2769 View
6 CommentResolvedComments enabled
In the category:
Undefined
The Translation Management Guidelines has a section on workflow, but the workflow it suggests may not be practical for each NRCs. We should expand it to discuss alternative workflows with their pros and cons.
These guidelines identify two major parties involved in the translation process: the translation project owner and the translation service provider. Furthermore, three major steps in the translation process are identified (Figure 1):
translation
translation review by the translation service provider (review 1)
translation review by the translation project owner (review 2).
Translation
Ideally, the translation should be carried out by professional translators with a health or social care background and/or health or social care personnel with a professional linguistic background. However, it may be difficult to find a sufficient number of people possessing all these qualifications.
Alternative models could include having the translation carried out either by authorized translators who have current access to consultants with a health or social care background (i.e. subject matter experts), or health and social care professionals who have been specially trained for the task and who may turn to professional translators for advice.
Translation review
Experience from existing translation projects indicates that a two-step review improves the quality of the translation. The first review is a kind of internal quality check performed by the translation service provider. The second review is an external review arranged by the translation project owner.
The competencies of the reviewers may vary, but they are often professional translators or health or social care professionals. All translated terms should at some point be reviewed by a health or social care professional who has been introduced to the structure of SNOMED CT® as well as to the rules of the linguistic guidelines applied in the target language. Ideally there should be a possibility for reviewers to address questions to subject matter experts.
The overall purpose of the reviews is to make sure that the preferred term reflects the underlying concept of the source language, that the term is relevant to the health and social care domain, and that the translation complies with the established linguistic guidelines of the target language.
The reviewers should also identify matters of principle and potential solutions to be presented to the editorial board for decisions.
Editing
Whatever the details of the translation workflow, an editorial board (or similar expert group) should play a part in the overall process and workflow. The editorial board should be interdisciplinary.
The board’s major task is to issue and maintain the linguistic guidelines and to resolve “difficult cases” and matters of principle based on the linguistic guidelines. The editorial board is responsible for the translation quality and issues the “accepted” terms. This means that the editorial board should check if the linguistic guidelines actually have been followed, and – if not – implement the corrections to translations that do not comply with the linguistic guidelines.
The process is nice and short, but seems to expect an awful lot of expertise from the reviewer. They should be a healthcare specialist and a professional translator - not many of those about... Also, most healthcare specialists have patients to treat so where are they going to find time to review 300,000 concepts? That question must be addressed before you start. For instance, we have hired a lot of physicians who have finished their MSc. and are waiting for their medical specialisation training.
It is probably (we haven't tried yet..) difficult to find translators that have both linguistic competence and healthcare competence. Also, even if the translators have healthcare competence, nobody can cover all domains of healtcare to the same degree. So I think the conclusion is - you have to choose what you priorities are at the different stages. E.g. one can choose to have domain knowledge as the highest priority for translator competence in the first stage. That way, you can ensure that the translations are useful and meaningful for healthcare professionals. Terminological/linguistic knowledge could be defined as the highest priority in the second review (which should be the responsibility of the NRC?). In that stage, one can consult healthcare specialist and an editorial board in difficult cases.
This portrays the Dutch workflow. The translation agency has their own 3-stage workflow. The NRC adds for each concept two reviews: semantic (is the meaning correct?) and clinical (is the translation recognisable for healthcare professionals?). Conflicts between FSN and definition, and duplicate concepts, are reported to Snomed Int.; highly specialised concepts are sent to an expert pool filled with healthcare specialists from different specialisations. The disadvantage is that it is, obviously, rather more complicated and more time-consuming as each concept is seen by at least three people (translator, semantic reviewer, clinical reviewer). The advantage is that it is much easier to find clinical validators and terminologists, than to find reviewers who are both healthcare professionals and terminologist. As Snomed contains upwards of 300,000 concepts, it is unrealistic to expect Snomed enthusiasts to review the entire thing, however motivated they are. When you start hiring reviewers, you had better make your requirements realistic.
This workflow was sketched at the end of last year. At first, we wanted all clinical validators to be clinical specialists. By now, we have compromised by hiring unspecialised doctors (with MD and an official registration as physician), training them as terminologists using the e-Learning environment, and then asking them to do the semantic and clinical review at once.
Clinical specialists would be better able to specify which translations are most recognisable, but as a rule they are simply too busy and not in sufficient need of cash... And it turned out that it was actually easier to find MD's than terminologists, if you don't mind that they stay only a few months.
I think it very difficult to separate Semantic review from Clinical validation. It should be a team effort (responsibility of the NRC?). For bigger projects, there may be a problem having sufficient resource in the NRC, in which case you would have to hire extra personell.
Who does the first translation/review? I think that depends a lot on the project. In case there is an implementation project with sufficient resources (external to the NRC), it could reqruit and manage the translator (based on requirements set by the NRC). In case there is no such project, the NRC would have to organise the translation project. In case only a few translations are to be made, and the NRC has sufficient competence/resources, the NRC itself can do the translation/review, consulting the editorial board or healthcare spesialist when needed.
6 Comments
Feikje Hielkema-Raadsveld
From the current guidelines:
These guidelines identify two major parties involved in the translation process: the translation project owner and the translation service provider. Furthermore, three major steps in the translation process are identified (Figure 1):
translation
translation review by the translation service provider (review 1)
translation review by the translation project owner (review 2).
Translation
Ideally, the translation should be carried out by professional translators with a health or social care background and/or health or social care personnel with a professional linguistic background. However, it may be difficult to find a sufficient number of people possessing all these qualifications.
Alternative models could include having the translation carried out either by authorized translators who have current access to consultants with a health or social care background (i.e. subject matter experts), or health and social care professionals who have been specially trained for the task and who may turn to professional translators for advice.
Translation review
Experience from existing translation projects indicates that a two-step review improves the quality of the translation. The first review is a kind of internal quality check performed by the translation service provider. The second review is an external review arranged by the translation project owner.
The competencies of the reviewers may vary, but they are often professional translators or health or social care professionals. All translated terms should at some point be reviewed by a health or social care professional who has been introduced to the structure of SNOMED CT® as well as to the rules of the linguistic guidelines applied in the target language. Ideally there should be a possibility for reviewers to address questions to subject matter experts.
The overall purpose of the reviews is to make sure that the preferred term reflects the underlying concept of the source language, that the term is relevant to the health and social care domain, and that the translation complies with the established linguistic guidelines of the target language.
The reviewers should also identify matters of principle and potential solutions to be presented to the editorial board for decisions.
Editing
Whatever the details of the translation workflow, an editorial board (or similar expert group) should play a part in the overall process and workflow. The editorial board should be interdisciplinary.
The board’s major task is to issue and maintain the linguistic guidelines and to resolve “difficult cases” and matters of principle based on the linguistic guidelines. The editorial board is responsible for the translation quality and issues the “accepted” terms. This means that the editorial board should check if the linguistic guidelines actually have been followed, and – if not – implement the corrections to translations that do not comply with the linguistic guidelines.
Feikje Hielkema-Raadsveld
The process is nice and short, but seems to expect an awful lot of expertise from the reviewer. They should be a healthcare specialist and a professional translator - not many of those about... Also, most healthcare specialists have patients to treat so where are they going to find time to review 300,000 concepts? That question must be addressed before you start. For instance, we have hired a lot of physicians who have finished their MSc. and are waiting for their medical specialisation training.
Jostein Ven
It is probably (we haven't tried yet..) difficult to find translators that have both linguistic competence and healthcare competence. Also, even if the translators have healthcare competence, nobody can cover all domains of healtcare to the same degree. So I think the conclusion is - you have to choose what you priorities are at the different stages. E.g. one can choose to have domain knowledge as the highest priority for translator competence in the first stage. That way, you can ensure that the translations are useful and meaningful for healthcare professionals. Terminological/linguistic knowledge could be defined as the highest priority in the second review (which should be the responsibility of the NRC?). In that stage, one can consult healthcare specialist and an editorial board in difficult cases.
Feikje Hielkema-Raadsveld
This portrays the Dutch workflow. The translation agency has their own 3-stage workflow. The NRC adds for each concept two reviews: semantic (is the meaning correct?) and clinical (is the translation recognisable for healthcare professionals?). Conflicts between FSN and definition, and duplicate concepts, are reported to Snomed Int.; highly specialised concepts are sent to an expert pool filled with healthcare specialists from different specialisations. The disadvantage is that it is, obviously, rather more complicated and more time-consuming as each concept is seen by at least three people (translator, semantic reviewer, clinical reviewer). The advantage is that it is much easier to find clinical validators and terminologists, than to find reviewers who are both healthcare professionals and terminologist. As Snomed contains upwards of 300,000 concepts, it is unrealistic to expect Snomed enthusiasts to review the entire thing, however motivated they are. When you start hiring reviewers, you had better make your requirements realistic.
Feikje Hielkema-Raadsveld
This workflow was sketched at the end of last year. At first, we wanted all clinical validators to be clinical specialists. By now, we have compromised by hiring unspecialised doctors (with MD and an official registration as physician), training them as terminologists using the e-Learning environment, and then asking them to do the semantic and clinical review at once.
Clinical specialists would be better able to specify which translations are most recognisable, but as a rule they are simply too busy and not in sufficient need of cash... And it turned out that it was actually easier to find MD's than terminologists, if you don't mind that they stay only a few months.
Jostein Ven
I think it very difficult to separate Semantic review from Clinical validation. It should be a team effort (responsibility of the NRC?). For bigger projects, there may be a problem having sufficient resource in the NRC, in which case you would have to hire extra personell.
Who does the first translation/review? I think that depends a lot on the project. In case there is an implementation project with sufficient resources (external to the NRC), it could reqruit and manage the translator (based on requirements set by the NRC). In case there is no such project, the NRC would have to organise the translation project. In case only a few translations are to be made, and the NRC has sufficient competence/resources, the NRC itself can do the translation/review, consulting the editorial board or healthcare spesialist when needed.