How to Calculate Drug Dose page; definition of Drug_Exposure quantity field

[MelaniePhilofsky]

EDIT: Please disregard the below statement, I misinterpreted the text.

The text:

"3. Quantified Drugs which are formulated as a concentration

Example: The Clinical Drug is Acetaminophen 250 mg/mL in a 5mL oral suspension. The Quantified Clinical Drug would have 1250 mg / 5 ml in the DRUG_STRENGTH table. Two suspensions are dispensed.
DRUG_STRENGTH The denominator_unit is either mg or mL. The denominator_value might be different from 1.
DRUG_EXPOSURE The quantity refers to a fraction or multiple of the pack.
In the example: 2
Ingredient dose= quantity x numerator_value [numerator_unit_concept_id]
Acetaminophen dose = 2 x 1250mg = 2500mg"

The example is 250 mg/ml, but the rest of the text states it is 1250mg. It should be 250mg. 1250 is an overdose and we don't want to accidentally give a toxic example for dose.

[MaximMoinat]

5 ml of a 250 mg/mL suspension gives you a total of 1250 mg in 5 ml. I assume that is what is done in this example, but it is confusing that it starts from the clinical drug. I propose to remove the first sentence and give a better example for the quantified clinical drug, e.g. 60 ML Acetaminophen 50 MG/ML Oral Suspension

And if we are writing this example, I would also add a separate calculation using just the Clinical Drug (i.e. where the quantity refers to number of ml given).

[MelaniePhilofsky]

Thank you for clarifying, @MaximMoinat!

I started rewriting the example only to discover the Drug_Strength.denominator_value is not always populated. I created an issue in the Vocab. GitHub. The denominator value is needed to determine dose when a person receives less than the full amount of drug in a container. example: quantity = 3ml of 1250 mg in 5 ml; then dose is 750mg. I think having a null value in denominator_value would be confusing to end users. I assume when the denominator is null for a liquid, then it should assumed to be = 1 but, we need the vocab team's input. I'll await a response from the vocal team before writing up an example. And I do agree with rewording and using a different drug example!

[MaximMoinat]

I believe if 3ml of a 5ml product is used, the quantity should be 0.6 (as in 3/5th of the product). Dose will simply be quantity * numerator_value, no denominator_value needed.

See also this paper for some calculation examples: https://pubmed.ncbi.nlm.nih.gov/38773798/

[MaximMoinat]

Tagging the author @tiozab for awareness.

[MelaniePhilofsky]

Doing it the way presented above, the denominator would be needed for the ETL since the data don't have 0.6 of drug product was taken.

Doctors prescribe the active ingredient, the dose and the form. The pharmacy gives the sig (aka instructions) based on the dispensed product and the prescribed dose. The source data generally state: take 2 tablets, take 2 drops, take 2 ml, etc.

The ETL doesn't utilize the Drug Strength table at this time. The Drug Strength table is utilized by those doing analysis. This would be a big change to the ETL. Something we need to think about very carefully. And if it is decided this is the way, we need to give much clearer instructions and examples. And we should create ETL code to ensure every ETL calculates quantity in the same manner.

[tiozab]

@MaximMoinat I am not the author of this page @clairblacketer I was contacting Clair in hope to update the page (but no green light yet)

[MaximMoinat]

Good to know you already contacted Clair for this same update, then at least we are on the same line in terms of the convention change.

(and I meant author of the article on calculating daily dose ;))

[MaximMoinat]

@MelaniePhilofsky I agree, for the mapping to the right drug concept we need the denominator unit and make sure the quantity matches this unit. We are need a set of matching examples both for mapping during ETL and the dose calculation during studies.

[tiozab]

@cgreich (quantity / denominator discussions)

[MelaniePhilofsky]

If we make this change, it would be a change to the definition of a field in the CDM. This change enhances the research potential of this field, but comes at a cost. This is a BIG change. All current ETLs will need to be updated; others might be using the data as it is currently defined; and Atlas utilizes this field. We need community input. @clairblacketer Let's put this topic on the agenda for a CDM WG meeting.

Note: I updated the issue name to reflect the current conversation.

[clairblacketer]

I agree @MelaniePhilofsky. The presentation about this at the CDM WG session on Apr 16th was very informative but we did not come to a decision at that time. Let's bring this up at our meeting this coming Tuesday.

[cgreich]

Friends: Hang on a sec. We (Theresa, Marcel, Alexander and others) are proposing a larger fix for this. I need to write up the proposal. Give me a sec.

[clairblacketer]

@cgreich no problem, we will hold off until we hear from you.

[clairblacketer]

We have multiple issues related to this topic, we will bundle them together and dedicate a specific meeting or meeting(s) to discuss.

[tiozab]

@clairblacketer, great, thanks. Please let me know if any of this is discussed in any meetings, I may not follow all the agendas.

[MaximMoinat]

@tiozab For your awareness, Christian has created a detailed proposal on the forums:
Drug Dose: Fixes required to allow a standardized calculation - Call for input