OHDSI/ETL-HealthVerityBuilder

Name: ETL-HealthVerityBuilder

Owner: Observational Health Data Sciences and Informatics

Description: ETL to take dataset from HealthVerity and convert to OMOP CDM v5 [in development, not for production use yet]

Created: 2017-09-08 17:43:45.0

Updated: 2017-09-08 18:29:24.0

Pushed: 2017-09-08 18:38:25.0

Homepage: null

Size: 631

Language: R

GitHub Committers

UserMost Recent Commit# Commits
Patrick Ryan2017-09-08 17:43:46.01
clairblacketer2017-09-08 18:38:24.07

Other Committers

UserEmailMost Recent Commit# Commits

README

ETL and builder for HealthVerity data

Note: This builder was written for a data cut containing data from an open claims datasource and a closed claims datasource and is still under development.

Dependencies

This package uses the DatabaseConnector package and the SqlRender package. It also requires an instance of the OMOP vocabulary.

How to run

The file extras/TestCode.R shows an example of how to run this builder.

  1. Using the DatabaseConnector package set the connection details to the server
  2. Set the variables cdmDatabaseSchema, sourceDatabaseSchema and vocabDatabaseSchema a. In this example the cdm schema, source schema and vocab schema all exist on the same server
  3. Run the chunkData() function. This will create a dataframe with a list of all the first digits of the HVIDs (set by nSubstring argument) in order to chunk the data to allow for more efficient processing
  4. Run the getIteraterNum() function. This will find how many chunks the code has to run
  5. Run the createVocabTables() function to create the vocabulary mapping tables.
  6. Run the createAllLookupTables() function to create the LOCATION, PERSON, CARE_SITE and PROVIDER tables
  7. Run the createObservationPeriodTable() function to create the OBSERVATION_PERIOD table. If createWebmdObsPeriod is set to TRUE then a separate OBSERVATION_PERIOD table will be created just for the time period the patients were in WebMD
  8. Run the createPayerPlanPeriodTable() function to create the PAYER_PLAN_PERIOD table.
  9. Run the createVisitOccurrenceTable() function to create the VISIT_OCCURRENCE table. Be sure to set runAllParts to TRUE
  10. Run the createSTEMTable() function to create the STEM table
  11. Run the createEventTables() function to create the following tables from the STEM table: CONDITION_OCCURRENCE, PROCEDURE_OCCURRENCE, DRUG_EXPOSURE, DEVICE_EXPOSURE, MEASUREMENT, OBSERVATION
This work is supported by the National Institutes of Health's National Center for Advancing Translational Sciences, Grant Number U24TR002306. This work is solely the responsibility of the creators and does not necessarily represent the official views of the National Institutes of Health.