This wiki has undergone a migration to Confluence found Here
Difference between revisions of "FHIR Specification Feedback (DSTU 2)"
Jump to navigation
Jump to search
Rene spronk (talk | contribs) |
(Handling duplicate data) |
||
| Line 9: | Line 9: | ||
*About '''Section 5.1.7 Merging records''' | *About '''Section 5.1.7 Merging records''' | ||
**If two Patient records have been linked somehow, then I'd expect a query for one of the Patient.identifiers of one of these Patient resources to return a response as if a query was executed against all of the linked patient records. A link of type '''seealso''' is likely to be another FHIR Server, so we may have to limit this behaviour to records that are linked and available on a single server. This is akin to some of the v2 implementations around linking patient records. The alternative is to burden the client with checking for links and executing multiple queries; to me this is functionality that should be taken care of by the server. [[User:Rene spronk|Rene spronk]] ([[User talk:Rene spronk|talk]]) 10:42, 29 October 2015 (EDT) | **If two Patient records have been linked somehow, then I'd expect a query for one of the Patient.identifiers of one of these Patient resources to return a response as if a query was executed against all of the linked patient records. A link of type '''seealso''' is likely to be another FHIR Server, so we may have to limit this behaviour to records that are linked and available on a single server. This is akin to some of the v2 implementations around linking patient records. The alternative is to burden the client with checking for links and executing multiple queries; to me this is functionality that should be taken care of by the server. [[User:Rene spronk|Rene spronk]] ([[User talk:Rene spronk|talk]]) 10:42, 29 October 2015 (EDT) | ||
| + | |||
| + | *About updates | ||
| + | I would like to see an update option similar to the conditional update but where a match would tell the server this is a duplicate so the server would do nothing. This would save processing time and logic on the server. Uploading duplicate data is going to be inevitable. | ||
Latest revision as of 09:29, 12 September 2016
This is the welcome page for feedback from DSTU
General Feedback
See the FHIR support page...
Specific feedback requested:
- About Section 5.1.7 Merging records
- If two Patient records have been linked somehow, then I'd expect a query for one of the Patient.identifiers of one of these Patient resources to return a response as if a query was executed against all of the linked patient records. A link of type seealso is likely to be another FHIR Server, so we may have to limit this behaviour to records that are linked and available on a single server. This is akin to some of the v2 implementations around linking patient records. The alternative is to burden the client with checking for links and executing multiple queries; to me this is functionality that should be taken care of by the server. Rene spronk (talk) 10:42, 29 October 2015 (EDT)
- About updates
I would like to see an update option similar to the conditional update but where a match would tell the server this is a duplicate so the server would do nothing. This would save processing time and logic on the server. Uploading duplicate data is going to be inevitable.
