Package 'qcdd'

Build a concept lookup from dictionary columns

Description

Combine paired concept ID and definition columns into a two-column tibble. Column pairs are matched by position and must have equal lengths.

Usage

add_concept_ids(df, concept_ids, definitions)

Arguments

df	A data frame or other object convertible to a tibble containing the dictionary columns.
concept_ids	Bare concept ID column names, supplied as a single name or combined with c().
definitions	Bare definition column names, supplied as a single name or combined with c(), in the same order as concept_ids.

Details

Identical ID-definition pairs and rows with missing concept IDs are removed after combining all column pairs. If an ID has conflicting definitions, a warning is issued and the first occurrence is retained, using column-pair order followed by input row order. Missing definitions are retained and may be selected as the first occurrence. When conflicts are found, the result is ordered by the concept ID groups.

Value

A tibble with concept_id and definitions columns and one row per non-missing concept ID. Conflicting definitions are stored in the CID_warning attribute and can be retrieved with cid_warning().

---

Retrieve conflicting concept definitions

Description

Retrieve the conflict report stored by add_concept_ids() without issuing another warning or modifying the lookup.

Usage

cid_warning(df)

Arguments

df	A concept lookup returned by add_concept_ids().

Value

A tibble with concept_id and definitions columns containing the distinct conflicting ID-definition pairs, including the definitions retained in the lookup. Returns NULL if the CID_warning attribute is absent, as when no conflicts were found.

---

Load the dictionary

Description

Load the dictionary

Usage

load_dictionary_file(path, ...)

Arguments

path	Path to the dictionary
...	additional parameters passed to rio::import and its underlying functions. For example the sheet name for an excel file

Value

a tibble containing the dictionary

---

Build a concept dictionary from a file

Description

Import a dictionary file with load_dictionary_file() and combine selected concept ID and definition columns with add_concept_ids().

Usage

make_dictionary(path, concept_ids, definitions, ...)

Arguments

path	Path to the dictionary file.
concept_ids	Bare concept ID column names, supplied as a single name or combined with c().
definitions	Bare definition column names, supplied as a single name or combined with c(), in the same order as concept_ids.
...	Additional arguments passed to load_dictionary_file() and ultimately to rio::import(), such as sheet and col_types for Excel.

Details

Column pairs must have equal lengths. Identical ID-definition pairs and rows with missing concept IDs are removed. Leading numeric value labels such as ⁠1 = ⁠ are stripped from definitions. Conflicting definitions issue a warning, and the first occurrence is retained, using column-pair order followed by input row order. Missing definitions are retained. See add_concept_ids() for details.

Value

A tibble with concept_id and definitions columns and one row per non-missing concept ID. When conflicts occur, the conflict report is stored in the CID_warning attribute and can be retrieved with cid_warning().

Examples

path <- tempfile(fileext = ".csv")
utils::write.csv(
  data.frame(
    concept_id_1 = c("72841936", "39508271"),
    definition_1 = c("Age", "Height"),
    concept_id_2 = c("86173409", "54296813"),
    definition_2 = c("Weight", "Blood pressure")
  ),
  path, row.names = FALSE
)
dictionary <- make_dictionary(
  path,
  concept_ids = c(concept_id_1, concept_id_2),
  definitions = c(definition_1, definition_2),
  colClasses = "character"
)
dictionary
unlink(path)

---