Again, OCL does not really provide a way to distribute the DRUG table for OpenMRS. The dropbox contains older versions of drug tables, and I have been adding to the CIEL drug table but currently there is no plan for distributing that through OCL at this time.
I don’t think that is such an issue as we already have a comprehensive data set for that, as it is the one area that was already digitised. It’s all in French, so I assume that if someone on the English locale uses it, because there will be no English translation it’ll just use the French data anyway.
But you need to have a link to the correct concept in the concept table. Are you not going to touch drug concepts in the concept table?
If I understand this correctly, the idea is that our list of drugs/supplies (or perhaps medical supplies would only sit in Odoo?) would be linked to a concept on a line by line basis?
Since you already have a custom drug table with quantity and dosage form, are you planning to treat OCL as a terminology layer only, or as the primary drug dictionary? That decision changes everything. In most Bahmni setups I’ve seen, teams map local drugs to OCL concepts instead of replacing their full drug master, mainly to avoid breaking existing workflows.
To be honest, I’m still trying to get my head correctly around how it all works/maps. Essentially, whatever is going to be least interruptive. I was working on the basis I was likely going to have to map our list we already have to the existing? What would happen if a user pulls up a drug that the OCL data provides, that we don’t have i.e., would we need to go through each item that doesn’t match and set it to not available or similar?
OCL has had some discussions about where the drugs and supplies should be managed, and it is likely not going to be in OCL when the stock/form/brands are concerned. However, OMRS does need the link between the drug row and the underlying concept. The underlying concept (usually an ingredient or multi-ingredient concept) has maps to SNOMED, RxNORM, etc, for things like CDSS. Right now, it looks like you have individual concepts for each drug, rather than a roll-up to ingredient/generic form in concept. You would need to make sure that each drug row has an appropriate map to a concept from OCL/CIEL/PIH etc. That mapping process could be facilitated by OCL, but would still primarily be a manual process. You would then need to work the other way and make sure that all OCL concepts are represented in the drug table if you plan to use those workflows. Not sure about the UI implications of concepts v. drugs in Bahmni these days….
1 Like
I have reviewed the drug table with the French names from Madagascar. I have added synonyms to CIEL’s ingredient concepts for those drugs which were possible and created new concepts were they were missing (for certain combinations). There are some herbal, OTC preparatations which I did not add. These will be available in the next CIEL release. It should make it possible to more easily find the correct standard drug ingredient/multi-ingredient concept to map a custom drug table row. Synching cloned CIEL/PIH concepts to local drug table rows will likely be the most challenging and that’s where having the enhanced indexes (and perhaps access to a standard, linked CIEL-drug table) might make this easier.
1 Like
Thanks Andrew. Does that mean it’s coming in the next release, and not available to use at this stage? Our implementation team have come back to me saying it has the basic names of the drugs only, which matches what you said e.g., Amoxicillin and clavulanic acid for which the dictionary already shows French name “Amoxicilline et acide clavulanique” but as drug we would be needing with strength like “Amoxicillin 125 mg + Clavulanic acid 31.25mg /5 ml” for which there is no French name.
I think you’ve covered this in saying, something like the above example, would still the hand cranking exercise to map each one back to the generic concept? Do you think there be a way that I could do it by having an export of data, manipulate/update that way, rather than clicking and linking 1 by 1 in the OpenMRS interface?
Correct. It will be in the next release. The mapping between drug and concept can be done using Initializer and so it could be possible to do the mapping offline and then import. This might be a use case to use the OCL Mapper (in Early access mode), or you can try a different method using the downloaded version of CIEL.
Noted, thank you.