Proposed Hierarchy Discussion Prior to 18.02.18 meeting

2019-Feb-18

Answers to questions posted for discussion and some background into potential answers

Comments

Nathan Billing (Unlicensed)

Feb 18, 2019

Nathan Billing (Unlicensed)

Feb 18, 2019

All,

Penni had shared this https://www.fda.gov/Food/GuidanceRegulation/GuidanceDocumentsRegulatoryInformation/DietarySupplements/ucm381189.htm which is a useful term. I think internationally "Beverage" , like Bill had mentioned is a social construct. Beverage looking at the FDA definitiion appears to be any nutrient containing liquid that is NOT a dietary supplement.

We may need to look at this within the context of the edible substance hierarchy to do this 762766007 |Edible substance (substance)| and then once this is in the approved hierarchy apply this to how we classify Nutrient intake of interest is that this has the following has the following children all of which are not yet fully specified terms and appear to address some of the complexity we have been re litigating in our meeting

226465004 |Drinks (substance)|
255620007 |Foods (substance)|
229979003 |General food and drink types (substance)|
440551000124102 |Bioactive substance (substance)|
735252001 |Protein and protein derivatives (substance)|
59533004 |Food additive (substance)|

Not sure that helps but if we were to work on sorting out these terms we could then use these "edible substances" as part of Dietary, oral and Nutrient intake.

Let me know what you think and how we can work on making the above more usable and fully specified

Thanks,

Nathan

ps will repost this in the discussion forum, sign up as watching get notified when others post and share thoughts asyncronously prior to next meeting

Former user (Deleted)

Feb 18, 2019

Nathan,

I'm concerned that looking to the substances may not yield the insight we need. For instance, "Water (substance)" is not included as an edible substance, yet "Drinks (substance)" are. We don't really have a good way of knowing who made that decision (or any other decisions about the substances hierarchy) or why. "Drinks" in SNOMED subsumes "Alcoholic beverages" and "Beverage whitener", but not "Medical beverage". I think reorganizing these substances could be a whole separate project, but may not be in the current scope of the CPG.

Piggybacking on my example from earlier today:

NCPT has the terms:

Carbohydrate intake
- Total estimated carbohydrate intake
- Total carbohydrate from diet
- Total carbohydrate from enteral nutrition
- Total carbohydrate from parenteral nutrition
- Total carbohydrate from intravenous fluids

As terms are included in NCPT based upon the requests of users of the language and approved by content experts, we assume that they have utility in nutrition practice.

My thinking is that we need to figure out how to represent this in a useable way in SNOMED.

In the currently published edition of SNOMED, here is the hierarchy:

Endocrine, nutritional and metabolic observable
- Nutritional observable
  - Dietary intake
    - Nutrient intake
      - Dietary carbohydrate intake

There are not currently terms in SNOMED representing:

Any total carbohydrate intake
carbohydrate from any route other than oral

So, we need to know:

1.Where does "Total..." go?

I would suggest that "Total" concepts reside under "Nutrient intake". In the current edition, that would be problematic because "Nutrient intake" is subsumed by "Dietary intake". Our request to change that has already been implemented for the next edition, so this is a good possibility

Endocrine, nutritional and metabolic observable
- Nutritional observable
  - Nutrient intake
    - Total estimated carbohydrate intake (new)
    - Dietary intake
      - Dietary carbohydrate intake

2. Where do the intake by route terms go?

Under previous editors, we have been disallowed from making terms that only serve the purpose of organizing other terms. I will need to follow up with Penni to see if that is still the case. Due to this, I wonder if we would be able to create terms like "Dietary intake" (i.e. Parenteral intake, Enteral intake, etc.) in order to create build the hierarchy like this:

Nutrition observable
- Nutrient Intake
  - Enteral intake (new)
    - Total carbohydrate from enteral nutrition (new)
  - Parenteral intake (new)
    - Total carbohydrate from parenteral nutrition (new)

Or it would end up looking more like this:

Nutrition observable
- Nutrient Intake
  - Total carbohydrate from oral nutrition (new)
    - Dietary carbohydrate intake
  - Total carbohydrate from enteral nutrition (new)
  - Total carbohydrate from parenteral nutrition (new)
  - Total carbohydrate from intravenous fluids (new)

I need to understand the consensus of the CPG before I start submitting these terms so that I can make sure we use Penni's time productively, and then return to the CPG to discuss her feedback.

Thanks,

Lee

(edited)

Nathan Billing (Unlicensed)

Feb 19, 2019

Lee,

Great to initiate dialogue. Look we do not need to work on normalizing ALL the details of the substances within the substance field. However they are there and terms we can use to better define what we are trying to do.

You raised a number of issues let me try respond to each in order,

Water is not an edible substance and this could be for a number of reasons,
- Water can be given via IV, used as flushes and other means
- It appears that it is defined in terms of its molecular structure i.e. as H20
  
  This does not mean it cannot be added as a Drink or Substance ,
  These terms are not fully defined or specified and by making use of them we will, help with the definition but also ensure consistency in our logic
Medical beverages are not subsumed by Drinks in Snomed as these are different products and fall under a different hierachy. i.e. not under the edible substance hierarchy but either the pharmaceutical/biological product hierarchy

OR by a medicinal product defined by therapeutic role such as a calorific agent or carbohydrate supplement

In short there are may ways to define supplements and each definition would depend on the context with which we are referring to them e.g as a supplement a patient is having as part of their diet history vs a supplement prescribed to meet an identified energy deficit
Now I am not suggesting that we need to re-organize items within the substance hierarchy, just that we make use of the ones that are there and make sense to help us define what we are working on. i.e making the links to better define the terminology we are putting in place
Re Carbohydrate intake and how to represent this in a meaningful way within SNOMED

Where does "Total..." go?
Total is not an observable entity it is a qualifier value it could go here somewhere
Or

This is I guess the point I am trying to make, it is not a list where we need totals under the Nutrient intake. There are better ways to model totals and values

The Nutrient intake is an observable that carbohydrate has been eaten, there is no scope to record how much this is where you need a quantifiable value and also why any total would be a quantifiable value

In terms of the route can be defined in the description of the food being eaten e.g. assuming a beverage is not consumed intravenously

This can be done in the diet order for a oral nutritional supplement (e.g. ensure) as part of setting the procedure or regimen therapy

by qualifying the route of administration such as part of a diet order

This will obviously be determined by how the patient is able to cope with the various consistencies of food observed

Or based on the notes captured in previous assessments

Look running out of time here and the Carbohydrate from any other route than oral could be a bit more tricky to calculate however I am assuming this is so you can record it as an observable entity or clinical finding

Hope that makes sense and helps

Regards,

Nathan