Convert ICD data from wide to long format

Convert ICD data from wide to long format

Note the distinction between labelling existing data with any classes which icd provides, and actually converting the structure of the data.

wide_to_long(
  x,
  visit_name = get_visit_name(x),
  icd_labels = NULL,
  icd_name = "icd_code",
  icd_regex = c("icd", "diag", "dx_", "dx")
)

Arguments

x	data.frame in wide format, i.e. one row per patient, and multiple columns containing ICD codes, empty strings or NA.
visit_name	The name of the column in the data frame which contains the patient or visit identifier. Typically this is the visit identifier, since patients come leave and enter hospital with different ICD-9 codes. It is a character vector of length one. If left empty, or NULL, then an attempt is made to guess which field has the ID for the patient encounter (not a patient ID, although this can of course be specified directly). The guesses proceed until a single match is made. Data frames may be wide with many matching fields, so to avoid false positives, anything but a single match is rejected. If there are no successful guesses, and visit_id was not specified, then the first column of the data frame is used.
icd_labels	vector of column names in which codes are found. If NULL, all columns matching the regular expression icd_regex will be included.
icd_name	The name of the column in the data.frame which contains the ICD codes. This is a character vector of length one. If it is NULL, icd9 will attempt to guess the column name, looking for progressively less likely possibilities until it matches a single column. Failing this, it will take the first column in the data frame. Specifying the column using this argument avoids the guesswork.
icd_regex	vector of character strings containing a regular expression to identify ICD-9 diagnosis columns to try (case-insensitive) in order. Default is c("icd", "diag", "dx_", "dx")

Value

data.frame with visit_name column named the same as input, and a column named by icd.name containing all the non-NA and non-empty codes found in the wide input data.

Details

Reshaping data is a common task, and is made easier here by knowing more about the underlying structure of the data. This function wraps the reshape function with specific behavior and checks related to ICD codes. Empty strings and NA values will be dropped, and everything else kept. No validation of the ICD codes is done.

Long and Wide Formats

As is common with many data sets, key variables can be concentrated in one column or spread over several. Tools format of clinical and administrative hospital data, we can perform the conversion efficiently and accurately, while keeping some metadata about the codes intact, e.g. whether they are ICD-9 or ICD-10.

Data structure

Long or wide format ICD data are all expected to be in a data frame. The data.frame itself does not carry any ICD classes at the top level, even if it only contains one type of code; whereas its constituent columns may have a class specified, e.g. icd9 or icd10who.

See also

Other ICD data conversion: comorbid_df_to_mat(), comorbid_mat_to_df(), convert, decimal_to_short(), long_to_wide(), short_to_decimal()

Examples

widedf <- data.frame(
  visit_name = c("a", "b", "c"),
  icd9_01 = c("441", "4424", "441"),
  icd9_02 = c(NA, "443", NA)
)
wide_to_long(widedf)
#>   visit_name icd_code
#> 1          a      441
#> 2          b     4424
#> 3          b      443
#> 4          c      441