tests/testthat/test-trauma_performance.R

testthat::test_that("trauma_performance handles invalid dataframe input", {
  testthat::expect_error(
    trauma_performance(NULL, Ps_col = Ps, outcome_col = death),
    regexp = "df.*must not be NULL"
  )
})

testthat::test_that("trauma_performance checks for binary outcome column", {
  df <- tibble::tibble(Ps = c(0.8, 0.9), outcome_col = c(100, 1))
  testthat::expect_error(
    trauma_performance(df, Ps_col = Ps, outcome_col = outcome_col),
    regexp = "outcome_col.*contains invalid values such as.*100.*Valid values are.*0, 1"
  )
})

testthat::test_that("trauma_performance correctly processes probabilities above 1", {
  df <- tibble::tibble(Ps = c(12, 20), death = c(1, 0))
  expect_error(
    trauma_performance(df, Ps_col = Ps, outcome_col = death),
    regexp = "Ps_col.*values must be contained within range.*0, 1.*Range of this input was.*12, 20"
  )
})

testthat::test_that("trauma_performance validates Ps column is numeric", {
  df <- tibble::tibble(Ps = c("high", "medium"), death = c(1, 0))
  testthat::expect_error(
    trauma_performance(df, Ps_col = Ps, outcome_col = death),
    "Ps_col.*must be.*numeric"
  )
})

testthat::test_that("trauma_performance checks if Ps values are between 0 and 100", {
  df <- tibble::tibble(Ps = c(101, -1), death = c(1, 0))
  testthat::expect_error(
    trauma_performance(df, Ps_col = Ps, outcome_col = death),
    regexp = "Ps_col.*values must be contained within range.*0, 1.*Range of this input was.*-1, 101"
  )
})

testthat::test_that("trauma_performance calculates W-score", {
  df <- tibble::tibble(Ps = c(0.8, 0.9), death = c(1, 0))
  result <- trauma_performance(df, Ps_col = Ps, outcome_col = death)
  testthat::expect_true("W_Score" %in% names(result))
})

testthat::test_that("trauma_performance calculates M-score correctly", {
  df <- tibble::tibble(Ps = c(0.8, 0.9), death = c(1, 0))
  result <- trauma_performance(df, Ps_col = Ps, outcome_col = death)
  testthat::expect_true("M_Score" %in% names(result))
})

testthat::test_that("trauma_performance calculates Z-score with survival method", {
  df <- tibble::tibble(Ps = c(0.8, 0.9), death = c(1, 0))
  result <- trauma_performance(
    df,
    Ps_col = Ps,
    outcome_col = death,
    z_method = "survival"
  )
  testthat::expect_true("Z_Score" %in% names(result))
})

testthat::test_that("trauma_performance calculates Z-score with mortality method", {
  df <- tibble::tibble(Ps = c(0.8, 0.9), death = c(1, 0))
  result <- trauma_performance(
    df,
    Ps_col = Ps,
    outcome_col = death,
    z_method = "mortality"
  )
  testthat::expect_true("Z_Score" %in% names(result))
})

testthat::test_that("trauma_performance works when outcome_col is TRUE/FALSE", {
  df <- tibble::tibble(Ps = c(0.8, 0.9), death = c(TRUE, FALSE))
  result <- trauma_performance(df, Ps_col = Ps, outcome_col = death)
  testthat::expect_true("W_Score" %in% names(result))
})

testthat::test_that("trauma_performance handles edge case of empty dataframe", {
  df <- tibble::tibble(Ps = numeric(0), death = logical(0))

  result <- trauma_performance(df, Ps_col = Ps, outcome_col = death)

  # Additionally, check if the result is a tibble with the correct structure (optional)
  testthat::expect_s3_class(result, "tbl_df") # Check that the result is a tibble
  testthat::expect_equal(nrow(result), 1) # Verify that 9 rows are in the result
  testthat::expect_equal(ncol(result), 9) # Verify that there are 2 columns
})

# Test missing Ps_col and outcome_col columns and logical binary data
testthat::test_that("trauma_performance correctly validates missing Ps_col and outcome_col columns and logical binary data", {
  set.seed(123)
  n_patients <- 100
  Ps <- c(0.85, NA, 0.75, 0.6, 0.91)
  survival_outcomes <- as.character(c(1, 0, 0, 1, 0))
  data <- data.frame(Ps = Ps, survival = survival_outcomes)

  testthat::expect_error(
    trauma_performance(data),
    regexp = "Both.*arguments must be provided"
  )

  testthat::expect_error(
    trauma_performance(data, Ps_col = Ps),
    regexp = "argument must be provided"
  )

  testthat::expect_error(
    trauma_performance(data, outcome_col = survival),
    regexp = "argument must be provided"
  )

  survival_outcomes <- c(1L, 2L, 2L, 1L, 2L)
  data <- data.frame(Ps = Ps, survival = survival_outcomes)

  testthat::expect_error(
    trauma_performance(data, Ps_col = Ps, outcome_col = survival),
    regexp = "outcome_col.*contains invalid values such as.*Valid values are"
  )

  Ps <- c(0.85, 1, 0.75, 0.6, 0.91)
  survival_outcomes <- c(T, F, T, F, T)
  data <- data.frame(Ps = Ps, survival = survival_outcomes)

  testthat::expect_s3_class(
    trauma_performance(data, Ps_col = Ps, outcome_col = survival),
    class = "tbl_df"
  )

  Ps <- c(0.85, NA, 0.75, 0.6, 0.91)
  survival_outcomes <- as.character(c(1, 0, 0, 1, 0))
  data <- data.frame(Ps = Ps, survival = survival_outcomes)

  testthat::expect_error(
    trauma_performance(data, Ps_col = Ps, outcome_col = survival),
    regexp = "outcome_col.*must be of class.*numeric, logical, integer"
  )
})

Try the traumar package in your browser

Any scripts or data that you put into this service are public.

traumar documentation built on Feb. 6, 2026, 1:06 a.m.