OHDSI Home | Forums | Wiki | Github

Multum Drug Vocabulary

(Clair Blacketer) #1

Has anyone used the Multum vocabulary successfully? I am trying the map the NHANES drugs and I was excited when I saw that they use Multum since we have it mapped to RxNorm. However, the codes NHANES uses to identify the drugs look completely different than what is in the vocab. Does anyone have an explanation as to why, or can anyone offer any guidance on how to use it?


(Christian Reich) #2


Oh God, Claire. I used to know this long ago. There is something about those. Can you post some examples?

(Clair Blacketer) #3

@Christian_Reich Sure!

Here is what it says on the NHANES website about the vocabulary they use:

NCHS used the Lexicon Plus®, a proprietary database of Cerner Multum, Inc. to assist with data collection, data editing, and release. The Lexicon Plus is a comprehensive database of all prescription and some nonprescription drug products available in the U.S. drug market. Multum distributes the Lexicon Plus in Microsoft Access format and posts regular updates on its website.

Taking insulin as an example, in the database it has generic drug code of d00262. There are 31 entries in our Multum vocabulary that contain the word insulin but the source codes range from 5068 to 13364 and none start with the letter ‘d’.

(Charles Bailey) #4

This is perhaps of limited use, since I understand the Multum details incompletely, but the two PEDSnet sites that use Multum as a drug info vendor report that Multum uses two “levels” of coding, the MMDC and the Multum Drug ID. It looks like NHANES has Multum Drug IDs, which are maybe close to an RxNorm IN or SCDF? I’m not sure whether OMOP has MMDCs or something else entirely, but I haven’t seen MDIDs. Clear as mud?

(Christian Reich) #5


We get the Multum codes from RxNorm. They use it to build RxNorm, it’s one of the source for Clinical and Branded Drugs (SCD, SBD). They used to be good to map data during ETL. If this is no longer the case (like it isn’t with Medi-Span and FDB-coded durgs), you will need a translation table from the codes provided to RxNorm, vs the codes you have in the data, unfortunately. You can read about what we do have here.

Maybe @Overhage can help. I’ll send him a separate note as well.

NHANES Multum Lexicon
(Hung Do) #6

Hi all, … fast forward 3 years… Did you get any progress on this topic, @Overhage @Christian_Reich @clairblacketer ? our hospital is also using Multum, for example, paracetamol = 20280. I hope that I won’t have to map from scratch… Are there any closely related vocabulary?
Really appreciate your help.

(Christian Reich) #7


Same old. Probably a good time to reach out to Cerner directly. Let me do that.

20280 paracetamol? We don’t have that. Which country is that? We are supporting the Multum drug repository in the US, where folks would not know what paracetamol is. It’s called Acetaminophen or Tylenol. Also, the Multum codes we have are on the drug product level, not the ingredient level. Let me know.

(Michael Kallfelz) #8

Dear Hung,
what you have found is a drug synonym ID. Multum is a little bit complex here and makes groupings of those synonyms based on “function” IDs, where for example ‘59’ is a Generic Product Name and ‘16’ is a Generic Drug Name. Again, it would be interesting to hear which country this is, as Multum has different “cultures” that they deliver to their clients. My guess is that we are talking UK?
In the US delivery (maybe in other cultures, too) in the client delivery package you will find a PDF about Multum’s Drug Names and Identifiers which presumably will be a good read.
In general, the base of everything in Multum is a DNUM (Drug Number) entry, a representation of a concept very close to ingredient level. The next most important entity is the MMDC (Main Multum Drug Code), which is a Generic Product and as such shares some similarity with the Semantic Clinical Drug (but is not a 100% match). DNUMs and MMDCs should be universally valid throughout all cultures (except for those, where the respective drugs are not available).
The data delivery is normally packaged in MS Access style database format. So, if you are lucky, in your culture you will also find a table called rxn_multum_map. If you search for your code 20280 in the drug synonym colum, you will find that it maps to the MMDC code 3826 and to the RxNorm RXCUI 198440. If you do not have this table, you might want to request this information from Multum to be added to your culture / delivery.
Please let us know, if you had any success.