It's almost unthinkable that just a few years ago many health systems were still using paper charts. As they began to roll out and adoption increased the types of available data sources we pretty much limited to the source systems themselves or text files that were shipped back and forth usually in a comma (or other delimiter) formatted file.
Today the sky is the limit. Literally. Open Cloud based data sources are spinning up faster than anyone can even keep track of them.
There are tons of articles on how to LOAD data from traditional sources. But what about some of the new data sources? One type of file format is really cool. It lets the originator output only the very minimal amount of information.
If you have 5 fields for a patient you can output something like this:
I know what you are thinking ... no magic there who cares.Let me finish typing before you start reading my data tables:
|2||Automobile||Foot driven engine|
Notice that this the records do not have to contain the same field attributes. This is not a simple cross tab type thing. If the fields are blank there is no need to transmit them at all. If the records are from differing sources without the same elements they can still be combined and generated.
But if you do a LOAD statement on this source in Qlik you are probably not surprised that you will get a table with three (3) fields:
The magic of Qlik's ETL still never ceases to amaze me. There is a simple keyword GENERIC that can be added to the LOAD statement and voila it will create an entity with First Name, Last Name, Age, Automobile, City and Occupation which has the values for the fields that were loaded. If you change the input file and new fields are added ... no problem ... the system will handle it for you.
Check it out in this short video clip: