Skip to end of metadata
Go to start of metadata

You are viewing an old version of this content. View the current version.

Compare with Current View Version History

« Previous Version 14 Current »

Instructions


What is a Catalogue?


A Catalogue is RDMP's representation of one of your datasets e.g. 'Hospital Admissions'. 

Having a Catalogue allows for continuity of cohort building and extraction projects long term regardless of underlying database activities e.g. moving datasets to a new server, schema changes, deprecating datasets etc.

A Catalogue consists of:

  • Human readable names/descriptions of what is in the dataset it is

  • A collection of items mapped to underlying columns in your database. 

  • Validation rules for each of the extractable items in the dataset

  • Graph definitions for viewing the contents of the dataset (and testing filters / cohorts built)

Attachments which help understand the dataset (e.g. a pdf file)

Each of these:

  • Can be extractable or not, or extractable only with Special Approval

  • Can involve a transform on the underlying column (E.g. hash on extraction, UPPER etc)

  • Have a human readable name/description of the column/transform

  • Can have curated WHERE filters defined on them which can be reused for project extraction/cohort generation etc

A Catalogue can be a part of project extraction configurations, used in cohort identification configurations.  They can be marked as Deprecated, Internal etc.

The separation of dataset and underlying table allows you to have multiple datasets both of which draw data from the same table.  It also makes it easier to handle moving a table/database (e.g. to a new server or database) / renaming etc.

If you expand a Catalogue (e.g. biochemistry) you can see the ‘Catalogue Items’ node.   These are the extractable columns in the dataset ‘biochemistry’.  If you expand a Catalogue Item, you can see two nodes:

  • the first is the Extraction Information logic for the column

  • the second is the underlying database column reference (ColumnInfo). 

If you are ever unsure what a node is you can use the ‘What is this’ function:

A full list of objects in RDMP is available through the Help=>Show Objects Help menu.

Finally if you are ever confused about the role of a feature or object you can check the Glossary.

In Summary


  • No labels