Importing and exporting codelists

In CodelistGenerator: Identify Relevant Clinical Codes and Evaluate Their Use

NOT_CRAN <- identical(tolower(Sys.getenv("NOT_CRAN")), "true")

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  out.width = "100%",
  eval = NOT_CRAN
)

Importing and exporting codelists

We will typically want to save the codelists we create for re-use. To show how we can do this, let's first create an empty folder to store our codelists in.

library(CodelistGenerator)

dir_codes <- file.path(tempdir(), "codelists")
dir.create(dir_codes)
list.files(dir_codes)

Now let's create a couple of codelists that we will save.

codelist <- list("codes1" = c(1L, 2L, 3L),
                 "codes2" = c(4L, 5L, 10L))
codelist <- newCodelist(codelist)

codelist 

We can use exportCodelist() to save these as two CSVs, one for each codelist.

exportCodelist(codelist, dir_codes, type = "csv")
list.files(dir_codes)

To import codelists, we have importCodelist(). Here we can see that we can easily import our codelists back into R.

importCodelist(dir_codes, type = "csv")

Importing concept sets

As we've seen in the previous vignettes, codelists can also be represented as concept set which is resolved against the OMOP CDM vocabulary. To import these we can use importConceptSetExpression().

Take this example concept set expression.

library(jsonlite)
concept_set_path <- system.file("concepts_for_mock/arthritis_with_excluded.json", 
                                package = "CodelistGenerator")
fromJSON(concept_set_path) |> toJSON(pretty = TRUE, auto_unbox = TRUE)

We can bring this into R as a concept set expression.

cse <- importConceptSetExpression(concept_set_path)
cse

And we can then resolve it to a concept set. Note, for this we will need to specify a cdm reference as the result will be tied to a given OMOP CDM vocabulary version which will specify the relevant descendants.

cdm <- mockVocabRef()
importConceptSetExpression(concept_set_path) |> 
  asCodelist(cdm)

CodelistGenerator documentation built on Dec. 17, 2025, 5:06 p.m.