Skip to contents

Paired Assay Agreement Table (Bias, Limits of Agreement, Lin's CCC)

Source: R/tbl_paired_assay_agreement.R
tbl_paired_assay_agreement.Rd

For each row of `pairs_df`, compares two measurement columns of the same specimens: the mean difference (bias) `method_1 - method_2`, the SD of the differences, Bland-Altman 95 percent limits of agreement (`bias +/- 1.96 SD`), Lin's concordance correlation coefficient and the percentage of rows missing each method.

Usage

tbl_paired_assay_agreement(
  data,
  pairs_df,
  id_col = NULL,
  time_col = NULL,
  metrics = c("bias", "sd_diff", "loa_lower", "loa_upper", "ccc", "pct_miss_1",
    "pct_miss_2"),
  digits = 2
)

Arguments

data

A data frame.

pairs_df

Data frame with columns `parameter`, `method_1` and `method_2` (names of numeric columns of `data`).

id_col

Optional subject column (adds `n_subjects`).

time_col

Optional time column; rows with a missing time are excluded, as they cannot be matched to a time point.

metrics

Statistics to return, any of `"bias"`, `"sd_diff"`, `"loa_lower"`, `"loa_upper"`, `"ccc"`, `"pct_miss_1"`, `"pct_miss_2"`. `parameter`, `method_1`, `method_2` and `n_pairs` are always returned.

digits

Decimals for rounding; `NULL` for none.

Value

A tibble with one row per pair. Statistics that cannot be computed (fewer than two pairs, constant data for the CCC) are `NA`.

Details

Only rows where both methods are present are paired. Repeated measurements of the same subject are treated as independent, so the limits are naive for longitudinal data; `id_col` only adds the number of subjects.

See also

[tbl_two_sample_compare()]

Examples

d <- data.frame(a = c(1, 2, 3, 4, 5), b = c(1.1, 2.3, 2.9, 4.2, 5.4))
tbl_paired_assay_agreement(d, data.frame(parameter = "x", method_1 = "a", method_2 = "b"))
#> # A tibble: 1 × 11
#>   parameter method_1 method_2 n_pairs  bias sd_diff loa_lower loa_upper   ccc
#>   <chr>     <chr>    <chr>      <int> <dbl>   <dbl>     <dbl>     <dbl> <dbl>
#> 1 x         a        b              5 -0.18    0.19     -0.56       0.2  0.99
#> # ℹ 2 more variables: pct_miss_1 <dbl>, pct_miss_2 <dbl>