Observable Entity
Information about a quality/property to be observed and how it will be observed
416320009 |Calcium volume (observable entity)|
276885007 |Core body temperature (observable entity)|
Use of Observable Entities
Observables entities may be used to code headers on a template, elements on a checklist, or to assign values to elements.
For example,
|Color of nail| is an observable entity. |Gray nails| is a finding.
Types of Observable Entities
There are four general types of observable entities for use in health care. Each has different representation requirements and patterns, i.e. the set of attributes will vary.
Quality. A characteristic, feature, or property that is inherent in someone or something.
For example, mass of a person, temperature of internal organs, concentration of sodium in plasma, angle of a joint
Disposition. A characteristic or feature that is not always realized in full.
For example, antibiotic susceptibility of a certain population
Function. The ability of a person, some part of a person, or a thing to perform activities or realize processes.
For example, ability to walk
Process. A process or outcome of a process
For example, secretion rate, heart rate, respiratory rate
Observable Entity vs. Evaluation Procedure
The observable entity and evaluation procedure hierarchies have some of the same attributes. While there should not be a one-to-one correspondence between concepts in these hierarchies, legacy implementations, as described below, may result in the occasional duplication of content.
SNOMED CT contains some concepts in the evaluation procedure hierarchy that logically belong in the observable entity hierarchy. This is a legacy issue that has been discussed extensively with the community of practice without resolution. At this time, the existing evaluation procedure hierarchy will be retained to support existing use cases. No new laboratory content under 122869004 |Measurement procedure (procedure)| will be added in the international release, as this domain is comprehensively covered by the LOINC extension to SNOMED CT. New laboratory content needed to satisfy existing use cases should be sourced from that extension. For non-laboratory content in the evaluation procedure hierarchy — such as clinical measurements performed directly on a patient rather than on a patient sample — new concepts may continue to be added where needed, although it is preferred that they be represented as observable entities.
While some users have indicated they require the use of a measurement procedure concept for ordering and an observable entity concept for reporting the result, this is not the preferred approach for the international release of SNOMED CT nor is it in line with other terminology standards such as LOINC and NPU. When a laboratory or clinical measurement procedure is ordered, there is an expectation that a value will be provided in association with that order. In these cases, measurement procedures are, for all intents and purposes, observables with a different semantic tag. The nature of their top-level parent (122869004 |Measurement procedure (procedure)|) implies that they require a value in order to be assessed and are therefore equivalent to observables.
Because both observable entities and evaluation procedures may be used for ordering, any analytics derived from order data must account for both hierarchy types to ensure complete and accurate results.
As for the progression of the completion of an assessment, that is related to the state diagram (i.e., status) of the progression of a procedure and should not be precoordinated, but handled by the information system in which orders are processed (it is dynamic, not static). The information system should be able to capture the status of a procedure (e.g., ordered, in process, completed). We would not expect the terminology to pre-coordinate this.
For example, LOINC recognizes three different aspects to an observable:
those that can serve as both an order and an observation (e.g. blood glucose level);
those that can be ordered but not directly resulted (e.g. urinalysis, which is a convenience order for multiple individual observations on urine); and
those that can only be resulted and not directly ordered (usually part of an automated system, such as computation of MCHC in hematology).
LOINC assigns this aspect with an attribute value. It is not one of the six main LOINC parts typically visible to users; however, it is included in the LOINC database.
Some areas of the observable entity hierarchy need clarification and remodeling. This includes upper level concepts and hierarchies such as 246464006 | Function (observable entity) | and 415178003 | Process (observable entity) | as well as intermediate primitive and leaf node concepts.
Last updated