Domain Concordance Input File (PCORnet)

A sample version of an PCORnet compliant input file for the Domain Concordance data quality module. This exact file is also included as a CSV in the package if the user wishes to use it, or the structure can be copied to produce a custom list of checks.

Usage

dcon_input_pcornet

Format

dcon_input_pcornet

A dataframe or CSV file with 10 columns

cohort_id

A short string describing the event in that row of the input file

check_id

A short string "code" used to identify the specific check (ex: cancer_oncologists). There should always be TWO rows with the SAME check_id, with each row representing one of the two input cohorts.

cohort_description

A longer description of the cohort (ex: cancer diagnoses, oncology specialist visits)

schema

The schema where the data is kept. Use cdm to use the pre-configured cdm_schema, result to use the preconfigured results_schema, or input the exact name of the schema.

table

The name of the CDM or pre-computed results table where the relevant data is kept

date_field

The date field to be used when time_between_events is not null to determine the length of time between each event occurrence

vocabulary_field

Typically only relevant for diagnoses and procedures, the field where the code vocabulary is kept (i.e. dx_type, px_type)

concept_field

The field with the concept codes that make up the valueset (ex: dx, provider_specialty_primary)

conceptset_name

OPTIONAL The string name of the concept set that will identify the concepts of interest, as it appears in the predefined file_subdirectory

filter_logic

OPTIONAL The logic that should be applied to the provided table in order to tailor the tables to the desired check assessment (ex: if you only want to assess billed diagnoses)

time_between_events

OPTIONAL An integer expressing the maximum number of days that should fall between the two events. Patients with events that occur too far apart will not be included in the combined cohort.

Details

If you choose to use this sample file, please be sure to set your file_subdirectory to system.file("sample_conceptsets", package = "ndq") so the functions know where to access the associated concept sets. You also have the option to download the sample concept sets to a local directory and point to that location.