Note - the QDM to QI-Core mapping that suggests using Observation.category and assigning a user-defined value of symptom should change to ignore the Observation.category element and not recommend a specific category code; just reference “Symptom” as an Observation.

The QDM User Group (10/21/2020) agreed with existing guidance in QDM documentation and with the existing QDM to QI-Core mapping for determination of symptoms.

The QDM User Group met on October 21, 2020 and discussed this issue. The group agreed that the QDM datatype "Symptom" currently contains guidance consistent with recommendations in the FHIR R4 and developing FHIR R5 Condition boundaries and relationships documentation. As stated the existence of a symptom would more effectively be retrieved using the QDM "Diagnosis" datatype (for long-term and/or as yet undiagnosed symptoms) and the QDM "Assessment, Performed" datatype for observations that indicate symptoms. As in the use case provided, a specific "Encounter, Performed" procedure code might be able to represent a specific encounter due to a symptom; thus, "Encounter, Performed" may also have value in capturing all possible information about symptoms.  The User Group did not support removing "Symptom" as a QDM datatype but there was support for general guidance that expressions should preferentially use "Assessment, Performed" and/or "Diagnosis" rather than "Symptom".  The QDM to QI-Core mapping is consistent with this approach as currently stated.  

Set for October 21, 2020 QDM User Group Meeting

• Note in the QDM definition – the word “indication” is probably incorrect – perhaps we need to say “Symptom is  one of several factors that may indicate a person has a condition or disease.”
• Basically, I expect that clinical software will indicate patient reported symptoms and clinician identified findings but all will be identified as observations unless the clinician determines the symptom requires long-term management whether or not it has clear causation. I think this is consistent with the recommendations in the FHIR Condition resource. 
• In existing QDM, a generic “observation” uses “Assessment, Performed”. And if the clinician has captured the information on a Problem list it MAY be identified as a condition (“Diagnosis” in QDM-speak).
• So it may be best to avoid the use of “Symptom” in QDM and I think that is what you have been thinking.  The discussion on QDM User Group will be important from the implementer and vendor perspective to learn what they would do with either “Symptom”, “Diagnosis”, or “Assessment, Performed” given your use case examples. You might need to Union “Assessment, Performed” and “Diagnosis” with the same value set for each.
• In QI-Core/FHIR, I’m re-thinking the Observation.category = symptom suggestion in the mapping. The relevant Observation_category value set has a SHOULD binding but symptom is NOT one of the 9 concepts. I’m sure Rob McClure will have a comment on this one and I’d like to hear what he has to say.
• Of course, if you use an evaluation tool indicating frailty based on a validated score consistent with a set of observations, “Assessment, Performed” in QDM, or Observation in QI-Core would be more straightforward. But I understand you are looking for more subtle, single observations.

This is definitely a good subject for the October QDM User Group discussion and I will add it to the agenda.

Thank you for submitting this question.

The QDM mapping to QI-Core is based on the FHIR R4.0.1 guidance in Boundaries and Relationships:

This resource is not typically used to record information about subjective and objective information that might lead to the recording of a Condition resource. Such signs and symptoms are typically captured using the Observation resource; although in some cases a persistent symptom, e.g. fever, headache may be captured as a condition before a definitive diagnosis can be discerned by a clinician. By contrast, headache may be captured as an Observation when it contributes to the establishment of a meningitis Condition.

Use the Observation resource when a symptom is resolved without long term management, tracking, or when a symptom contributes to the establishment of a condition.

Use Condition when a symptom requires long term management, tracking, or is used as a proxy for a diagnosis or problem that is not yet determined.

When the diagnosis is related to an allergy or intolerance, the Condition and AllergyIntolerance resources can both be used. However, to be actionable for decision support, using Condition alone is not sufficient as the allergy or intolerance condition needs to be represented as an AllergyIntolerance.

http://hl7.org/fhir/condition.html

FHIR R5 has the same language:

This resource is not typically used to record information about subjective and objective information that might lead to the recording of a Condition resource. Such signs and symptoms are typically captured using the Observation resource; although in some cases a persistent symptom, e.g. fever, headache may be captured as a condition before a definitive diagnosis can be discerned by a clinician. By contrast, headache may be captured as an Observation when it contributes to the establishment of a meningitis Condition.

Use the Observation resource when a symptom is resolved without long term management, tracking, or when a symptom contributes to the establishment of a condition.

Use Condition when a symptom requires long term management, tracking, or is used as a proxy for a diagnosis or problem that is not yet determined.

When the diagnosis is related to an allergy or intolerance, the Condition and AllergyIntolerance resources can both be used. However, to be actionable for decision support, using Condition alone is not sufficient as the allergy or intolerance condition needs to be represented as an AllergyIntolerance.

http://hl7.org/fhir/2020Sep/condition.html

Note that FHIR R5 has a new resource (maturity level 0) ConditionDefinition – a set of system properties for a particular condition. It uses the same Observation.category (Observation_category value set as a preferred binding). This new resource doesn’t seem to use symptom as a specific factor leading to ConditionDefinition except as an Observation even though the category would be unclear (unless activity, social-history, or survey).

CURRENT QDM to QI-Core mapping states:

QDM defines Symptom as an indication that a person has a condition or disease. Some examples include headache, fever, fatigue, nausea, vomiting, and pain. Symptoms are subjective manifestations of the disease perceived by the patient. As an example to differentiate symptom from finding, the patient’s subjective symptom of fever is distinguished from the temperature (a finding). For a finding, there is either a source of a temperature-measuring device together with a recorder of the device (electronically) or an individual (healthcare provider, patient, etc.).

Note: Definitions regarding symptom on the FHIR condition resource Boundaries and Relationships (Section 9.2.2: http://hl7.org/fhir/condition.html):

• [The Condition] resource is not typically used to record information about subjective and objective information that might lead to the recording of a Condition resource. Such signs and symptoms are typically captured using the Observation resource; although in some cases a persistent symptom, e.g. fever, headache may be captured as a condition before a definitive diagnosis can be discerned by a clinician. By contrast, headache may be captured as an Observation when it contributes to the establishment of a meningitis Condition.
• Use the Observation resource when a symptom is resolved without long term management, tracking, or when a symptom contributes to the establishment of a condition.
• Use Condition when a symptom requires long term management, tracking, or is used as a proxy for a diagnosis or problem that is not yet determined.

Section 8.22 http://hl7.org/fhir/us/qicore/qdm-to-qicore.html