A sample version of the file structure expected for the evp_variable_file
parameter in the evp_process function. The user should recreate this file
structure and include their own variables.
Format
evp_variable_file_pcornet
A data frame with 6 columns
- variable
A string label for the variable of interest
- domain_tbl
The name of the CDM table where the variable can be found
- concept_field
The field in the default_tbl that should be used to join to the codeset
- date_field
The date field in the default_tbl that should be used to filter the dataset to the cohort period and for longitudinal analyses
- vocabulary_field
field in the
default_tblthat defines the vocabulary type of the concept (i.e. dx_type)- codeset_name
The name of the codeset as found in the specs directory; file extension should not be included
- filter_logic
optional; a string to be parsed as logic to filter the default_tbl and better identify the variable of interest
Details
Please note that the codesets should be stored in the file_subdirectory established
when ssdqa.gen::initialize_dq_session is executed.
Examples of appropriately structured codeset files are attached to the ssdqa.gen
package and can be accessed with ssdqa.gen::. Please note that if the vocabulary_field
is not NULL (i.e. for diagnosis or procedures concepts), the vocabulary_id field in the codeset
must be included.