2 Introduction

In this vignette, we will explore the OmopSketch functions designed to provide information about the number of counts of concepts in tables. Specifically, there are two key functions that facilitate this, summariseConceptIdCounts() and tableConceptIdCounts(). The former one creates a summary statistics results with the number of counts per each concept in the clinical table, and the latter one displays the result in a table.

library(duckdb)
library(OmopSketch)
library(dplyr)


cdm <- mockOmopSketch()

cdm
#> 
#> ── # OMOP CDM reference (duckdb) of mockOmopSketch ─────────────────────────────
#> • omop tables: person, observation_period, cdm_source, concept, vocabulary,
#> concept_relationship, concept_synonym, concept_ancestor, drug_strength,
#> condition_occurrence, death, drug_exposure, measurement, observation,
#> procedure_occurrence, visit_occurrence, device_exposure
#> • cohort tables: -
#> • achilles tables: -
#> • other tables: -

3 Summarise concept id counts

We now use the summariseConceptIdCounts() function from the OmopSketch package to retrieve counts for each concept id and name, as well as for each source concept id and name, across the clinical tables.

summariseConceptIdCounts(cdm, omopTableName = "drug_exposure") |>
  select(group_level, variable_name, variable_level, estimate_name, estimate_value, additional_name, additional_level) |>
  glimpse()
#> Rows: 216
#> Columns: 7
#> $ group_level      <chr> "drug_exposure", "drug_exposure", "drug_exposure", "d…
#> $ variable_name    <chr> "Diclofenac Sodium 75 MG Delayed Release Oral Tablet"…
#> $ variable_level   <chr> "40162359", "40213227", "40213281", "782043", "111951…
#> $ estimate_name    <chr> "count_records", "count_records", "count_records", "c…
#> $ estimate_value   <chr> "100", "100", "100", "100", "100", "100", "100", "100…
#> $ additional_name  <chr> "source_concept_id &&& source_concept_name", "source_…
#> $ additional_level <chr> "0 &&& No matching concept", "0 &&& No matching conce…

By default, the function returns the number of records (estimate_name == "count_records") for each concept_id. To include counts by person, you can set the countBy argument to "person" or to c("record", "person") to obtain both record and person counts.

summariseConceptIdCounts(cdm,
  omopTableName = "drug_exposure",
  countBy = c("record", "person")
) |>
  select( variable_name, estimate_name, estimate_value) 
#> # A tibble: 432 × 3
#>    variable_name                                    estimate_name estimate_value
#>    <chr>                                            <chr>         <chr>         
#>  1 Diclofenac Sodium 75 MG Delayed Release Oral Ta… count_records 100           
#>  2 Diclofenac Sodium 75 MG Delayed Release Oral Ta… count_subjec… 65            
#>  3 tetanus and diphtheria toxoids, adsorbed, prese… count_records 100           
#>  4 tetanus and diphtheria toxoids, adsorbed, prese… count_subjec… 65            
#>  5 diphtheria, tetanus toxoids and acellular pertu… count_records 100           
#>  6 diphtheria, tetanus toxoids and acellular pertu… count_subjec… 66            
#>  7 Isoflurane                                       count_records 100           
#>  8 Isoflurane                                       count_subjec… 65            
#>  9 Dextromethorphan                                 count_records 100           
#> 10 Dextromethorphan                                 count_subjec… 64            
#> # ℹ 422 more rows

Further stratification can be applied using the interval, sex, and ageGroup arguments. The interval argument supports “overall” (no time stratification), “years”, “quarters”, or “months”.

summariseConceptIdCounts(cdm,
  omopTableName = "condition_occurrence",
  countBy = "person",
  interval = "years",
  sex = TRUE,
  ageGroup = list("<=50" = c(0, 50), ">50" = c(51, Inf))
) |>
  select(group_level, strata_level, variable_name, estimate_name, additional_level) |>
  glimpse()
#> Rows: 17,628
#> Columns: 5
#> $ group_level      <chr> "condition_occurrence", "condition_occurrence", "cond…
#> $ strata_level     <chr> "overall", "overall", "overall", "overall", "overall"…
#> $ variable_name    <chr> "Diverticular disease", "Injury of medial collateral …
#> $ estimate_name    <chr> "count_subjects", "count_subjects", "count_subjects",…
#> $ additional_level <chr> "0 &&& No matching concept", "0 &&& No matching conce…

We can also filter the clinical table to a specific time window by setting the dateRange argument.

summarisedResult <- summariseConceptIdCounts(cdm,
                                             omopTableName = "condition_occurrence",
                                             dateRange = as.Date(c("1990-01-01", "2010-01-01"))) 
summarisedResult |>
  omopgenerics::settings()|>
  glimpse()
#> Rows: 1
#> Columns: 10
#> $ result_id          <int> 1
#> $ result_type        <chr> "summarise_concept_id_counts"
#> $ package_name       <chr> "OmopSketch"
#> $ package_version    <chr> "0.5.1"
#> $ group              <chr> "omop_table"
#> $ strata             <chr> ""
#> $ additional         <chr> "source_concept_id &&& source_concept_name"
#> $ min_cell_count     <chr> "0"
#> $ study_period_end   <chr> "2010-01-01"
#> $ study_period_start <chr> "1990-01-01"

Finally, you can summarise concept counts on a subset of records by specifying the sample argument.

summariseConceptIdCounts(cdm,
                         omopTableName = "condition_occurrence",
                         sample = 50) |>
  select(group_level, variable_name, estimate_name) |>
  glimpse()
#> Rows: 36
#> Columns: 3
#> $ group_level   <chr> "condition_occurrence", "condition_occurrence", "conditi…
#> $ variable_name <chr> "Escherichia coli urinary tract infection", "First degre…
#> $ estimate_name <chr> "count_records", "count_records", "count_records", "coun…

result <- summariseConceptIdCounts(cdm,
  omopTableName = "measurement",
  countBy = "record"
) 
tableConceptIdCounts(result, type = "reactable")

tableConceptIdCounts(result, type = "datatable")

tableConceptIdCounts(result, display = "overall")

tableConceptIdCounts(result, display = "standard")

tableConceptIdCounts(result, display = "source")
#> Warning: Values from `estimate_value` are not uniquely identified; output will contain
#> list-cols.
#> • Use `values_fn = list` to suppress this warning.
#> • Use `values_fn = {summary_fun}` to summarise duplicates.
#> • Use the following dplyr code to identify duplicates.
#>   {data} |>
#>   dplyr::summarise(n = dplyr::n(), .by = c(cdm_name, group_level,
#>   source_concept_name, source_concept_id, result_id, group_name, estimate_type,
#>   estimate_name)) |>
#>   dplyr::filter(n > 1L)

tableConceptIdCounts(result, display = "missing source")

tableConceptIdCounts(result, display = "missing standard")
#> Warning: `result` does not contain any `summarise_concept_id_counts` data.

result <- summariseConceptIdCounts(cdm,
  omopTableName = "drug_exposure",
  countBy = "record"
) 
tableTopConceptCounts(result, type = "gt")

tableTopConceptCounts(result, top = 5)

result <- summariseConceptIdCounts(cdm,
  omopTableName = "drug_exposure",
  countBy = c("record", "person")
) 
tableTopConceptCounts(result, countBy = "person")