parseRPDR (version 1.1.2)

convert_enc: Searches columns for given diseases defined by ICD codes.

Description

Analyzes encounter data loaded using load_enc. Converts columns with ICD codes and text to simple ICD codes. If requested, the data.table is returned with new columns corresponding to boolean values, whether given group of diagnoses are present in the given columns. If collapse is given, then the information is aggregated based-on the collapse column and the earliest of latest time of the given diagnosis is provided.

Usage

convert_enc(
  d,
  code = c("enc_diag_admit", "enc_diag_princ", paste0("enc_diag_", 1:10)),
  keep = FALSE,
  codes_to_find = NULL,
  collapse = NULL,
  code_time = "time_enc_admit",
  aggr_type = "earliest",
  nThread = parallel::detectCores() - 1
)

Value

data.table, with formatted ICD code columns and possibly indicator columns if provided. If collapse is present, then only unique ID and the summary columns are returned.

Arguments

d

data.table, database containing encounter information data loaded using the load_enc function.

code

string vector, an array of column names to convert to simple ICD codes. The new column names will be the old one with ICD_ added to the beginning of it.

keep

boolean, whether to keep original columns that were converted. Defaults to FALSE.

codes_to_find

list, a list of arrays corresponding to sets of ICD codes. The function searches the columns in code and new boolean columns with the name of each list element will be created. These columns are indicators whether the given disease is present in the set of ICD codes or not.

collapse

string, a column name on which to collapse the data.table. Used in case we wish to assess whether given diagnoses are present within all the same instances of collapse. See vignette for details.

code_time

string, column name of the time column. Defaults to time_enc_admit. Used in case collapse is present to provide the earliest or latest instance of diagnosing the given disease.

aggr_type

string, if multiple diagnoses are present within the same case of collapse, which timepoint to return. Supported are: "earliest" or "latest". Defaults to earliest.

nThread

integer, number of threads to use for parallelization. If it is set to 1, then no parallel backends are created and the function is executed sequentially.

Examples

if (FALSE) {
#Parse encounter ICD columns and keep original ones as well
data_enc_parse <- convert_enc(d = data_enc, keep = TRUE, nThread = 2)

#Parse encounter ICD columns and discard original ones,
#and create indicator variable for the following diseases
diseases <- list(HT = c("I10"), Stroke = c("434.91", "I63.50"))
data_enc_disease <-  convert_enc(d = data_enc, keep = FALSE,
codes_to_find = diseases, nThread = 2)

#Parse encounter ICD columns and discard original ones
#and create indicator variables for the following diseases and summarize per patient,
#whether there are any encounters where the given diseases were registered
diseases <- list(HT = c("I10"), Stroke = c("434.91", "I63.50"))
data_enc_disease <-  convert_enc(d = data_enc, keep = FALSE,
codes_to_find = diseases, nThread = 2, collapse = "ID_MERGE")
}