Chart Review Table

USC Biomedical Student Research Seminar

library(readxl)
library(dplyr)
library(purrr)
library(tibble)
library(gtsummary)
library(gt)
library(ggplot2)

1 Introduction

This code was generated for the 2026 Biomedical Research Seminar at the University of Southern California by Jonathan Nelson. It creates a descriptive summary table comparing patient, perioperative, pathologic, and postoperative outcomes between surgical approaches, and includes p-values and statistical test labels for each comparison.

The analysis uses a de-identified dataset provided by Sammie Rosen. The dataset includes demographic, clinical, operative, histologic, and follow-up variables collected for patients undergoing extraperitoneal versus transvesical simple prostatectomy.

1.1 Download folder with files for analysis

Download the Folder

1.2 Load in the Dataset

df <- read_excel("MASTER EP vs TV simples - deidentified.xlsx", sheet = "Sheet1")

df

1.3 Name and Factor Table

df <- read_excel("MASTER EP vs TV simples - deidentified.xlsx", sheet = "Sheet1") %>%
  mutate(
    approach = factor(
      `Surgical Approach (1=EP or 2=TV)`,
      levels = c(1, 2),
      labels = c("Extraperitoneal", "Transvesical")
    ),

    `HTN status (0=No or 1=Yes)` = factor(`HTN status (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `DM2 status (0=No or 1=Yes)` = factor(`DM2 status (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `HLD (0=No or 1=Yes)` = factor(`HLD (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `prior Inguinal Hernia/pelvic surgery (0=No or 1=Yes)` = factor(`prior Inguinal Hernia/pelvic surgery (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `Prior Abdominal Surgery (0=No or 1=Yes)` = factor(`Prior Abdominal Surgery (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),

    `Pre-Op Biopsy Done (0=No 1=Yes)` = factor(`Pre-Op Biopsy Done (0=No 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `Pre-Op Biopsy Result (0=Neg 1=Pos)` = factor(`Pre-Op Biopsy Result (0=Neg 1=Pos)`, levels = c(0, 1), labels = c("Negative", "Positive")),
    `Pre-Op Foley Present (0=No 1=Yes)` = factor(`Pre-Op Foley Present (0=No 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),

    `Lymph Node Dissection (0=No or 1=Yes)` = factor(`Lymph Node Dissection (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `Conversion (0=No or 1=Yes)` = factor(`Conversion (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `Drain placement (0=No or 1=Yes)` = factor(`Drain placement (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `pRBCs given (0=No 1=Yes)` = factor(`pRBCs given (0=No 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),

    `LOS (0=Same Day 1=Inpatient)` = factor(`LOS (0=Same Day 1=Inpatient)`, levels = c(0, 1), labels = c("Same Day", "Inpatient")),
    `Narcotics (PO day 0-1) (0 = no, 1 = yes)` = factor(`Narcotics (PO day 0-1) (0 = no, 1 = yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `30-Day Readmission (0=No or 1=Yes)` = factor(`30-Day Readmission (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `Early Continence at 6 Weeks (0=No or 1=Yes) continent = no pads or 1 protection pad` =
      factor(`Early Continence at 6 Weeks (0=No or 1=Yes) continent = no pads or 1 protection pad`,
             levels = c(0, 1), labels = c("No", "Yes")),
    `Continence 3 Months (0=No or 1=Yes)` = factor(`Continence 3 Months (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `30 day Complications (0=no 1=Yes)` = factor(`30 day Complications (0=no 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),

    `ASA (1-5)` = factor(`ASA (1-5)`)
  )

df 

1.4 Let’s Look at One of the Variables

1.4.1 Age at Time of Surgery

df$`Age (at time of surgery)`

##  [1] 72 73 61 83 63 69 67 75 72 73 76 71 74 84 62 70 64 67 88 75 82 79 76 67 74
## [26] 59 71 74 80 74 76 72 65 62 74 76 80 72 78 81 65 68 68 77 85 80 70 75 67 72
## [51] 71 75 64 58 74 64 66 58

1.4.1.1 Calculate the Statistic

df %>%
  group_by(approach) %>%
  summarise(
    mean_age = mean(`Age (at time of surgery)`, na.rm = TRUE),
    sd_age   = sd(`Age (at time of surgery)`, na.rm = TRUE),
    .groups = "drop"
  )

wilcox.test(`Age (at time of surgery)` ~ approach, data = df)

## 
##  Wilcoxon rank sum test with continuity correction
## 
## data:  Age (at time of surgery) by approach
## W = 441.5, p-value = 0.654
## alternative hypothesis: true location shift is not equal to 0

summary_df <- df %>%
  group_by(approach) %>%
  summarise(
    mean_age = mean(`Age (at time of surgery)`, na.rm = TRUE),
    .groups = "drop"
  )

ggplot(df, aes(x = approach, y = `Age (at time of surgery)`)) +
  geom_jitter(width = 0.15, alpha = 0.6, size = 4) +
  geom_point(
    data = summary_df,
    aes(x = approach, y = mean_age),
    color = "red",
    size = 6
  ) +
  labs(
    x = "Approach",
    y = "Age (at time of surgery)",
    title = "Age Distribution by Surgical Approach (with Means)"
  ) +
  theme_classic()

1.4.1.2 Putting It Into a Table

# Summary stats
age_summary <- df %>%
  group_by(approach) %>%
  summarise(
    mean = mean(`Age (at time of surgery)`, na.rm = TRUE),
    sd   = sd(`Age (at time of surgery)`, na.rm = TRUE),
    .groups = "drop"
  )

# Format into one row
age_tbl <- tibble(
  Variable = "Age (yr)",
  Extraperitoneal = sprintf("%.1f (%.1f)",
                            age_summary$mean[age_summary$approach == "Extraperitoneal"],
                            age_summary$sd[age_summary$approach == "Extraperitoneal"]),
  Transvesical = sprintf("%.1f (%.1f)",
                         age_summary$mean[age_summary$approach == "Transvesical"],
                         age_summary$sd[age_summary$approach == "Transvesical"]),
  `P-value` = formatC(
    wilcox.test(`Age (at time of surgery)` ~ approach, data = df)$p.value,
    format = "f", digits = 3
  ),
  Test = "Wilcoxon rank-sum"
)

age_tbl %>%
  gt()

Variable	Extraperitoneal	Transvesical	P-value	Test
Age (yr)	72.4 (6.8)	71.2 (7.2)	0.654	Wilcoxon rank-sum

1.5 Working Table

df <- read_excel("MASTER EP vs TV simples - deidentified.xlsx", sheet = "Sheet1") %>%
  mutate(
    approach = factor(
      `Surgical Approach (1=EP or 2=TV)`,
      levels = c(1, 2),
      labels = c("Extraperitoneal", "Transvesical")
    ),

    `HTN status (0=No or 1=Yes)` = factor(`HTN status (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `DM2 status (0=No or 1=Yes)` = factor(`DM2 status (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `HLD (0=No or 1=Yes)` = factor(`HLD (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `prior Inguinal Hernia/pelvic surgery (0=No or 1=Yes)` = factor(`prior Inguinal Hernia/pelvic surgery (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `Prior Abdominal Surgery (0=No or 1=Yes)` = factor(`Prior Abdominal Surgery (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),

    `Pre-Op Biopsy Done (0=No 1=Yes)` = factor(`Pre-Op Biopsy Done (0=No 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `Pre-Op Biopsy Result (0=Neg 1=Pos)` = factor(`Pre-Op Biopsy Result (0=Neg 1=Pos)`, levels = c(0, 1), labels = c("Negative", "Positive")),
    `Pre-Op Foley Present (0=No 1=Yes)` = factor(`Pre-Op Foley Present (0=No 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),

    `Lymph Node Dissection (0=No or 1=Yes)` = factor(`Lymph Node Dissection (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `Conversion (0=No or 1=Yes)` = factor(`Conversion (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `Drain placement (0=No or 1=Yes)` = factor(`Drain placement (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `pRBCs given (0=No 1=Yes)` = factor(`pRBCs given (0=No 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),

    `LOS (0=Same Day 1=Inpatient)` = factor(`LOS (0=Same Day 1=Inpatient)`, levels = c(0, 1), labels = c("Same Day", "Inpatient")),
    `Narcotics (PO day 0-1) (0 = no, 1 = yes)` = factor(`Narcotics (PO day 0-1) (0 = no, 1 = yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `30-Day Readmission (0=No or 1=Yes)` = factor(`30-Day Readmission (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `Early Continence at 6 Weeks (0=No or 1=Yes) continent = no pads or 1 protection pad` =
      factor(`Early Continence at 6 Weeks (0=No or 1=Yes) continent = no pads or 1 protection pad`,
             levels = c(0, 1), labels = c("No", "Yes")),
    `Continence 3 Months (0=No or 1=Yes)` = factor(`Continence 3 Months (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `30 day Complications (0=no 1=Yes)` = factor(`30 day Complications (0=no 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),

    `ASA (1-5)` = factor(`ASA (1-5)`)
  )

baseline_labels <- list(
  `Age (at time of surgery)` ~ "Age (yr)",
  BMI ~ "BMI (kg/m^2)",
  CCI ~ "CCI",
  `ASA (1-5)` ~ "ASA physical status grading",
  `HTN status (0=No or 1=Yes)` ~ "History of hypertension",
  `DM2 status (0=No or 1=Yes)` ~ "History of type 2 diabetes mellitus",
  `HLD (0=No or 1=Yes)` ~ "History of hyperlipidemia",
  `prior Inguinal Hernia/pelvic surgery (0=No or 1=Yes)` ~ "History of prior inguinal/pelvic surgery",
  `Prior Abdominal Surgery (0=No or 1=Yes)` ~ "History of prior abdominal surgery",
  `Pre-op Hgb` ~ "Pre-op Hgb",
  `Pre-op PSA` ~ "Pre-op PSA (ng/mL)",
  `Pre-op Prostate Volume (g)` ~ "Prostate volume (ml)"
)

preop_labels <- list(
  `SHIM Score (pre-op)` ~ "SHIM",
  `IPSS (pre-op)` ~ "IPSS",
  `QOL (pre-op)` ~ "QoL",
  `Pre-Op Biopsy Done (0=No 1=Yes)` ~ "Biopsy done",
  `Pre-Op Biopsy Result (0=Neg 1=Pos)` ~ "Malignancy on biopsy",
  `Pre-Op Foley Present (0=No 1=Yes)` ~ "Pre-op foley placed"
)

intraop_labels <- list(
  `Console Time (minutes)` ~ "Console time (minutes)",
  `Operative Time (minutes)` ~ "Operative time (minutes)",
  `EBL (cc)` ~ "Estimated blood loss (mL)",
  `Conversion (0=No or 1=Yes)` ~ "Conversion to multiport",
  `Lymph Node Dissection (0=No or 1=Yes)` ~ "Lymph node dissection",
  `Drain placement (0=No or 1=Yes)` ~ "Drain placement",
  `pRBCs given (0=No 1=Yes)` ~ "Blood units given",
  `delta Hgb` ~ "dHgb (%)"
)

histology_labels <- list(
  `Specimen Weight (g)` ~ "Specimen weight (g)",
  `Specimen weight/pre-op estmated weight` ~ "Specimen wt/pre-op estimated wt",
  `Pathologic Diagnosis (incidental malignancy) 0= benign 1= malignant` ~ "Incidental malignancy finding"
)

postop_care_labels <- list(
  `LOS (0=Same Day 1=Inpatient)` ~ "Length of stay <24 hours",
  `LOS (hours) (look at note times)` ~ "Average length of stay (hours)",
  `Narcotics (PO day 0-1) (0 = no, 1 = yes)` ~ "Narcotics use post-op day 0-1",
  `Postop Pain Score (PACU)` ~ "Post Op Pain Score",
  `Foley Duration (Days post-op)` ~ "Foley catheter duration (days)"
)

outcome_labels <- list(
  `30 day Complications (0=no 1=Yes)` ~ "30 day postoperative complications (Clavien >=2 only)",
  `30-Day Readmission (0=No or 1=Yes)` ~ "30 day readmission",
  `Post-op PVR` ~ "Post-op PVR",
  `Early Continence at 6 Weeks (0=No or 1=Yes) continent = no pads or 1 protection pad` ~ "Early Continence at 4 weeks",
  `Continence 3 Months (0=No or 1=Yes)` ~ "Continence at 3 months"
)

format_p <- function(x) {
  dplyr::case_when(
    is.na(x) ~ NA_character_,
    x < 0.001 ~ "<0.001",
    TRUE ~ formatC(x, format = "f", digits = 3)
  )
}

safe_pvalue <- function(data, group_var, var, continuous = FALSE) {
  x <- data[[var]]
  g <- data[[group_var]]
  keep <- complete.cases(x, g)
  x <- x[keep]
  g <- g[keep]

  if (length(unique(g)) < 2) return(NA_real_)

  tryCatch({
    if (continuous) {
      wilcox.test(x ~ g)$p.value
    } else {
      tab <- table(x, g)
      if (any(dim(tab) < 2)) return(NA_real_)
      if (all(dim(tab) == c(2, 2))) {
        fisher.test(tab)$p.value
      } else {
        suppressWarnings(chisq.test(tab, correct = FALSE)$p.value)
      }
    }
  }, error = function(e) NA_real_)
}

calc_manual_pvals <- function(data, group_var, vars, continuous_vars) {
  map_dfr(vars, function(v) {
    tibble(
      variable = v,
      manual_p = safe_pvalue(data, group_var, v, continuous = v %in% continuous_vars)
    )
  })
}

attach_manual_pvals <- function(tbl, pvals) {
  tbl %>%
    modify_table_body(
      ~ .x %>%
        left_join(pvals, by = "variable") %>%
        mutate(
          p.value = if_else(
            row_type == "label",
            format_p(manual_p),
            NA_character_
          )
        ) %>%
        select(-manual_p)
    ) %>%
    modify_header(p.value ~ "**P-value**") %>%
    modify_column_unhide(p.value)
}

make_tbl <- function(data, vars, labels, statistic, continuous_vars) {
  pvals <- calc_manual_pvals(data, "approach", vars, continuous_vars)

  data %>%
    select(approach, all_of(vars)) %>%
    tbl_summary(
      by = approach,
      label = labels,
      statistic = statistic,
      digits = all_continuous() ~ 1,
      missing = "ifany"
    ) %>%
    attach_manual_pvals(pvals)
}

baseline_vars <- c(
  "Age (at time of surgery)", "BMI", "CCI", "ASA (1-5)",
  "HTN status (0=No or 1=Yes)", "DM2 status (0=No or 1=Yes)", "HLD (0=No or 1=Yes)",
  "prior Inguinal Hernia/pelvic surgery (0=No or 1=Yes)",
  "Prior Abdominal Surgery (0=No or 1=Yes)",
  "Pre-op Hgb", "Pre-op PSA", "Pre-op Prostate Volume (g)"
)
baseline_cont <- c(
  "Age (at time of surgery)", "BMI", "CCI",
  "Pre-op Hgb", "Pre-op PSA", "Pre-op Prostate Volume (g)"
)

preop_vars <- c(
  "SHIM Score (pre-op)", "IPSS (pre-op)", "QOL (pre-op)",
  "Pre-Op Biopsy Done (0=No 1=Yes)",
  "Pre-Op Biopsy Result (0=Neg 1=Pos)",
  "Pre-Op Foley Present (0=No 1=Yes)"
)
preop_cont <- c("SHIM Score (pre-op)", "IPSS (pre-op)", "QOL (pre-op)")

intraop_vars <- c(
  "Console Time (minutes)", "Operative Time (minutes)", "EBL (cc)",
  "Conversion (0=No or 1=Yes)",
  "Lymph Node Dissection (0=No or 1=Yes)",
  "Drain placement (0=No or 1=Yes)",
  "pRBCs given (0=No 1=Yes)",
  "delta Hgb"
)
intraop_cont <- c("Console Time (minutes)", "Operative Time (minutes)", "EBL (cc)", "delta Hgb")

histology_vars <- c(
  "Specimen Weight (g)",
  "Specimen weight/pre-op estmated weight",
  "Pathologic Diagnosis (incidental malignancy) 0= benign 1= malignant"
)
histology_cont <- c("Specimen Weight (g)", "Specimen weight/pre-op estmated weight")

postop_care_vars <- c(
  "LOS (0=Same Day 1=Inpatient)",
  "LOS (hours) (look at note times)",
  "Narcotics (PO day 0-1) (0 = no, 1 = yes)",
  "Postop Pain Score (PACU)",
  "Foley Duration (Days post-op)"
)
postop_care_cont <- c("LOS (hours) (look at note times)", "Postop Pain Score (PACU)", "Foley Duration (Days post-op)")

outcome_vars <- c(
  "30 day Complications (0=no 1=Yes)",
  "30-Day Readmission (0=No or 1=Yes)",
  "Post-op PVR",
  "Early Continence at 6 Weeks (0=No or 1=Yes) continent = no pads or 1 protection pad",
  "Continence 3 Months (0=No or 1=Yes)"
)
outcome_cont <- c("Post-op PVR")

baseline_tbl <- make_tbl(
  df,
  vars = baseline_vars,
  labels = baseline_labels,
  statistic = list(
    all_continuous() ~ "{mean} ({sd})",
    `ASA (1-5)` ~ "{n} ({p}%)",
    all_categorical() ~ "{n} ({p}%)"
  ),
  continuous_vars = baseline_cont
)

preop_tbl <- make_tbl(
  df,
  vars = preop_vars,
  labels = preop_labels,
  statistic = list(
    c(`SHIM Score (pre-op)`, `IPSS (pre-op)`, `QOL (pre-op)`) ~ "{median} ({p25}, {p75})",
    all_categorical() ~ "{n} ({p}%)"
  ),
  continuous_vars = preop_cont
)

intraop_tbl <- make_tbl(
  df,
  vars = intraop_vars,
  labels = intraop_labels,
  statistic = list(
    all_continuous() ~ "{mean} ({sd})",
    all_categorical() ~ "{n} ({p}%)"
  ),
  continuous_vars = intraop_cont
)

histology_tbl <- make_tbl(
  df,
  vars = histology_vars,
  labels = histology_labels,
  statistic = list(
    all_continuous() ~ "{mean} ({sd})",
    all_categorical() ~ "{n} ({p}%)"
  ),
  continuous_vars = histology_cont
)

postop_care_tbl <- make_tbl(
  df,
  vars = postop_care_vars,
  labels = postop_care_labels,
  statistic = list(
    `LOS (hours) (look at note times)` ~ "{mean} ({sd})",
    `Postop Pain Score (PACU)` ~ "{mean} ({sd})",
    `Foley Duration (Days post-op)` ~ "{median} ({p25}, {p75})",
    all_categorical() ~ "{n} ({p}%)"
  ),
  continuous_vars = postop_care_cont
)

outcomes_tbl <- make_tbl(
  df,
  vars = outcome_vars,
  labels = outcome_labels,
  statistic = list(
    `Post-op PVR` ~ "{median} ({p25}, {p75})",
    all_categorical() ~ "{n} ({p}%)"
  ),
  continuous_vars = outcome_cont
)

final_tbl <- tbl_stack(
  tbls = list(
    baseline_tbl,
    preop_tbl,
    intraop_tbl,
    histology_tbl,
    postop_care_tbl,
    outcomes_tbl
  ),
  group_header = c(
    "Baseline characteristics",
    "Pre-operative variables",
    "Intraoperative parameters",
    "Histopathology",
    "Postoperative care",
    "Postoperative outcomes"
  )
) %>%
  modify_column_unhide(p.value) %>%
  bold_labels()

final_tbl

Characteristic	Extraperitoneal N = 331	Transvesical N = 251	P-value
Baseline characteristics
Age (yr)	72.4 (6.8)	71.2 (7.2)	0.654
BMI (kg/m^2)	27.6 (5.2)	26.9 (4.6)	0.668
Unknown	2	0
CCI			0.452
1	1 (3.1%)	2 (8.0%)
2	6 (19%)	6 (24%)
3	16 (50%)	12 (48%)
4	9 (28%)	2 (8.0%)
5	0 (0%)	2 (8.0%)
6	0 (0%)	1 (4.0%)
Unknown	1	0
ASA physical status grading			0.457
2	17 (55%)	17 (68%)
3	13 (42%)	8 (32%)
4	1 (3.2%)	0 (0%)
Unknown	2	0
History of hypertension	21 (64%)	11 (44%)	0.185
History of type 2 diabetes mellitus	6 (18%)	4 (16%)	1.000
History of hyperlipidemia	22 (67%)	13 (52%)	0.290
History of prior inguinal/pelvic surgery	5 (15%)	4 (16%)	1.000
History of prior abdominal surgery	16 (48%)	2 (8.0%)	0.001
Pre-op Hgb	13.6 (1.4)	12.9 (1.5)	0.097
Unknown	7	0
Pre-op PSA (ng/mL)	7.4 (7.9)	6.5 (5.1)	1.000
Unknown	8	8
Prostate volume (ml)	125.2 (49.4)	120.3 (37.4)	0.975
Pre-operative variables
SHIM	12.0 (3.0, 16.0)	15.0 (7.5, 19.5)	0.384
Unknown	8	9
IPSS	20.0 (12.0, 25.0)	19.0 (15.0, 24.0)	0.811
Unknown	4	8
QoL			0.307
1	2 (13%)	1 (6.3%)
2	1 (6.3%)	2 (13%)
3	3 (19%)	4 (25%)
4	2 (13%)	6 (38%)
5	2 (13%)	1 (6.3%)
6	6 (38%)	2 (13%)
Unknown	17	9
Biopsy done	6 (18%)	6 (24%)	0.745
Malignancy on biopsy			1.000
Negative	6 (100%)	6 (86%)
Positive	0 (0%)	1 (14%)
Unknown	27	18
Pre-op foley placed	13 (39%)	14 (56%)	0.289
Intraoperative parameters
Console time (minutes)	146.1 (55.2)	138.1 (51.4)	0.701
Unknown	18	5
Operative time (minutes)	181.2 (57.3)	190.6 (45.2)	0.192
Unknown	14	0
Estimated blood loss (mL)			0.027
25	1 (3.1%)	0 (0%)
50	10 (31%)	4 (16%)
75	5 (16%)	1 (4.0%)
100	9 (28%)	7 (28%)
125	1 (3.1%)	2 (8.0%)
150	2 (6.3%)	7 (28%)
200	1 (3.1%)	4 (16%)
300	2 (6.3%)	0 (0%)
500	1 (3.1%)	0 (0%)
Unknown	1	0
Conversion to multiport	0 (0%)	0 (0%)	1.000
Lymph node dissection	0 (0%)	0 (0%)	1.000
Drain placement	1 (3.0%)	0 (0%)	1.000
Blood units given	0 (0%)	0 (0%)	1.000
dHgb (%)	-0.8 (0.7)	-0.6 (1.1)	0.612
Unknown	14	8
Histopathology
Specimen weight (g)	74.7 (40.6)	64.2 (28.8)	0.504
Unknown	2	0
Specimen wt/pre-op estimated wt	0.6 (0.1)	0.5 (0.2)	0.255
Unknown	2	0
Incidental malignancy finding	5 (16%)	3 (12%)	1.000
Unknown	1	0
Postoperative care
Length of stay <24 hours			0.377
Same Day	11 (55%)	10 (40%)
Inpatient	9 (45%)	15 (60%)
Unknown	13	0
Average length of stay (hours)	18.4 (14.3)	18.4 (9.7)	0.423
Unknown	13	0
Narcotics use post-op day 0-1	1 (4.0%)	5 (20%)	0.189
Unknown	8	0
Post Op Pain Score			0.009
0	1 (6.3%)	8 (35%)
1	2 (13%)	2 (8.7%)
2	1 (6.3%)	2 (8.7%)
3	0 (0%)	2 (8.7%)
4	3 (19%)	2 (8.7%)
5	1 (6.3%)	5 (22%)
6	1 (6.3%)	1 (4.3%)
7	2 (13%)	0 (0%)
8	5 (31%)	1 (4.3%)
Unknown	17	2
Foley catheter duration (days)	9.0 (6.0, 10.0)	7.0 (6.0, 9.0)	0.204
Unknown	7	2
Postoperative outcomes
30 day postoperative complications (Clavien >=2 only)	4 (15%)	4 (17%)	1.000
Unknown	6	2
30 day readmission	1 (4.0%)	1 (4.2%)	1.000
Unknown	8	1
Post-op PVR	9.0 (0.0, 19.0)	5.0 (0.0, 18.0)	0.846
Unknown	6	18
Early Continence at 4 weeks	27 (90%)	21 (95%)	0.629
Unknown	3	3
Continence at 3 months	20 (100%)	22 (100%)	1.000
Unknown	13	3
1 Mean (SD); n (%)

1.6 Breaking Down the Code

This script:
1. Reads the data from Excel
2. Recodes raw 0/1 variables into readable categories
3. Groups variables into clinical sections
4. Builds summary statistics for each section
5. Computes p-values comparing the two surgical approaches
6. Combines everything into one polished table

1.6.1 Read the Excel File and Create New Variable

df <- read_excel(...) %>%
  mutate(...)

1.6.2 Create Label Lists

baseline_labels
preop_labels
intraop_labels
histology_labels
postop_care_labels
outcome_labels

`Age (at time of surgery)` ~ "Age (yr)"

1.6.3 Format the p-value

format_p <- function(x) { ... }

1.6.4 Compute p-value safely

This is the section that was the most challenging.

1.6.4.1 Wilcox for continuous variable

1.6.4.2 Fisher/Chi-squared for categorical variable

safe_pvalue <- function(data, group_var, var, continuous = FALSE) { ... }

wilcox.test(x ~ g)$p.value

tab <- table(x, g)

1.6.5 Build a p-value table for many variables

calc_manual_pvals <- function(data, group_var, vars, continuous_vars) { ... }

1.6.6 Attach p-values into the gtsummary table

attach_manual_pvals <- function(tbl, pvals) { ... }

1.6.7 Create a reusable table-making function

make_tbl <- function(data, vars, labels, statistic, continuous_vars) { ... }

1.6.8 Define which variables belong in each section

make_tbl <- function(data, vars, labels, statistic, continuous_vars) { ... }

1.6.9 Create each summary table section

baseline_tbl <- make_tbl(...)

1.6.10 Stack all sections into one final table

final_tbl <- tbl_stack(
  tbls = list(...)
  group_header = c(...)
)

1.6.11 Display the final table

final_tbl)

1.7 Table with Statistics Info

I wanted an indication of which statistic test was used for each comparison.

df <- read_excel("MASTER EP vs TV simples - deidentified.xlsx", sheet = "Sheet1") %>%
  mutate(
    approach = factor(
      `Surgical Approach (1=EP or 2=TV)`,
      levels = c(1, 2),
      labels = c("Extraperitoneal", "Transvesical")
    ),

    `HTN status (0=No or 1=Yes)` = factor(`HTN status (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `DM2 status (0=No or 1=Yes)` = factor(`DM2 status (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `HLD (0=No or 1=Yes)` = factor(`HLD (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `prior Inguinal Hernia/pelvic surgery (0=No or 1=Yes)` = factor(`prior Inguinal Hernia/pelvic surgery (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `Prior Abdominal Surgery (0=No or 1=Yes)` = factor(`Prior Abdominal Surgery (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),

    `Pre-Op Biopsy Done (0=No 1=Yes)` = factor(`Pre-Op Biopsy Done (0=No 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `Pre-Op Biopsy Result (0=Neg 1=Pos)` = factor(`Pre-Op Biopsy Result (0=Neg 1=Pos)`, levels = c(0, 1), labels = c("Negative", "Positive")),
    `Pre-Op Foley Present (0=No 1=Yes)` = factor(`Pre-Op Foley Present (0=No 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),

    `Lymph Node Dissection (0=No or 1=Yes)` = factor(`Lymph Node Dissection (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `Conversion (0=No or 1=Yes)` = factor(`Conversion (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `Drain placement (0=No or 1=Yes)` = factor(`Drain placement (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `pRBCs given (0=No 1=Yes)` = factor(`pRBCs given (0=No 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),

    `LOS (0=Same Day 1=Inpatient)` = factor(`LOS (0=Same Day 1=Inpatient)`, levels = c(0, 1), labels = c("Same Day", "Inpatient")),
    `Narcotics (PO day 0-1) (0 = no, 1 = yes)` = factor(`Narcotics (PO day 0-1) (0 = no, 1 = yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `30-Day Readmission (0=No or 1=Yes)` = factor(`30-Day Readmission (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `Early Continence at 6 Weeks (0=No or 1=Yes) continent = no pads or 1 protection pad` =
      factor(`Early Continence at 6 Weeks (0=No or 1=Yes) continent = no pads or 1 protection pad`,
             levels = c(0, 1), labels = c("No", "Yes")),
    `Continence 3 Months (0=No or 1=Yes)` = factor(`Continence 3 Months (0=No or 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),
    `30 day Complications (0=no 1=Yes)` = factor(`30 day Complications (0=no 1=Yes)`, levels = c(0, 1), labels = c("No", "Yes")),

    `ASA (1-5)` = factor(`ASA (1-5)`)
  )

baseline_labels <- list(
  `Age (at time of surgery)` ~ "Age (yr)",
  BMI ~ "BMI (kg/m^2)",
  CCI ~ "CCI",
  `ASA (1-5)` ~ "ASA physical status grading",
  `HTN status (0=No or 1=Yes)` ~ "History of hypertension",
  `DM2 status (0=No or 1=Yes)` ~ "History of type 2 diabetes mellitus",
  `HLD (0=No or 1=Yes)` ~ "History of hyperlipidemia",
  `prior Inguinal Hernia/pelvic surgery (0=No or 1=Yes)` ~ "History of prior inguinal/pelvic surgery",
  `Prior Abdominal Surgery (0=No or 1=Yes)` ~ "History of prior abdominal surgery",
  `Pre-op Hgb` ~ "Pre-op Hgb",
  `Pre-op PSA` ~ "Pre-op PSA (ng/mL)",
  `Pre-op Prostate Volume (g)` ~ "Prostate volume (ml)"
)

preop_labels <- list(
  `SHIM Score (pre-op)` ~ "SHIM",
  `IPSS (pre-op)` ~ "IPSS",
  `QOL (pre-op)` ~ "QoL",
  `Pre-Op Biopsy Done (0=No 1=Yes)` ~ "Biopsy done",
  `Pre-Op Biopsy Result (0=Neg 1=Pos)` ~ "Malignancy on biopsy",
  `Pre-Op Foley Present (0=No 1=Yes)` ~ "Pre-op foley placed"
)

intraop_labels <- list(
  `Console Time (minutes)` ~ "Console time (minutes)",
  `Operative Time (minutes)` ~ "Operative time (minutes)",
  `EBL (cc)` ~ "Estimated blood loss (mL)",
  `Conversion (0=No or 1=Yes)` ~ "Conversion to multiport",
  `Lymph Node Dissection (0=No or 1=Yes)` ~ "Lymph node dissection",
  `Drain placement (0=No or 1=Yes)` ~ "Drain placement",
  `pRBCs given (0=No 1=Yes)` ~ "Blood units given",
  `delta Hgb` ~ "dHgb (%)"
)

histology_labels <- list(
  `Specimen Weight (g)` ~ "Specimen weight (g)",
  `Specimen weight/pre-op estmated weight` ~ "Specimen wt/pre-op estimated wt",
  `Pathologic Diagnosis (incidental malignancy) 0= benign 1= malignant` ~ "Incidental malignancy finding"
)

postop_care_labels <- list(
  `LOS (0=Same Day 1=Inpatient)` ~ "Length of stay <24 hours",
  `LOS (hours) (look at note times)` ~ "Average length of stay (hours)",
  `Narcotics (PO day 0-1) (0 = no, 1 = yes)` ~ "Narcotics use post-op day 0-1",
  `Postop Pain Score (PACU)` ~ "Post Op Pain Score",
  `Foley Duration (Days post-op)` ~ "Foley catheter duration (days)"
)

outcome_labels <- list(
  `30 day Complications (0=no 1=Yes)` ~ "30 day postoperative complications (Clavien >=2 only)",
  `30-Day Readmission (0=No or 1=Yes)` ~ "30 day readmission",
  `Post-op PVR` ~ "Post-op PVR",
  `Early Continence at 6 Weeks (0=No or 1=Yes) continent = no pads or 1 protection pad` ~ "Early Continence at 4 weeks",
  `Continence 3 Months (0=No or 1=Yes)` ~ "Continence at 3 months"
)

format_p <- function(x) {
  dplyr::case_when(
    is.na(x) ~ NA_character_,
    x < 0.001 ~ "<0.001",
    TRUE ~ formatC(x, format = "f", digits = 3)
  )
}

safe_test <- function(data, group_var, var, continuous = FALSE) {
  x <- data[[var]]
  g <- data[[group_var]]
  keep <- complete.cases(x, g)
  x <- x[keep]
  g <- g[keep]

  if (length(unique(g)) < 2) {
    return(list(p = NA_real_, test = NA_character_))
  }

  tryCatch({
    if (continuous) {
      p <- wilcox.test(x ~ g)$p.value
      list(p = p, test = "Wilcoxon rank-sum")
    } else {
      tab <- table(x, g)
      if (any(dim(tab) < 2)) {
        return(list(p = NA_real_, test = NA_character_))
      }
      if (all(dim(tab) == c(2, 2))) {
        p <- fisher.test(tab)$p.value
        list(p = p, test = "Fisher's exact")
      } else {
        p <- suppressWarnings(chisq.test(tab, correct = FALSE)$p.value)
        list(p = p, test = "Chi-square")
      }
    }
  }, error = function(e) {
    list(p = NA_real_, test = NA_character_)
  })
}

calc_manual_stats <- function(data, group_var, vars, continuous_vars) {
  map_dfr(vars, function(v) {
    res <- safe_test(data, group_var, v, continuous = v %in% continuous_vars)
    tibble(
      variable = v,
      manual_p = res$p,
      manual_test = res$test
    )
  })
}

attach_manual_stats <- function(tbl, stats) {
  tbl %>%
    modify_table_body(
      ~ .x %>%
        left_join(stats, by = "variable") %>%
        mutate(
          p.value = if_else(row_type == "label", format_p(manual_p), NA_character_),
          stat_test = if_else(row_type == "label", manual_test, NA_character_)
        ) %>%
        select(-manual_p, -manual_test)
    ) %>%
    modify_header(
      p.value ~ "**P-value**",
      stat_test ~ "**Test used**"
    ) %>%
    modify_column_unhide(c(p.value, stat_test))
}

make_tbl <- function(data, vars, labels, statistic, continuous_vars) {
  stats <- calc_manual_stats(data, "approach", vars, continuous_vars)

  data %>%
    select(approach, all_of(vars)) %>%
    tbl_summary(
      by = approach,
      label = labels,
      statistic = statistic,
      digits = all_continuous() ~ 1,
      missing = "ifany"
    ) %>%
    attach_manual_stats(stats)
}

baseline_vars <- c(
  "Age (at time of surgery)", "BMI", "CCI", "ASA (1-5)",
  "HTN status (0=No or 1=Yes)", "DM2 status (0=No or 1=Yes)", "HLD (0=No or 1=Yes)",
  "prior Inguinal Hernia/pelvic surgery (0=No or 1=Yes)",
  "Prior Abdominal Surgery (0=No or 1=Yes)",
  "Pre-op Hgb", "Pre-op PSA", "Pre-op Prostate Volume (g)"
)
baseline_cont <- c(
  "Age (at time of surgery)", "BMI", "CCI",
  "Pre-op Hgb", "Pre-op PSA", "Pre-op Prostate Volume (g)"
)

preop_vars <- c(
  "SHIM Score (pre-op)", "IPSS (pre-op)", "QOL (pre-op)",
  "Pre-Op Biopsy Done (0=No 1=Yes)",
  "Pre-Op Biopsy Result (0=Neg 1=Pos)",
  "Pre-Op Foley Present (0=No 1=Yes)"
)
preop_cont <- c("SHIM Score (pre-op)", "IPSS (pre-op)", "QOL (pre-op)")

intraop_vars <- c(
  "Console Time (minutes)", "Operative Time (minutes)", "EBL (cc)",
  "Conversion (0=No or 1=Yes)",
  "Lymph Node Dissection (0=No or 1=Yes)",
  "Drain placement (0=No or 1=Yes)",
  "pRBCs given (0=No 1=Yes)",
  "delta Hgb"
)
intraop_cont <- c("Console Time (minutes)", "Operative Time (minutes)", "EBL (cc)", "delta Hgb")

histology_vars <- c(
  "Specimen Weight (g)",
  "Specimen weight/pre-op estmated weight",
  "Pathologic Diagnosis (incidental malignancy) 0= benign 1= malignant"
)
histology_cont <- c("Specimen Weight (g)", "Specimen weight/pre-op estmated weight")

postop_care_vars <- c(
  "LOS (0=Same Day 1=Inpatient)",
  "LOS (hours) (look at note times)",
  "Narcotics (PO day 0-1) (0 = no, 1 = yes)",
  "Postop Pain Score (PACU)",
  "Foley Duration (Days post-op)"
)
postop_care_cont <- c("LOS (hours) (look at note times)", "Postop Pain Score (PACU)", "Foley Duration (Days post-op)")

outcome_vars <- c(
  "30 day Complications (0=no 1=Yes)",
  "30-Day Readmission (0=No or 1=Yes)",
  "Post-op PVR",
  "Early Continence at 6 Weeks (0=No or 1=Yes) continent = no pads or 1 protection pad",
  "Continence 3 Months (0=No or 1=Yes)"
)
outcome_cont <- c("Post-op PVR")

baseline_tbl <- make_tbl(
  df,
  vars = baseline_vars,
  labels = baseline_labels,
  statistic = list(
    all_continuous() ~ "{mean} ({sd})",
    `ASA (1-5)` ~ "{n} ({p}%)",
    all_categorical() ~ "{n} ({p}%)"
  ),
  continuous_vars = baseline_cont
)

preop_tbl <- make_tbl(
  df,
  vars = preop_vars,
  labels = preop_labels,
  statistic = list(
    c(`SHIM Score (pre-op)`, `IPSS (pre-op)`, `QOL (pre-op)`) ~ "{median} ({p25}, {p75})",
    all_categorical() ~ "{n} ({p}%)"
  ),
  continuous_vars = preop_cont
)

intraop_tbl <- make_tbl(
  df,
  vars = intraop_vars,
  labels = intraop_labels,
  statistic = list(
    all_continuous() ~ "{mean} ({sd})",
    all_categorical() ~ "{n} ({p}%)"
  ),
  continuous_vars = intraop_cont
)

histology_tbl <- make_tbl(
  df,
  vars = histology_vars,
  labels = histology_labels,
  statistic = list(
    all_continuous() ~ "{mean} ({sd})",
    all_categorical() ~ "{n} ({p}%)"
  ),
  continuous_vars = histology_cont
)

postop_care_tbl <- make_tbl(
  df,
  vars = postop_care_vars,
  labels = postop_care_labels,
  statistic = list(
    `LOS (hours) (look at note times)` ~ "{mean} ({sd})",
    `Postop Pain Score (PACU)` ~ "{mean} ({sd})",
    `Foley Duration (Days post-op)` ~ "{median} ({p25}, {p75})",
    all_categorical() ~ "{n} ({p}%)"
  ),
  continuous_vars = postop_care_cont
)

outcomes_tbl <- make_tbl(
  df,
  vars = outcome_vars,
  labels = outcome_labels,
  statistic = list(
    `Post-op PVR` ~ "{median} ({p25}, {p75})",
    all_categorical() ~ "{n} ({p}%)"
  ),
  continuous_vars = outcome_cont
)

final_tbl_2 <- tbl_stack(
  tbls = list(
    baseline_tbl,
    preop_tbl,
    intraop_tbl,
    histology_tbl,
    postop_care_tbl,
    outcomes_tbl
  ),
  group_header = c(
    "Baseline characteristics",
    "Pre-operative variables",
    "Intraoperative parameters",
    "Histopathology",
    "Postoperative care",
    "Postoperative outcomes"
  )
) %>%
  modify_column_unhide(c(p.value, stat_test)) %>%
  bold_labels()

final_tbl_2

Characteristic	Extraperitoneal N = 331	Transvesical N = 251	P-value	Test used
Baseline characteristics
Age (yr)	72.4 (6.8)	71.2 (7.2)	0.654	Wilcoxon rank-sum
BMI (kg/m^2)	27.6 (5.2)	26.9 (4.6)	0.668	Wilcoxon rank-sum
Unknown	2	0
CCI			0.452	Wilcoxon rank-sum
1	1 (3.1%)	2 (8.0%)
2	6 (19%)	6 (24%)
3	16 (50%)	12 (48%)
4	9 (28%)	2 (8.0%)
5	0 (0%)	2 (8.0%)
6	0 (0%)	1 (4.0%)
Unknown	1	0
ASA physical status grading			0.457	Chi-square
2	17 (55%)	17 (68%)
3	13 (42%)	8 (32%)
4	1 (3.2%)	0 (0%)
Unknown	2	0
History of hypertension	21 (64%)	11 (44%)	0.185	Fisher's exact
History of type 2 diabetes mellitus	6 (18%)	4 (16%)	1.000	Fisher's exact
History of hyperlipidemia	22 (67%)	13 (52%)	0.290	Fisher's exact
History of prior inguinal/pelvic surgery	5 (15%)	4 (16%)	1.000	Fisher's exact
History of prior abdominal surgery	16 (48%)	2 (8.0%)	0.001	Fisher's exact
Pre-op Hgb	13.6 (1.4)	12.9 (1.5)	0.097	Wilcoxon rank-sum
Unknown	7	0
Pre-op PSA (ng/mL)	7.4 (7.9)	6.5 (5.1)	1.000	Wilcoxon rank-sum
Unknown	8	8
Prostate volume (ml)	125.2 (49.4)	120.3 (37.4)	0.975	Wilcoxon rank-sum
Pre-operative variables
SHIM	12.0 (3.0, 16.0)	15.0 (7.5, 19.5)	0.384	Wilcoxon rank-sum
Unknown	8	9
IPSS	20.0 (12.0, 25.0)	19.0 (15.0, 24.0)	0.811	Wilcoxon rank-sum
Unknown	4	8
QoL			0.307	Wilcoxon rank-sum
1	2 (13%)	1 (6.3%)
2	1 (6.3%)	2 (13%)
3	3 (19%)	4 (25%)
4	2 (13%)	6 (38%)
5	2 (13%)	1 (6.3%)
6	6 (38%)	2 (13%)
Unknown	17	9
Biopsy done	6 (18%)	6 (24%)	0.745	Fisher's exact
Malignancy on biopsy			1.000	Fisher's exact
Negative	6 (100%)	6 (86%)
Positive	0 (0%)	1 (14%)
Unknown	27	18
Pre-op foley placed	13 (39%)	14 (56%)	0.289	Fisher's exact
Intraoperative parameters
Console time (minutes)	146.1 (55.2)	138.1 (51.4)	0.701	Wilcoxon rank-sum
Unknown	18	5
Operative time (minutes)	181.2 (57.3)	190.6 (45.2)	0.192	Wilcoxon rank-sum
Unknown	14	0
Estimated blood loss (mL)			0.027	Wilcoxon rank-sum
25	1 (3.1%)	0 (0%)
50	10 (31%)	4 (16%)
75	5 (16%)	1 (4.0%)
100	9 (28%)	7 (28%)
125	1 (3.1%)	2 (8.0%)
150	2 (6.3%)	7 (28%)
200	1 (3.1%)	4 (16%)
300	2 (6.3%)	0 (0%)
500	1 (3.1%)	0 (0%)
Unknown	1	0
Conversion to multiport	0 (0%)	0 (0%)	1.000	Fisher's exact
Lymph node dissection	0 (0%)	0 (0%)	1.000	Fisher's exact
Drain placement	1 (3.0%)	0 (0%)	1.000	Fisher's exact
Blood units given	0 (0%)	0 (0%)	1.000	Fisher's exact
dHgb (%)	-0.8 (0.7)	-0.6 (1.1)	0.612	Wilcoxon rank-sum
Unknown	14	8
Histopathology
Specimen weight (g)	74.7 (40.6)	64.2 (28.8)	0.504	Wilcoxon rank-sum
Unknown	2	0
Specimen wt/pre-op estimated wt	0.6 (0.1)	0.5 (0.2)	0.255	Wilcoxon rank-sum
Unknown	2	0
Incidental malignancy finding	5 (16%)	3 (12%)	1.000	Fisher's exact
Unknown	1	0
Postoperative care
Length of stay <24 hours			0.377	Fisher's exact
Same Day	11 (55%)	10 (40%)
Inpatient	9 (45%)	15 (60%)
Unknown	13	0
Average length of stay (hours)	18.4 (14.3)	18.4 (9.7)	0.423	Wilcoxon rank-sum
Unknown	13	0
Narcotics use post-op day 0-1	1 (4.0%)	5 (20%)	0.189	Fisher's exact
Unknown	8	0
Post Op Pain Score			0.009	Wilcoxon rank-sum
0	1 (6.3%)	8 (35%)
1	2 (13%)	2 (8.7%)
2	1 (6.3%)	2 (8.7%)
3	0 (0%)	2 (8.7%)
4	3 (19%)	2 (8.7%)
5	1 (6.3%)	5 (22%)
6	1 (6.3%)	1 (4.3%)
7	2 (13%)	0 (0%)
8	5 (31%)	1 (4.3%)
Unknown	17	2
Foley catheter duration (days)	9.0 (6.0, 10.0)	7.0 (6.0, 9.0)	0.204	Wilcoxon rank-sum
Unknown	7	2
Postoperative outcomes
30 day postoperative complications (Clavien >=2 only)	4 (15%)	4 (17%)	1.000	Fisher's exact
Unknown	6	2
30 day readmission	1 (4.0%)	1 (4.2%)	1.000	Fisher's exact
Unknown	8	1
Post-op PVR	9.0 (0.0, 19.0)	5.0 (0.0, 18.0)	0.846	Wilcoxon rank-sum
Unknown	6	18
Early Continence at 4 weeks	27 (90%)	21 (95%)	0.629	Fisher's exact
Unknown	3	3
Continence at 3 months	20 (100%)	22 (100%)	1.000	Fisher's exact
Unknown	13	3
1 Mean (SD); n (%)

1.8 Improving Aesthetics of Table

final_tbl %>%
  as_gt() %>%
  gt::tab_style(
    style = list(
      gt::cell_text(weight = "bold"),
      gt::cell_borders(
        sides = c("top", "bottom"),
        color = "black",
        weight = gt::px(2)
      )
    ),
    locations = gt::cells_row_groups()
  ) %>%
  gt::tab_options(
    row_group.as_column = FALSE
  )

Characteristic	Extraperitoneal N = 331	Transvesical N = 251	P-value
Baseline characteristics
Age (yr)	72.4 (6.8)	71.2 (7.2)	0.654
BMI (kg/m^2)	27.6 (5.2)	26.9 (4.6)	0.668
Unknown	2	0
CCI			0.452
1	1 (3.1%)	2 (8.0%)
2	6 (19%)	6 (24%)
3	16 (50%)	12 (48%)
4	9 (28%)	2 (8.0%)
5	0 (0%)	2 (8.0%)
6	0 (0%)	1 (4.0%)
Unknown	1	0
ASA physical status grading			0.457
2	17 (55%)	17 (68%)
3	13 (42%)	8 (32%)
4	1 (3.2%)	0 (0%)
Unknown	2	0
History of hypertension	21 (64%)	11 (44%)	0.185
History of type 2 diabetes mellitus	6 (18%)	4 (16%)	1.000
History of hyperlipidemia	22 (67%)	13 (52%)	0.290
History of prior inguinal/pelvic surgery	5 (15%)	4 (16%)	1.000
History of prior abdominal surgery	16 (48%)	2 (8.0%)	0.001
Pre-op Hgb	13.6 (1.4)	12.9 (1.5)	0.097
Unknown	7	0
Pre-op PSA (ng/mL)	7.4 (7.9)	6.5 (5.1)	1.000
Unknown	8	8
Prostate volume (ml)	125.2 (49.4)	120.3 (37.4)	0.975
Pre-operative variables
SHIM	12.0 (3.0, 16.0)	15.0 (7.5, 19.5)	0.384
Unknown	8	9
IPSS	20.0 (12.0, 25.0)	19.0 (15.0, 24.0)	0.811
Unknown	4	8
QoL			0.307
1	2 (13%)	1 (6.3%)
2	1 (6.3%)	2 (13%)
3	3 (19%)	4 (25%)
4	2 (13%)	6 (38%)
5	2 (13%)	1 (6.3%)
6	6 (38%)	2 (13%)
Unknown	17	9
Biopsy done	6 (18%)	6 (24%)	0.745
Malignancy on biopsy			1.000
Negative	6 (100%)	6 (86%)
Positive	0 (0%)	1 (14%)
Unknown	27	18
Pre-op foley placed	13 (39%)	14 (56%)	0.289
Intraoperative parameters
Console time (minutes)	146.1 (55.2)	138.1 (51.4)	0.701
Unknown	18	5
Operative time (minutes)	181.2 (57.3)	190.6 (45.2)	0.192
Unknown	14	0
Estimated blood loss (mL)			0.027
25	1 (3.1%)	0 (0%)
50	10 (31%)	4 (16%)
75	5 (16%)	1 (4.0%)
100	9 (28%)	7 (28%)
125	1 (3.1%)	2 (8.0%)
150	2 (6.3%)	7 (28%)
200	1 (3.1%)	4 (16%)
300	2 (6.3%)	0 (0%)
500	1 (3.1%)	0 (0%)
Unknown	1	0
Conversion to multiport	0 (0%)	0 (0%)	1.000
Lymph node dissection	0 (0%)	0 (0%)	1.000
Drain placement	1 (3.0%)	0 (0%)	1.000
Blood units given	0 (0%)	0 (0%)	1.000
dHgb (%)	-0.8 (0.7)	-0.6 (1.1)	0.612
Unknown	14	8
Histopathology
Specimen weight (g)	74.7 (40.6)	64.2 (28.8)	0.504
Unknown	2	0
Specimen wt/pre-op estimated wt	0.6 (0.1)	0.5 (0.2)	0.255
Unknown	2	0
Incidental malignancy finding	5 (16%)	3 (12%)	1.000
Unknown	1	0
Postoperative care
Length of stay <24 hours			0.377
Same Day	11 (55%)	10 (40%)
Inpatient	9 (45%)	15 (60%)
Unknown	13	0
Average length of stay (hours)	18.4 (14.3)	18.4 (9.7)	0.423
Unknown	13	0
Narcotics use post-op day 0-1	1 (4.0%)	5 (20%)	0.189
Unknown	8	0
Post Op Pain Score			0.009
0	1 (6.3%)	8 (35%)
1	2 (13%)	2 (8.7%)
2	1 (6.3%)	2 (8.7%)
3	0 (0%)	2 (8.7%)
4	3 (19%)	2 (8.7%)
5	1 (6.3%)	5 (22%)
6	1 (6.3%)	1 (4.3%)
7	2 (13%)	0 (0%)
8	5 (31%)	1 (4.3%)
Unknown	17	2
Foley catheter duration (days)	9.0 (6.0, 10.0)	7.0 (6.0, 9.0)	0.204
Unknown	7	2
Postoperative outcomes
30 day postoperative complications (Clavien >=2 only)	4 (15%)	4 (17%)	1.000
Unknown	6	2
30 day readmission	1 (4.0%)	1 (4.2%)	1.000
Unknown	8	1
Post-op PVR	9.0 (0.0, 19.0)	5.0 (0.0, 18.0)	0.846
Unknown	6	18
Early Continence at 4 weeks	27 (90%)	21 (95%)	0.629
Unknown	3	3
Continence at 3 months	20 (100%)	22 (100%)	1.000
Unknown	13	3
1 Mean (SD); n (%)

1.9 Revise the Table Your Way

1. Make the categories with significant differences red?
2. Bold significant P values?
3. Fix the kg/m^2 to actually be superscript in table?

2 Session Info

Sys.time()

## [1] "2026-04-25 21:49:49 PDT"

sessionInfo()

## R version 4.5.1 (2025-06-13 ucrt)
## Platform: x86_64-w64-mingw32/x64
## Running under: Windows 11 x64 (build 26200)
## 
## Matrix products: default
##   LAPACK version 3.12.1
## 
## locale:
## [1] LC_COLLATE=English_United States.utf8 
## [2] LC_CTYPE=English_United States.utf8   
## [3] LC_MONETARY=English_United States.utf8
## [4] LC_NUMERIC=C                          
## [5] LC_TIME=English_United States.utf8    
## 
## time zone: America/Los_Angeles
## tzcode source: internal
## 
## attached base packages:
## [1] stats     graphics  grDevices utils     datasets  methods   base     
## 
## other attached packages:
## [1] ggplot2_4.0.1   gt_1.1.0        gtsummary_2.5.0 tibble_3.3.0   
## [5] purrr_1.1.0     dplyr_1.1.4     readxl_1.4.5   
## 
## loaded via a namespace (and not attached):
##  [1] gtable_0.3.6       jsonlite_2.0.0     compiler_4.5.1     tidyselect_1.2.1  
##  [5] xml2_1.4.0         dichromat_2.0-0.1  tidyr_1.3.1        jquerylib_0.1.4   
##  [9] scales_1.4.0       yaml_2.3.10        fastmap_1.2.0      R6_2.6.1          
## [13] commonmark_2.0.0   cards_0.7.1        labeling_0.4.3     generics_0.1.4    
## [17] knitr_1.50         RColorBrewer_1.1-3 bslib_0.9.0        pillar_1.11.0     
## [21] rlang_1.1.6        cachem_1.1.0       litedown_0.7       xfun_0.53         
## [25] S7_0.2.0           fs_1.6.6           sass_0.4.10        cli_3.6.5         
## [29] withr_3.0.2        magrittr_2.0.3     digest_0.6.37      grid_4.5.1        
## [33] rstudioapi_0.17.1  markdown_2.0       lifecycle_1.0.5    vctrs_0.6.5       
## [37] evaluate_1.0.5     glue_1.8.0         farver_2.1.2       cellranger_1.1.0  
## [41] rmarkdown_2.29     tools_4.5.1        pkgconfig_2.0.3    htmltools_0.5.8.1