Abstract
Background/Aims The purpose of the presentation is to demonstrate how we use SAS and Oracle to load VDW_Utilization, VDW_DX, and VDW_PX tables from Clarity at the Kaiser Permanente Northern California (KPNC) Division of Research (DOR) site.
Methods DOR uses the best of Oracle PL/ SQL and SAS capabilities in building Extract Transform and Load (ETL) processes. These processes extract patient encounter, diagnosis, and procedure data from Teradata-based Clarity. The data is then transformed to fit HMORN’s VDW definitions of the table. This data is then loaded into the Oracle-based VDW table on DOR’s research database and then finally a copy of the table is also created as a SAS dataset.
Results DOR builds robust and efficient ETL processes that refresh VDW Utilization table on a monthly basis processing millions of records/observations. The ETL processes have the capability to identify daily changes in Clarity and update the VDW tables on a daily basis.
Conclusions KPNC DOR combines the best of both Oracle and SAS worlds to build ETL processes that load the data into VDW Utilization tables efficiently.
