Do FHIR and CSV files contain the same information #1588
-
|
For my application, I am generating a large amount of data. When the number of patients is large, it could be difficult to handle numerous single-patient FHIR files. So I generated CSV files as well, hence my question: do the CSV files contain all the information of FHIR ones, although in a different template, or do they contain less information? Many thanks for helping! |
Beta Was this translation helpful? Give feedback.
Replies: 2 comments
-
They are very close, but not exactly the same. The data is transformed, HL7 FHIR resources have some internal hierarchy that does not exist in the tables. It is possible translation is lossy, but I haven't done an extensive study (nor am I going to do so). But for most purposes, especially if you want to do statistical analysis on the CSV, it is fine. I often use CSV, and turn off FHIR, if I'm just using the output for data analysis.
The advantage of HL7 FHIR is that is an international standard, to the extent that you care about standards compliance, whereas the CSV is not... it is just something we made up and is not supported anywhere else. |
Beta Was this translation helpful? Give feedback.
-
|
Also note that the Bulk Data export provides data in new-line delimited files (NDJSON), which are like CSV; except with dynamic columns. Many databases that support JSON column types (Postgres, Mongo, etc) can import the files directly, and do automated column extraction. |
Beta Was this translation helpful? Give feedback.
They are very close, but not exactly the same. The data is transformed, HL7 FHIR resources have some internal hierarchy that does not exist in the tables. It is possible translation is lossy, but I haven't done an extensive study (nor am I going to do so). But for most purposes, especially if you want to do statistical analysis on the CSV, it is fine. I often use CSV, and turn off FHIR, if I'm just using the output for data analysis.
The advantage of HL7 FHIR is that is an international standar…