WAS_A - Editorial Guide

Current usage:

The purpose of this association type has been to record the complete set of direct parent supertype(s) that were assigned to the inactivated concept in the last release in which it was still active. It's use indicates that there is currently, and unlikely ever to be, a concept or combination of concepts which exactly represents the semantic meaning of the concept to be inactivated.

Providing this information to users/implementers enables the retrieval of historical data based upon the concepts original position in the hierarchy and can now be derived automatically from the RF2 release.

Proposed changes:

While these relationships can now be automatically generated it remains the case that the original inactivated concept will always be related to the original supertypes. If the inactivated concept was either inadequately modeled or subsequent changes have occurred to either the super or subtypes these would not be reflected in the automatically generated WAS_A relationships. For this reason it is proposed that the WAS_A relationships should be explicitly stated and it should be possible to manually change or add WAS_A relationships to more completely and accurately model the inactivation.

The use of WAS_A should be an 'association of last resort' and should only be used where no active clinically useful concept exists that alone or in combination with other(s) could provide semantic equivalence. Potential uses include inactivation of 'classification concepts' such as those which are of the form 'XXX NOS' and XXX otherwise specified' etc. where it is not possible or desirable to create true semantic equivalence and inactivation of groupers which are considered to be of little or no clinical use. In these instances there would also be an option to provide a REPLACED_BY target either in combination with or in place of the WAS_A relationships.

Combinatorial Logic

Whenever an already stated WAS_A target itself also becomes inactive - whether at the same release or later - the combinatorial logic of associations should be:

(A) WAS_A (B AND C)  and  (B) SAME_AS D implies (A) WAS_A (C AND D)
(A) WAS_A (B AND C)  and  (B) REPLACED_BY D implies (A) WAS_A (C AND D)
(A^IntEd) WAS_A (B^IntEd AND C^IntEd)  and  (B^IntEd) MOVED_TO (D^NRC) implies (A^IntEd) WAS_A (C^IntEd) - ? need to manually look for another WAS_A
A WAS_A (B AND C)  and  E MOVED_FROM B implies (A) WAS_A (C AND E)
(A) WAS_A (B AND C)  and  (B) POSSIBLY_EQUIVALENT_TO (D OR E) implies ((A) WAS_A (C)) AND ((A) WAS_A (D OR E))
(A) WAS_A (B AND C)  and  (B) WAS_A (D AND E) implies (A) WAS_A (C AND D AND E)