This application has been migrated to Atlassian Cloud.
Any changes made here may be lost. Please use the above link for all work going forward.
This works similar to how PatientView operates now. Most of the record is re-sent each time with the exception of the LabOrders, Observations and Procedures sections.
Within these nodes Start and End date attributes can be supplied to indicate that the sections only contain the part of the patients record that falls between those two dates.
If the attributes are supplied then any items of that type that have been received from the unit between those dates are deleted before the items received in the file are loaded.
If the patient's record has been updated or added to then an extract should be generated to send to the UKRDC. This should include notifying us if a patient has died and any amendments that are made to a patient's record after they are deceased.
The default is that every extract should include all data that is held about patient even if these items are unchanged. The reason for this is that the way the UKRDC works as standard is to delete all the existing records of a particular type and then replace them with the data held in the incoming file.
There exceptions to this are -
Where in order to reduce the size of the update files only those records that fall within a configurable time range need to be supplied, with the range being communicated via Start & End attributes on those sections in the XML file. If these attributes are present then the UKRDC's behaviour is to delete only those existing records of each type when the date on which they occured falls between the date range. Any records in the file for that date range are then inserted.
When this method is being used an empty section should still be included in the file if there are no records which fall within the date range.
The date span should be wide enough to cover the longest typical period between an activity (such as a blood sample being taken) and the data last being loaded amended on the system (such as when the result is returned from the lab).- for example the time taken for result to be confirmed. The date values used for the Start and End attributes should be the date range which was used by the extraction tool when querying the database and NOT the earliest and latest dates of items it found. The later behaviour is seen in some current PV feeds.