My Research Work
library(readxl)
df_final <- read_excel("C:/Users/pc/Documents/my_thesis/research_data/r_EDA/final_EDA.xlsx")df <- read_excel("C:/Users/pc/Documents/my_thesis/research_data/r_EDA/final_LOS_EDA.xlsx")# Keep the first occurrence of each MRN
df_final <- df_final[!duplicated(df_final$MRN), ]
# Display duplicate removal results
cat("Remaining records:", nrow(df_final), "\n")## Remaining records: 5515cat("Removed records:", 6309 - nrow(df_final), "\n")## Removed records: 794cat("Remaining duplicated MRNs:", sum(duplicated(df_final$MRN)), "\n")## Remaining duplicated MRNs: 0clinical_limits <- list(
hr = c(30, 250),
sbp = c(30, 300),
dbp = c(20, 180),
rr = c(5, 80),
temp = c(30, 45),
spo2 = c(0, 101),
rbs = c(20, 700),
Pain_score = c(0, 10),
sodium = c(100, 180),
potassium = c(1.5, 10),
wbc = c(0.1, 100),
rbc = c(0.5, 10),
basophil = c(0, 20),
eosinophil = c(0, 40),
monocyte = c(0, 30),
neutrophil = c(0, 100),
lymphocyte = c(0, 100),
hemoglobin = c(2, 25),
hematocrit = c(5, 80),
mcv = c(50, 150),
mch = c(10, 50),
mchc = c(20, 45),
mpv = c(5, 20),
plt = c(5, 2000),
ast_sgot = c(0, 5000),
alt_sgpt = c(0, 5000),
creatinine = c(0.1, 20),
oxy_use = c(0, 100),
Age_years = c(13, 120)
)df_original <- df_finalcat("Replacing values outside clinical limits...\n")## Replacing values outside clinical limits...for (col in names(clinical_limits)) {
low <- clinical_limits[[col]][1]
high <- clinical_limits[[col]][2]
outliers <- !is.na(df_final[[col]]) &
(df_final[[col]] < low | df_final[[col]] > high)
n_outliers <- sum(outliers)
df_final[[col]][outliers] <- NA
cat(col, ":", n_outliers, "values replaced with NA\n")
}## hr : 3 values replaced with NA
## sbp : 23 values replaced with NA
## dbp : 23 values replaced with NA
## rr : 2 values replaced with NA
## temp : 73 values replaced with NA
## spo2 : 183 values replaced with NA
## rbs : 5 values replaced with NA
## Pain_score : 0 values replaced with NA
## sodium : 33 values replaced with NA
## potassium : 14 values replaced with NA
## wbc : 2 values replaced with NA
## rbc : 15 values replaced with NA
## basophil : 0 values replaced with NA
## eosinophil : 1 values replaced with NA
## monocyte : 15 values replaced with NA
## neutrophil : 7 values replaced with NA
## lymphocyte : 1 values replaced with NA
## hemoglobin : 5 values replaced with NA
## hematocrit : 7 values replaced with NA
## mcv : 21 values replaced with NA
## mch : 29 values replaced with NA
## mchc : 15 values replaced with NA
## mpv : 82 values replaced with NA
## plt : 4 values replaced with NA
## ast_sgot : 6 values replaced with NA
## alt_sgpt : 0 values replaced with NA
## creatinine : 11 values replaced with NA
## oxy_use : 0 values replaced with NA
## Age_years : 0 values replaced with NAlibrary(dplyr)
library(DT)## Warning: package 'DT' was built under R version 4.4.3numeric_cols <- c(
"hr", "sbp", "dbp", "rr", "temp", "spo2",
"rbs", "Pain_score", "sodium", "potassium",
"wbc", "rbc", "basophil", "eosinophil",
"monocyte", "neutrophil", "lymphocyte",
"hemoglobin", "hematocrit", "mcv", "mch",
"mchc", "mpv", "plt", "ast_sgot", "alt_sgpt",
"creatinine", "oxy_use", "Age_years"
)calculate_skewness <- function(x) {
x <- x[!is.na(x)]
n <- length(x)
if (n < 3) return(NA)
m <- mean(x)
s <- sd(x)
if (s == 0) return(0)
skew <- (n / ((n - 1) * (n - 2))) *
sum(((x - m) / s)^3)
return(skew)
}summary <- data.frame(
Variable = numeric_cols,
Count = sapply(
df_final[numeric_cols],
function(x) sum(!is.na(x))
),
Missing = sapply(
df_final[numeric_cols],
function(x) sum(is.na(x))
),
`Missing (%)` = sapply(
df_final[numeric_cols],
function(x) mean(is.na(x)) * 100
),
Mean = sapply(
df_final[numeric_cols],
function(x) mean(x, na.rm = TRUE)
),
Std = sapply(
df_final[numeric_cols],
function(x) sd(x, na.rm = TRUE)
),
Min = sapply(
df_final[numeric_cols],
function(x) min(x, na.rm = TRUE)
),
`25%` = sapply(
df_final[numeric_cols],
function(x) quantile(x, 0.25, na.rm = TRUE)
),
Median = sapply(
df_final[numeric_cols],
function(x) median(x, na.rm = TRUE)
),
`75%` = sapply(
df_final[numeric_cols],
function(x) quantile(x, 0.75, na.rm = TRUE)
),
Max = sapply(
df_final[numeric_cols],
function(x) max(x, na.rm = TRUE)
),
Skewness = sapply(
df_final[numeric_cols],
calculate_skewness
)
)
summary[-1] <- round(summary[-1], 2)
summary# ============================================================
# AGE CATEGORIZATION
# ============================================================
df <- df %>%
mutate(
Age_group = case_when(
Age_years <= 24 ~ "≤24",
Age_years >= 25 & Age_years <= 29 ~ "25–29",
Age_years >= 30 & Age_years <= 40 ~ "30–40",
Age_years >= 41 & Age_years <= 50 ~ "41–50",
Age_years > 51 ~ ">51",
TRUE ~ NA_character_
)
)
df_final <- df_final %>%
mutate(
Age_group = case_when(
Age_years <= 24 ~ "≤24",
Age_years >= 25 & Age_years <= 29 ~ "25–29",
Age_years >= 30 & Age_years <= 40 ~ "30–40",
Age_years >= 41 & Age_years <= 50 ~ "41–50",
Age_years > 51 ~ ">51",
TRUE ~ NA_character_
)
)
# ============================================================
# SET ORDER OF AGE CATEGORIES
# ============================================================
df$Age_group <- factor(
df$Age_group,
levels = c("≤24", "25–29", "30–40", "41–50", ">51")
)
df_final$Age_group <- factor(
df_final$Age_group,
levels = c("≤24", "25–29", "30–40", "41–50", ">51")
)
table(df$Age_group, useNA = "ifany")##
## ≤24 25–29 30–40 41–50 >51 <NA>
## 198 199 405 427 2198 55table(df_final$Age_group, useNA = "ifany")##
## ≤24 25–29 30–40 41–50 >51 <NA>
## 286 247 556 680 3663 83datatable(
summary,
options = list(
pageLength = 10,
scrollX = TRUE,
searching = TRUE,
ordering = TRUE
),
rownames = FALSE
)selected_vars <- c(
"Age_years",
"hr",
"sbp",
"rr",
"temp",
"spo2",
"rbs",
"hemoglobin",
"wbc"
)library(tidyr)
library(ggplot2)## Warning: package 'ggplot2' was built under R version 4.4.3hist_data <- df_final %>%
select(all_of(selected_vars)) %>%
pivot_longer(
cols = everything(),
names_to = "Variable",
values_to = "Value"
)hist_data$Variable <- factor(
hist_data$Variable,
levels = selected_vars,
labels = c(
"Age (years)",
"Heart rate (bpm)",
"Systolic BP (mmHg)",
"Respiratory rate (breaths/min)",
"Temperature (\u00B0C)",
"SpO\u2082 (%)",
"Random blood sugar",
"Hemoglobin (g/dL)",
"White blood cell count"
)
)p <- ggplot(
hist_data,
aes(x = Value)
) +
geom_histogram(
aes(y = after_stat(density)),
bins = 30,
fill = "steelblue",
color = "white",
alpha = 0.85
) +
geom_density(
linewidth = 0.9,
color = "black",
na.rm = TRUE
) +
facet_wrap(
~ Variable,
scales = "free",
ncol = 2
) +
labs(
x = NULL,
y = "Density"
) +
theme_classic(base_size = 11) +
theme(
strip.background = element_rect(
fill = "grey95",
color = "grey50",
linewidth = 0.5
),
strip.text = element_text(
face = "bold",
size = 10
),
axis.text = element_text(
color = "black"
),
axis.title.y = element_text(
face = "bold"
),
panel.spacing = unit(1.1, "lines"),
plot.margin = margin(10, 10, 10, 10)
)
p## Warning: Removed 4377 rows containing non-finite outside the scale range
## (`stat_bin()`).
corrilation analysisnumeric_var <- c(
"hr",
"sbp",
"dbp",
"rr",
"temp",
"spo2",
"rbs",
"Pain_score",
"sodium",
"potassium",
"wbc",
"rbc",
"basophil",
"eosinophil",
"monocyte",
"neutrophil",
"lymphocyte",
"hemoglobin",
"hematocrit",
"mcv",
"mch",
"mchc",
"mpv",
"plt",
"ast_sgot",
"alt_sgpt",
"creatinine",
"Age_years"
)
numeric_var <- numeric_var[
numeric_var %in% names(df_final)
]
corr <- cor(
df_final[, numeric_var, drop = FALSE],
use = "pairwise.complete.obs",
method = "pearson"
)
corr_check <- corr
diag(corr_check) <- NA
keep <- apply(
abs(corr_check) >= 0.40,
1,
function(x) any(x, na.rm = TRUE)
)
corr_selected <- corr[
keep,
keep,
drop = FALSE
]
if (nrow(corr_selected) >= 2) {
library(corrplot)
corrplot(
corr_selected,
method = "color",
type = "upper",
order = "hclust",
diag = FALSE,
addCoef.col = "black",
number.cex = 0.65,
tl.col = "black",
tl.cex = 0.95,
tl.srt = 90,
col = colorRampPalette(
c("blue", "white", "red")
)(200)
)
} else {
message(
"Fewer than two variables met the |r| >= 0.40 criterion."
)
}## Warning: package 'corrplot' was built under R version 4.4.3## corrplot 0.95 loaded
# 1. Gender
gender_map <- c(
"0" = "Female",
"1" = "Male"
)
df$Gender <- unname(
gender_map[as.character(df$Gender)]
)
# 2. Admission Source
admission_map <- c(
"1" = "Emergency",
"2" = "Medical Ward",
"3" = "Surgical Ward",
"4" = "Pediatrics Ward",
"5" = "Maternity Ward",
"6" = "Orthopedic Ward",
"7" = "HDU",
"8" = "Other"
)
df$Admission_source <- unname(
admission_map[as.character(df$Admission_source)]
)
# 3. Diagnosis Superclass
diag_map <- c(
"1" = "Cardiovascular Disease",
"2" = "Respiratory Disease",
"3" = "Neurological Disease",
"4" = "Sepsis / Infection",
"5" = "Gastrointestinal / Hepatic Disease",
"6" = "Trauma / Toxicology",
"7" = "Endocrine / Metabolic Disease",
"8" = "Renal Disease",
"9" = "Thromboembolism",
"10" = "Malignancy",
"11" = "Multi-organ Failure",
"12" = "Obstetric / Gynecology",
"13" = "Autoimmune / Systemic Disease",
"14" = "Other"
)
df$diag_superclass <- unname(
diag_map[as.character(df$diag_superclass)]
)Q1 <- quantile(df$LOS_days, 0.25, na.rm = TRUE)
Q3 <- quantile(df$LOS_days, 0.75, na.rm = TRUE)
IQR_value <- Q3 - Q1
upper_bound <- Q3 + 1.5 * IQR_value
df <- df[
!is.na(df$LOS_days) &
df$LOS_days >= 1 &
df$LOS_days <= upper_bound,
]df$LOS_binary <- factor(
ifelse(
df$LOS_days <= 7,
"<=7_days",
">7_days"
),
levels = c("<=7_days", ">7_days")
)
table(df$LOS_binary, useNA = "ifany")##
## <=7_days >7_days
## 1383 1870# ============================================================
# LOS DESCRIPTIVE TABLE
# Gender + Admission source + Diagnosis
# ============================================================
library(dplyr)
library(tidyr)
library(gt)
# ============================================================
# 1. Restore / convert Gender safely
# ============================================================
df$Gender <- case_when(
as.character(df$Gender) == "0" ~ "Female",
as.character(df$Gender) == "1" ~ "Male",
as.character(df$Gender) == "Female" ~ "Female",
as.character(df$Gender) == "Male" ~ "Male",
TRUE ~ NA_character_
)
# ============================================================
# 2. Restore / convert Admission source safely
# ============================================================
df$Admission_source <- case_when(
as.character(df$Admission_source) == "1" ~ "Emergency",
as.character(df$Admission_source) == "2" ~ "Medical Ward",
as.character(df$Admission_source) == "3" ~ "Surgical Ward",
as.character(df$Admission_source) == "4" ~ "Pediatrics Ward",
as.character(df$Admission_source) == "5" ~ "Maternity Ward",
as.character(df$Admission_source) == "6" ~ "Orthopedic Ward",
as.character(df$Admission_source) == "7" ~ "HDU",
as.character(df$Admission_source) == "8" ~ "Other",
as.character(df$Admission_source) == "Emergency" ~ "Emergency",
as.character(df$Admission_source) == "Medical Ward" ~ "Medical Ward",
as.character(df$Admission_source) == "Surgical Ward" ~ "Surgical Ward",
as.character(df$Admission_source) == "Pediatrics Ward" ~ "Pediatrics Ward",
as.character(df$Admission_source) == "Maternity Ward" ~ "Maternity Ward",
as.character(df$Admission_source) == "Orthopedic Ward" ~ "Orthopedic Ward",
as.character(df$Admission_source) == "HDU" ~ "HDU",
as.character(df$Admission_source) == "Other" ~ "Other",
as.character(df$Admission_source) == "Other Specialty Wards" ~ "Other",
TRUE ~ NA_character_
)
# ============================================================
# 3. Restore / convert Diagnosis safely
# ============================================================
df$diag_superclass <- case_when(
as.character(df$diag_superclass) == "1" ~ "Cardiovascular Disease",
as.character(df$diag_superclass) == "2" ~ "Respiratory Disease",
as.character(df$diag_superclass) == "3" ~ "Neurological Disease",
as.character(df$diag_superclass) == "4" ~ "Sepsis / Infection",
as.character(df$diag_superclass) == "5" ~ "Gastrointestinal / Hepatic Disease",
as.character(df$diag_superclass) == "6" ~ "Trauma / Toxicology",
as.character(df$diag_superclass) == "7" ~ "Endocrine / Metabolic Disease",
as.character(df$diag_superclass) == "8" ~ "Renal Disease",
as.character(df$diag_superclass) == "9" ~ "Thromboembolism",
as.character(df$diag_superclass) == "10" ~ "Malignancy",
as.character(df$diag_superclass) == "11" ~ "Multi-organ Failure",
as.character(df$diag_superclass) == "12" ~ "Obstetric / Gynecology",
as.character(df$diag_superclass) == "13" ~ "Autoimmune / Systemic Disease",
as.character(df$diag_superclass) == "14" ~ "Other",
as.character(df$diag_superclass) == "Cardiovascular Disease" ~ "Cardiovascular Disease",
as.character(df$diag_superclass) == "Respiratory Disease" ~ "Respiratory Disease",
as.character(df$diag_superclass) == "Neurological Disease" ~ "Neurological Disease",
as.character(df$diag_superclass) == "Sepsis / Infection" ~ "Sepsis / Infection",
as.character(df$diag_superclass) == "Gastrointestinal / Hepatic Disease" ~ "Gastrointestinal / Hepatic Disease",
as.character(df$diag_superclass) == "Trauma / Toxicology" ~ "Trauma / Toxicology",
as.character(df$diag_superclass) == "Endocrine / Metabolic Disease" ~ "Endocrine / Metabolic Disease",
as.character(df$diag_superclass) == "Renal Disease" ~ "Renal Disease",
as.character(df$diag_superclass) == "Thromboembolism" ~ "Thromboembolism",
as.character(df$diag_superclass) == "Malignancy" ~ "Malignancy",
as.character(df$diag_superclass) == "Multi-organ Failure" ~ "Multi-organ Failure",
as.character(df$diag_superclass) == "Obstetric / Gynecology" ~ "Obstetric / Gynecology",
as.character(df$diag_superclass) == "Autoimmune / Systemic Disease" ~ "Autoimmune / Systemic Disease",
as.character(df$diag_superclass) == "Other" ~ "Other",
TRUE ~ NA_character_
)
# ============================================================
# 4. Make sure LOS_binary exists correctly
# ============================================================
df$LOS_binary <- factor(
ifelse(
is.na(df$LOS_days),
NA_character_,
ifelse(
df$LOS_days <= 7,
"<=7_days",
">7_days"
)
),
levels = c("<=7_days", ">7_days")
)
# ============================================================
# 5. Create clean analysis datasets
# ============================================================
gender_data <- df %>%
filter(
!is.na(Gender),
!is.na(LOS_binary)
)
admission_data <- df %>%
filter(
!is.na(Admission_source),
!is.na(LOS_binary)
)
diagnosis_data <- df %>%
filter(
!is.na(diag_superclass),
!is.na(LOS_binary)
)
# ============================================================
# 6. Safe p-value function
# ============================================================
get_pvalue <- function(x, y) {
keep <- !is.na(x) & !is.na(y)
x <- x[keep]
y <- y[keep]
tab <- table(x, y)
if (nrow(tab) < 2 || ncol(tab) < 2) {
return(NA_real_)
}
chi <- suppressWarnings(
chisq.test(tab)
)
if (any(chi$expected < 5)) {
suppressWarnings(
chisq.test(
tab,
simulate.p.value = TRUE,
B = 10000
)$p.value
)
} else {
chi$p.value
}
}
# ============================================================
# 7. P-values
# ============================================================
gender_p <- get_pvalue(
gender_data$Gender,
gender_data$LOS_binary
)
admission_p <- get_pvalue(
admission_data$Admission_source,
admission_data$LOS_binary
)
diagnosis_p <- get_pvalue(
diagnosis_data$diag_superclass,
diagnosis_data$LOS_binary
)
# ============================================================
# 8. Total LOS groups
# ============================================================
short_total <- sum(
df$LOS_binary == "<=7_days",
na.rm = TRUE
)
long_total <- sum(
df$LOS_binary == ">7_days",
na.rm = TRUE
)
# ============================================================
# 9. Function to create each variable table
# ============================================================
make_los_table <- function(data, variable, variable_label, p_value) {
data %>%
count(
.data[[variable]],
LOS_binary
) %>%
pivot_wider(
names_from = LOS_binary,
values_from = n,
values_fill = 0
) %>%
rename(
Category = all_of(variable)
) %>%
mutate(
Variable = variable_label,
Total = `<=7_days` + `>7_days`,
`<=7_pct` = ifelse(
short_total > 0,
`<=7_days` / short_total * 100,
0
),
`>7_pct` = ifelse(
long_total > 0,
`>7_days` / long_total * 100,
0
),
p_value = p_value
)
}
# ============================================================
# 10. Create the three tables
# ============================================================
gender_result <- make_los_table(
gender_data,
"Gender",
"Gender",
gender_p
)
admission_result <- make_los_table(
admission_data,
"Admission_source",
"Admission source",
admission_p
)
diagnosis_result <- make_los_table(
diagnosis_data,
"diag_superclass",
"Diagnosis",
diagnosis_p
)
# ============================================================
# 11. Combine
# ============================================================
result_los <- bind_rows(
gender_result,
admission_result,
diagnosis_result
)
# ============================================================
# 12. Format p-values
# ============================================================
result_los <- result_los %>%
mutate(
`p-value` = case_when(
is.na(p_value) ~ "",
p_value < 0.001 ~ "<0.001",
TRUE ~ sprintf("%.3f", p_value)
)
)
# ============================================================
# 13. Show variable and p-value once
# ============================================================
result_los <- result_los %>%
group_by(Variable) %>%
mutate(
Variable_display = if_else(
row_number() == 1,
Variable,
""
),
p_value_display = if_else(
row_number() == 1,
`p-value`,
""
)
) %>%
ungroup()
# ============================================================
# 14. Final data for gt
# ============================================================
result_los <- result_los %>%
select(
Variable_display,
Category,
Total,
`<=7_days`,
`<=7_pct`,
`>7_days`,
`>7_pct`,
p_value_display
)
# ============================================================
# 15. Publication table
# ============================================================
publication_table_los <- result_los %>%
gt() %>%
cols_label(
Variable_display = "Variable",
Category = "Category",
Total = "Total",
`<=7_days` = "n",
`<=7_pct` = "%",
`>7_days` = "n",
`>7_pct` = "%",
p_value_display = "p-value"
) %>%
tab_spanner(
label = "≤7 days",
columns = c(
`<=7_days`,
`<=7_pct`
)
) %>%
tab_spanner(
label = ">7 days",
columns = c(
`>7_days`,
`>7_pct`
)
) %>%
tab_style(
style = cell_text(weight = "bold"),
locations = cells_body(
columns = Variable_display,
rows = Variable_display != ""
)
) %>%
tab_style(
style = cell_text(indent = px(15)),
locations = cells_body(
columns = Category
)
) %>%
tab_style(
style = cell_borders(
sides = "top",
color = "black",
weight = px(1)
),
locations = cells_body(
rows = Variable_display != ""
)
) %>%
cols_align(
align = "left",
columns = c(
Variable_display,
Category
)
) %>%
cols_align(
align = "center",
columns = c(
Total,
`<=7_days`,
`<=7_pct`,
`>7_days`,
`>7_pct`,
p_value_display
)
) %>%
fmt_number(
columns = c(
`<=7_pct`,
`>7_pct`
),
decimals = 1
) %>%
tab_options(
table.border.top.color = "black",
table.border.bottom.color = "black",
table.border.top.width = px(1),
table.border.bottom.width = px(1),
column_labels.border.top.width = px(1),
column_labels.border.bottom.width = px(1),
table.font.size = px(12),
data_row.padding = px(5)
)
publication_table_los| Variable | Category | Total |
≤7 days
|
>7 days
|
p-value | ||
|---|---|---|---|---|---|---|---|
| n | % | n | % | ||||
| Gender | Female | 1386 | 581 | 42.0 | 805 | 43.0 | 0.578 |
| Male | 1867 | 802 | 42.0 | 1065 | 43.0 | ||
| Admission source | Emergency | 2051 | 845 | 61.1 | 1206 | 64.5 | 0.275 |
| HDU | 12 | 5 | 61.1 | 7 | 64.5 | ||
| Maternity Ward | 59 | 32 | 61.1 | 27 | 64.5 | ||
| Medical Ward | 469 | 212 | 61.1 | 257 | 64.5 | ||
| Orthopedic Ward | 20 | 7 | 61.1 | 13 | 64.5 | ||
| Other | 74 | 28 | 61.1 | 46 | 64.5 | ||
| Pediatrics Ward | 6 | 2 | 61.1 | 4 | 64.5 | ||
| Surgical Ward | 562 | 252 | 61.1 | 310 | 64.5 | ||
| Diagnosis | Autoimmune / Systemic Disease | 10 | 7 | 0.5 | 3 | 0.2 | <0.001 |
| Cardiovascular Disease | 841 | 404 | 0.5 | 437 | 0.2 | ||
| Endocrine / Metabolic Disease | 140 | 56 | 0.5 | 84 | 0.2 | ||
| Gastrointestinal / Hepatic Disease | 237 | 113 | 0.5 | 124 | 0.2 | ||
| Malignancy | 76 | 30 | 0.5 | 46 | 0.2 | ||
| Multi-organ Failure | 33 | 17 | 0.5 | 16 | 0.2 | ||
| Neurological Disease | 549 | 211 | 0.5 | 338 | 0.2 | ||
| Obstetric / Gynecology | 34 | 18 | 0.5 | 16 | 0.2 | ||
| Other | 134 | 54 | 0.5 | 80 | 0.2 | ||
| Renal Disease | 101 | 43 | 0.5 | 58 | 0.2 | ||
| Respiratory Disease | 495 | 216 | 0.5 | 279 | 0.2 | ||
| Sepsis / Infection | 366 | 137 | 0.5 | 229 | 0.2 | ||
| Thromboembolism | 68 | 17 | 0.5 | 51 | 0.2 | ||
| Trauma / Toxicology | 169 | 60 | 0.5 | 109 | 0.2 | ||
# ============================================================
# PUBLICATION TABLE
# MORTALITY + LENGTH OF STAY
# INCLUDING AGE GROUP
# ============================================================
library(dplyr)
library(tidyr)
library(gt)
# ============================================================
# 1. CREATE DISPLAY VERSIONS
# ============================================================
# ------------------------------------------------------------
# MORTALITY DATA — df_final
# ------------------------------------------------------------
df_final_display <- df_final %>%
mutate(
# --------------------------------------------------------
# GENDER
# --------------------------------------------------------
Gender_display = case_when(
as.character(Gender) == "0" ~ "Female",
as.character(Gender) == "1" ~ "Male",
as.character(Gender) == "Female" ~ "Female",
as.character(Gender) == "Male" ~ "Male",
TRUE ~ NA_character_
),
# --------------------------------------------------------
# AGE GROUP
# --------------------------------------------------------
Age_group = case_when(
!is.na(Age_years) & Age_years <= 24 ~ "<=24",
!is.na(Age_years) & Age_years >= 25 & Age_years <= 29 ~ "25-29",
!is.na(Age_years) & Age_years >= 30 & Age_years <= 40 ~ "30-40",
!is.na(Age_years) & Age_years >= 41 & Age_years <= 50 ~ "41-50",
!is.na(Age_years) & Age_years > 50 ~ ">51",
TRUE ~ NA_character_
),
# --------------------------------------------------------
# ADMISSION SOURCE
# --------------------------------------------------------
Admission_display = case_when(
as.character(Admission_source) == "1" ~ "Emergency",
as.character(Admission_source) == "2" ~ "Medical Ward",
as.character(Admission_source) == "3" ~ "Surgical Ward",
as.character(Admission_source) == "4" ~ "Pediatrics Ward",
as.character(Admission_source) == "5" ~ "Maternity Ward",
as.character(Admission_source) == "6" ~ "Orthopedic Ward",
as.character(Admission_source) == "7" ~ "HDU",
as.character(Admission_source) == "8" ~ "Other",
as.character(Admission_source) == "Emergency" ~ "Emergency",
as.character(Admission_source) == "Medical Ward" ~ "Medical Ward",
as.character(Admission_source) == "Surgical Ward" ~ "Surgical Ward",
as.character(Admission_source) == "Pediatrics Ward" ~ "Pediatrics Ward",
as.character(Admission_source) == "Maternity Ward" ~ "Maternity Ward",
as.character(Admission_source) == "Orthopedic Ward" ~ "Orthopedic Ward",
as.character(Admission_source) == "HDU" ~ "HDU",
as.character(Admission_source) == "Other" ~ "Other",
as.character(Admission_source) == "Other Specialty Wards" ~ "Other",
TRUE ~ NA_character_
),
# --------------------------------------------------------
# DIAGNOSIS
# --------------------------------------------------------
Diagnosis_display = case_when(
as.character(diag_superclass) == "1" ~
"Cardiovascular Disease",
as.character(diag_superclass) == "2" ~
"Respiratory Disease",
as.character(diag_superclass) == "3" ~
"Neurological Disease",
as.character(diag_superclass) == "4" ~
"Sepsis / Infection",
as.character(diag_superclass) == "5" ~
"Gastrointestinal / Hepatic Disease",
as.character(diag_superclass) == "6" ~
"Trauma / Toxicology",
as.character(diag_superclass) == "7" ~
"Endocrine / Metabolic Disease",
as.character(diag_superclass) == "8" ~
"Renal Disease",
as.character(diag_superclass) == "9" ~
"Thromboembolism",
as.character(diag_superclass) == "10" ~
"Malignancy",
as.character(diag_superclass) == "11" ~
"Multi-organ Failure",
as.character(diag_superclass) == "12" ~
"Obstetric / Gynecology",
as.character(diag_superclass) == "13" ~
"Autoimmune / Systemic Disease",
as.character(diag_superclass) == "14" ~
"Other",
as.character(diag_superclass) == "Cardiovascular Disease" ~
"Cardiovascular Disease",
as.character(diag_superclass) == "Respiratory Disease" ~
"Respiratory Disease",
as.character(diag_superclass) == "Neurological Disease" ~
"Neurological Disease",
as.character(diag_superclass) == "Sepsis / Infection" ~
"Sepsis / Infection",
as.character(diag_superclass) ==
"Gastrointestinal / Hepatic Disease" ~
"Gastrointestinal / Hepatic Disease",
as.character(diag_superclass) == "Trauma / Toxicology" ~
"Trauma / Toxicology",
as.character(diag_superclass) ==
"Endocrine / Metabolic Disease" ~
"Endocrine / Metabolic Disease",
as.character(diag_superclass) == "Renal Disease" ~
"Renal Disease",
as.character(diag_superclass) == "Thromboembolism" ~
"Thromboembolism",
as.character(diag_superclass) == "Malignancy" ~
"Malignancy",
as.character(diag_superclass) == "Multi-organ Failure" ~
"Multi-organ Failure",
as.character(diag_superclass) ==
"Obstetric / Gynecology" ~
"Obstetric / Gynecology",
as.character(diag_superclass) ==
"Autoimmune / Systemic Disease" ~
"Autoimmune / Systemic Disease",
as.character(diag_superclass) == "Other" ~ "Other",
TRUE ~ NA_character_
)
)
# ------------------------------------------------------------
# LOS DATA — df
# ------------------------------------------------------------
df_display <- df %>%
mutate(
# --------------------------------------------------------
# GENDER
# --------------------------------------------------------
Gender_display = case_when(
as.character(Gender) == "0" ~ "Female",
as.character(Gender) == "1" ~ "Male",
as.character(Gender) == "Female" ~ "Female",
as.character(Gender) == "Male" ~ "Male",
TRUE ~ NA_character_
),
# --------------------------------------------------------
# AGE GROUP
# --------------------------------------------------------
Age_group = case_when(
!is.na(Age_years) & Age_years <= 24 ~ "<=24",
!is.na(Age_years) & Age_years >= 25 & Age_years <= 29 ~ "25-29",
!is.na(Age_years) & Age_years >= 30 & Age_years <= 40 ~ "30-40",
!is.na(Age_years) & Age_years >= 41 & Age_years <= 50 ~ "41-50",
!is.na(Age_years) & Age_years > 50 ~ ">51",
TRUE ~ NA_character_
),
# --------------------------------------------------------
# ADMISSION SOURCE
# --------------------------------------------------------
Admission_display = case_when(
as.character(Admission_source) == "1" ~ "Emergency",
as.character(Admission_source) == "2" ~ "Medical Ward",
as.character(Admission_source) == "3" ~ "Surgical Ward",
as.character(Admission_source) == "4" ~ "Pediatrics Ward",
as.character(Admission_source) == "5" ~ "Maternity Ward",
as.character(Admission_source) == "6" ~ "Orthopedic Ward",
as.character(Admission_source) == "7" ~ "HDU",
as.character(Admission_source) == "8" ~ "Other",
as.character(Admission_source) == "Emergency" ~ "Emergency",
as.character(Admission_source) == "Medical Ward" ~ "Medical Ward",
as.character(Admission_source) == "Surgical Ward" ~ "Surgical Ward",
as.character(Admission_source) == "Pediatrics Ward" ~ "Pediatrics Ward",
as.character(Admission_source) == "Maternity Ward" ~ "Maternity Ward",
as.character(Admission_source) == "Orthopedic Ward" ~ "Orthopedic Ward",
as.character(Admission_source) == "HDU" ~ "HDU",
as.character(Admission_source) == "Other" ~ "Other",
as.character(Admission_source) == "Other Specialty Wards" ~ "Other",
TRUE ~ NA_character_
),
# --------------------------------------------------------
# DIAGNOSIS
# --------------------------------------------------------
Diagnosis_display = case_when(
as.character(diag_superclass) == "1" ~
"Cardiovascular Disease",
as.character(diag_superclass) == "2" ~
"Respiratory Disease",
as.character(diag_superclass) == "3" ~
"Neurological Disease",
as.character(diag_superclass) == "4" ~
"Sepsis / Infection",
as.character(diag_superclass) == "5" ~
"Gastrointestinal / Hepatic Disease",
as.character(diag_superclass) == "6" ~
"Trauma / Toxicology",
as.character(diag_superclass) == "7" ~
"Endocrine / Metabolic Disease",
as.character(diag_superclass) == "8" ~
"Renal Disease",
as.character(diag_superclass) == "9" ~
"Thromboembolism",
as.character(diag_superclass) == "10" ~
"Malignancy",
as.character(diag_superclass) == "11" ~
"Multi-organ Failure",
as.character(diag_superclass) == "12" ~
"Obstetric / Gynecology",
as.character(diag_superclass) == "13" ~
"Autoimmune / Systemic Disease",
as.character(diag_superclass) == "14" ~
"Other",
as.character(diag_superclass) == "Cardiovascular Disease" ~
"Cardiovascular Disease",
as.character(diag_superclass) == "Respiratory Disease" ~
"Respiratory Disease",
as.character(diag_superclass) == "Neurological Disease" ~
"Neurological Disease",
as.character(diag_superclass) == "Sepsis / Infection" ~
"Sepsis / Infection",
as.character(diag_superclass) ==
"Gastrointestinal / Hepatic Disease" ~
"Gastrointestinal / Hepatic Disease",
as.character(diag_superclass) == "Trauma / Toxicology" ~
"Trauma / Toxicology",
as.character(diag_superclass) ==
"Endocrine / Metabolic Disease" ~
"Endocrine / Metabolic Disease",
as.character(diag_superclass) == "Renal Disease" ~
"Renal Disease",
as.character(diag_superclass) == "Thromboembolism" ~
"Thromboembolism",
as.character(diag_superclass) == "Malignancy" ~
"Malignancy",
as.character(diag_superclass) == "Multi-organ Failure" ~
"Multi-organ Failure",
as.character(diag_superclass) ==
"Obstetric / Gynecology" ~
"Obstetric / Gynecology",
as.character(diag_superclass) ==
"Autoimmune / Systemic Disease" ~
"Autoimmune / Systemic Disease",
as.character(diag_superclass) == "Other" ~ "Other",
TRUE ~ NA_character_
)
)
# ============================================================
# 2. OUTCOME ORDER
# ============================================================
df_final_display$stat <- factor(
df_final_display$stat,
levels = c("Survived", "Died")
)
df_display$LOS_binary <- factor(
df_display$LOS_binary,
levels = c("<=7_days", ">7_days")
)
# ============================================================
# 3. SAFE P-VALUE FUNCTION
# ============================================================
get_pvalue <- function(x, y) {
keep <- !is.na(x) & !is.na(y)
x <- x[keep]
y <- y[keep]
tab <- table(x, y)
# No observations or only one level
if (sum(tab) == 0 ||
nrow(tab) < 2 ||
ncol(tab) < 2) {
return(NA_real_)
}
chi <- suppressWarnings(
chisq.test(tab)
)
if (any(chi$expected < 5)) {
return(
suppressWarnings(
chisq.test(
tab,
simulate.p.value = TRUE,
B = 10000
)$p.value
)
)
} else {
return(chi$p.value)
}
}
# ============================================================
# 4. P-VALUES — MORTALITY
# ============================================================
mortality_p_gender <- get_pvalue(
df_final_display$Gender_display,
df_final_display$stat
)
mortality_p_age <- get_pvalue(
df_final_display$Age_group,
df_final_display$stat
)
mortality_p_admission <- get_pvalue(
df_final_display$Admission_display,
df_final_display$stat
)
mortality_p_diagnosis <- get_pvalue(
df_final_display$Diagnosis_display,
df_final_display$stat
)
# ============================================================
# 5. P-VALUES — LOS
# ============================================================
los_p_gender <- get_pvalue(
df_display$Gender_display,
df_display$LOS_binary
)
los_p_age <- get_pvalue(
df_display$Age_group,
df_display$LOS_binary
)
los_p_admission <- get_pvalue(
df_display$Admission_display,
df_display$LOS_binary
)
los_p_diagnosis <- get_pvalue(
df_display$Diagnosis_display,
df_display$LOS_binary
)
# ============================================================
# 6. FORMAT P-VALUE
# ============================================================
format_p <- function(p) {
if (is.na(p)) {
return("")
}
if (p < 0.001) {
return("<0.001")
}
sprintf("%.3f", p)
}
# ============================================================
# 7. FUNCTION TO ENSURE MORTALITY COLUMNS EXIST
# ============================================================
complete_mortality_columns <- function(data) {
if (!"Survived" %in% names(data)) {
data$Survived <- 0L
}
if (!"Died" %in% names(data)) {
data$Died <- 0L
}
data
}
# ============================================================
# 8. FUNCTION TO ENSURE LOS COLUMNS EXIST
# ============================================================
complete_los_columns <- function(data) {
if (!"<=7_days" %in% names(data)) {
data[["<=7_days"]] <- 0L
}
if (!">7_days" %in% names(data)) {
data[[">7_days"]] <- 0L
}
data
}
# ============================================================
# 9. MORTALITY — GENDER
# ============================================================
mortality_gender <- df_final_display %>%
filter(
!is.na(Gender_display),
!is.na(stat)
) %>%
count(
Category = Gender_display,
stat
) %>%
pivot_wider(
names_from = stat,
values_from = n,
values_fill = 0
) %>%
complete_mortality_columns() %>%
mutate(
Variable = "Gender",
Category = as.character(Category)
)
# ============================================================
# 10. MORTALITY — AGE GROUP
# ============================================================
mortality_age <- df_final_display %>%
filter(
!is.na(Age_group),
!is.na(stat)
) %>%
count(
Category = Age_group,
stat
) %>%
pivot_wider(
names_from = stat,
values_from = n,
values_fill = 0
) %>%
complete_mortality_columns() %>%
mutate(
Variable = "Age group",
Category = as.character(Category)
)
# ============================================================
# 11. MORTALITY — ADMISSION SOURCE
# ============================================================
mortality_admission <- df_final_display %>%
filter(
!is.na(Admission_display),
!is.na(stat)
) %>%
count(
Category = Admission_display,
stat
) %>%
pivot_wider(
names_from = stat,
values_from = n,
values_fill = 0
) %>%
complete_mortality_columns() %>%
mutate(
Variable = "Admission source",
Category = as.character(Category)
)
# ============================================================
# 12. MORTALITY — DIAGNOSIS
# ============================================================
mortality_diagnosis <- df_final_display %>%
filter(
!is.na(Diagnosis_display),
!is.na(stat)
) %>%
count(
Category = Diagnosis_display,
stat
) %>%
pivot_wider(
names_from = stat,
values_from = n,
values_fill = 0
) %>%
complete_mortality_columns() %>%
mutate(
Variable = "Diagnosis",
Category = as.character(Category)
)
# ============================================================
# 13. MORTALITY COUNTS + WITHIN-CATEGORY PERCENTAGES
# ============================================================
mortality_result <- bind_rows(
mortality_gender,
mortality_age,
mortality_admission,
mortality_diagnosis
) %>%
mutate(
Category = as.character(Category),
Survived = coalesce(Survived, 0L),
Died = coalesce(Died, 0L),
Mortality_Total = Survived + Died,
Survived_pct = if_else(
Mortality_Total > 0,
Survived / Mortality_Total * 100,
NA_real_
),
Died_pct = if_else(
Mortality_Total > 0,
Died / Mortality_Total * 100,
NA_real_
)
) %>%
select(
Variable,
Category,
Mortality_Total,
Survived,
Survived_pct,
Died,
Died_pct
)
# ============================================================
# 14. LOS — GENDER
# ============================================================
los_gender <- df_display %>%
filter(
!is.na(Gender_display),
!is.na(LOS_binary)
) %>%
count(
Category = Gender_display,
LOS_binary
) %>%
pivot_wider(
names_from = LOS_binary,
values_from = n,
values_fill = 0
) %>%
complete_los_columns() %>%
mutate(
Variable = "Gender",
Category = as.character(Category)
)
# ============================================================
# 15. LOS — AGE GROUP
# ============================================================
los_age <- df_display %>%
filter(
!is.na(Age_group),
!is.na(LOS_binary)
) %>%
count(
Category = Age_group,
LOS_binary
) %>%
pivot_wider(
names_from = LOS_binary,
values_from = n,
values_fill = 0
) %>%
complete_los_columns() %>%
mutate(
Variable = "Age group",
Category = as.character(Category)
)
# ============================================================
# 16. LOS — ADMISSION SOURCE
# ============================================================
los_admission <- df_display %>%
filter(
!is.na(Admission_display),
!is.na(LOS_binary)
) %>%
count(
Category = Admission_display,
LOS_binary
) %>%
pivot_wider(
names_from = LOS_binary,
values_from = n,
values_fill = 0
) %>%
complete_los_columns() %>%
mutate(
Variable = "Admission source",
Category = as.character(Category)
)
# ============================================================
# 17. LOS — DIAGNOSIS
# ============================================================
los_diagnosis <- df_display %>%
filter(
!is.na(Diagnosis_display),
!is.na(LOS_binary)
) %>%
count(
Category = Diagnosis_display,
LOS_binary
) %>%
pivot_wider(
names_from = LOS_binary,
values_from = n,
values_fill = 0
) %>%
complete_los_columns() %>%
mutate(
Variable = "Diagnosis",
Category = as.character(Category)
)
# ============================================================
# 18. LOS COUNTS + WITHIN-CATEGORY PERCENTAGES
# ============================================================
los_result <- bind_rows(
los_gender,
los_age,
los_admission,
los_diagnosis
) %>%
mutate(
Category = as.character(Category),
`<=7_days` = coalesce(`<=7_days`, 0L),
`>7_days` = coalesce(`>7_days`, 0L),
LOS_Total = `<=7_days` + `>7_days`,
`<=7_pct` = if_else(
LOS_Total > 0,
`<=7_days` / LOS_Total * 100,
NA_real_
),
`>7_pct` = if_else(
LOS_Total > 0,
`>7_days` / LOS_Total * 100,
NA_real_
)
) %>%
select(
Variable,
Category,
LOS_Total,
`<=7_days`,
`<=7_pct`,
`>7_days`,
`>7_pct`
)
# ============================================================
# 19. MERGE MORTALITY + LOS
# ============================================================
result_combined <- full_join(
mortality_result,
los_result,
by = c("Variable", "Category")
)
# ============================================================
# 20. ADD P-VALUES
# ============================================================
result_combined <- result_combined %>%
mutate(
`Mortality p-value` = case_when(
Variable == "Gender" ~ format_p(mortality_p_gender),
Variable == "Age group" ~ format_p(mortality_p_age),
Variable == "Admission source" ~ format_p(mortality_p_admission),
Variable == "Diagnosis" ~ format_p(mortality_p_diagnosis),
TRUE ~ ""
),
`LOS p-value` = case_when(
Variable == "Gender" ~ format_p(los_p_gender),
Variable == "Age group" ~ format_p(los_p_age),
Variable == "Admission source" ~ format_p(los_p_admission),
Variable == "Diagnosis" ~ format_p(los_p_diagnosis),
TRUE ~ ""
)
)
# ============================================================
# 21. FORCE VARIABLE ORDER
# ============================================================
result_combined$Variable <- factor(
result_combined$Variable,
levels = c(
"Gender",
"Age group",
"Admission source",
"Diagnosis"
)
)
result_combined <- result_combined %>%
arrange(Variable)
# ============================================================
# 22. SHOW VARIABLE + P-VALUE ONLY ONCE
# ============================================================
result_combined <- result_combined %>%
group_by(Variable) %>%
mutate(
Variable_display = if_else(
row_number() == 1,
as.character(Variable),
""
),
Mortality_p_display = if_else(
row_number() == 1,
`Mortality p-value`,
""
),
LOS_p_display = if_else(
row_number() == 1,
`LOS p-value`,
""
)
) %>%
ungroup()
# ============================================================
# 23. FINAL TABLE DATA
# ============================================================
result_combined <- result_combined %>%
select(
Variable_display,
Category,
Mortality_Total,
Survived,
Survived_pct,
Died,
Died_pct,
Mortality_p_display,
LOS_Total,
`<=7_days`,
`<=7_pct`,
`>7_days`,
`>7_pct`,
LOS_p_display
)
# ============================================================
# 24. PUBLICATION TABLE
# ============================================================
publishable_table_los <- result_combined %>%
gt() %>%
# ----------------------------------------------------------
# COLUMN LABELS
# ----------------------------------------------------------
cols_label(
Variable_display = "Variable",
Category = "Category",
# TOTAL COLUMN — BLANK IN FINAL DISPLAY
Mortality_Total = "",
# SURVIVED
Survived = "",
Survived_pct = "%",
# DIED
Died = "",
Died_pct = "%",
Mortality_p_display = "p-value",
# TOTAL COLUMN — BLANK IN FINAL DISPLAY
LOS_Total = "",
# ≤7 DAYS
`<=7_days` = "",
`<=7_pct` = "%",
# >7 DAYS
`>7_days` = "",
`>7_pct` = "%",
LOS_p_display = "p-value"
) %>%
# ----------------------------------------------------------
# SURVIVED
# ----------------------------------------------------------
tab_spanner(
label = "Survived",
columns = c(
Survived,
Survived_pct
),
id = "survived"
) %>%
# ----------------------------------------------------------
# DIED
# ----------------------------------------------------------
tab_spanner(
label = "Died",
columns = c(
Died,
Died_pct
),
id = "died"
) %>%
# ----------------------------------------------------------
# ≤7 DAYS
# ----------------------------------------------------------
tab_spanner(
label = "≤7 days",
columns = c(
`<=7_days`,
`<=7_pct`
),
id = "short_los"
) %>%
# ----------------------------------------------------------
# >7 DAYS
# ----------------------------------------------------------
tab_spanner(
label = ">7 days",
columns = c(
`>7_days`,
`>7_pct`
),
id = "long_los"
) %>%
# ----------------------------------------------------------
# BOLD VARIABLE
# ----------------------------------------------------------
tab_style(
style = cell_text(
weight = "bold"
),
locations = cells_body(
columns = Variable_display,
rows = Variable_display != ""
)
) %>%
# ----------------------------------------------------------
# INDENT CATEGORY
# ----------------------------------------------------------
tab_style(
style = cell_text(
indent = px(15)
),
locations = cells_body(
columns = Category
)
) %>%
# ----------------------------------------------------------
# LINE BEFORE EACH VARIABLE
# ----------------------------------------------------------
tab_style(
style = cell_borders(
sides = "top",
color = "black",
weight = px(1)
),
locations = cells_body(
rows = Variable_display != ""
)
) %>%
# ----------------------------------------------------------
# LEFT ALIGNMENT
# ----------------------------------------------------------
cols_align(
align = "left",
columns = c(
Variable_display,
Category
)
) %>%
# ----------------------------------------------------------
# CENTER NUMERICAL COLUMNS
# ----------------------------------------------------------
cols_align(
align = "center",
columns = c(
Mortality_Total,
Survived,
Survived_pct,
Died,
Died_pct,
Mortality_p_display,
LOS_Total,
`<=7_days`,
`<=7_pct`,
`>7_days`,
`>7_pct`,
LOS_p_display
)
) %>%
# ----------------------------------------------------------
# PERCENTAGES — ONE DECIMAL
# ----------------------------------------------------------
fmt_number(
columns = c(
Survived_pct,
Died_pct,
`<=7_pct`,
`>7_pct`
),
decimals = 1
) %>%
# ----------------------------------------------------------
# TABLE STYLE
# ----------------------------------------------------------
tab_options(
table.border.top.color = "black",
table.border.bottom.color = "black",
table.border.top.width = px(1),
table.border.bottom.width = px(1),
column_labels.border.top.width = px(1),
column_labels.border.bottom.width = px(1),
table.font.size = px(12),
data_row.padding = px(5)
)
# ============================================================
# 25. DISPLAY FINAL TABLE
# ============================================================
publishable_table_los| Variable | Category |
Survived
|
Died
|
p-value |
≤7 days
|
>7 days
|
p-value | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| % | % | % | % | ||||||||||
| Gender | Female | NA | NA | NA | NA | NA | 1386 | 581 | 41.9 | 805 | 58.1 | 0.578 | |
| Male | NA | NA | NA | NA | NA | 1867 | 802 | 43.0 | 1065 | 57.0 | |||
| Age group | 25-29 | NA | NA | NA | NA | NA | 184 | 103 | 56.0 | 81 | 44.0 | <0.001 | |
| 30-40 | NA | NA | NA | NA | NA | 370 | 175 | 47.3 | 195 | 52.7 | |||
| 41-50 | NA | NA | NA | NA | NA | 408 | 165 | 40.4 | 243 | 59.6 | |||
| <=24 | NA | NA | NA | NA | NA | 188 | 96 | 51.1 | 92 | 48.9 | |||
| >51 | NA | NA | NA | NA | NA | 2102 | 844 | 40.2 | 1258 | 59.8 | |||
| Admission source | Emergency | NA | NA | NA | NA | NA | 2051 | 845 | 41.2 | 1206 | 58.8 | 0.275 | |
| HDU | NA | NA | NA | NA | NA | 12 | 5 | 41.7 | 7 | 58.3 | |||
| Maternity Ward | NA | NA | NA | NA | NA | 59 | 32 | 54.2 | 27 | 45.8 | |||
| Medical Ward | NA | NA | NA | NA | NA | 469 | 212 | 45.2 | 257 | 54.8 | |||
| Orthopedic Ward | NA | NA | NA | NA | NA | 20 | 7 | 35.0 | 13 | 65.0 | |||
| Other | NA | NA | NA | NA | NA | 74 | 28 | 37.8 | 46 | 62.2 | |||
| Pediatrics Ward | NA | NA | NA | NA | NA | 6 | 2 | 33.3 | 4 | 66.7 | |||
| Surgical Ward | NA | NA | NA | NA | NA | 562 | 252 | 44.8 | 310 | 55.2 | |||
| Diagnosis | Autoimmune / Systemic Disease | NA | NA | NA | NA | NA | 10 | 7 | 70.0 | 3 | 30.0 | <0.001 | |
| Cardiovascular Disease | NA | NA | NA | NA | NA | 841 | 404 | 48.0 | 437 | 52.0 | |||
| Endocrine / Metabolic Disease | NA | NA | NA | NA | NA | 140 | 56 | 40.0 | 84 | 60.0 | |||
| Gastrointestinal / Hepatic Disease | NA | NA | NA | NA | NA | 237 | 113 | 47.7 | 124 | 52.3 | |||
| Malignancy | NA | NA | NA | NA | NA | 76 | 30 | 39.5 | 46 | 60.5 | |||
| Multi-organ Failure | NA | NA | NA | NA | NA | 33 | 17 | 51.5 | 16 | 48.5 | |||
| Neurological Disease | NA | NA | NA | NA | NA | 549 | 211 | 38.4 | 338 | 61.6 | |||
| Obstetric / Gynecology | NA | NA | NA | NA | NA | 34 | 18 | 52.9 | 16 | 47.1 | |||
| Other | NA | NA | NA | NA | NA | 134 | 54 | 40.3 | 80 | 59.7 | |||
| Renal Disease | NA | NA | NA | NA | NA | 101 | 43 | 42.6 | 58 | 57.4 | |||
| Respiratory Disease | NA | NA | NA | NA | NA | 495 | 216 | 43.6 | 279 | 56.4 | |||
| Sepsis / Infection | NA | NA | NA | NA | NA | 366 | 137 | 37.4 | 229 | 62.6 | |||
| Thromboembolism | NA | NA | NA | NA | NA | 68 | 17 | 25.0 | 51 | 75.0 | |||
| Trauma / Toxicology | NA | NA | NA | NA | NA | 169 | 60 | 35.5 | 109 | 64.5 | |||
# ============================================================
# PUBLICATION TABLE
# MORTALITY + LENGTH OF STAY
# INCLUDING AGE GROUP
# ============================================================
library(dplyr)
library(tidyr)
library(gt)
# ============================================================
# 1. CREATE DISPLAY VERSIONS
# ============================================================
# ------------------------------------------------------------
# MORTALITY DATA — df_final
# ------------------------------------------------------------
df_final_display <- df_final %>%
mutate(
# --------------------------------------------------------
# GENDER
# --------------------------------------------------------
Gender_display = case_when(
as.character(Gender) == "0" ~ "Female",
as.character(Gender) == "1" ~ "Male",
as.character(Gender) == "Female" ~ "Female",
as.character(Gender) == "Male" ~ "Male",
TRUE ~ NA_character_
),
# --------------------------------------------------------
# AGE GROUP
# --------------------------------------------------------
Age_group = case_when(
!is.na(Age_years) & Age_years <= 24 ~ "<=24",
!is.na(Age_years) & Age_years >= 25 & Age_years <= 29 ~ "25-29",
!is.na(Age_years) & Age_years >= 30 & Age_years <= 40 ~ "30-40",
!is.na(Age_years) & Age_years >= 41 & Age_years <= 50 ~ "41-50",
!is.na(Age_years) & Age_years > 50 ~ ">51",
TRUE ~ NA_character_
),
# --------------------------------------------------------
# ADMISSION SOURCE
# --------------------------------------------------------
Admission_display = case_when(
as.character(Admission_source) == "1" ~ "Emergency",
as.character(Admission_source) == "2" ~ "Medical Ward",
as.character(Admission_source) == "3" ~ "Surgical Ward",
as.character(Admission_source) == "4" ~ "Pediatrics Ward",
as.character(Admission_source) == "5" ~ "Maternity Ward",
as.character(Admission_source) == "6" ~ "Orthopedic Ward",
as.character(Admission_source) == "7" ~ "HDU",
as.character(Admission_source) == "8" ~ "Other",
as.character(Admission_source) == "Emergency" ~ "Emergency",
as.character(Admission_source) == "Medical Ward" ~ "Medical Ward",
as.character(Admission_source) == "Surgical Ward" ~ "Surgical Ward",
as.character(Admission_source) == "Pediatrics Ward" ~ "Pediatrics Ward",
as.character(Admission_source) == "Maternity Ward" ~ "Maternity Ward",
as.character(Admission_source) == "Orthopedic Ward" ~ "Orthopedic Ward",
as.character(Admission_source) == "HDU" ~ "HDU",
as.character(Admission_source) == "Other" ~ "Other",
as.character(Admission_source) == "Other Specialty Wards" ~ "Other",
TRUE ~ NA_character_
),
# --------------------------------------------------------
# DIAGNOSIS
# --------------------------------------------------------
Diagnosis_display = case_when(
as.character(diag_superclass) == "1" ~
"Cardiovascular Disease",
as.character(diag_superclass) == "2" ~
"Respiratory Disease",
as.character(diag_superclass) == "3" ~
"Neurological Disease",
as.character(diag_superclass) == "4" ~
"Sepsis / Infection",
as.character(diag_superclass) == "5" ~
"Gastrointestinal / Hepatic Disease",
as.character(diag_superclass) == "6" ~
"Trauma / Toxicology",
as.character(diag_superclass) == "7" ~
"Endocrine / Metabolic Disease",
as.character(diag_superclass) == "8" ~
"Renal Disease",
as.character(diag_superclass) == "9" ~
"Thromboembolism",
as.character(diag_superclass) == "10" ~
"Malignancy",
as.character(diag_superclass) == "11" ~
"Multi-organ Failure",
as.character(diag_superclass) == "12" ~
"Obstetric / Gynecology",
as.character(diag_superclass) == "13" ~
"Autoimmune / Systemic Disease",
as.character(diag_superclass) == "14" ~
"Other",
as.character(diag_superclass) == "Cardiovascular Disease" ~
"Cardiovascular Disease",
as.character(diag_superclass) == "Respiratory Disease" ~
"Respiratory Disease",
as.character(diag_superclass) == "Neurological Disease" ~
"Neurological Disease",
as.character(diag_superclass) == "Sepsis / Infection" ~
"Sepsis / Infection",
as.character(diag_superclass) ==
"Gastrointestinal / Hepatic Disease" ~
"Gastrointestinal / Hepatic Disease",
as.character(diag_superclass) == "Trauma / Toxicology" ~
"Trauma / Toxicology",
as.character(diag_superclass) ==
"Endocrine / Metabolic Disease" ~
"Endocrine / Metabolic Disease",
as.character(diag_superclass) == "Renal Disease" ~
"Renal Disease",
as.character(diag_superclass) == "Thromboembolism" ~
"Thromboembolism",
as.character(diag_superclass) == "Malignancy" ~
"Malignancy",
as.character(diag_superclass) == "Multi-organ Failure" ~
"Multi-organ Failure",
as.character(diag_superclass) ==
"Obstetric / Gynecology" ~
"Obstetric / Gynecology",
as.character(diag_superclass) ==
"Autoimmune / Systemic Disease" ~
"Autoimmune / Systemic Disease",
as.character(diag_superclass) == "Other" ~
"Other",
TRUE ~ NA_character_
)
)
# ------------------------------------------------------------
# LOS DATA — df
# ------------------------------------------------------------
df_display <- df %>%
mutate(
# --------------------------------------------------------
# GENDER
# --------------------------------------------------------
Gender_display = case_when(
as.character(Gender) == "0" ~ "Female",
as.character(Gender) == "1" ~ "Male",
as.character(Gender) == "Female" ~ "Female",
as.character(Gender) == "Male" ~ "Male",
TRUE ~ NA_character_
),
# --------------------------------------------------------
# AGE GROUP
# --------------------------------------------------------
Age_group = case_when(
!is.na(Age_years) & Age_years <= 24 ~ "<=24",
!is.na(Age_years) & Age_years >= 25 & Age_years <= 29 ~ "25-29",
!is.na(Age_years) & Age_years >= 30 & Age_years <= 40 ~ "30-40",
!is.na(Age_years) & Age_years >= 41 & Age_years <= 50 ~ "41-50",
!is.na(Age_years) & Age_years > 50 ~ ">51",
TRUE ~ NA_character_
),
# --------------------------------------------------------
# ADMISSION SOURCE
# --------------------------------------------------------
Admission_display = case_when(
as.character(Admission_source) == "1" ~ "Emergency",
as.character(Admission_source) == "2" ~ "Medical Ward",
as.character(Admission_source) == "3" ~ "Surgical Ward",
as.character(Admission_source) == "4" ~ "Pediatrics Ward",
as.character(Admission_source) == "5" ~ "Maternity Ward",
as.character(Admission_source) == "6" ~ "Orthopedic Ward",
as.character(Admission_source) == "7" ~ "HDU",
as.character(Admission_source) == "8" ~ "Other",
as.character(Admission_source) == "Emergency" ~ "Emergency",
as.character(Admission_source) == "Medical Ward" ~ "Medical Ward",
as.character(Admission_source) == "Surgical Ward" ~ "Surgical Ward",
as.character(Admission_source) == "Pediatrics Ward" ~ "Pediatrics Ward",
as.character(Admission_source) == "Maternity Ward" ~ "Maternity Ward",
as.character(Admission_source) == "Orthopedic Ward" ~ "Orthopedic Ward",
as.character(Admission_source) == "HDU" ~ "HDU",
as.character(Admission_source) == "Other" ~ "Other",
as.character(Admission_source) == "Other Specialty Wards" ~ "Other",
TRUE ~ NA_character_
),
# --------------------------------------------------------
# DIAGNOSIS
# --------------------------------------------------------
Diagnosis_display = case_when(
as.character(diag_superclass) == "1" ~
"Cardiovascular Disease",
as.character(diag_superclass) == "2" ~
"Respiratory Disease",
as.character(diag_superclass) == "3" ~
"Neurological Disease",
as.character(diag_superclass) == "4" ~
"Sepsis / Infection",
as.character(diag_superclass) == "5" ~
"Gastrointestinal / Hepatic Disease",
as.character(diag_superclass) == "6" ~
"Trauma / Toxicology",
as.character(diag_superclass) == "7" ~
"Endocrine / Metabolic Disease",
as.character(diag_superclass) == "8" ~
"Renal Disease",
as.character(diag_superclass) == "9" ~
"Thromboembolism",
as.character(diag_superclass) == "10" ~
"Malignancy",
as.character(diag_superclass) == "11" ~
"Multi-organ Failure",
as.character(diag_superclass) == "12" ~
"Obstetric / Gynecology",
as.character(diag_superclass) == "13" ~
"Autoimmune / Systemic Disease",
as.character(diag_superclass) == "14" ~
"Other",
as.character(diag_superclass) == "Cardiovascular Disease" ~
"Cardiovascular Disease",
as.character(diag_superclass) == "Respiratory Disease" ~
"Respiratory Disease",
as.character(diag_superclass) == "Neurological Disease" ~
"Neurological Disease",
as.character(diag_superclass) == "Sepsis / Infection" ~
"Sepsis / Infection",
as.character(diag_superclass) ==
"Gastrointestinal / Hepatic Disease" ~
"Gastrointestinal / Hepatic Disease",
as.character(diag_superclass) == "Trauma / Toxicology" ~
"Trauma / Toxicology",
as.character(diag_superclass) ==
"Endocrine / Metabolic Disease" ~
"Endocrine / Metabolic Disease",
as.character(diag_superclass) == "Renal Disease" ~
"Renal Disease",
as.character(diag_superclass) == "Thromboembolism" ~
"Thromboembolism",
as.character(diag_superclass) == "Malignancy" ~
"Malignancy",
as.character(diag_superclass) == "Multi-organ Failure" ~
"Multi-organ Failure",
as.character(diag_superclass) ==
"Obstetric / Gynecology" ~
"Obstetric / Gynecology",
as.character(diag_superclass) ==
"Autoimmune / Systemic Disease" ~
"Autoimmune / Systemic Disease",
as.character(diag_superclass) == "Other" ~
"Other",
TRUE ~ NA_character_
)
)
# ============================================================
# 2. CREATE A ROBUST MORTALITY OUTCOME
# ============================================================
# IMPORTANT:
# Do NOT overwrite the original stat variable.
# This safely recognizes the existing Survived/Died coding.
df_final_display <- df_final_display %>%
mutate(
Mortality_outcome = case_when(
as.character(stat) %in%
c("Survived", "survived", "Alive", "alive", "0") ~
"Survived",
as.character(stat) %in%
c("Died", "died", "Dead", "dead", "1") ~
"Died",
TRUE ~ NA_character_
)
)
# ============================================================
# 3. LOS OUTCOME ORDER
# ============================================================
df_display <- df_display %>%
mutate(
LOS_outcome = case_when(
as.character(LOS_binary) %in%
c("<=7_days", "≤7 days", "<=7 days", "0") ~
"<=7_days",
as.character(LOS_binary) %in%
c(">7_days", ">7 days", "1") ~
">7_days",
TRUE ~ NA_character_
)
)
# ============================================================
# 4. SAFE P-VALUE FUNCTION
# ============================================================
get_pvalue <- function(x, y) {
keep <- !is.na(x) & !is.na(y)
x <- x[keep]
y <- y[keep]
tab <- table(x, y)
if (length(tab) == 0 ||
sum(tab) == 0 ||
nrow(tab) < 2 ||
ncol(tab) < 2) {
return(NA_real_)
}
chi <- suppressWarnings(
chisq.test(tab)
)
if (any(chi$expected < 5)) {
return(
suppressWarnings(
chisq.test(
tab,
simulate.p.value = TRUE,
B = 10000
)$p.value
)
)
} else {
return(chi$p.value)
}
}
# ============================================================
# 5. MORTALITY P-VALUES
# ============================================================
mortality_p_gender <- get_pvalue(
df_final_display$Gender_display,
df_final_display$Mortality_outcome
)
mortality_p_age <- get_pvalue(
df_final_display$Age_group,
df_final_display$Mortality_outcome
)
mortality_p_admission <- get_pvalue(
df_final_display$Admission_display,
df_final_display$Mortality_outcome
)
mortality_p_diagnosis <- get_pvalue(
df_final_display$Diagnosis_display,
df_final_display$Mortality_outcome
)
# ============================================================
# 6. LOS P-VALUES
# ============================================================
los_p_gender <- get_pvalue(
df_display$Gender_display,
df_display$LOS_outcome
)
los_p_age <- get_pvalue(
df_display$Age_group,
df_display$LOS_outcome
)
los_p_admission <- get_pvalue(
df_display$Admission_display,
df_display$LOS_outcome
)
los_p_diagnosis <- get_pvalue(
df_display$Diagnosis_display,
df_display$LOS_outcome
)
# ============================================================
# 7. FORMAT P-VALUE
# ============================================================
format_p <- function(p) {
if (is.na(p)) {
return("")
}
if (p < 0.001) {
return("<0.001")
}
sprintf("%.3f", p)
}
# ============================================================
# 8. MORTALITY — GENDER
# ============================================================
mortality_gender <- df_final_display %>%
filter(
!is.na(Gender_display),
!is.na(Mortality_outcome)
) %>%
count(
Category = Gender_display,
Mortality_outcome
) %>%
pivot_wider(
names_from = Mortality_outcome,
values_from = n,
values_fill = 0
) %>%
mutate(
Category = as.character(Category)
)
if (!"Survived" %in% names(mortality_gender)) {
mortality_gender$Survived <- 0L
}
if (!"Died" %in% names(mortality_gender)) {
mortality_gender$Died <- 0L
}
mortality_gender <- mortality_gender %>%
mutate(
Variable = "Gender"
)
# ============================================================
# 9. MORTALITY — AGE GROUP
# ============================================================
mortality_age <- df_final_display %>%
filter(
!is.na(Age_group),
!is.na(Mortality_outcome)
) %>%
count(
Category = Age_group,
Mortality_outcome
) %>%
pivot_wider(
names_from = Mortality_outcome,
values_from = n,
values_fill = 0
) %>%
mutate(
Category = as.character(Category)
)
if (!"Survived" %in% names(mortality_age)) {
mortality_age$Survived <- 0L
}
if (!"Died" %in% names(mortality_age)) {
mortality_age$Died <- 0L
}
mortality_age <- mortality_age %>%
mutate(
Variable = "Age group"
)
# ============================================================
# 10. MORTALITY — ADMISSION SOURCE
# ============================================================
mortality_admission <- df_final_display %>%
filter(
!is.na(Admission_display),
!is.na(Mortality_outcome)
) %>%
count(
Category = Admission_display,
Mortality_outcome
) %>%
pivot_wider(
names_from = Mortality_outcome,
values_from = n,
values_fill = 0
) %>%
mutate(
Category = as.character(Category)
)
if (!"Survived" %in% names(mortality_admission)) {
mortality_admission$Survived <- 0L
}
if (!"Died" %in% names(mortality_admission)) {
mortality_admission$Died <- 0L
}
mortality_admission <- mortality_admission %>%
mutate(
Variable = "Admission source"
)
# ============================================================
# 11. MORTALITY — DIAGNOSIS
# ============================================================
mortality_diagnosis <- df_final_display %>%
filter(
!is.na(Diagnosis_display),
!is.na(Mortality_outcome)
) %>%
count(
Category = Diagnosis_display,
Mortality_outcome
) %>%
pivot_wider(
names_from = Mortality_outcome,
values_from = n,
values_fill = 0
) %>%
mutate(
Category = as.character(Category)
)
if (!"Survived" %in% names(mortality_diagnosis)) {
mortality_diagnosis$Survived <- 0L
}
if (!"Died" %in% names(mortality_diagnosis)) {
mortality_diagnosis$Died <- 0L
}
mortality_diagnosis <- mortality_diagnosis %>%
mutate(
Variable = "Diagnosis"
)
# ============================================================
# 12. MORTALITY RESULT
# ============================================================
mortality_result <- bind_rows(
mortality_gender,
mortality_age,
mortality_admission,
mortality_diagnosis
) %>%
mutate(
Category = as.character(Category),
Survived = coalesce(Survived, 0L),
Died = coalesce(Died, 0L),
Mortality_Total = Survived + Died,
# --------------------------------------------------------
# WITHIN-CATEGORY PERCENTAGES
# --------------------------------------------------------
Survived_pct = if_else(
Mortality_Total > 0,
Survived / Mortality_Total * 100,
NA_real_
),
Died_pct = if_else(
Mortality_Total > 0,
Died / Mortality_Total * 100,
NA_real_
)
) %>%
select(
Variable,
Category,
Mortality_Total,
Survived,
Survived_pct,
Died,
Died_pct
)
# ============================================================
# 13. LOS — GENDER
# ============================================================
los_gender <- df_display %>%
filter(
!is.na(Gender_display),
!is.na(LOS_outcome)
) %>%
count(
Category = Gender_display,
LOS_outcome
) %>%
pivot_wider(
names_from = LOS_outcome,
values_from = n,
values_fill = 0
) %>%
mutate(
Category = as.character(Category)
)
if (!"<=7_days" %in% names(los_gender)) {
los_gender[["<=7_days"]] <- 0L
}
if (!">7_days" %in% names(los_gender)) {
los_gender[[">7_days"]] <- 0L
}
los_gender <- los_gender %>%
mutate(
Variable = "Gender"
)
# ============================================================
# 14. LOS — AGE GROUP
# ============================================================
los_age <- df_display %>%
filter(
!is.na(Age_group),
!is.na(LOS_outcome)
) %>%
count(
Category = Age_group,
LOS_outcome
) %>%
pivot_wider(
names_from = LOS_outcome,
values_from = n,
values_fill = 0
) %>%
mutate(
Category = as.character(Category)
)
if (!"<=7_days" %in% names(los_age)) {
los_age[["<=7_days"]] <- 0L
}
if (!">7_days" %in% names(los_age)) {
los_age[[">7_days"]] <- 0L
}
los_age <- los_age %>%
mutate(
Variable = "Age group"
)
# ============================================================
# 15. LOS — ADMISSION SOURCE
# ============================================================
los_admission <- df_display %>%
filter(
!is.na(Admission_display),
!is.na(LOS_outcome)
) %>%
count(
Category = Admission_display,
LOS_outcome
) %>%
pivot_wider(
names_from = LOS_outcome,
values_from = n,
values_fill = 0
) %>%
mutate(
Category = as.character(Category)
)
if (!"<=7_days" %in% names(los_admission)) {
los_admission[["<=7_days"]] <- 0L
}
if (!">7_days" %in% names(los_admission)) {
los_admission[[">7_days"]] <- 0L
}
los_admission <- los_admission %>%
mutate(
Variable = "Admission source"
)
# ============================================================
# 16. LOS — DIAGNOSIS
# ============================================================
los_diagnosis <- df_display %>%
filter(
!is.na(Diagnosis_display),
!is.na(LOS_outcome)
) %>%
count(
Category = Diagnosis_display,
LOS_outcome
) %>%
pivot_wider(
names_from = LOS_outcome,
values_from = n,
values_fill = 0
) %>%
mutate(
Category = as.character(Category)
)
if (!"<=7_days" %in% names(los_diagnosis)) {
los_diagnosis[["<=7_days"]] <- 0L
}
if (!">7_days" %in% names(los_diagnosis)) {
los_diagnosis[[">7_days"]] <- 0L
}
los_diagnosis <- los_diagnosis %>%
mutate(
Variable = "Diagnosis"
)
# ============================================================
# 17. LOS RESULT
# ============================================================
los_result <- bind_rows(
los_gender,
los_age,
los_admission,
los_diagnosis
) %>%
mutate(
Category = as.character(Category),
`<=7_days` = coalesce(`<=7_days`, 0L),
`>7_days` = coalesce(`>7_days`, 0L),
LOS_Total = `<=7_days` + `>7_days`,
# --------------------------------------------------------
# WITHIN-CATEGORY PERCENTAGES
# --------------------------------------------------------
`<=7_pct` = if_else(
LOS_Total > 0,
`<=7_days` / LOS_Total * 100,
NA_real_
),
`>7_pct` = if_else(
LOS_Total > 0,
`>7_days` / LOS_Total * 100,
NA_real_
)
) %>%
select(
Variable,
Category,
LOS_Total,
`<=7_days`,
`<=7_pct`,
`>7_days`,
`>7_pct`
)
# ============================================================
# 18. MERGE MORTALITY + LOS
# ============================================================
result_combined <- full_join(
mortality_result,
los_result,
by = c("Variable", "Category")
)
# ============================================================
# 19. ADD P-VALUES
# ============================================================
result_combined <- result_combined %>%
mutate(
`Mortality p-value` = case_when(
Variable == "Gender" ~ format_p(mortality_p_gender),
Variable == "Age group" ~ format_p(mortality_p_age),
Variable == "Admission source" ~ format_p(mortality_p_admission),
Variable == "Diagnosis" ~ format_p(mortality_p_diagnosis),
TRUE ~ ""
),
`LOS p-value` = case_when(
Variable == "Gender" ~ format_p(los_p_gender),
Variable == "Age group" ~ format_p(los_p_age),
Variable == "Admission source" ~ format_p(los_p_admission),
Variable == "Diagnosis" ~ format_p(los_p_diagnosis),
TRUE ~ ""
)
)
# ============================================================
# 20. VARIABLE ORDER
# ============================================================
result_combined$Variable <- factor(
result_combined$Variable,
levels = c(
"Gender",
"Age group",
"Admission source",
"Diagnosis"
)
)
result_combined <- result_combined %>%
arrange(Variable)
# ============================================================
# 21. SHOW VARIABLE + P-VALUE ONLY ONCE
# ============================================================
result_combined <- result_combined %>%
group_by(Variable) %>%
mutate(
Variable_display = if_else(
row_number() == 1,
as.character(Variable),
""
),
Mortality_p_display = if_else(
row_number() == 1,
`Mortality p-value`,
""
),
LOS_p_display = if_else(
row_number() == 1,
`LOS p-value`,
""
)
) %>%
ungroup()
# ============================================================
# 22. FINAL TABLE DATA
# ============================================================
result_combined <- result_combined %>%
select(
Variable_display,
Category,
Mortality_Total,
Survived,
Survived_pct,
Died,
Died_pct,
Mortality_p_display,
LOS_Total,
`<=7_days`,
`<=7_pct`,
`>7_days`,
`>7_pct`,
LOS_p_display
)
# ============================================================
# 23. PUBLICATION TABLE
# ============================================================
publishable_table_los <- result_combined %>%
gt() %>%
# ----------------------------------------------------------
# COLUMN LABELS
# ----------------------------------------------------------
cols_label(
Variable_display = "Variable",
Category = "Category",
# Total values are intentionally NOT displayed
Mortality_Total = "",
# Survived
Survived = "",
Survived_pct = "%",
# Died
Died = "",
Died_pct = "%",
Mortality_p_display = "p-value",
# Total values are intentionally NOT displayed
LOS_Total = "",
# ≤7 days
`<=7_days` = "",
`<=7_pct` = "%",
# >7 days
`>7_days` = "",
`>7_pct` = "%",
LOS_p_display = "p-value"
) %>%
# ----------------------------------------------------------
# SURVIVED
# ----------------------------------------------------------
tab_spanner(
label = "Survived",
columns = c(
Survived,
Survived_pct
),
id = "survived"
) %>%
# ----------------------------------------------------------
# DIED
# ----------------------------------------------------------
tab_spanner(
label = "Died",
columns = c(
Died,
Died_pct
),
id = "died"
) %>%
# ----------------------------------------------------------
# ≤7 DAYS
# ----------------------------------------------------------
tab_spanner(
label = "≤7 days",
columns = c(
`<=7_days`,
`<=7_pct`
),
id = "short_los"
) %>%
# ----------------------------------------------------------
# >7 DAYS
# ----------------------------------------------------------
tab_spanner(
label = ">7 days",
columns = c(
`>7_days`,
`>7_pct`
),
id = "long_los"
) %>%
# ----------------------------------------------------------
# BOLD VARIABLE
# ----------------------------------------------------------
tab_style(
style = cell_text(
weight = "bold"
),
locations = cells_body(
columns = Variable_display,
rows = Variable_display != ""
)
) %>%
# ----------------------------------------------------------
# INDENT CATEGORY
# ----------------------------------------------------------
tab_style(
style = cell_text(
indent = px(15)
),
locations = cells_body(
columns = Category
)
) %>%
# ----------------------------------------------------------
# LINE BEFORE EACH VARIABLE
# ----------------------------------------------------------
tab_style(
style = cell_borders(
sides = "top",
color = "black",
weight = px(1)
),
locations = cells_body(
rows = Variable_display != ""
)
) %>%
# ----------------------------------------------------------
# LEFT ALIGNMENT
# ----------------------------------------------------------
cols_align(
align = "left",
columns = c(
Variable_display,
Category
)
) %>%
# ----------------------------------------------------------
# CENTER NUMERICAL COLUMNS
# ----------------------------------------------------------
cols_align(
align = "center",
columns = c(
Mortality_Total,
Survived,
Survived_pct,
Died,
Died_pct,
Mortality_p_display,
LOS_Total,
`<=7_days`,
`<=7_pct`,
`>7_days`,
`>7_pct`,
LOS_p_display
)
) %>%
# ----------------------------------------------------------
# PERCENTAGES — ONE DECIMAL
# ----------------------------------------------------------
fmt_number(
columns = c(
Survived_pct,
Died_pct,
`<=7_pct`,
`>7_pct`
),
decimals = 1
) %>%
# ----------------------------------------------------------
# TABLE STYLE
# ----------------------------------------------------------
tab_options(
table.border.top.color = "black",
table.border.bottom.color = "black",
table.border.top.width = px(1),
table.border.bottom.width = px(1),
column_labels.border.top.width = px(1),
column_labels.border.bottom.width = px(1),
table.font.size = px(12),
data_row.padding = px(5)
)
# ============================================================
# 24. DISPLAY FINAL TABLE
# ============================================================
publishable_table_los| Variable | Category |
Survived
|
Died
|
p-value |
≤7 days
|
>7 days
|
p-value | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| % | % | % | % | ||||||||||
| Gender | Female | 2342 | 1467 | 62.6 | 875 | 37.4 | 0.043 | 1386 | 581 | 41.9 | 805 | 58.1 | 0.578 |
| Male | 3173 | 1901 | 59.9 | 1272 | 40.1 | 1867 | 802 | 43.0 | 1065 | 57.0 | |||
| Age group | 25-29 | 247 | 72 | 29.1 | 175 | 70.9 | <0.001 | 184 | 103 | 56.0 | 81 | 44.0 | <0.001 |
| 30-40 | 556 | 206 | 37.1 | 350 | 62.9 | 370 | 175 | 47.3 | 195 | 52.7 | |||
| 41-50 | 680 | 396 | 58.2 | 284 | 41.8 | 408 | 165 | 40.4 | 243 | 59.6 | |||
| <=24 | 286 | 114 | 39.9 | 172 | 60.1 | 188 | 96 | 51.1 | 92 | 48.9 | |||
| >51 | 3742 | 2580 | 68.9 | 1162 | 31.1 | 2102 | 844 | 40.2 | 1258 | 59.8 | |||
| Admission source | Emergency | 3491 | 2177 | 62.4 | 1314 | 37.6 | <0.001 | 2051 | 845 | 41.2 | 1206 | 58.8 | 0.272 |
| HDU | 15 | 12 | 80.0 | 3 | 20.0 | 12 | 5 | 41.7 | 7 | 58.3 | |||
| Maternity Ward | 111 | 87 | 78.4 | 24 | 21.6 | 59 | 32 | 54.2 | 27 | 45.8 | |||
| Medical Ward | 839 | 389 | 46.4 | 450 | 53.6 | 469 | 212 | 45.2 | 257 | 54.8 | |||
| Orthopedic Ward | 32 | 17 | 53.1 | 15 | 46.9 | 20 | 7 | 35.0 | 13 | 65.0 | |||
| Other | 150 | 59 | 39.3 | 91 | 60.7 | 74 | 28 | 37.8 | 46 | 62.2 | |||
| Pediatrics Ward | 13 | 11 | 84.6 | 2 | 15.4 | 6 | 2 | 33.3 | 4 | 66.7 | |||
| Surgical Ward | 864 | 616 | 71.3 | 248 | 28.7 | 562 | 252 | 44.8 | 310 | 55.2 | |||
| Diagnosis | Autoimmune / Systemic Disease | 18 | 9 | 50.0 | 9 | 50.0 | <0.001 | 10 | 7 | 70.0 | 3 | 30.0 | <0.001 |
| Cardiovascular Disease | 1385 | 655 | 47.3 | 730 | 52.7 | 841 | 404 | 48.0 | 437 | 52.0 | |||
| Endocrine / Metabolic Disease | 226 | 190 | 84.1 | 36 | 15.9 | 140 | 56 | 40.0 | 84 | 60.0 | |||
| Gastrointestinal / Hepatic Disease | 440 | 339 | 77.0 | 101 | 23.0 | 237 | 113 | 47.7 | 124 | 52.3 | |||
| Malignancy | 126 | 102 | 81.0 | 24 | 19.0 | 76 | 30 | 39.5 | 46 | 60.5 | |||
| Multi-organ Failure | 76 | 0 | 0.0 | 76 | 100.0 | 33 | 17 | 51.5 | 16 | 48.5 | |||
| Neurological Disease | 923 | 576 | 62.4 | 347 | 37.6 | 549 | 211 | 38.4 | 338 | 61.6 | |||
| Obstetric / Gynecology | 51 | 36 | 70.6 | 15 | 29.4 | 34 | 18 | 52.9 | 16 | 47.1 | |||
| Other | 239 | 164 | 68.6 | 75 | 31.4 | 134 | 54 | 40.3 | 80 | 59.7 | |||
| Renal Disease | 151 | 81 | 53.6 | 70 | 46.4 | 101 | 43 | 42.6 | 58 | 57.4 | |||
| Respiratory Disease | 873 | 489 | 56.0 | 384 | 44.0 | 495 | 216 | 43.6 | 279 | 56.4 | |||
| Sepsis / Infection | 626 | 456 | 72.8 | 170 | 27.2 | 366 | 137 | 37.4 | 229 | 62.6 | |||
| Thromboembolism | 133 | 88 | 66.2 | 45 | 33.8 | 68 | 17 | 25.0 | 51 | 75.0 | |||
| Trauma / Toxicology | 248 | 183 | 73.8 | 65 | 26.2 | 169 | 60 | 35.5 | 109 | 64.5 | |||