Diet findings

Hi  Elaine Wooler and everyone,

Here are some additional analytical review items for the group to consider when reviewing the spreadsheet.

The Summary tab on the spreadsheet contains a description of the content on each of the tabs in the spreadsheet.

There were 58 potential duplicates that were identified through both automated and manual review of the hierarchy. The analysis was unable to be completed using solely automated analytical means because the concepts in the finding hierarchy contain ambiguous words such as "low" and "restricted" which were represented using the word "decreased" in regime/therapy. 

< 182922004 |Dietary regime (regime/therapy) |

• 97.4% (266 out of 273 total concepts) were created after 2010 which means this is a newer hierarchy that is better aligned to current day editorial guidance
• Two concepts out of the entire hierarchy contain the prefix “provision of”.
  • The CRG may want to consider creating an editorial policy regarding the use of "provision of" as a prefix in regard to a regime/therapy
• One concept contains the article “a” in the FSN. This should be removed as it is prohibited under current editorial policy
• The hierarchy does contain diets that are not prescribed “clinical” or by a clinician (e.g. 765024005 |Atkins diet (regime/therapy) |)

< 41829006 |Dietary finding (finding)|

• 2.6 % (9 out of 339 total concepts) were created after 2010 which means this is an older hierarchy which is more likely to be out-of-date with current editorial guidance.
• Concept descriptions contain the words “low” and/or “restricted” that have been replaced by the word “decreased” in the regime/therapy hierarchy
• Concept descriptions contain the words “high” and/or “rich” that have been replaced by the word “increased” in the regime/therapy hierarchy
• Concepts represent ambiguous diet names with no indication of increase or decrease of a specific nutrient. (e.g. 111176008 |Calcium diet (finding)| )
• Concept descriptions precoordinated a numeric range for a nutrient with the nutrient itself. (e.g. 41112006 |150 milligram calcium diet (finding)|). These are not currently found in the Regime/Therapy hierarchy, so a decision is needed to determine the future of these concepts.
• Concept descriptions that precoordinated a medicinal product with a reason for giving the product represented as a diet (e.g. 34394003 |Hyperalimentation formula for burn hypermetabolism (finding)|)
• Potentially excessive precoordination with ambiguous meaning (e.g. 86433007 |Infant diet 0-12 months: human milk, vitamin D and C, fluoride and iron (finding)|)
• Concept descriptions that precoordinated a diet restriction provided during a given period of time (e.g. 52659007 |Vitamin K restricted diet during anticoagulation therapy (finding)|)

Both subhierarchies

• Concept descriptions contain inconsistent use of the hyphen as part of the description (e.g. 302322006 |Egg-free diet (finding)| versus 1254995006 |Egg free diet (regime/therapy) |). The Nutrition CRG should define editorial policy around the use of the hyphen in the representing a dietary regime in SNOMED CT and then the terminology cleaned up to align with the editorial policy.
• The Nutrition CRG should consider creating editorial guidance around the proper representation of the word “free” in reference to the nutrient content of a diet and/or food (e.g. 160671006 |Gluten-free diet (finding)|). This is not necessarily a diet devoid of <x> but may indicate a diet consisting of foods that are below an arbitrary threshold of detection as set forth by a regulatory agency.
• The Nutrition CRG should consider creating editorial guidance around the proper application of reserved terms used in the nutrition labeling of food products, as defined by regulatory agencies, in reference to a dietary regime/therapy name (e.g. Low sodium, very low sodium, low fat, fat-free).

I hope this helps the CRG in reviewing the spreadsheet to determine the work to be done moving forward.

Best Regards,

John