Comment:

Condition.clinicalStatus should not be a limited to a code from the condition-code value set. Why?

(1) If Condition really can be used to represent a concern, as its definition states, we expect it to support cases such as "can't pay for medication" and "potentially abusive environment," which don't have clinical status values. The purpose of this property seem to indicate whether the item is "of concern" or "not of concern," thereby supporting application decisions regarding when to display the content. "Of concern" may subsume states that are asymptomatic, which might be classified as "inactive" from a physiological perspective.

(2) The state of the disorder may be a fine-grained clinical concept, possibly specific to the disorder, including not only "well-controlled" and "remission," but also "latent," "sub-clinical," "symptomatic," or any number of domain-specific values. These clinical values should be given an value set that depends on the specific Condition.

Recommend approach is to separate into two attributes:

1. concernStatus: codeableConcept, few values (e.g., active & inactive)

2. clinicalStatus: codeableConcept, many values

It may be possible to infer an appropriate concernStatus from a clinicalStatus, given some set of contextual assumptions.

Current FHIR list:

active - ambiguous. Symptomatic, or Of concern?

recurrence - possibly orthogonal to state axis; difficult to distinguish from relapse

relapse - possibly orthogonal to state axis; difficult to distinguish from recurrence

well-controlled - clinically precise but disorder-specific

poorly-controlled - clinically precise but disorder-specific

inactive ambiguous. Asymptomatic, or Not of concern?

remission - clinically precise but disorder-specific

resolved - possibly useful in both places

Summary:

Condition.clinicalStatus should not be a limited to a code from the condition-code value set