Something that has become clear in CIMI, and will affect openEHR, 13606 and most likely any archetype developer is that acknowledgements of 3rd party copyrights and trademarks need to be made. The most obvious common one is likely to be for SNOMED CT codes in archetype bindings (Stan Huff at Intermountain is still working on whether such acknowledgements are needed for LOINC codes). However, it could be for anything, e.g. rights to use a scale like Barthel or Waterlow.
At the moment there is no dedicated place in the model for this particular meta-data. It could just go in ‘other_details’ but I suspect that we need to be more precise than that. Consider for example, the openEHR Barthel scale archetype - it currently carries this text in the ‘Use’ section:
Note:
The Maryland State Medical Society holds the copyright for the Barthel Index. It may be used freely for non-commercial purposes with the following citation:
Mahoney FI, Barthel D. “Functional evaluation: the Barthel Index.”
Maryland State Med Journal 1965;14:56-61. Used with permission.
Permission is required to modify the Barthel Index or to use it for commercial purposes.
This seems less than optimal, and is certainly not going to be reliably tool-separable from the main ‘Use’ content, since the word ‘Note:’ and the placement of this text are purely local choices.
There is another issue here. The acknowledgement text actually included in the archetype needs to be minimal, and as far as legally possible not contain volatile elements that can change. Therefore, I think the general approach needs to be as is typically done with open source licences: not including the whole text, but including a reliable URL to the licence text either from the issuer (e.g. Creative Commons CC-BY page) or an agreement between the publisher and the licensor (e.g. between IHTSDO and CIMI for the use of SNOMED CT, and details of that use).
I have updated the meta-data page on the wiki to indicate what I think is the requirement - see end of the main table.
I am increasingly of the feeling that we need to act on this soon.
It is actually relevant to have this information in the ‘Use’ as it is a reasonable place to look to see constraints on use and how it should be implemented in systems.
No reason why it can’t be in a dedicated/purpose-built place as well.
My worry about having a dedicated attribute is that the requirements
may vary pretty widely from simple reference to wordy attribution (as
per the Barthel example).
This also raises an issue about whether the Foundation should seek a
similar arrangement with IHTSDO re use of SNOMED CT terms.
Hello!
We are using LOINC in Austria for coding lab results on a national scale.
As far as I know nobody needs to pay anything to Regenstrief to do so.
I am not aware of any "must mention Regenstrief" requirements, but I may miss something.
Greetings from Vienna,
Stefan
Stefan Sauermann
Program Director
Biomedical Engineering Sciences (Master)
University of Applied Sciences Technikum Wien
Hoechstaedtplatz 5, 1200 Vienna, Austria
P: +43 1 333 40 77 - 988
M: +43 664 6192555
E: stefan.sauermann@technikum-wien.at
My interpretation of the use attribute is that it refers to the use of archetype from the clinical perspective, not from the legal one. I would not mix both things.
Moreover, one of the new metadata requirements is to include the license of the archetype itself (not only the copyright). So probably this license and the third party licenses could be represented in the same place.
We also have the other_details references.
If only a reference is required this could be the place for it, if a wordy attribution with constraints on how the archetype can be used is required, this could be in ‘Use’.
References could of course be made a dedicated field, rather than one that is used by convention in other_details.
So we all would probably benefit from creating some copy and paste
examples for common 3rd party attribution that can be easily
incorporated into archetypes / resources.
Ian
Dr Ian McNicoll
office +44 (0)1536 414 994
fax +44 (0)1536 516317
mobile +44 (0)775 209 7859
skype ianmcnicoll
ian.mcnicoll@oceaninformatics.com
Clinical Modelling Consultant, Ocean Informatics, UK
Director openEHR Foundation www.openehr.org/knowledge
Honorary Senior Research Associate, CHIME, UCL
SCIMP Working Group, NHS Scotland
BCS Primary Health Care www.phcsg.org
If the 3rd party attribution will be always the same, should not be
stored in the repository or make that it is somewhat assumed for
archetypes in a given domain instead of repeating it on each
archetype?
And as David says, a new metadata attribute was included to deal with
this if still it is needed to be put (the barthel scale use case)
I did wonder about the idea of repository-level attribution but in
practice the archetypes will often be disconnected and shared in local
repositories etc, so I suspect we are stuck with attribution for each
archetype.
I have looked at Thomas's proposal on the wiki and now think that is
probably the best answer. It keeps 'use' and 'references' for their
original intention and might make it easier to automate the import of
boilerplate term of use etc. e.g. The repository manager could check
for SNOMED CT and LOINC bindings on upload and offer to add the
requisite text.
Ian
Dr Ian McNicoll
mobile +44 (0)775 209 7859
office +44 (0)1536 414994
skype: ianmcnicoll
email: ian@freshehr.com
twitter: @ianmcnicoll
Director, freshEHR Clinical Informatics
Director, openEHR Foundation
Director, HANDIHealth CIC
Hon. Senior Research Associate, CHIME, UCL
right - that’s a good point because there may be some clinically relevant statement to make, especially like in the Barthel case. But I believe there needs to be an easily findable legal form of statement that recognises each 3rd party IP implicated in the archetype - and I think that the latter statements are more likely to be legal-ish words + URLs. - thomas
Well, my point here isn't to discuss whether LOINC needs to be mentioned, it's to determine the way for general recognition of 3rd party IP used in the artefact.
I've had discussions with IHTSDO about needing a formal statement about SCT that should be placed in archetypes and this will be identical to that which they are currently working on with CIMI.
