Historical Analytics - update EHR (Data Healing)
Approach
In this approach, rather than changing the version of SNOMED that is being considered, the underlying data is instead brought up to date with the most recent version available, using historical associations.
Pros / Cons
Pro | Con |
|---|---|
Most useful as a "one shot" solution for a given set of data to answer questions at a single point of time. | Can only be done in the context of the analysis being performed. We would not wish to modify the original record |
SNOMED continually update the historical associations for inactive concepts, so this task can be performed with only the most recent SNOMED CT release. | Potentially millions of records to update, TB of data, compared to 124K historically inactive SNOMED concepts. |
More applicable when attributes are used as part of query as modelling will be as up to date as possible. | Only possible where historical associations properly populated. |
Not useful for real time decision support system | |
Becomes more complex if records are successively updated eg X → Y → Z if an audit trail of changes are required. |
Use Cases
Most likely used in a data warehouse situation where records are brought up to date as they are imported into the system.
Copyright © 2026, SNOMED International