Task 1: MIMIC-IV Hospitality Mortality after ICU Admission

Inspecting the Data from MIMIC-IV dataset

We begin by loading and inspecting the required data from MIMIC-IV:

# load the mortality data
patient_ids <- id_col(load_concepts("adm_episode", "miiv", 
                                    verbose = FALSE)[adm_episode == 1])
dat <- load_concepts(c("acu_24", "diag", "age", "sex", "charlson",
                       "lact_24", "pafi_24", "ast_24",
                       "race", "death"), "miiv", patient_ids = patient_ids,
                     verbose = FALSE)
dat <- dat[race %in% c("Caucasian", "African American")]
dat[, c(index_var(dat)) := NULL]
imp_lst <- list(
  age = 65,
  acu_24 = 0,
  charlson = 0,
  lact_24 = 1,
  ast_24 = 20,
  pafi_24 = 500,
  death = FALSE
)

for (i in seq_len(ncol(dat))) {

  var <- names(dat)[i]
  if (any(is.na(dat[[var]])) & !is.null(imp_lst[[var]]))
    dat[is.na(get(var)), c(var) := imp_lst[[var]]]
}

knitr::kable(head(dat), caption = "MIMIC-IV Mortality data.")

MIMIC-IV Mortality data.

stay_id	acu_24	diag	age	sex	charlson	lact_24	pafi_24	ast_24	race	death
30000153	3	TRAUM	61	Male	0	1.7	430.0000	20	Caucasian	FALSE
30001396	3	MED	40	Male	1	1.8	361.9048	23	Caucasian	FALSE
30001446	11	MED	56	Male	1	1.7	466.6667	114	Caucasian	FALSE
30001656	3	NMED	68	Female	0	1.0	500.0000	20	Caucasian	FALSE
30001947	2	SURG	43	Male	1	0.7	535.0000	36	Caucasian	FALSE
30002415	4	CSURG	72	Female	0	1.0	240.0000	20	Caucasian	FALSE

We consider the cohort of all patients in the database admitted to the ICU (and we only consider the first admission of each patient). We also load information on the SOFA score, admission diagnosis, age, sex, Charlson comorbidity index, worst values of lactate, PaO2/FiO2, and AST in first 24 hours. We further load the race information (protected attribute \(X\)) and the death indicator (outcome \(Y\)). We want to investigate the disparities in outcome between "Caucasian" and "African American" groups.

Constructing the SFM

We next construct the Standard Fairness Model, with also a decision \(D\):

# constructing the SFM
X <- "race"
Z <- c("age", "sex")
W <- c("acu_24", "diag", "charlson", "lact_24", "pafi_24", "ast_24")
Y <- "death"

Decomposing the Disparity

fcb <- fairness_cookbook(
  data = dat, X = X, Z = Z, W = W, Y = Y,
  x1 = "Caucasian", x0 = "African American"
)

2.10843373493976% of extreme P(x | z) or P(x | z, w) probabilities.
Estimates likely biased.

We can then inspect the decomposition by calling autoplot() on the fcb object which is an S3 class of type faircause:

autoplot(fcb) +
  labs(title="Causal decomposition of mortality difference: MIMIC-IV race effect",
       y="Mortality difference (%)") +
  theme_minimal() +
  scale_x_discrete(labels = c("Total Variation", "Direct", "Indirect",
                                "Confounded"), name = "Pathway") +
  scale_fill_discrete(labels = c("Total Variation", "Direct", "Indirect",
                                 "Confounded"), name = "Pathway")

Zooming-in on the Spurious Effect

To better understand the spurious effect, we plot the density of the age distributions between groups: