Natural experiments for translation in diabetes (NEXT-D)

This is code to support the NEXT-D study.

Inital code by Al'ona Furmanchuk with Dan Connolly. All contributors:

See also:

Implemenation overview

NEXT-D query code targets PCORNET CDM implementations, originally developed against a SQLServer (see SQLServer_impl) CDM by Al'ona Furmanchuk and then ported to Oracle (see Oracle_impl).

The SQLServer implementation using local temp tables. These are not implemented by Oracle which uses global temp tables. The Oracle implementation therefore has "init_Oracle_temp_tables_ddl.sql" to define these tables. The definitions of these tables (but note their data contents) will persist in the Oracle schema used.

Oracle sites may wish to create a separate NEXT-D schema with select priveleges on their CDM schema to segregate NEXT-D specific work.

Reference code sets

Common code references for lab and medication are in ref_code_table_data and will need to be loaded into their corresponding tables in the NEXT-D/CDM schema when required by subsquent data analysis and extract steps.

These data are not required for the "Table1" extract due 2017-09-05 (see Ticket:545)

Table 1 subset extraction (due 2017-09-05)

SQLServer code has not been updated yet for this.

Oracle code is in Oracle_impl/NextD_table1.sql.

  • First run init_Oracle_temp_tables_ddl.sql (errors from TRUNCATE and DROP can be ignored on an initial run)
  • Either modify NextD_table1.sql to reference your specific CDM schema (replacing all references to "&&PCORNET_CDM""), or rely on SqlPlus variable substition if that is your Oracle client of choice.
  • Run NextD_table1.sql
  • Extract the "SubTable1_for_export" result table and upload to REDCap.

Study Info

  • Principal Investigators:
    • Bernard S. Black, JD, MA
    • Abel N. Kho, MD
  • Co-Investigators:
    • Laura J. Rasmussen-Torvik, PhD, MPH, FAHA
    • John Meurer, MD, MBA
    • Russ Waitman, PhD
    • Mei Liu, PhD
  • Nov 2014 study protocol