact with as much of the body surface as possible, typically through a mattress that contains a large volume of constantly moving media, e.g., water, air, or mud that lifts the patient to simulate a floating effect. It is used in cases of decubitus ulcers or where a patient has little remaining body fat and the displacement of body weight is vital for treatment and/or comfort. It can also be used for the treatment badly burned patients and/or to aid circulation. The device may allow for the regulation of mattress temperature to enhance treatment. See also: Air-fluidized bed; Low-air-loss bed

As you can see at the end of the description it also gives alternatives to consider.

Despite what Geoff says this seems to be the way ahead as more and more countries and manufacturers are adopting the GMDN to name their medical devices.

I have found the people at the GMDN to very helpful in trying to trace codes when I have been stuck in finding the correct nomenclature in the past.

If you have any queries on finding other GMDN codes I would be only too happy to try and help you out if you email me. smile

Posted By: Geoff Hannis Re: GMDN - 16/02/12 11:40 AM

There's nothing wrong with classifying and categorizing the kit, John ... and I say that having spent many hours (hundreds, thousands?) over the years doing just that ... just as long as the folk doing the classifying make a good job of it. smile

And I write as one who has suffered the consequences (usually in the form of extra - unnecessary - work) of it being done badly as well. Back in the army, for example (they were forever plonking lab kit under "general medical" ... and once the numbers had been assigned it was nigh-on impossible to get them corrected). And also on various overseas projects (they almost always got it wrong ... and we were forever having to "cross refer" to more sensible lists).

For those who have never tried it, I would advise that (just like writing specs) it's not as easy as it may first appear.

Don't forget that others have had a fair crack at it over the years. ECRI for example.

So, questions that come to mind are:-

1) Why not use the ECRI codes (they are in use worldwide)?
2) Who are these GMDN people anyway?
3) Are they a government agency (or a money spinning company)?
4) Do you want to rely on them (to be beholden to them)?
5) Can we be sure that they'll always be around?
6) How much does it all cost?
7) Why can't the National Health Service do this sort of thing for themselves (and have their own codes)?

Others tuning in may also be wondering:-

1) What are these guys talking about?
2) Why does the kit need to have codified anyway?
3) What purpose does it serve?
4) Does it relate to tech support and maintenance (or just "property control")?
5) Is it that important that (lots of) time needs to be found to carry it through?

... and stuff like that! think

Lastly (and it sounds like you have already done this, John) I would suggest that folk always retain a mechanism to be able to assign their own codes in cases where the "official" ones don't really fit ... or are just plain wrong!
Posted By: John Stewart Re: GMDN - 16/02/12 12:14 PM

1 ECRI are moving towards adopting GMDN codes
2 See GMDN website
3 No/This non-profit company, “GMDN Agency”, acting as the Maintenance Agency Secretariat (MAS), functions as the hub in the running and maintenance of the GMDN, providing services and information for access to the GMDN data through this present Internet site or other means. To ensure continuing permanency of the GMDN, revenues may be generated through the licensing or sale of GMDN Agency products and services, or by direct funding allocated by relevant global regulatory bodies or other parties.
4 As much as any other agency
5 Nothing is permanent
6 See GMDN website
7 The whole point is to create a Global Medical Device Nomenclature.

Posted By: biomedbill Re: GMDN - 16/02/12 5:01 PM
6) approx. 800€ p.a. for a limited search (100 devices).

What information can we get from GMDN? I had a look at the website and came away with a sore head from trying to work out all the acronyms.
Posted By: Geoff Hannis Re: GMDN - 16/02/12 8:33 PM

I had better not say anything else, Bill, lest I get into "trouble" again with the Moderators. frown
Posted By: Neil Porter Re: GMDN - 17/02/12 8:04 AM
Just ask a nurse is it a Patient bed, Electric, a patient bed, mechanical! they don't care, when they report a bed faulty it is just a bed!! Our asset management system (home grown) will tell the tech if it is mechanical or not. No need for all the variations of beds, ultrasound machines etc, if your database can tell you the difference. One last thing will there be a different code for different software versions?
Posted By: RoJo Re: GMDN - 21/02/12 8:58 AM
When it comes to fault reporting you are right the "bed" is broken plus the assset number is all that is needed. But when it comes to planning equipment replacement or contracts you need to know exactly what you have.

I have been going through GMDN and they seem to have got rid of the code for a simple ripple mattress e.g. Huntleigh Nimbus etc. There are overlays and specialist mattresses like low air loss and fluidise. I went to report this as needing a new number and realised they would charge me 200euros for the privilege.
Am I right?
Posted By: Neil Porter Re: GMDN - 21/02/12 12:43 PM
Planning equipment replacements here is done through proper specifications not the name.
Posted By: Geoff Hannis Re: GMDN - 21/02/12 1:23 PM

@Robert: please refer to the other current thread regarding "scams"! whistle

(hint:- just bung in your own codes)
Posted By: John Stewart Re: GMDN - 21/02/12 1:35 PM

I think this one should do the job...

1. Alternating-pressure bed mattress overlay system
Categories: 04 Electro mechanical medical devices , 11 Assistive products for persons with disability
Alternating-pressure bed mattress overlay system P An assembly of devices consisting of an alternating-pressure bed mattress overlay and a dedicated pump, which also functions as the control unit, designed to actively alternate a bed occupant's bed-contact points, typically to relieve pressure points for comfort and to prevent pressure sores. It is commonly used for elderly immobilized (especially in cases of decubitus ulcers), patients with disabilities, or patients with low body fat. See also: Alternating-pressure pump; Alternating-pressure chair cushion system; Alternating-pressure bed mattress overlay, single-use; Alternating-pressure bed mattress overlay, reusable


p.s. any new codes should be paid for by the manufacturers.
Posted By: Neil Porter Re: GMDN - 22/02/12 12:18 PM
I will remember to put that on my purchase requisition form next time I need a mattress
Posted By: John Stewart Re: GMDN - 23/02/12 10:30 AM
If you want only a mattress you should use this one Neil or you will get the wrong device!!!

14. Basic bed mattress
Categories: 05 Hospital hardware , 11 Assistive products for persons with disability
Basic bed mattress P A large, durable, fabric case filled with a resilient material designed to be placed on a bed mattress support platform to provide a comfortable surface on which an occupant can lie and rest/sleep. The device provides no special therapeutic effect.

Posted By: Neil Porter Re: GMDN - 23/02/12 1:01 PM
John, I will just order using the part number thank you.
Posted By: Geoff Hannis Re: GMDN - 24/02/12 3:10 PM

Just before the weekend sees this tread sinking below the horizon ... a question if I may:-

What is the maximum length of a GMDN code? And is any special format employed? think

I shall probably be doing a bit of work on a Hospital Equipping Project database over the coming few days and may need to consider adding a field to allow the user to link to GMDN codes if need be.

GMDN codes that I have seen have not appeared particularly elegant, but perhaps I was looking at obsolete stuff. Two formats seem to be possible:- AA9999 and just 9999.

On a "historical note" in the past I have linked to:-

1) ADB (Activity Data Base) : AAA999
2) CSI (can't remember - Chartered Surveyors Institute, perhaps?)
3) ECRI (ECRI) : 99-9999
4) JSN (Joint Services Number - USA) : A9999
5) MOH (Saudi MoH) : A999 - although other formats were also used
6) UMDNS (UMDNS - based on ECRI) : 9999

... to name but six (and I am aware of others as well, but thankfully never had to use them).

From my notes seven characters have always been enough. What I may do is just bung in a blank (free-form) field, ten characters wide, for users to er, use as they see fit. After all, we never know what "new" codes will be introduced the week after next, do we? smile
Posted By: RoJo Re: GMDN - 24/02/12 4:01 PM
Dear All
What a thread has started.

I did find the overlay code but it is an overlay and not a mattress it specifically says somewhere about putting on a conventional mattress. And the conventional mattress is just that. As for equipment replacement it is for planning purposes. If yuo want to know how many alternating pressure mattresses you have for example you just search on the GMDN code it your database tells you irrespecive of manufacturer model etc.

You say the manufacturer should pay, but remember the number is for an equipment type, so who is going to jump first then let all their competitors use it? If you are a subsciber once the number has been created you can use it.

Geoff as for the format used, it is infact very simple. The number is the next one in order when they decide on the device type. In other words absolutely no logic at all!!!!!!!! I was hoping for a structured system e.g. Patient diagnosis > Monitor > Multiparemeter monitor etc. etc. But no, there is no obvious logic at all. And the collective term codes (for those in the know) are not much help either as one device can come under several collective terms. So it is not very hierachical.

Why are we using it? Because it is now the "official" universal worldwide code and our group collective of trusts has decided to go for it.
The ECRI UNMDS numbers are a subset of GMDN.
Posted By: Geoff Hannis Re: GMDN - 24/02/12 4:12 PM

Thanks for that, Robert. smile

Much as I had suspected then. To be honest ... apart from what we used to call "keeping up with Joneses"*, I can't see any point in adopting anything new if it isn't an improvement over what we had before! frown

I thought (or rather, had hoped) that the whole idea was to take what was good from what has been done in the past by others, shake it all up, and devise a logical system based upon hierarchies that could be developed (grown), rather than changing the codes every other week. Or am I missing something of value here? think

I believe that a proper classification system should be so well conceived and robust that it could serve equipment of all types (including, for example, building fixtures and fittings, motor vehicles ... and everything else in between). For a classic example, see the FSN (the U.S. Federal cataloguing system).

If a system is limited to "medical" equipment only - where do you draw the line, even within a hospital?

* And that's not something that I find particularly appealing, I must admit.
Posted By: RoJo Re: GMDN - 27/02/12 8:45 AM
The problem is it might not be "nice", easy or logical but it is "the one" so like it or not we have to go with the flow.
Posted By: Geoff Hannis Re: GMDN - 27/02/12 2:00 PM

Yeah ... that's the way progress is made. frown

Good luck with all that Robert, I'll leave you to it Mate. smile
Posted By: Geoff Hannis Re: GMDN - 25/07/16 12:10 PM

Does anyone have any information about the Italian *CIVAB system of equipment codes? think

- Basic structure?
- Outline of the idea?
- Anything special about them?
- Are they still being used?
- Does the Si3C system use these codes?

And (whilst I'm on here), I heard a few (five or more) years ago that "the WHO is working towards a unified nomenclature system that can be used globally":-

- Has anyone heard anything more about that one?
- Was (is) it GMDN (or UMDNS)?

* Developed, I believe, for the Italian Ministry of Health some years ago by TBS - and (possibly) based upon an earlier coding system known as ACMAGEST (?)

** Yet Another Global Medical Device Nomenclature System
Posted By: Geoff Hannis Re: GMDN - 25/07/16 1:10 PM

OK guys; I've found what I was looking for (the basics, anyway).

It seems that a CIVAB code is an "eight-character string" that identifies a model of biomedical equipment. The code comprises:-

1) Equipment class (3 chars)
2) Manufacturer (3 chars)
3) Model (2 chars)

For example:- DEF HPI 20 indicates "Defibrillator, Hewlett-Packard, 43120-A"

And yes, the Si3C system does (or certainly did) use such codes as these. smile

But what is not clear is that when the complete CIVAD code needs to be quoted, is it given as:-

1) DEF HPI 20 (ie, with spaces)
2) DEF-HPI-20 (10 chars)
3) DEFHPI20 (and parsed by the system)
Posted By: Geoff Hannis Re: GMDN - 09/01/17 8:49 PM

Whilst looking for something else, I've just stumbled across this list.

It's about "Expected Equipment Lifespans", and uses both GMDN and UMDNS codes. smile

Although I must say that it all looks a bit "random" to me! The table would have been so much more useful if it had included the Risk Factor (as mentioned in the premable) for each item, as well as the "formula" used to arrive at it.
Posted By: Neil Porter Re: GMDN - 12/01/17 5:03 AM
Geoff, Whilst looking at your list I did notice that the last update to the list was 2004, and that the list contained instruments and other stuff not related to BME. I have found a different list, does not say when it was last updated but does only contain medical equipment.

Posted By: Geoff Hannis Re: GMDN - 12/01/17 3:25 PM

Very good, Neil (and I'm guessing 2011). smile

But ... no Codes!

And no clues as to the criteria used to arrive at the life expectancy figures. frown
Posted By: Malcolm Re: GMDN - 16/01/17 7:52 AM
Hi Geoff,as a Si3C user the "name" is presented as:HEARTSTREAM XL M4735A within the model search facility, as long as you use the % in the string the seach returns any similar item.
Unfortunatly, I believe that CIVAB is no longer being updated.
As an aside what data base does ITAL GB use??

Posted By: Geoff Hannis Re: GMDN - 16/01/17 12:40 PM

Cheers for that, Malcolm. smile

Are you saying that Si3C doesn't use "codes", then (CIVAB, nor any other) ... but just relies on a Model string?

How does it get on with our (oft-cited) "Dinamap" problem? In other words, how well does the system discriminate between different names (strings) for the same equipment? think

I have also wondered about TBS-GB ... but have no information on that score (athough someone I knew there did once admit to using Ansur).
Posted By: Malcolm Re: GMDN - 17/01/17 8:51 AM
Hi Geoff, I'm loosing the ability to write in english these days(not that my italian is any better!!!!)
Si3C is firmly (obsessively) based on CIVAB,as it is no longer kept uptodate?? we are seeing some holes appearing,however for the older equipment someone has gone to the trouble to input almost all the models available in europe(for example searching the db for "dinamap returns 69 different models split between Monitor and NIBP types.
I would assume Dr John can respond to the type of db the TBS-GB uses.
As for ANSUR I met a few years ago on a course hosted by SLT Milan (Fluke distributor for Italy) an engineer from TBS GB who had been given a project(cue Sisyphus) to convert the word based PM protocols to run in ANSUR........

Posted By: Geoff Hannis Re: GMDN - 17/01/17 1:16 PM

The plot thickens! smile

But I'm starting another thread on this one, Malcolm ... as we're getting a bit (way) off "GMDN" here.
Posted By: Angus Re: GMDN - 18/04/20 3:32 PM
ECRI used to allow a free download of the current UMDNS. Now I see this is by paid subscription only.
Posted By: Geoff Hannis Re: GMDN - 18/04/20 4:19 PM
What are you actually in need of, Angus ... how deep do you need to go?

By the way, you do know that GMDN and UMDNS are different, I hope?

When it comes to allocating UMDNS codes, I believe that the equipment manufacturer is supposed to do that; but don't quote me.

As I have made clear in various threads on this and similar topics over the years - I am not a great fan of any of the so-called "global" codes. frown

Yes; you're right - ECRI used to be "not for profit".
Posted By: Geoff Hannis Re: GMDN - 18/04/20 4:36 PM
If you are being asked to go deeper (beyond basic codes), rather than reinventing the wheel, you could try using ARTG* codes.

See this site ... try searching for electrocardiograph for example. Note that you can save (export) search results to .xls.

Meanwhile, how about yet another code-base ... take a look at UNSPSC** ... and try entering electrocardiograph in the Code Name field.

* Australian Register of Therapeutic Goods
** United Nations Standard Products and Services Code

Note that UNSPSC is associated with GS1 (formerly the Uniform Code Council) - another one that has become popular. Yes; there's money in codes!

But UNSPSC is not too bad, as (at least) it uses an hierarchical (8-digit) structure.
Posted By: Angus Re: GMDN - 19/04/20 2:08 PM
Thanks for the info Geoff. Your input is, as always, invaluable.
Posted By: Geoff Hannis Re: GMDN - 19/04/20 3:47 PM
You're very kind, Angus ... but I'm really just pleased that you haven't asked about EUDAMED (or even, EDMA).

That's the thing about codes ... like ideas, they're everywhere; you just need to choose the one most suited to your needs.

On that tack, perhaps if you offered a clue or two about what it is you're trying to achieve, then maybe I (and/or others) may be able to assist you further!

But bear in mind that "GMDN, the de facto global standard for identifying the world’s millions of medical devices, has been made freely available to all users for the first time on 1st April 2019" (see the GMDN Agency website) ... and that the MHRA recommends its use!
© EBME Forums: Biomedical and Clinical Engineering Discussion Forums.