Skip to contents

Compute a Composite Clinical or Biomarker Index

Source: R/composite_index.R
compute_composite_index.Rd

Combines multiple numeric variables into a single composite score using standardized z-scores, first principal component (PCA), or weighted sums.

Usage

compute_composite_index(
  data,
  cols,
  method = c("zscore", "pca", "sum"),
  weights = NULL,
  index_name = "composite_index"
)

Arguments

data

A data frame or tibble.

cols

Character vector of numeric column names to include in the index.

method

Index construction method: `"zscore"` (default, mean of z-scores), `"pca"` (first principal component score, scaled so higher values indicate greater severity), or `"sum"` (weighted sum).

weights

Optional numeric vector of weights matching `cols`. If `NULL`, equal weights are used.

index_name

Character string naming the output index column (default `"composite_index"`).

Value

The input data frame augmented with the new composite index column.

See also

[index_variable_table()], [center()]

Examples

df <- data.frame(sbp = c(120, 140, 160), dbp = c(80, 90, 100), bmi = c(22, 28, 35))
compute_composite_index(df, cols = c("sbp", "dbp", "bmi"), method = "zscore")
#>   sbp dbp bmi composite_index
#> 1 120  80  22     -0.99113316
#> 2 140  90  28     -0.01707718
#> 3 160 100  35      1.00821035