Unmapped Concepts — check

This function will evaluate the count and proportion of unmapped concepts associated with the fact type of interest. If produce_mapped_list is set to TRUE, a summary of the source values associated with unmapped concepts will also be produced to help identify areas where mappings could potentially be improved. This analysis can be also be executed longitudinally by year.

Usage

check_uc(
  uc_tbl,
  by_year = FALSE,
  produce_mapped_list = TRUE,
  unmapped_values = c(44814650L, 0L, 44814653L, 44814649L),
  check_string = "uc"
)

Arguments

uc_tbl

tabular input || required

The primary input table that contains descriptive information about the checks to be executed by the function. It should include definitions for the fields that should be evaluated for missingness and other relevant metadata. see ?uc_input_omop or ?uc_input_pcornet for examples of the input structure

by_year

boolean || defaults to FALSE

A boolean indicating whether the analysis should be conducted longitudinally by year

Note: the mapped list functionality is NOT available for the longitudinal analysis

produce_mapped_list

boolean || defaults to TRUE

A boolean indicating whether a list of source values associated with unmapped concepts should be produced in addition to the primary output. This table will be iteratively output to the database backend of choice and will NOT be stored locally. Only source values with > 10 occurrences are included.

unmapped_values

string / vector || defaults to 44814650, 0, 44814653, & 44814649

A string or vector listing the concept(s) that indicate an unmapped value. The function will check for NULL values by default, so that does not need to be defined here.

check_string

string || defaults to uc

An abbreviated identifier that will be used to label all output from this module

Value

If by_year is FALSE, this function will produce a dataframe with the total row count, the unmapped row count, the proportion of unmapped values, and some additional descriptive metadata for each user-defined check

If by_year is TRUE, this function will produce a dataframe with the total row count, the unmapped row count, the proportion of unmapped values, and some additional descriptive metadata for each user-defined check stratified by each year present in the fact table

If produce_mapped_list is TRUE, then a table with name uc_grpd that includes the source values (with > 10 appearances) and counts of those values associated with unmapped concepts will be produced in addition to the primary tabular output

Examples

# First create input file with desired checks to be executed
# You can access examples for both OMOP & PCORnet here:
ndq::uc_input_omop
#> # A tibble: 9 × 8
#>   check_id check_description       schema table concept_field source_value_field
#>   <chr>    <chr>                   <chr>  <chr> <chr>         <chr>             
#> 1 dr       all drugs               cdm    drug… drug_concept… drug_source_value 
#> 2 di       inpatient administrati… cdm    drug… drug_concept… drug_source_value 
#> 3 dp       prescription drugs      cdm    drug… drug_concept… drug_source_value 
#> 4 du       drug dose unit          cdm    drug… dose_unit_co… dose_unit_source_…
#> 5 drt      drug route              cdm    drug… route_concep… route_source_value
#> 6 co       conditions              cdm    cond… condition_co… condition_source_…
#> 7 co_scid  condition source        cdm    cond… condition_so… condition_source_…
#> 8 po       ordered procedures      cdm    proc… procedure_co… procedure_source_…
#> 9 pb       billed procedures       cdm    proc… procedure_co… procedure_source_…
#> # ℹ 2 more variables: date_field <chr>, filter_logic <chr>
ndq::uc_input_pcornet
#> # A tibble: 7 × 8
#>   check_id check_description       schema table concept_field source_value_field
#>   <chr>    <chr>                   <chr>  <chr> <chr>         <chr>             
#> 1 di       inpatient administrati… cdm    med_… medadmin_code raw_medadmin_med_…
#> 2 dp       prescription drugs      cdm    pres… rxnorm_cui    raw_rx_med_name   
#> 3 du       rx dose unit            cdm    pres… rx_dose_orde… raw_rx_dose_order…
#> 4 drt      rx route                cdm    pres… rx_route      raw_rx_route      
#> 5 co       diagnoses               cdm    diag… dx            raw_dx            
#> 6 po       ordered procedures      cdm    proc… px            raw_px            
#> 7 pb       billed procedures       cdm    proc… px            raw_px            
#> # ℹ 2 more variables: date_field <chr>, filter_logic <chr>

# Use this as your input to the UC function
## Overall
if (FALSE) { # \dontrun{
my_uc_rslt <- check_uc(uc_tbl = ndq::uc_input_omop,
                       by_year = FALSE,
                       produce_mapped_list = TRUE,
                       unmapped_values = c(44814650L,0L,
                                           44814653L, 44814649L),
                       omop_or_pcornet = 'omop',
                       check_string = 'uc')
} # }

## By Year
if (FALSE) { # \dontrun{
my_uc_rslt <- check_uc(uc_tbl = ndq::uc_input_omop,
                       by_year = TRUE,
                       produce_mapped_list = FALSE,
                       unmapped_values = c(44814650L,0L,
                                           44814653L, 44814649L),
                       omop_or_pcornet = 'omop',
                       check_string = 'uc')
} # }