Harmonize clinical codes across sites

Usage

mapClinicalCodes(
  x,
  mapping,
  code_col = "scale_name",
  mapped_col = "scale_name_std",
  from_col = "from",
  to_col = "to",
  unmatched = c("keep", "na", "drop")
)

Arguments

x

Clinical metadata data frame.

mapping

Code mapping, either:

named character vector (names = source code, values = target code), or
data frame with columns defined by from_col and to_col.

code_col

Source code column in x.

mapped_col

Output mapped code column.

from_col

Source column in mapping data.frame.

to_col

Target column in mapping data.frame.

unmatched

Behavior for unmatched codes: "keep", "na", or "drop".

Value

Data frame with mapped_col appended.

References

Goldberger AL, et al. (2000). "PhysioBank, PhysioToolkit, and PhysioNet: components of a new research resource for complex physiologic signals." Circulation, 101(23), e215-e220. doi:10.1161/01.CIR.101.23.e215

Examples

df <- data.frame(scale_name = c("fim_total", "Berg"), stringsAsFactors = FALSE)
map <- c(fim_total = "FIM", Berg = "BBS")
mapClinicalCodes(df, map, code_col = "scale_name")
#>   scale_name scale_name_std
#> 1  fim_total            FIM
#> 2       Berg            BBS