WNV Dataset Comparison: Old vs New (February 2026 Review)

Setup

# ── Packages ──────────────────────────────────────────────────────────────────
if (!requireNamespace("pacman", quietly = TRUE)) install.packages("pacman")
pacman::p_load(
  readxl, tableone, lubridate, gt, flextable, officer, fs,
  dplyr, tidyr, ggplot2, broom, forcats, glue, scales, gridExtra, knitr
)

knitr::opts_chunk$set(echo = TRUE, message = FALSE, warning = FALSE, fig.align = "center")

# ── Source helper functions ───────────────────────────────────────────────────
if (file.exists(params$helper_path)) {
  source(params$helper_path)
  message("Helper functions loaded from: ", params$helper_path)
} else {
  stop("Helper file not found at: ", params$helper_path)
}

# ── Create output directory ──────────────────────────────────────────────────
dir.create(params$export_dir, showWarnings = FALSE, recursive = TRUE)
message("Outputs will be saved to: ", params$export_dir)

Load Both Datasets

# ── Read OLD dataset ─────────────────────────────────────────────────────────
raw_old <- read_excel(params$old_data_path, sheet = params$sheet_name)
message(glue("OLD dataset loaded: {params$old_data_path}"))
message(glue("  Dimensions: {nrow(raw_old)} rows x {ncol(raw_old)} cols"))

# ── Read NEW dataset ─────────────────────────────────────────────────────────
raw_new <- read_excel(params$new_data_path, sheet = params$sheet_name)
message(glue("NEW dataset loaded: {params$new_data_path}"))
message(glue("  Dimensions: {nrow(raw_new)} rows x {ncol(raw_new)} cols"))

# ── Trim trailing rows (if applicable) ───────────────────────────────────────
if (!is.null(params$trim_rows) && length(params$trim_rows) > 0) {
  # Only trim if the dataset has enough rows
  trim_old <- params$trim_rows[params$trim_rows <= nrow(raw_old)]
  trim_new <- params$trim_rows[params$trim_rows <= nrow(raw_new)]

  if (length(trim_old) > 0) {
    raw_old <- raw_old[-trim_old, ]
    message(glue("OLD: Trimmed rows {paste(trim_old, collapse=',')}. New dims: {nrow(raw_old)} x {ncol(raw_old)}"))
  }
  if (length(trim_new) > 0) {
    raw_new <- raw_new[-trim_new, ]
    message(glue("NEW: Trimmed rows {paste(trim_new, collapse=',')}. New dims: {nrow(raw_new)} x {ncol(raw_new)}"))
  }
}

# ── Quick dimension comparison ────────────────────────────────────────────────
dim_compare <- tibble(
  Dataset  = c("OLD", "NEW", "Delta"),
  Rows     = c(nrow(raw_old), nrow(raw_new), nrow(raw_new) - nrow(raw_old)),
  Columns  = c(ncol(raw_old), ncol(raw_new), ncol(raw_new) - ncol(raw_old))
)
knitr::kable(dim_compare, caption = "Dataset Dimension Comparison")

Dataset Dimension Comparison

Dataset	Rows	Columns
OLD	1304	41
NEW	1304	46
Delta	0	5

Raw Data Comparison

# ── Column differences ────────────────────────────────────────────────────────
cols_old <- names(raw_old)
cols_new <- names(raw_new)

cols_added   <- setdiff(cols_new, cols_old)
cols_removed <- setdiff(cols_old, cols_new)
cols_shared  <- intersect(cols_old, cols_new)

message(glue("Shared columns: {length(cols_shared)}"))
if (length(cols_added) > 0)   message(glue("Columns ADDED in NEW: {paste(cols_added, collapse=', ')}"))
if (length(cols_removed) > 0) message(glue("Columns REMOVED from OLD: {paste(cols_removed, collapse=', ')}"))
if (length(cols_added) == 0 && length(cols_removed) == 0) message("Column names are IDENTICAL between datasets.")

# ── Per-column NA comparison (shared columns only) ────────────────────────────
na_old <- sapply(raw_old[cols_shared], function(x) sum(is.na(x)))
na_new <- sapply(raw_new[cols_shared], function(x) sum(is.na(x)))

na_compare <- tibble(
  Column     = cols_shared,
  NA_Old     = na_old,
  NA_New     = na_new,
  Delta_NA   = na_new - na_old
) %>% dplyr::filter(Delta_NA != 0)  # Show only columns where NAs changed

if (nrow(na_compare) > 0) {
  knitr::kable(na_compare, caption = "Columns Where Missing Value Count Changed")
} else {
  message("No changes in missing value counts across shared columns.")
}

# ── Per-column unique-value comparison ────────────────────────────────────────
uniq_old <- sapply(raw_old[cols_shared], function(x) dplyr::n_distinct(x, na.rm = TRUE))
uniq_new <- sapply(raw_new[cols_shared], function(x) dplyr::n_distinct(x, na.rm = TRUE))

uniq_compare <- tibble(
  Column       = cols_shared,
  Unique_Old   = uniq_old,
  Unique_New   = uniq_new,
  Delta_Unique = uniq_new - uniq_old
) %>% dplyr::filter(Delta_Unique != 0)

if (nrow(uniq_compare) > 0) {
  knitr::kable(uniq_compare, caption = "Columns Where Unique Value Count Changed")
} else {
  message("No changes in unique value counts across shared columns.")
}

Data Wrangling

# ── Apply factor conversions and derived variables to BOTH datasets ───────────
data_old <- convert_factors_and_derive(raw_old)
message(glue("data_old created. Dims: {nrow(data_old)} x {ncol(data_old)}"))

data_new <- convert_factors_and_derive(raw_new)
message(glue("data_new created. Dims: {nrow(data_new)} x {ncol(data_new)}"))

# ── Derive _ordered variables (needed for Tables 3 & 4) ──────────────────────
# These are derived from the _rslt factor columns:
# In unified_helpers.R, NA values → "Not Sent"
derive_ordered_vars <- function(df) {
  df$csfpcr_ordered <- factor(
    ifelse(df$csfpcr_rslt == "Not Sent", 0, 1),
    levels = c(0, 1), labels = c("CSF PCR Not Ordered", "CSF PCR Ordered")
  )
  df$csfigm_ordered <- factor(
    ifelse(df$csfigm_rslt == "Not Sent", 0, 1),
    levels = c(0, 1), labels = c("CSF IgM Not Ordered", "CSF IgM Ordered")
  )
  df$serumigm_ordered <- factor(
    ifelse(df$serumigm_rslt == "Not Sent", 0, 1),
    levels = c(0, 1), labels = c("Serum IgM Not Ordered", "Serum IgM Ordered")
  )
  return(df)
}

data_old <- derive_ordered_vars(data_old)
data_new <- derive_ordered_vars(data_new)

# ── Verify factor level consistency ───────────────────────────────────────────
key_factors <- c("sex", "race", "dead", "medicaid", "site", "service",
                 "immunosup", "transplant", "bcellmed", "guideline2",
                 "wnvpos", "csftp60", "csftnc5", "csflymph50", "inseasontest")

factor_issues <- character(0)
for (fv in key_factors) {
  if (fv %in% names(data_old) && fv %in% names(data_new)) {
    lvl_old <- levels(data_old[[fv]])
    lvl_new <- levels(data_new[[fv]])
    if (!identical(lvl_old, lvl_new)) {
      factor_issues <- c(factor_issues,
        glue("{fv}: OLD levels = [{paste(lvl_old, collapse=', ')}] vs NEW = [{paste(lvl_new, collapse=', ')}]"))
    }
  }
}

if (length(factor_issues) > 0) {
  message("WARNING: Factor level mismatches detected:")
  for (fi in factor_issues) message("  ", fi)
} else {
  message("All key factor variables have identical levels between OLD and NEW datasets.")
}

Table 1: Demographics by Guideline Adherence

# ── Common definitions ────────────────────────────────────────────────────────
table1_vars <- c(
  "sex", "csfageyr", "race", "nonwhite", "dead", "medicaid",
  "site", "service", "immunosup", "transplant", "bcellmed", "csfadmit_day",
  "inseasontest", "serumwbc", "csftp", "csftp60", "csfgluc", "csftnc",
  "csftnc5", "csflymph", "csflymph50"
)
continuous_vars <- c("csfageyr", "serumwbc", "csftp", "csfgluc", "csftnc", "csflymph", "csfadmit_day")
factor_vars_t1 <- setdiff(table1_vars, continuous_vars)

# ── Helper: build a simple TableOne dataframe ─────────────────────────────────
build_tableone_df <- function(data, vars, strata, cont_vars, fac_vars) {
  # Ensure continuous vars are numeric
  data[cont_vars] <- lapply(data[cont_vars], function(x) as.numeric(as.character(x)))

  tobj <- tableone::CreateTableOne(
    vars = vars, strata = strata, data = data,
    factorVars = fac_vars, addOverall = TRUE
  )
  tmat <- print(tobj, nonnormal = cont_vars, showAllLevels = TRUE,
                printToggle = FALSE, noSpaces = TRUE)
  tdf <- as.data.frame(tmat, stringsAsFactors = FALSE)

  # Preserve unique rownames as row_id BEFORE any modifications
  tdf$row_id <- rownames(tdf)

  colnames(tdf)[1] <- "level"
  tdf$level <- trimws(tdf$level)

  # Copy p-values
  fld <- lapply(fac_vars, function(x) levels(data[[x]]))
  names(fld) <- fac_vars
  tdf <- suppressMessages(copyPvalueToAllLevels(tdf, fld))

  # Remove first row (n) and test column
  tdf <- tdf[tdf$level != "n", ]
  if ("test" %in% colnames(tdf)) tdf <- dplyr::select(tdf, -test)
  return(tdf)
}

# ── Helper: build a highlighted delta gt table from two TableOne DFs ──────────
build_delta_gt <- function(df_old, df_new, title_text = "Delta Comparison") {
  # Join by row_id (unique rownames from TableOne) — NOT by level text
  delta <- dplyr::inner_join(
    df_old %>% dplyr::select(row_id, Characteristic = level, Overall_OLD = Overall, p_OLD = p),
    df_new %>% dplyr::select(row_id, Overall_NEW = Overall, p_NEW = p),
    by = "row_id"
  ) %>%
    dplyr::mutate(
      overall_changed = (Overall_OLD != Overall_NEW),
      p_changed = (!is.na(p_OLD) & !is.na(p_NEW) & p_OLD != "" & p_NEW != "" & p_OLD != p_NEW)
    ) %>%
    dplyr::select(Characteristic, Overall_OLD, p_OLD, Overall_NEW, p_NEW, overall_changed, p_changed)

  # Build gt with highlighting
  gt_out <- delta %>%
    dplyr::select(-overall_changed, -p_changed) %>%
    gt::gt() %>%
    gt::tab_header(title = gt::md(paste0("**", title_text, "**"))) %>%
    gt::cols_label(
      Characteristic = "",
      Overall_OLD = gt::md("**Overall (OLD)**"),
      p_OLD       = gt::md("**P (OLD)**"),
      Overall_NEW = gt::md("**Overall (NEW)**"),
      p_NEW       = gt::md("**P (NEW)**")
    ) %>%
    gt::tab_spanner(label = gt::md("**OLD Dataset**"), columns = c(Overall_OLD, p_OLD)) %>%
    gt::tab_spanner(label = gt::md("**NEW Dataset**"), columns = c(Overall_NEW, p_NEW)) %>%
    gt::cols_align(align = "center", columns = c(Overall_OLD, p_OLD, Overall_NEW, p_NEW)) %>%
    gt::cols_align(align = "left", columns = Characteristic) %>%
    gt::tab_options(table.width = gt::pct(100), table.font.size = gt::px(12))

  # Highlight rows where Overall changed
  changed_rows <- which(delta$overall_changed)
  if (length(changed_rows) > 0) {
    gt_out <- gt_out %>%
      gt::tab_style(
        style = gt::cell_fill(color = "#FFFFCC"),
        locations = gt::cells_body(columns = c(Overall_OLD, Overall_NEW), rows = changed_rows)
      )
  }

  # Highlight rows where p-value changed
  p_rows <- which(delta$p_changed)
  if (length(p_rows) > 0) {
    gt_out <- gt_out %>%
      gt::tab_style(
        style = gt::cell_fill(color = "#FFE0B2"),
        locations = gt::cells_body(columns = c(p_OLD, p_NEW), rows = p_rows)
      )
  }

  # Return both the gt object and the raw delta dataframe (for Word export)
  return(list(gt = gt_out, data = delta))
}

# ── Build Table 1 for OLD and NEW ────────────────────────────────────────────
t1_old <- build_tableone_df(data_old, table1_vars, "guideline2", continuous_vars, factor_vars_t1)

## [1] "--- Entering copyPvalueToAllLevels function ---"
## [1] "Processing variable: sex"
## [1] "  Factor levels for sex : Male, Female"
## [1] "  Rows in df_tableone for these levels: 2, 3"
## [1] "  Rows with p-value in factor levels of sex : 2"
## [1] "  P-value found for sex : 0.612"
## [1] "  P-value copied to rows: 2, 3"
## [1] "Processing variable: race"
## [1] "  Factor levels for race : White, Black, Other/Multiple, Unknown"
## [1] "  Rows in df_tableone for these levels: 5, 6, 7, 8, 9"
## [1] "  Rows with p-value in factor levels of race : 5, 9"
## [1] "  P-value found for race : 0.032"
## [1] "  P-value copied to rows: 5, 6, 7, 8, 9"
## [1] "Processing variable: nonwhite"
## [1] "  Factor levels for nonwhite : White, Non-White"
## [1] "  Rows in df_tableone for these levels: 5, 9, 10"
## [1] "  Rows with p-value in factor levels of nonwhite : 5, 9"
## [1] "  P-value found for nonwhite : 0.032"
## [1] "  P-value copied to rows: 5, 9, 10"
## [1] "Processing variable: dead"
## [1] "  Factor levels for dead : Not Reported Dead, Reported Dead"
## [1] "  Rows in df_tableone for these levels: 11, 12"
## [1] "  Rows with p-value in factor levels of dead : 11"
## [1] "  P-value found for dead : 0.172"
## [1] "  P-value copied to rows: 11, 12"
## [1] "Processing variable: medicaid"
## [1] "  Factor levels for medicaid : Other Insurance, Medicaid/Safety Net"
## [1] "  Rows in df_tableone for these levels: 13, 14"
## [1] "  Rows with p-value in factor levels of medicaid : 13"
## [1] "  P-value found for medicaid : 0.246"
## [1] "  P-value copied to rows: 13, 14"
## [1] "Processing variable: site"
## [1] "  Factor levels for site : Mass General Hospital, Brigham and Women's Hospital"
## [1] "  Rows in df_tableone for these levels: 15, 16"
## [1] "  Rows with p-value in factor levels of site : 15"
## [1] "  P-value found for site : <0.001"
## [1] "  P-value copied to rows: 15, 16"
## [1] "Processing variable: service"
## [1] "  Factor levels for service : Neurology, Medicine/Subspecialties, Oncology, Other"
## [1] "  Rows in df_tableone for these levels: 17, 18, 19, 20"
## [1] "  Rows with p-value in factor levels of service : 17"
## [1] "  P-value found for service : <0.001"
## [1] "  P-value copied to rows: 17, 18, 19, 20"
## [1] "Processing variable: immunosup"
## [1] "  Factor levels for immunosup : Not Immunosuppressed, Immunosuppressed"
## [1] "  Rows in df_tableone for these levels: 21, 22"
## [1] "  Rows with p-value in factor levels of immunosup : 21"
## [1] "  P-value found for immunosup : <0.001"
## [1] "  P-value copied to rows: 21, 22"
## [1] "Processing variable: transplant"
## [1] "  Factor levels for transplant : No Transplant History, History of Transplant"
## [1] "  Rows in df_tableone for these levels: 23, 24"
## [1] "  Rows with p-value in factor levels of transplant : 23"
## [1] "  P-value found for transplant : <0.001"
## [1] "  P-value copied to rows: 23, 24"
## [1] "Processing variable: bcellmed"
## [1] "  Factor levels for bcellmed : No B-cell Depleting Meds, B-cell Depletion"
## [1] "  Rows in df_tableone for these levels: 25, 26"
## [1] "  Rows with p-value in factor levels of bcellmed : 25"
## [1] "  P-value found for bcellmed : 0.005"
## [1] "  P-value copied to rows: 25, 26"
## [1] "Processing variable: inseasontest"
## [1] "  Factor levels for inseasontest : Test Sent Off-Season, Test Sent June-Nov"
## [1] "  Rows in df_tableone for these levels: 28, 29"
## [1] "  Rows with p-value in factor levels of inseasontest : 28"
## [1] "  P-value found for inseasontest : 0.369"
## [1] "  P-value copied to rows: 28, 29"
## [1] "Processing variable: csftp60"
## [1] "  Factor levels for csftp60 : CSF TP < 60, CSF TP ≥ 60"
## [1] "  Rows in df_tableone for these levels: 32, 33"
## [1] "  Rows with p-value in factor levels of csftp60 : 32"
## [1] "  P-value found for csftp60 : <0.001"
## [1] "  P-value copied to rows: 32, 33"
## [1] "Processing variable: csftnc5"
## [1] "  Factor levels for csftnc5 : CSF TNC < 5, CSF TNC ≥ 5"
## [1] "  Rows in df_tableone for these levels: 36, 37"
## [1] "  Rows with p-value in factor levels of csftnc5 : 36"
## [1] "  P-value found for csftnc5 : <0.001"
## [1] "  P-value copied to rows: 36, 37"
## [1] "Processing variable: csflymph50"
## [1] "  Factor levels for csflymph50 : Normal CSF Cell Count (CSF TNC < 5), Pleocytosis (Lymph ≥ 50%), Pleocytosis (Lymph < 50%)"
## [1] "  Rows in df_tableone for these levels: 39, 40, 41"
## [1] "  Rows with p-value in factor levels of csflymph50 : 39"
## [1] "  P-value found for csflymph50 : <0.001"
## [1] "  P-value copied to rows: 39, 40, 41"
## [1] "--- Exiting copyPvalueToAllLevels function ---"
## [1] "--- End Debugging ---"

t1_new <- build_tableone_df(data_new, table1_vars, "guideline2", continuous_vars, factor_vars_t1)

## [1] "--- Entering copyPvalueToAllLevels function ---"
## [1] "Processing variable: sex"
## [1] "  Factor levels for sex : Male, Female"
## [1] "  Rows in df_tableone for these levels: 2, 3"
## [1] "  Rows with p-value in factor levels of sex : 2"
## [1] "  P-value found for sex : 0.646"
## [1] "  P-value copied to rows: 2, 3"
## [1] "Processing variable: race"
## [1] "  Factor levels for race : White, Black, Other/Multiple, Unknown"
## [1] "  Rows in df_tableone for these levels: 5, 6, 7, 8, 9"
## [1] "  Rows with p-value in factor levels of race : 5, 9"
## [1] "  P-value found for race : 0.052"
## [1] "  P-value copied to rows: 5, 6, 7, 8, 9"
## [1] "Processing variable: nonwhite"
## [1] "  Factor levels for nonwhite : White, Non-White"
## [1] "  Rows in df_tableone for these levels: 5, 9, 10"
## [1] "  Rows with p-value in factor levels of nonwhite : 5, 9"
## [1] "  P-value found for nonwhite : 0.052"
## [1] "  P-value copied to rows: 5, 9, 10"
## [1] "Processing variable: dead"
## [1] "  Factor levels for dead : Not Reported Dead, Reported Dead"
## [1] "  Rows in df_tableone for these levels: 11, 12"
## [1] "  Rows with p-value in factor levels of dead : 11"
## [1] "  P-value found for dead : 0.213"
## [1] "  P-value copied to rows: 11, 12"
## [1] "Processing variable: medicaid"
## [1] "  Factor levels for medicaid : Other Insurance, Medicaid/Safety Net"
## [1] "  Rows in df_tableone for these levels: 13, 14"
## [1] "  Rows with p-value in factor levels of medicaid : 13"
## [1] "  P-value found for medicaid : 0.143"
## [1] "  P-value copied to rows: 13, 14"
## [1] "Processing variable: site"
## [1] "  Factor levels for site : Mass General Hospital, Brigham and Women's Hospital"
## [1] "  Rows in df_tableone for these levels: 15, 16"
## [1] "  Rows with p-value in factor levels of site : 15"
## [1] "  P-value found for site : <0.001"
## [1] "  P-value copied to rows: 15, 16"
## [1] "Processing variable: service"
## [1] "  Factor levels for service : Neurology, Medicine/Subspecialties, Oncology, Other"
## [1] "  Rows in df_tableone for these levels: 17, 18, 19, 20"
## [1] "  Rows with p-value in factor levels of service : 17"
## [1] "  P-value found for service : <0.001"
## [1] "  P-value copied to rows: 17, 18, 19, 20"
## [1] "Processing variable: immunosup"
## [1] "  Factor levels for immunosup : Not Immunosuppressed, Immunosuppressed"
## [1] "  Rows in df_tableone for these levels: 21, 22"
## [1] "  Rows with p-value in factor levels of immunosup : 21"
## [1] "  P-value found for immunosup : <0.001"
## [1] "  P-value copied to rows: 21, 22"
## [1] "Processing variable: transplant"
## [1] "  Factor levels for transplant : No Transplant History, History of Transplant"
## [1] "  Rows in df_tableone for these levels: 23, 24"
## [1] "  Rows with p-value in factor levels of transplant : 23"
## [1] "  P-value found for transplant : <0.001"
## [1] "  P-value copied to rows: 23, 24"
## [1] "Processing variable: bcellmed"
## [1] "  Factor levels for bcellmed : No B-cell Depleting Meds, B-cell Depletion"
## [1] "  Rows in df_tableone for these levels: 25, 26"
## [1] "  Rows with p-value in factor levels of bcellmed : 25"
## [1] "  P-value found for bcellmed : 0.005"
## [1] "  P-value copied to rows: 25, 26"
## [1] "Processing variable: inseasontest"
## [1] "  Factor levels for inseasontest : Test Sent Off-Season, Test Sent June-Nov"
## [1] "  Rows in df_tableone for these levels: 28, 29"
## [1] "  Rows with p-value in factor levels of inseasontest : 28"
## [1] "  P-value found for inseasontest : 0.308"
## [1] "  P-value copied to rows: 28, 29"
## [1] "Processing variable: csftp60"
## [1] "  Factor levels for csftp60 : CSF TP < 60, CSF TP ≥ 60"
## [1] "  Rows in df_tableone for these levels: 32, 33"
## [1] "  Rows with p-value in factor levels of csftp60 : 32"
## [1] "  P-value found for csftp60 : <0.001"
## [1] "  P-value copied to rows: 32, 33"
## [1] "Processing variable: csftnc5"
## [1] "  Factor levels for csftnc5 : CSF TNC < 5, CSF TNC ≥ 5"
## [1] "  Rows in df_tableone for these levels: 36, 37"
## [1] "  Rows with p-value in factor levels of csftnc5 : 36"
## [1] "  P-value found for csftnc5 : <0.001"
## [1] "  P-value copied to rows: 36, 37"
## [1] "Processing variable: csflymph50"
## [1] "  Factor levels for csflymph50 : Normal CSF Cell Count (CSF TNC < 5), Pleocytosis (Lymph ≥ 50%), Pleocytosis (Lymph < 50%)"
## [1] "  Rows in df_tableone for these levels: 39, 40, 41"
## [1] "  Rows with p-value in factor levels of csflymph50 : 39"
## [1] "  P-value found for csflymph50 : <0.001"
## [1] "  P-value copied to rows: 39, 40, 41"
## [1] "--- Exiting copyPvalueToAllLevels function ---"
## [1] "--- End Debugging ---"

# ── Display both tables ──────────────────────────────────────────────────────
cat("### Table 1 — OLD Dataset\n")

## ### Table 1 — OLD Dataset

knitr::kable(t1_old, caption = "Table 1 (OLD): Demographics & Lab Values by Guideline Adherence")

Table 1 (OLD): Demographics & Lab Values by Guideline Adherence

	level	Overall	Adherent	Nonadherent	p	row_id
n		1304	333	971		n
sex….	Male	703 (53.9)	184 (55.3)	519 (53.5)	0.612	sex….
X	Female	601 (46.1)	149 (44.7)	452 (46.5)	0.612	X
csfageyr..median..IQR..		63.00 [48.00, 73.00]	62.00 [44.00, 74.00]	64.00 [50.00, 73.00]	0.072	csfageyr..median..IQR..
race….	White	994 (76.2)	254 (76.3)	740 (76.2)	0.032	race….
X.1	Black	105 (8.1)	16 (4.8)	89 (9.2)	0.032	X.1
X.2	Other/Multiple	129 (9.9)	39 (11.7)	90 (9.3)	0.032	X.2
X.3	Unknown	76 (5.8)	24 (7.2)	52 (5.4)	0.032	X.3
nonwhite….	White	1070 (82.1)	278 (83.5)	792 (81.6)	0.032	nonwhite….
X.4	Non-White	234 (17.9)	55 (16.5)	179 (18.4)	0.032	X.4
dead….	Not Reported Dead	915 (70.2)	244 (73.3)	671 (69.1)	0.172	dead….
X.5	Reported Dead	389 (29.8)	89 (26.7)	300 (30.9)	0.172	X.5
medicaid….	Other Insurance	933 (71.5)	247 (74.2)	686 (70.6)	0.246	medicaid….
X.6	Medicaid/Safety Net	371 (28.5)	86 (25.8)	285 (29.4)	0.246	X.6
site….	Mass General Hospital	706 (54.1)	231 (69.4)	475 (48.9)	<0.001	site….
X.7	Brigham and Women’s Hospital	598 (45.9)	102 (30.6)	496 (51.1)	<0.001	X.7
service….	Neurology	584 (44.8)	210 (63.1)	374 (38.5)	<0.001	service….
X.8	Medicine/Subspecialties	456 (35.0)	87 (26.1)	369 (38.0)	<0.001	X.8
X.9	Oncology	178 (13.7)	22 (6.6)	156 (16.1)	<0.001	X.9
X.10	Other	86 (6.6)	14 (4.2)	72 (7.4)	<0.001	X.10
immunosup….	Not Immunosuppressed	1133 (86.9)	320 (96.1)	813 (83.7)	<0.001	immunosup….
X.11	Immunosuppressed	171 (13.1)	13 (3.9)	158 (16.3)	<0.001	X.11
transplant….	No Transplant History	1160 (89.0)	321 (96.4)	839 (86.4)	<0.001	transplant….
X.12	History of Transplant	144 (11.0)	12 (3.6)	132 (13.6)	<0.001	X.12
bcellmed….	No B-cell Depleting Meds	1271 (97.5)	332 (99.7)	939 (96.7)	0.005	bcellmed….
X.13	B-cell Depletion	33 (2.5)	1 (0.3)	32 (3.3)	0.005	X.13
csfadmit_day..median..IQR..		1.00 [1.00, 4.00]	2.00 [1.00, 4.00]	1.00 [1.00, 4.00]	0.756	csfadmit_day..median..IQR..
inseasontest….	Test Sent Off-Season	479 (36.7)	115 (34.5)	364 (37.5)	0.369	inseasontest….
X.14	Test Sent June-Nov	825 (63.3)	218 (65.5)	607 (62.5)	0.369	X.14
serumwbc..median..IQR..		8.17 [5.68, 11.40]	8.46 [6.38, 11.29]	8.07 [5.47, 11.47]	0.171	serumwbc..median..IQR..
csftp..median..IQR..		47.50 [31.70, 78.05]	57.50 [37.00, 92.60]	45.00 [30.00, 69.90]	<0.001	csftp..median..IQR..
csftp60….	CSF TP < 60	834 (64.0)	172 (51.7)	662 (68.2)	<0.001	csftp60….
X.15	CSF TP ≥ 60	469 (36.0)	161 (48.3)	308 (31.8)	<0.001	X.15
csfgluc..median..IQR..		68.00 [58.00, 85.00]	66.00 [56.00, 81.00]	69.00 [59.00, 86.00]	0.001	csfgluc..median..IQR..
csftnc..median..IQR..		3.00 [1.00, 21.00]	13.00 [2.00, 77.00]	2.00 [1.00, 10.00]	<0.001	csftnc..median..IQR..
csftnc5….	CSF TNC < 5	769 (59.0)	125 (37.5)	644 (66.3)	<0.001	csftnc5….
X.16	CSF TNC ≥ 5	535 (41.0)	208 (62.5)	327 (33.7)	<0.001	X.16
csflymph..median..IQR..		66.00 [25.00, 86.00]	74.00 [39.00, 89.00]	62.00 [20.00, 85.00]	<0.001	csflymph..median..IQR..
csflymph50….	Normal CSF Cell Count (CSF TNC < 5)	769 (59.0)	125 (37.5)	644 (66.3)	<0.001	csflymph50….
X.17	Pleocytosis (Lymph ≥ 50%)	350 (26.8)	153 (45.9)	197 (20.3)	<0.001	X.17
X.18	Pleocytosis (Lymph < 50%)	185 (14.2)	55 (16.5)	130 (13.4)	<0.001	X.18

cat("\n### Table 1 — NEW Dataset\n")

## 
## ### Table 1 — NEW Dataset

knitr::kable(t1_new, caption = "Table 1 (NEW): Demographics & Lab Values by Guideline Adherence")

Table 1 (NEW): Demographics & Lab Values by Guideline Adherence

	level	Overall	Adherent	Nonadherent	p	row_id
n		1304	330	974		n
sex….	Male	703 (53.9)	182 (55.2)	521 (53.5)	0.646	sex….
X	Female	601 (46.1)	148 (44.8)	453 (46.5)	0.646	X
csfageyr..median..IQR..		63.00 [48.00, 73.00]	62.00 [44.75, 74.00]	64.00 [49.00, 73.00]	0.116	csfageyr..median..IQR..
race….	White	994 (76.2)	254 (77.0)	740 (76.0)	0.052	race….
X.1	Black	105 (8.1)	16 (4.8)	89 (9.1)	0.052	X.1
X.2	Other/Multiple	129 (9.9)	36 (10.9)	93 (9.5)	0.052	X.2
X.3	Unknown	76 (5.8)	24 (7.3)	52 (5.3)	0.052	X.3
nonwhite….	White	1070 (82.1)	278 (84.2)	792 (81.3)	0.052	nonwhite….
X.4	Non-White	234 (17.9)	52 (15.8)	182 (18.7)	0.052	X.4
dead….	Not Reported Dead	915 (70.2)	241 (73.0)	674 (69.2)	0.213	dead….
X.5	Reported Dead	389 (29.8)	89 (27.0)	300 (30.8)	0.213	X.5
medicaid….	Other Insurance	933 (71.5)	247 (74.8)	686 (70.4)	0.143	medicaid….
X.6	Medicaid/Safety Net	371 (28.5)	83 (25.2)	288 (29.6)	0.143	X.6
site….	Mass General Hospital	706 (54.1)	229 (69.4)	477 (49.0)	<0.001	site….
X.7	Brigham and Women’s Hospital	598 (45.9)	101 (30.6)	497 (51.0)	<0.001	X.7
service….	Neurology	584 (44.8)	208 (63.0)	376 (38.6)	<0.001	service….
X.8	Medicine/Subspecialties	456 (35.0)	86 (26.1)	370 (38.0)	<0.001	X.8
X.9	Oncology	178 (13.7)	22 (6.7)	156 (16.0)	<0.001	X.9
X.10	Other	86 (6.6)	14 (4.2)	72 (7.4)	<0.001	X.10
immunosup….	Not Immunosuppressed	1114 (85.4)	317 (96.1)	797 (81.8)	<0.001	immunosup….
X.11	Immunosuppressed	190 (14.6)	13 (3.9)	177 (18.2)	<0.001	X.11
transplant….	No Transplant History	1160 (89.0)	318 (96.4)	842 (86.4)	<0.001	transplant….
X.12	History of Transplant	144 (11.0)	12 (3.6)	132 (13.6)	<0.001	X.12
bcellmed….	No B-cell Depleting Meds	1271 (97.5)	329 (99.7)	942 (96.7)	0.005	bcellmed….
X.13	B-cell Depletion	33 (2.5)	1 (0.3)	32 (3.3)	0.005	X.13
csfadmit_day..median..IQR..		1.00 [1.00, 4.00]	2.00 [1.00, 4.00]	1.00 [1.00, 4.00]	0.819	csfadmit_day..median..IQR..
inseasontest….	Test Sent Off-Season	479 (36.7)	113 (34.2)	366 (37.6)	0.308	inseasontest….
X.14	Test Sent June-Nov	825 (63.3)	217 (65.8)	608 (62.4)	0.308	X.14
serumwbc..median..IQR..		8.17 [5.68, 11.40]	8.50 [6.43, 11.30]	8.07 [5.46, 11.46]	0.117	serumwbc..median..IQR..
csftp..median..IQR..		47.50 [31.70, 78.05]	57.05 [36.78, 92.60]	45.00 [30.00, 70.00]	<0.001	csftp..median..IQR..
csftp60….	CSF TP < 60	834 (64.0)	172 (52.1)	662 (68.0)	<0.001	csftp60….
X.15	CSF TP ≥ 60	469 (36.0)	158 (47.9)	311 (32.0)	<0.001	X.15
csfgluc..median..IQR..		68.00 [58.00, 85.00]	66.00 [56.00, 81.00]	69.00 [59.00, 86.00]	0.003	csfgluc..median..IQR..
csftnc..median..IQR..		3.00 [1.00, 21.00]	13.00 [2.00, 74.75]	2.00 [1.00, 10.00]	<0.001	csftnc..median..IQR..
csftnc5….	CSF TNC < 5	769 (59.0)	124 (37.6)	645 (66.2)	<0.001	csftnc5….
X.16	CSF TNC ≥ 5	535 (41.0)	206 (62.4)	329 (33.8)	<0.001	X.16
csflymph..median..IQR..		66.00 [25.00, 86.00]	74.00 [39.00, 89.00]	62.00 [20.00, 85.00]	<0.001	csflymph..median..IQR..
csflymph50….	Normal CSF Cell Count (CSF TNC < 5)	769 (59.0)	124 (37.6)	645 (66.2)	<0.001	csflymph50….
X.17	Pleocytosis (Lymph ≥ 50%)	350 (26.8)	151 (45.8)	199 (20.4)	<0.001	X.17
X.18	Pleocytosis (Lymph < 50%)	185 (14.2)	55 (16.7)	130 (13.3)	<0.001	X.18

# ── Delta table with highlighting ─────────────────────────────────────────────
delta_t1_result <- build_delta_gt(t1_old, t1_new, "Table 1 Delta: Demographics by Guideline Adherence")
delta_t1 <- delta_t1_result$data
delta_t1_result$gt

Table 1 Delta: Demographics by Guideline Adherence
	OLD Dataset		NEW Dataset
	Overall (OLD)	P (OLD)	Overall (NEW)	P (NEW)
	1304		1304
Male	703 (53.9)	0.612	703 (53.9)	0.646
Female	601 (46.1)	0.612	601 (46.1)	0.646
	63.00 [48.00, 73.00]	0.072	63.00 [48.00, 73.00]	0.116
White	994 (76.2)	0.032	994 (76.2)	0.052
Black	105 (8.1)	0.032	105 (8.1)	0.052
Other/Multiple	129 (9.9)	0.032	129 (9.9)	0.052
Unknown	76 (5.8)	0.032	76 (5.8)	0.052
White	1070 (82.1)	0.032	1070 (82.1)	0.052
Non-White	234 (17.9)	0.032	234 (17.9)	0.052
Not Reported Dead	915 (70.2)	0.172	915 (70.2)	0.213
Reported Dead	389 (29.8)	0.172	389 (29.8)	0.213
Other Insurance	933 (71.5)	0.246	933 (71.5)	0.143
Medicaid/Safety Net	371 (28.5)	0.246	371 (28.5)	0.143
Mass General Hospital	706 (54.1)	<0.001	706 (54.1)	<0.001
Brigham and Women's Hospital	598 (45.9)	<0.001	598 (45.9)	<0.001
Neurology	584 (44.8)	<0.001	584 (44.8)	<0.001
Medicine/Subspecialties	456 (35.0)	<0.001	456 (35.0)	<0.001
Oncology	178 (13.7)	<0.001	178 (13.7)	<0.001
Other	86 (6.6)	<0.001	86 (6.6)	<0.001
Not Immunosuppressed	1133 (86.9)	<0.001	1114 (85.4)	<0.001
Immunosuppressed	171 (13.1)	<0.001	190 (14.6)	<0.001
No Transplant History	1160 (89.0)	<0.001	1160 (89.0)	<0.001
History of Transplant	144 (11.0)	<0.001	144 (11.0)	<0.001
No B-cell Depleting Meds	1271 (97.5)	0.005	1271 (97.5)	0.005
B-cell Depletion	33 (2.5)	0.005	33 (2.5)	0.005
	1.00 [1.00, 4.00]	0.756	1.00 [1.00, 4.00]	0.819
Test Sent Off-Season	479 (36.7)	0.369	479 (36.7)	0.308
Test Sent June-Nov	825 (63.3)	0.369	825 (63.3)	0.308
	8.17 [5.68, 11.40]	0.171	8.17 [5.68, 11.40]	0.117
	47.50 [31.70, 78.05]	<0.001	47.50 [31.70, 78.05]	<0.001
CSF TP < 60	834 (64.0)	<0.001	834 (64.0)	<0.001
CSF TP ≥ 60	469 (36.0)	<0.001	469 (36.0)	<0.001
	68.00 [58.00, 85.00]	0.001	68.00 [58.00, 85.00]	0.003
	3.00 [1.00, 21.00]	<0.001	3.00 [1.00, 21.00]	<0.001
CSF TNC < 5	769 (59.0)	<0.001	769 (59.0)	<0.001
CSF TNC ≥ 5	535 (41.0)	<0.001	535 (41.0)	<0.001
	66.00 [25.00, 86.00]	<0.001	66.00 [25.00, 86.00]	<0.001
Normal CSF Cell Count (CSF TNC < 5)	769 (59.0)	<0.001	769 (59.0)	<0.001
Pleocytosis (Lymph ≥ 50%)	350 (26.8)	<0.001	350 (26.8)	<0.001
Pleocytosis (Lymph < 50%)	185 (14.2)	<0.001	185 (14.2)	<0.001

Table New: Demographics by Immune Status

# ── Build for OLD and NEW stratified by immunosup ─────────────────────────────
tn_old <- build_tableone_df(data_old, table1_vars, "immunosup", continuous_vars, factor_vars_t1)

## [1] "--- Entering copyPvalueToAllLevels function ---"
## [1] "Processing variable: sex"
## [1] "  Factor levels for sex : Male, Female"
## [1] "  Rows in df_tableone for these levels: 2, 3"
## [1] "  Rows with p-value in factor levels of sex : 2"
## [1] "  P-value found for sex : 0.090"
## [1] "  P-value copied to rows: 2, 3"
## [1] "Processing variable: race"
## [1] "  Factor levels for race : White, Black, Other/Multiple, Unknown"
## [1] "  Rows in df_tableone for these levels: 5, 6, 7, 8, 9"
## [1] "  Rows with p-value in factor levels of race : 5, 9"
## [1] "  P-value found for race : 0.074"
## [1] "  P-value copied to rows: 5, 6, 7, 8, 9"
## [1] "Processing variable: nonwhite"
## [1] "  Factor levels for nonwhite : White, Non-White"
## [1] "  Rows in df_tableone for these levels: 5, 9, 10"
## [1] "  Rows with p-value in factor levels of nonwhite : 5, 9"
## [1] "  P-value found for nonwhite : 0.074"
## [1] "  P-value copied to rows: 5, 9, 10"
## [1] "Processing variable: dead"
## [1] "  Factor levels for dead : Not Reported Dead, Reported Dead"
## [1] "  Rows in df_tableone for these levels: 11, 12"
## [1] "  Rows with p-value in factor levels of dead : 11"
## [1] "  P-value found for dead : <0.001"
## [1] "  P-value copied to rows: 11, 12"
## [1] "Processing variable: medicaid"
## [1] "  Factor levels for medicaid : Other Insurance, Medicaid/Safety Net"
## [1] "  Rows in df_tableone for these levels: 13, 14"
## [1] "  Rows with p-value in factor levels of medicaid : 13"
## [1] "  P-value found for medicaid : 0.696"
## [1] "  P-value copied to rows: 13, 14"
## [1] "Processing variable: site"
## [1] "  Factor levels for site : Mass General Hospital, Brigham and Women's Hospital"
## [1] "  Rows in df_tableone for these levels: 15, 16"
## [1] "  Rows with p-value in factor levels of site : 15"
## [1] "  P-value found for site : <0.001"
## [1] "  P-value copied to rows: 15, 16"
## [1] "Processing variable: service"
## [1] "  Factor levels for service : Neurology, Medicine/Subspecialties, Oncology, Other"
## [1] "  Rows in df_tableone for these levels: 17, 18, 19, 20"
## [1] "  Rows with p-value in factor levels of service : 17"
## [1] "  P-value found for service : <0.001"
## [1] "  P-value copied to rows: 17, 18, 19, 20"
## [1] "Processing variable: immunosup"
## [1] "  Factor levels for immunosup : Not Immunosuppressed, Immunosuppressed"
## [1] "  Rows in df_tableone for these levels: 21, 22"
## [1] "  Rows with p-value in factor levels of immunosup : 21"
## [1] "  P-value found for immunosup : <0.001"
## [1] "  P-value copied to rows: 21, 22"
## [1] "Processing variable: transplant"
## [1] "  Factor levels for transplant : No Transplant History, History of Transplant"
## [1] "  Rows in df_tableone for these levels: 23, 24"
## [1] "  Rows with p-value in factor levels of transplant : 23"
## [1] "  P-value found for transplant : <0.001"
## [1] "  P-value copied to rows: 23, 24"
## [1] "Processing variable: bcellmed"
## [1] "  Factor levels for bcellmed : No B-cell Depleting Meds, B-cell Depletion"
## [1] "  Rows in df_tableone for these levels: 25, 26"
## [1] "  Rows with p-value in factor levels of bcellmed : 25"
## [1] "  P-value found for bcellmed : <0.001"
## [1] "  P-value copied to rows: 25, 26"
## [1] "Processing variable: inseasontest"
## [1] "  Factor levels for inseasontest : Test Sent Off-Season, Test Sent June-Nov"
## [1] "  Rows in df_tableone for these levels: 28, 29"
## [1] "  Rows with p-value in factor levels of inseasontest : 28"
## [1] "  P-value found for inseasontest : 0.255"
## [1] "  P-value copied to rows: 28, 29"
## [1] "Processing variable: csftp60"
## [1] "  Factor levels for csftp60 : CSF TP < 60, CSF TP ≥ 60"
## [1] "  Rows in df_tableone for these levels: 32, 33"
## [1] "  Rows with p-value in factor levels of csftp60 : 32"
## [1] "  P-value found for csftp60 : 0.645"
## [1] "  P-value copied to rows: 32, 33"
## [1] "Processing variable: csftnc5"
## [1] "  Factor levels for csftnc5 : CSF TNC < 5, CSF TNC ≥ 5"
## [1] "  Rows in df_tableone for these levels: 36, 37"
## [1] "  Rows with p-value in factor levels of csftnc5 : 36"
## [1] "  P-value found for csftnc5 : 0.658"
## [1] "  P-value copied to rows: 36, 37"
## [1] "Processing variable: csflymph50"
## [1] "  Factor levels for csflymph50 : Normal CSF Cell Count (CSF TNC < 5), Pleocytosis (Lymph ≥ 50%), Pleocytosis (Lymph < 50%)"
## [1] "  Rows in df_tableone for these levels: 39, 40, 41"
## [1] "  Rows with p-value in factor levels of csflymph50 : 39"
## [1] "  P-value found for csflymph50 : 0.651"
## [1] "  P-value copied to rows: 39, 40, 41"
## [1] "--- Exiting copyPvalueToAllLevels function ---"
## [1] "--- End Debugging ---"

tn_new <- build_tableone_df(data_new, table1_vars, "immunosup", continuous_vars, factor_vars_t1)

## [1] "--- Entering copyPvalueToAllLevels function ---"
## [1] "Processing variable: sex"
## [1] "  Factor levels for sex : Male, Female"
## [1] "  Rows in df_tableone for these levels: 2, 3"
## [1] "  Rows with p-value in factor levels of sex : 2"
## [1] "  P-value found for sex : 0.040"
## [1] "  P-value copied to rows: 2, 3"
## [1] "Processing variable: race"
## [1] "  Factor levels for race : White, Black, Other/Multiple, Unknown"
## [1] "  Rows in df_tableone for these levels: 5, 6, 7, 8, 9"
## [1] "  Rows with p-value in factor levels of race : 5, 9"
## [1] "  P-value found for race : 0.615"
## [1] "  P-value copied to rows: 5, 6, 7, 8, 9"
## [1] "Processing variable: nonwhite"
## [1] "  Factor levels for nonwhite : White, Non-White"
## [1] "  Rows in df_tableone for these levels: 5, 9, 10"
## [1] "  Rows with p-value in factor levels of nonwhite : 5, 9"
## [1] "  P-value found for nonwhite : 0.615"
## [1] "  P-value copied to rows: 5, 9, 10"
## [1] "Processing variable: dead"
## [1] "  Factor levels for dead : Not Reported Dead, Reported Dead"
## [1] "  Rows in df_tableone for these levels: 11, 12"
## [1] "  Rows with p-value in factor levels of dead : 11"
## [1] "  P-value found for dead : <0.001"
## [1] "  P-value copied to rows: 11, 12"
## [1] "Processing variable: medicaid"
## [1] "  Factor levels for medicaid : Other Insurance, Medicaid/Safety Net"
## [1] "  Rows in df_tableone for these levels: 13, 14"
## [1] "  Rows with p-value in factor levels of medicaid : 13"
## [1] "  P-value found for medicaid : 0.262"
## [1] "  P-value copied to rows: 13, 14"
## [1] "Processing variable: site"
## [1] "  Factor levels for site : Mass General Hospital, Brigham and Women's Hospital"
## [1] "  Rows in df_tableone for these levels: 15, 16"
## [1] "  Rows with p-value in factor levels of site : 15"
## [1] "  P-value found for site : <0.001"
## [1] "  P-value copied to rows: 15, 16"
## [1] "Processing variable: service"
## [1] "  Factor levels for service : Neurology, Medicine/Subspecialties, Oncology, Other"
## [1] "  Rows in df_tableone for these levels: 17, 18, 19, 20"
## [1] "  Rows with p-value in factor levels of service : 17"
## [1] "  P-value found for service : <0.001"
## [1] "  P-value copied to rows: 17, 18, 19, 20"
## [1] "Processing variable: immunosup"
## [1] "  Factor levels for immunosup : Not Immunosuppressed, Immunosuppressed"
## [1] "  Rows in df_tableone for these levels: 21, 22"
## [1] "  Rows with p-value in factor levels of immunosup : 21"
## [1] "  P-value found for immunosup : <0.001"
## [1] "  P-value copied to rows: 21, 22"
## [1] "Processing variable: transplant"
## [1] "  Factor levels for transplant : No Transplant History, History of Transplant"
## [1] "  Rows in df_tableone for these levels: 23, 24"
## [1] "  Rows with p-value in factor levels of transplant : 23"
## [1] "  P-value found for transplant : <0.001"
## [1] "  P-value copied to rows: 23, 24"
## [1] "Processing variable: bcellmed"
## [1] "  Factor levels for bcellmed : No B-cell Depleting Meds, B-cell Depletion"
## [1] "  Rows in df_tableone for these levels: 25, 26"
## [1] "  Rows with p-value in factor levels of bcellmed : 25"
## [1] "  P-value found for bcellmed : <0.001"
## [1] "  P-value copied to rows: 25, 26"
## [1] "Processing variable: inseasontest"
## [1] "  Factor levels for inseasontest : Test Sent Off-Season, Test Sent June-Nov"
## [1] "  Rows in df_tableone for these levels: 28, 29"
## [1] "  Rows with p-value in factor levels of inseasontest : 28"
## [1] "  P-value found for inseasontest : 0.210"
## [1] "  P-value copied to rows: 28, 29"
## [1] "Processing variable: csftp60"
## [1] "  Factor levels for csftp60 : CSF TP < 60, CSF TP ≥ 60"
## [1] "  Rows in df_tableone for these levels: 32, 33"
## [1] "  Rows with p-value in factor levels of csftp60 : 32"
## [1] "  P-value found for csftp60 : 0.931"
## [1] "  P-value copied to rows: 32, 33"
## [1] "Processing variable: csftnc5"
## [1] "  Factor levels for csftnc5 : CSF TNC < 5, CSF TNC ≥ 5"
## [1] "  Rows in df_tableone for these levels: 36, 37"
## [1] "  Rows with p-value in factor levels of csftnc5 : 36"
## [1] "  P-value found for csftnc5 : 0.582"
## [1] "  P-value copied to rows: 36, 37"
## [1] "Processing variable: csflymph50"
## [1] "  Factor levels for csflymph50 : Normal CSF Cell Count (CSF TNC < 5), Pleocytosis (Lymph ≥ 50%), Pleocytosis (Lymph < 50%)"
## [1] "  Rows in df_tableone for these levels: 39, 40, 41"
## [1] "  Rows with p-value in factor levels of csflymph50 : 39"
## [1] "  P-value found for csflymph50 : 0.767"
## [1] "  P-value copied to rows: 39, 40, 41"
## [1] "--- Exiting copyPvalueToAllLevels function ---"
## [1] "--- End Debugging ---"

cat("### Table New — OLD Dataset\n")

## ### Table New — OLD Dataset

knitr::kable(tn_old, caption = "Table New (OLD): Demographics & Lab Values by Immune Status")

Table New (OLD): Demographics & Lab Values by Immune Status

	level	Overall	Not Immunosuppressed	Immunosuppressed	p	row_id
n		1304	1133	171		n
sex….	Male	703 (53.9)	600 (53.0)	103 (60.2)	0.090	sex….
X	Female	601 (46.1)	533 (47.0)	68 (39.8)	0.090	X
csfageyr..median..IQR..		63.00 [48.00, 73.00]	63.00 [47.00, 73.15]	65.00 [57.50, 71.50]	0.108	csfageyr..median..IQR..
race….	White	994 (76.2)	850 (75.0)	144 (84.2)	0.074	race….
X.1	Black	105 (8.1)	96 (8.5)	9 (5.3)	0.074	X.1
X.2	Other/Multiple	129 (9.9)	118 (10.4)	11 (6.4)	0.074	X.2
X.3	Unknown	76 (5.8)	69 (6.1)	7 (4.1)	0.074	X.3
nonwhite….	White	1070 (82.1)	919 (81.1)	151 (88.3)	0.074	nonwhite….
X.4	Non-White	234 (17.9)	214 (18.9)	20 (11.7)	0.074	X.4
dead….	Not Reported Dead	915 (70.2)	838 (74.0)	77 (45.0)	<0.001	dead….
X.5	Reported Dead	389 (29.8)	295 (26.0)	94 (55.0)	<0.001	X.5
medicaid….	Other Insurance	933 (71.5)	808 (71.3)	125 (73.1)	0.696	medicaid….
X.6	Medicaid/Safety Net	371 (28.5)	325 (28.7)	46 (26.9)	0.696	X.6
site….	Mass General Hospital	706 (54.1)	655 (57.8)	51 (29.8)	<0.001	site….
X.7	Brigham and Women’s Hospital	598 (45.9)	478 (42.2)	120 (70.2)	<0.001	X.7
service….	Neurology	584 (44.8)	564 (49.8)	20 (11.7)	<0.001	service….
X.8	Medicine/Subspecialties	456 (35.0)	403 (35.6)	53 (31.0)	<0.001	X.8
X.9	Oncology	178 (13.7)	96 (8.5)	82 (48.0)	<0.001	X.9
X.10	Other	86 (6.6)	70 (6.2)	16 (9.4)	<0.001	X.10
immunosup….	Not Immunosuppressed	1133 (86.9)	1133 (100.0)	0 (0.0)	<0.001	immunosup….
X.11	Immunosuppressed	171 (13.1)	0 (0.0)	171 (100.0)	<0.001	X.11
transplant….	No Transplant History	1160 (89.0)	1133 (100.0)	27 (15.8)	<0.001	transplant….
X.12	History of Transplant	144 (11.0)	0 (0.0)	144 (84.2)	<0.001	X.12
bcellmed….	No B-cell Depleting Meds	1271 (97.5)	1133 (100.0)	138 (80.7)	<0.001	bcellmed….
X.13	B-cell Depletion	33 (2.5)	0 (0.0)	33 (19.3)	<0.001	X.13
csfadmit_day..median..IQR..		1.00 [1.00, 4.00]	1.00 [0.00, 3.00]	4.00 [1.00, 11.00]	<0.001	csfadmit_day..median..IQR..
inseasontest….	Test Sent Off-Season	479 (36.7)	409 (36.1)	70 (40.9)	0.255	inseasontest….
X.14	Test Sent June-Nov	825 (63.3)	724 (63.9)	101 (59.1)	0.255	X.14
serumwbc..median..IQR..		8.17 [5.68, 11.40]	8.57 [6.17, 11.49]	5.10 [3.15, 8.38]	<0.001	serumwbc..median..IQR..
csftp..median..IQR..		47.50 [31.70, 78.05]	48.00 [31.00, 78.00]	46.55 [33.92, 80.03]	0.935	csftp..median..IQR..
csftp60….	CSF TP < 60	834 (64.0)	722 (63.7)	112 (65.9)	0.645	csftp60….
X.15	CSF TP ≥ 60	469 (36.0)	411 (36.3)	58 (34.1)	0.645	X.15
csfgluc..median..IQR..		68.00 [58.00, 85.00]	68.00 [58.00, 84.50]	67.00 [57.50, 88.50]	0.890	csfgluc..median..IQR..
csftnc..median..IQR..		3.00 [1.00, 21.00]	3.00 [1.00, 23.00]	3.00 [1.00, 12.50]	0.463	csftnc..median..IQR..
csftnc5….	CSF TNC < 5	769 (59.0)	665 (58.7)	104 (60.8)	0.658	csftnc5….
X.16	CSF TNC ≥ 5	535 (41.0)	468 (41.3)	67 (39.2)	0.658	X.16
csflymph..median..IQR..		66.00 [25.00, 86.00]	65.00 [25.00, 86.00]	66.00 [20.50, 87.00]	0.691	csflymph..median..IQR..
csflymph50….	Normal CSF Cell Count (CSF TNC < 5)	769 (59.0)	665 (58.7)	104 (60.8)	0.651	csflymph50….
X.17	Pleocytosis (Lymph ≥ 50%)	350 (26.8)	309 (27.3)	41 (24.0)	0.651	X.17
X.18	Pleocytosis (Lymph < 50%)	185 (14.2)	159 (14.0)	26 (15.2)	0.651	X.18

cat("\n### Table New — NEW Dataset\n")

## 
## ### Table New — NEW Dataset

knitr::kable(tn_new, caption = "Table New (NEW): Demographics & Lab Values by Immune Status")

Table New (NEW): Demographics & Lab Values by Immune Status

	level	Overall	Not Immunosuppressed	Immunosuppressed	p	row_id
n		1304	1114	190		n
sex….	Male	703 (53.9)	587 (52.7)	116 (61.1)	0.040	sex….
X	Female	601 (46.1)	527 (47.3)	74 (38.9)	0.040	X
csfageyr..median..IQR..		63.00 [48.00, 73.00]	63.00 [47.00, 74.00]	65.00 [55.24, 71.00]	0.345	csfageyr..median..IQR..
race….	White	994 (76.2)	843 (75.7)	151 (79.5)	0.615	race….
X.1	Black	105 (8.1)	90 (8.1)	15 (7.9)	0.615	X.1
X.2	Other/Multiple	129 (9.9)	113 (10.1)	16 (8.4)	0.615	X.2
X.3	Unknown	76 (5.8)	68 (6.1)	8 (4.2)	0.615	X.3
nonwhite….	White	1070 (82.1)	911 (81.8)	159 (83.7)	0.615	nonwhite….
X.4	Non-White	234 (17.9)	203 (18.2)	31 (16.3)	0.615	X.4
dead….	Not Reported Dead	915 (70.2)	826 (74.1)	89 (46.8)	<0.001	dead….
X.5	Reported Dead	389 (29.8)	288 (25.9)	101 (53.2)	<0.001	X.5
medicaid….	Other Insurance	933 (71.5)	804 (72.2)	129 (67.9)	0.262	medicaid….
X.6	Medicaid/Safety Net	371 (28.5)	310 (27.8)	61 (32.1)	0.262	X.6
site….	Mass General Hospital	706 (54.1)	641 (57.5)	65 (34.2)	<0.001	site….
X.7	Brigham and Women’s Hospital	598 (45.9)	473 (42.5)	125 (65.8)	<0.001	X.7
service….	Neurology	584 (44.8)	558 (50.1)	26 (13.7)	<0.001	service….
X.8	Medicine/Subspecialties	456 (35.0)	391 (35.1)	65 (34.2)	<0.001	X.8
X.9	Oncology	178 (13.7)	95 (8.5)	83 (43.7)	<0.001	X.9
X.10	Other	86 (6.6)	70 (6.3)	16 (8.4)	<0.001	X.10
immunosup….	Not Immunosuppressed	1114 (85.4)	1114 (100.0)	0 (0.0)	<0.001	immunosup….
X.11	Immunosuppressed	190 (14.6)	0 (0.0)	190 (100.0)	<0.001	X.11
transplant….	No Transplant History	1160 (89.0)	1114 (100.0)	46 (24.2)	<0.001	transplant….
X.12	History of Transplant	144 (11.0)	0 (0.0)	144 (75.8)	<0.001	X.12
bcellmed….	No B-cell Depleting Meds	1271 (97.5)	1114 (100.0)	157 (82.6)	<0.001	bcellmed….
X.13	B-cell Depletion	33 (2.5)	0 (0.0)	33 (17.4)	<0.001	X.13
csfadmit_day..median..IQR..		1.00 [1.00, 4.00]	1.00 [0.00, 3.00]	3.00 [1.00, 10.00]	<0.001	csfadmit_day..median..IQR..
inseasontest….	Test Sent Off-Season	479 (36.7)	401 (36.0)	78 (41.1)	0.210	inseasontest….
X.14	Test Sent June-Nov	825 (63.3)	713 (64.0)	112 (58.9)	0.210	X.14
serumwbc..median..IQR..		8.17 [5.68, 11.40]	8.60 [6.19, 11.48]	5.14 [3.14, 8.46]	<0.001	serumwbc..median..IQR..
csftp..median..IQR..		47.50 [31.70, 78.05]	47.95 [31.00, 78.00]	47.10 [33.90, 86.00]	0.721	csftp..median..IQR..
csftp60….	CSF TP < 60	834 (64.0)	712 (63.9)	122 (64.6)	0.931	csftp60….
X.15	CSF TP ≥ 60	469 (36.0)	402 (36.1)	67 (35.4)	0.931	X.15
csfgluc..median..IQR..		68.00 [58.00, 85.00]	68.00 [58.00, 84.25]	67.00 [56.25, 88.75]	0.935	csfgluc..median..IQR..
csftnc..median..IQR..		3.00 [1.00, 21.00]	3.00 [1.00, 24.00]	3.00 [1.00, 11.75]	0.377	csftnc..median..IQR..
csftnc5….	CSF TNC < 5	769 (59.0)	653 (58.6)	116 (61.1)	0.582	csftnc5….
X.16	CSF TNC ≥ 5	535 (41.0)	461 (41.4)	74 (38.9)	0.582	X.16
csflymph..median..IQR..		66.00 [25.00, 86.00]	65.00 [25.00, 86.00]	66.50 [23.50, 87.00]	0.976	csflymph..median..IQR..
csflymph50….	Normal CSF Cell Count (CSF TNC < 5)	769 (59.0)	653 (58.6)	116 (61.1)	0.767	csflymph50….
X.17	Pleocytosis (Lymph ≥ 50%)	350 (26.8)	303 (27.2)	47 (24.7)	0.767	X.17
X.18	Pleocytosis (Lymph < 50%)	185 (14.2)	158 (14.2)	27 (14.2)	0.767	X.18

# ── Delta table with highlighting ─────────────────────────────────────────────
delta_tn_result <- build_delta_gt(tn_old, tn_new, "Table New Delta: Demographics by Immune Status")
delta_tn <- delta_tn_result$data
delta_tn_result$gt

Table New Delta: Demographics by Immune Status
	OLD Dataset		NEW Dataset
	Overall (OLD)	P (OLD)	Overall (NEW)	P (NEW)
	1304		1304
Male	703 (53.9)	0.090	703 (53.9)	0.040
Female	601 (46.1)	0.090	601 (46.1)	0.040
	63.00 [48.00, 73.00]	0.108	63.00 [48.00, 73.00]	0.345
White	994 (76.2)	0.074	994 (76.2)	0.615
Black	105 (8.1)	0.074	105 (8.1)	0.615
Other/Multiple	129 (9.9)	0.074	129 (9.9)	0.615
Unknown	76 (5.8)	0.074	76 (5.8)	0.615
White	1070 (82.1)	0.074	1070 (82.1)	0.615
Non-White	234 (17.9)	0.074	234 (17.9)	0.615
Not Reported Dead	915 (70.2)	<0.001	915 (70.2)	<0.001
Reported Dead	389 (29.8)	<0.001	389 (29.8)	<0.001
Other Insurance	933 (71.5)	0.696	933 (71.5)	0.262
Medicaid/Safety Net	371 (28.5)	0.696	371 (28.5)	0.262
Mass General Hospital	706 (54.1)	<0.001	706 (54.1)	<0.001
Brigham and Women's Hospital	598 (45.9)	<0.001	598 (45.9)	<0.001
Neurology	584 (44.8)	<0.001	584 (44.8)	<0.001
Medicine/Subspecialties	456 (35.0)	<0.001	456 (35.0)	<0.001
Oncology	178 (13.7)	<0.001	178 (13.7)	<0.001
Other	86 (6.6)	<0.001	86 (6.6)	<0.001
Not Immunosuppressed	1133 (86.9)	<0.001	1114 (85.4)	<0.001
Immunosuppressed	171 (13.1)	<0.001	190 (14.6)	<0.001
No Transplant History	1160 (89.0)	<0.001	1160 (89.0)	<0.001
History of Transplant	144 (11.0)	<0.001	144 (11.0)	<0.001
No B-cell Depleting Meds	1271 (97.5)	<0.001	1271 (97.5)	<0.001
B-cell Depletion	33 (2.5)	<0.001	33 (2.5)	<0.001
	1.00 [1.00, 4.00]	<0.001	1.00 [1.00, 4.00]	<0.001
Test Sent Off-Season	479 (36.7)	0.255	479 (36.7)	0.210
Test Sent June-Nov	825 (63.3)	0.255	825 (63.3)	0.210
	8.17 [5.68, 11.40]	<0.001	8.17 [5.68, 11.40]	<0.001
	47.50 [31.70, 78.05]	0.935	47.50 [31.70, 78.05]	0.721
CSF TP < 60	834 (64.0)	0.645	834 (64.0)	0.931
CSF TP ≥ 60	469 (36.0)	0.645	469 (36.0)	0.931
	68.00 [58.00, 85.00]	0.890	68.00 [58.00, 85.00]	0.935
	3.00 [1.00, 21.00]	0.463	3.00 [1.00, 21.00]	0.377
CSF TNC < 5	769 (59.0)	0.658	769 (59.0)	0.582
CSF TNC ≥ 5	535 (41.0)	0.658	535 (41.0)	0.582
	66.00 [25.00, 86.00]	0.691	66.00 [25.00, 86.00]	0.976
Normal CSF Cell Count (CSF TNC < 5)	769 (59.0)	0.651	769 (59.0)	0.767
Pleocytosis (Lymph ≥ 50%)	350 (26.8)	0.651	350 (26.8)	0.767
Pleocytosis (Lymph < 50%)	185 (14.2)	0.651	185 (14.2)	0.767

Table 2: Demographics by WNV Positivity

# ── Build for OLD and NEW stratified by wnvpos ────────────────────────────────
t2_old <- build_tableone_df(data_old, table1_vars, "wnvpos", continuous_vars, factor_vars_t1)

## [1] "--- Entering copyPvalueToAllLevels function ---"
## [1] "Processing variable: sex"
## [1] "  Factor levels for sex : Male, Female"
## [1] "  Rows in df_tableone for these levels: 2, 3"
## [1] "  Rows with p-value in factor levels of sex : 2"
## [1] "  P-value found for sex : 0.848"
## [1] "  P-value copied to rows: 2, 3"
## [1] "Processing variable: race"
## [1] "  Factor levels for race : White, Black, Other/Multiple, Unknown"
## [1] "  Rows in df_tableone for these levels: 5, 6, 7, 8, 9"
## [1] "  Rows with p-value in factor levels of race : 5, 9"
## [1] "  P-value found for race : 0.446"
## [1] "  P-value copied to rows: 5, 6, 7, 8, 9"
## [1] "Processing variable: nonwhite"
## [1] "  Factor levels for nonwhite : White, Non-White"
## [1] "  Rows in df_tableone for these levels: 5, 9, 10"
## [1] "  Rows with p-value in factor levels of nonwhite : 5, 9"
## [1] "  P-value found for nonwhite : 0.446"
## [1] "  P-value copied to rows: 5, 9, 10"
## [1] "Processing variable: dead"
## [1] "  Factor levels for dead : Not Reported Dead, Reported Dead"
## [1] "  Rows in df_tableone for these levels: 11, 12"
## [1] "  Rows with p-value in factor levels of dead : 11"
## [1] "  P-value found for dead : 0.587"
## [1] "  P-value copied to rows: 11, 12"
## [1] "Processing variable: medicaid"
## [1] "  Factor levels for medicaid : Other Insurance, Medicaid/Safety Net"
## [1] "  Rows in df_tableone for these levels: 13, 14"
## [1] "  Rows with p-value in factor levels of medicaid : 13"
## [1] "  P-value found for medicaid : 0.694"
## [1] "  P-value copied to rows: 13, 14"
## [1] "Processing variable: site"
## [1] "  Factor levels for site : Mass General Hospital, Brigham and Women's Hospital"
## [1] "  Rows in df_tableone for these levels: 15, 16"
## [1] "  Rows with p-value in factor levels of site : 15"
## [1] "  P-value found for site : 0.011"
## [1] "  P-value copied to rows: 15, 16"
## [1] "Processing variable: service"
## [1] "  Factor levels for service : Neurology, Medicine/Subspecialties, Oncology, Other"
## [1] "  Rows in df_tableone for these levels: 17, 18, 19, 20"
## [1] "  Rows with p-value in factor levels of service : 17"
## [1] "  P-value found for service : 0.146"
## [1] "  P-value copied to rows: 17, 18, 19, 20"
## [1] "Processing variable: immunosup"
## [1] "  Factor levels for immunosup : Not Immunosuppressed, Immunosuppressed"
## [1] "  Rows in df_tableone for these levels: 21, 22"
## [1] "  Rows with p-value in factor levels of immunosup : 21"
## [1] "  P-value found for immunosup : 0.262"
## [1] "  P-value copied to rows: 21, 22"
## [1] "Processing variable: transplant"
## [1] "  Factor levels for transplant : No Transplant History, History of Transplant"
## [1] "  Rows in df_tableone for these levels: 23, 24"
## [1] "  Rows with p-value in factor levels of transplant : 23"
## [1] "  P-value found for transplant : 0.386"
## [1] "  P-value copied to rows: 23, 24"
## [1] "Processing variable: bcellmed"
## [1] "  Factor levels for bcellmed : No B-cell Depleting Meds, B-cell Depletion"
## [1] "  Rows in df_tableone for these levels: 25, 26"
## [1] "  Rows with p-value in factor levels of bcellmed : 25"
## [1] "  P-value found for bcellmed : 0.842"
## [1] "  P-value copied to rows: 25, 26"
## [1] "Processing variable: inseasontest"
## [1] "  Factor levels for inseasontest : Test Sent Off-Season, Test Sent June-Nov"
## [1] "  Rows in df_tableone for these levels: 28, 29"
## [1] "  Rows with p-value in factor levels of inseasontest : 28"
## [1] "  P-value found for inseasontest : <0.001"
## [1] "  P-value copied to rows: 28, 29"
## [1] "Processing variable: csftp60"
## [1] "  Factor levels for csftp60 : CSF TP < 60, CSF TP ≥ 60"
## [1] "  Rows in df_tableone for these levels: 32, 33"
## [1] "  Rows with p-value in factor levels of csftp60 : 32"
## [1] "  P-value found for csftp60 : 0.001"
## [1] "  P-value copied to rows: 32, 33"
## [1] "Processing variable: csftnc5"
## [1] "  Factor levels for csftnc5 : CSF TNC < 5, CSF TNC ≥ 5"
## [1] "  Rows in df_tableone for these levels: 36, 37"
## [1] "  Rows with p-value in factor levels of csftnc5 : 36"
## [1] "  P-value found for csftnc5 : <0.001"
## [1] "  P-value copied to rows: 36, 37"
## [1] "Processing variable: csflymph50"
## [1] "  Factor levels for csflymph50 : Normal CSF Cell Count (CSF TNC < 5), Pleocytosis (Lymph ≥ 50%), Pleocytosis (Lymph < 50%)"
## [1] "  Rows in df_tableone for these levels: 39, 40, 41"
## [1] "  Rows with p-value in factor levels of csflymph50 : 39"
## [1] "  P-value found for csflymph50 : <0.001"
## [1] "  P-value copied to rows: 39, 40, 41"
## [1] "--- Exiting copyPvalueToAllLevels function ---"
## [1] "--- End Debugging ---"

t2_new <- build_tableone_df(data_new, table1_vars, "wnvpos", continuous_vars, factor_vars_t1)

## [1] "--- Entering copyPvalueToAllLevels function ---"
## [1] "Processing variable: sex"
## [1] "  Factor levels for sex : Male, Female"
## [1] "  Rows in df_tableone for these levels: 2, 3"
## [1] "  Rows with p-value in factor levels of sex : 2"
## [1] "  P-value found for sex : 0.848"
## [1] "  P-value copied to rows: 2, 3"
## [1] "Processing variable: race"
## [1] "  Factor levels for race : White, Black, Other/Multiple, Unknown"
## [1] "  Rows in df_tableone for these levels: 5, 6, 7, 8, 9"
## [1] "  Rows with p-value in factor levels of race : 5, 9"
## [1] "  P-value found for race : 0.446"
## [1] "  P-value copied to rows: 5, 6, 7, 8, 9"
## [1] "Processing variable: nonwhite"
## [1] "  Factor levels for nonwhite : White, Non-White"
## [1] "  Rows in df_tableone for these levels: 5, 9, 10"
## [1] "  Rows with p-value in factor levels of nonwhite : 5, 9"
## [1] "  P-value found for nonwhite : 0.446"
## [1] "  P-value copied to rows: 5, 9, 10"
## [1] "Processing variable: dead"
## [1] "  Factor levels for dead : Not Reported Dead, Reported Dead"
## [1] "  Rows in df_tableone for these levels: 11, 12"
## [1] "  Rows with p-value in factor levels of dead : 11"
## [1] "  P-value found for dead : 0.587"
## [1] "  P-value copied to rows: 11, 12"
## [1] "Processing variable: medicaid"
## [1] "  Factor levels for medicaid : Other Insurance, Medicaid/Safety Net"
## [1] "  Rows in df_tableone for these levels: 13, 14"
## [1] "  Rows with p-value in factor levels of medicaid : 13"
## [1] "  P-value found for medicaid : 0.694"
## [1] "  P-value copied to rows: 13, 14"
## [1] "Processing variable: site"
## [1] "  Factor levels for site : Mass General Hospital, Brigham and Women's Hospital"
## [1] "  Rows in df_tableone for these levels: 15, 16"
## [1] "  Rows with p-value in factor levels of site : 15"
## [1] "  P-value found for site : 0.011"
## [1] "  P-value copied to rows: 15, 16"
## [1] "Processing variable: service"
## [1] "  Factor levels for service : Neurology, Medicine/Subspecialties, Oncology, Other"
## [1] "  Rows in df_tableone for these levels: 17, 18, 19, 20"
## [1] "  Rows with p-value in factor levels of service : 17"
## [1] "  P-value found for service : 0.146"
## [1] "  P-value copied to rows: 17, 18, 19, 20"
## [1] "Processing variable: immunosup"
## [1] "  Factor levels for immunosup : Not Immunosuppressed, Immunosuppressed"
## [1] "  Rows in df_tableone for these levels: 21, 22"
## [1] "  Rows with p-value in factor levels of immunosup : 21"
## [1] "  P-value found for immunosup : 0.199"
## [1] "  P-value copied to rows: 21, 22"
## [1] "Processing variable: transplant"
## [1] "  Factor levels for transplant : No Transplant History, History of Transplant"
## [1] "  Rows in df_tableone for these levels: 23, 24"
## [1] "  Rows with p-value in factor levels of transplant : 23"
## [1] "  P-value found for transplant : 0.386"
## [1] "  P-value copied to rows: 23, 24"
## [1] "Processing variable: bcellmed"
## [1] "  Factor levels for bcellmed : No B-cell Depleting Meds, B-cell Depletion"
## [1] "  Rows in df_tableone for these levels: 25, 26"
## [1] "  Rows with p-value in factor levels of bcellmed : 25"
## [1] "  P-value found for bcellmed : 0.842"
## [1] "  P-value copied to rows: 25, 26"
## [1] "Processing variable: inseasontest"
## [1] "  Factor levels for inseasontest : Test Sent Off-Season, Test Sent June-Nov"
## [1] "  Rows in df_tableone for these levels: 28, 29"
## [1] "  Rows with p-value in factor levels of inseasontest : 28"
## [1] "  P-value found for inseasontest : <0.001"
## [1] "  P-value copied to rows: 28, 29"
## [1] "Processing variable: csftp60"
## [1] "  Factor levels for csftp60 : CSF TP < 60, CSF TP ≥ 60"
## [1] "  Rows in df_tableone for these levels: 32, 33"
## [1] "  Rows with p-value in factor levels of csftp60 : 32"
## [1] "  P-value found for csftp60 : 0.001"
## [1] "  P-value copied to rows: 32, 33"
## [1] "Processing variable: csftnc5"
## [1] "  Factor levels for csftnc5 : CSF TNC < 5, CSF TNC ≥ 5"
## [1] "  Rows in df_tableone for these levels: 36, 37"
## [1] "  Rows with p-value in factor levels of csftnc5 : 36"
## [1] "  P-value found for csftnc5 : <0.001"
## [1] "  P-value copied to rows: 36, 37"
## [1] "Processing variable: csflymph50"
## [1] "  Factor levels for csflymph50 : Normal CSF Cell Count (CSF TNC < 5), Pleocytosis (Lymph ≥ 50%), Pleocytosis (Lymph < 50%)"
## [1] "  Rows in df_tableone for these levels: 39, 40, 41"
## [1] "  Rows with p-value in factor levels of csflymph50 : 39"
## [1] "  P-value found for csflymph50 : <0.001"
## [1] "  P-value copied to rows: 39, 40, 41"
## [1] "--- Exiting copyPvalueToAllLevels function ---"
## [1] "--- End Debugging ---"

cat("### Table 2 — OLD Dataset\n")

## ### Table 2 — OLD Dataset

knitr::kable(t2_old, caption = "Table 2 (OLD): Demographics & Lab Values by WNV Positivity")

Table 2 (OLD): Demographics & Lab Values by WNV Positivity

	level	Overall	Positive WNV	No Positive Result	p	row_id
n		1304	26	1278		n
sex….	Male	703 (53.9)	15 (57.7)	688 (53.8)	0.848	sex….
X	Female	601 (46.1)	11 (42.3)	590 (46.2)	0.848	X
csfageyr..median..IQR..		63.00 [48.00, 73.00]	73.00 [62.00, 79.71]	63.00 [48.00, 73.00]	0.008	csfageyr..median..IQR..
race….	White	994 (76.2)	22 (84.6)	972 (76.1)	0.446	race….
X.1	Black	105 (8.1)	0 (0.0)	105 (8.2)	0.446	X.1
X.2	Other/Multiple	129 (9.9)	2 (7.7)	127 (9.9)	0.446	X.2
X.3	Unknown	76 (5.8)	2 (7.7)	74 (5.8)	0.446	X.3
nonwhite….	White	1070 (82.1)	24 (92.3)	1046 (81.8)	0.446	nonwhite….
X.4	Non-White	234 (17.9)	2 (7.7)	232 (18.2)	0.446	X.4
dead….	Not Reported Dead	915 (70.2)	20 (76.9)	895 (70.0)	0.587	dead….
X.5	Reported Dead	389 (29.8)	6 (23.1)	383 (30.0)	0.587	X.5
medicaid….	Other Insurance	933 (71.5)	20 (76.9)	913 (71.4)	0.694	medicaid….
X.6	Medicaid/Safety Net	371 (28.5)	6 (23.1)	365 (28.6)	0.694	X.6
site….	Mass General Hospital	706 (54.1)	21 (80.8)	685 (53.6)	0.011	site….
X.7	Brigham and Women’s Hospital	598 (45.9)	5 (19.2)	593 (46.4)	0.011	X.7
service….	Neurology	584 (44.8)	10 (38.5)	574 (44.9)	0.146	service….
X.8	Medicine/Subspecialties	456 (35.0)	14 (53.8)	442 (34.6)	0.146	X.8
X.9	Oncology	178 (13.7)	2 (7.7)	176 (13.8)	0.146	X.9
X.10	Other	86 (6.6)	0 (0.0)	86 (6.7)	0.146	X.10
immunosup….	Not Immunosuppressed	1133 (86.9)	25 (96.2)	1108 (86.7)	0.262	immunosup….
X.11	Immunosuppressed	171 (13.1)	1 (3.8)	170 (13.3)	0.262	X.11
transplant….	No Transplant History	1160 (89.0)	25 (96.2)	1135 (88.8)	0.386	transplant….
X.12	History of Transplant	144 (11.0)	1 (3.8)	143 (11.2)	0.386	X.12
bcellmed….	No B-cell Depleting Meds	1271 (97.5)	26 (100.0)	1245 (97.4)	0.842	bcellmed….
X.13	B-cell Depletion	33 (2.5)	0 (0.0)	33 (2.6)	0.842	X.13
csfadmit_day..median..IQR..		1.00 [1.00, 4.00]	1.00 [0.25, 3.75]	1.00 [1.00, 4.00]	0.484	csfadmit_day..median..IQR..
inseasontest….	Test Sent Off-Season	479 (36.7)	0 (0.0)	479 (37.5)	<0.001	inseasontest….
X.14	Test Sent June-Nov	825 (63.3)	26 (100.0)	799 (62.5)	<0.001	X.14
serumwbc..median..IQR..		8.17 [5.68, 11.40]	9.04 [7.88, 11.23]	8.13 [5.65, 11.40]	0.128	serumwbc..median..IQR..
csftp..median..IQR..		47.50 [31.70, 78.05]	69.50 [54.00, 113.25]	47.00 [31.20, 77.80]	0.001	csftp..median..IQR..
csftp60….	CSF TP < 60	834 (64.0)	8 (30.8)	826 (64.7)	0.001	csftp60….
X.15	CSF TP ≥ 60	469 (36.0)	18 (69.2)	451 (35.3)	0.001	X.15
csfgluc..median..IQR..		68.00 [58.00, 85.00]	60.50 [53.25, 70.00]	68.00 [58.00, 85.00]	0.025	csfgluc..median..IQR..
csftnc..median..IQR..		3.00 [1.00, 21.00]	49.50 [12.25, 99.50]	3.00 [1.00, 18.00]	<0.001	csftnc..median..IQR..
csftnc5….	CSF TNC < 5	769 (59.0)	4 (15.4)	765 (59.9)	<0.001	csftnc5….
X.16	CSF TNC ≥ 5	535 (41.0)	22 (84.6)	513 (40.1)	<0.001	X.16
csflymph..median..IQR..		66.00 [25.00, 86.00]	56.50 [26.25, 75.25]	66.00 [25.00, 86.75]	0.598	csflymph..median..IQR..
csflymph50….	Normal CSF Cell Count (CSF TNC < 5)	769 (59.0)	4 (15.4)	765 (59.9)	<0.001	csflymph50….
X.17	Pleocytosis (Lymph ≥ 50%)	350 (26.8)	12 (46.2)	338 (26.4)	<0.001	X.17
X.18	Pleocytosis (Lymph < 50%)	185 (14.2)	10 (38.5)	175 (13.7)	<0.001	X.18

cat("\n### Table 2 — NEW Dataset\n")

## 
## ### Table 2 — NEW Dataset

knitr::kable(t2_new, caption = "Table 2 (NEW): Demographics & Lab Values by WNV Positivity")

Table 2 (NEW): Demographics & Lab Values by WNV Positivity

	level	Overall	Positive WNV	No Positive Result	p	row_id
n		1304	26	1278		n
sex….	Male	703 (53.9)	15 (57.7)	688 (53.8)	0.848	sex….
X	Female	601 (46.1)	11 (42.3)	590 (46.2)	0.848	X
csfageyr..median..IQR..		63.00 [48.00, 73.00]	73.00 [62.00, 79.71]	63.00 [48.00, 73.00]	0.008	csfageyr..median..IQR..
race….	White	994 (76.2)	22 (84.6)	972 (76.1)	0.446	race….
X.1	Black	105 (8.1)	0 (0.0)	105 (8.2)	0.446	X.1
X.2	Other/Multiple	129 (9.9)	2 (7.7)	127 (9.9)	0.446	X.2
X.3	Unknown	76 (5.8)	2 (7.7)	74 (5.8)	0.446	X.3
nonwhite….	White	1070 (82.1)	24 (92.3)	1046 (81.8)	0.446	nonwhite….
X.4	Non-White	234 (17.9)	2 (7.7)	232 (18.2)	0.446	X.4
dead….	Not Reported Dead	915 (70.2)	20 (76.9)	895 (70.0)	0.587	dead….
X.5	Reported Dead	389 (29.8)	6 (23.1)	383 (30.0)	0.587	X.5
medicaid….	Other Insurance	933 (71.5)	20 (76.9)	913 (71.4)	0.694	medicaid….
X.6	Medicaid/Safety Net	371 (28.5)	6 (23.1)	365 (28.6)	0.694	X.6
site….	Mass General Hospital	706 (54.1)	21 (80.8)	685 (53.6)	0.011	site….
X.7	Brigham and Women’s Hospital	598 (45.9)	5 (19.2)	593 (46.4)	0.011	X.7
service….	Neurology	584 (44.8)	10 (38.5)	574 (44.9)	0.146	service….
X.8	Medicine/Subspecialties	456 (35.0)	14 (53.8)	442 (34.6)	0.146	X.8
X.9	Oncology	178 (13.7)	2 (7.7)	176 (13.8)	0.146	X.9
X.10	Other	86 (6.6)	0 (0.0)	86 (6.7)	0.146	X.10
immunosup….	Not Immunosuppressed	1114 (85.4)	25 (96.2)	1089 (85.2)	0.199	immunosup….
X.11	Immunosuppressed	190 (14.6)	1 (3.8)	189 (14.8)	0.199	X.11
transplant….	No Transplant History	1160 (89.0)	25 (96.2)	1135 (88.8)	0.386	transplant….
X.12	History of Transplant	144 (11.0)	1 (3.8)	143 (11.2)	0.386	X.12
bcellmed….	No B-cell Depleting Meds	1271 (97.5)	26 (100.0)	1245 (97.4)	0.842	bcellmed….
X.13	B-cell Depletion	33 (2.5)	0 (0.0)	33 (2.6)	0.842	X.13
csfadmit_day..median..IQR..		1.00 [1.00, 4.00]	1.00 [0.25, 3.75]	1.00 [1.00, 4.00]	0.484	csfadmit_day..median..IQR..
inseasontest….	Test Sent Off-Season	479 (36.7)	0 (0.0)	479 (37.5)	<0.001	inseasontest….
X.14	Test Sent June-Nov	825 (63.3)	26 (100.0)	799 (62.5)	<0.001	X.14
serumwbc..median..IQR..		8.17 [5.68, 11.40]	9.04 [7.88, 11.23]	8.13 [5.65, 11.40]	0.128	serumwbc..median..IQR..
csftp..median..IQR..		47.50 [31.70, 78.05]	69.50 [54.00, 113.25]	47.00 [31.20, 77.80]	0.001	csftp..median..IQR..
csftp60….	CSF TP < 60	834 (64.0)	8 (30.8)	826 (64.7)	0.001	csftp60….
X.15	CSF TP ≥ 60	469 (36.0)	18 (69.2)	451 (35.3)	0.001	X.15
csfgluc..median..IQR..		68.00 [58.00, 85.00]	60.50 [53.25, 70.00]	68.00 [58.00, 85.00]	0.025	csfgluc..median..IQR..
csftnc..median..IQR..		3.00 [1.00, 21.00]	49.50 [12.25, 99.50]	3.00 [1.00, 18.00]	<0.001	csftnc..median..IQR..
csftnc5….	CSF TNC < 5	769 (59.0)	4 (15.4)	765 (59.9)	<0.001	csftnc5….
X.16	CSF TNC ≥ 5	535 (41.0)	22 (84.6)	513 (40.1)	<0.001	X.16
csflymph..median..IQR..		66.00 [25.00, 86.00]	56.50 [26.25, 75.25]	66.00 [25.00, 86.75]	0.598	csflymph..median..IQR..
csflymph50….	Normal CSF Cell Count (CSF TNC < 5)	769 (59.0)	4 (15.4)	765 (59.9)	<0.001	csflymph50….
X.17	Pleocytosis (Lymph ≥ 50%)	350 (26.8)	12 (46.2)	338 (26.4)	<0.001	X.17
X.18	Pleocytosis (Lymph < 50%)	185 (14.2)	10 (38.5)	175 (13.7)	<0.001	X.18

# ── Delta table with highlighting ─────────────────────────────────────────────
delta_t2_result <- build_delta_gt(t2_old, t2_new, "Table 2 Delta: Demographics by WNV Positivity")
delta_t2 <- delta_t2_result$data
delta_t2_result$gt

Table 2 Delta: Demographics by WNV Positivity
	OLD Dataset		NEW Dataset
	Overall (OLD)	P (OLD)	Overall (NEW)	P (NEW)
	1304		1304
Male	703 (53.9)	0.848	703 (53.9)	0.848
Female	601 (46.1)	0.848	601 (46.1)	0.848
	63.00 [48.00, 73.00]	0.008	63.00 [48.00, 73.00]	0.008
White	994 (76.2)	0.446	994 (76.2)	0.446
Black	105 (8.1)	0.446	105 (8.1)	0.446
Other/Multiple	129 (9.9)	0.446	129 (9.9)	0.446
Unknown	76 (5.8)	0.446	76 (5.8)	0.446
White	1070 (82.1)	0.446	1070 (82.1)	0.446
Non-White	234 (17.9)	0.446	234 (17.9)	0.446
Not Reported Dead	915 (70.2)	0.587	915 (70.2)	0.587
Reported Dead	389 (29.8)	0.587	389 (29.8)	0.587
Other Insurance	933 (71.5)	0.694	933 (71.5)	0.694
Medicaid/Safety Net	371 (28.5)	0.694	371 (28.5)	0.694
Mass General Hospital	706 (54.1)	0.011	706 (54.1)	0.011
Brigham and Women's Hospital	598 (45.9)	0.011	598 (45.9)	0.011
Neurology	584 (44.8)	0.146	584 (44.8)	0.146
Medicine/Subspecialties	456 (35.0)	0.146	456 (35.0)	0.146
Oncology	178 (13.7)	0.146	178 (13.7)	0.146
Other	86 (6.6)	0.146	86 (6.6)	0.146
Not Immunosuppressed	1133 (86.9)	0.262	1114 (85.4)	0.199
Immunosuppressed	171 (13.1)	0.262	190 (14.6)	0.199
No Transplant History	1160 (89.0)	0.386	1160 (89.0)	0.386
History of Transplant	144 (11.0)	0.386	144 (11.0)	0.386
No B-cell Depleting Meds	1271 (97.5)	0.842	1271 (97.5)	0.842
B-cell Depletion	33 (2.5)	0.842	33 (2.5)	0.842
	1.00 [1.00, 4.00]	0.484	1.00 [1.00, 4.00]	0.484
Test Sent Off-Season	479 (36.7)	<0.001	479 (36.7)	<0.001
Test Sent June-Nov	825 (63.3)	<0.001	825 (63.3)	<0.001
	8.17 [5.68, 11.40]	0.128	8.17 [5.68, 11.40]	0.128
	47.50 [31.70, 78.05]	0.001	47.50 [31.70, 78.05]	0.001
CSF TP < 60	834 (64.0)	0.001	834 (64.0)	0.001
CSF TP ≥ 60	469 (36.0)	0.001	469 (36.0)	0.001
	68.00 [58.00, 85.00]	0.025	68.00 [58.00, 85.00]	0.025
	3.00 [1.00, 21.00]	<0.001	3.00 [1.00, 21.00]	<0.001
CSF TNC < 5	769 (59.0)	<0.001	769 (59.0)	<0.001
CSF TNC ≥ 5	535 (41.0)	<0.001	535 (41.0)	<0.001
	66.00 [25.00, 86.00]	0.598	66.00 [25.00, 86.00]	0.598
Normal CSF Cell Count (CSF TNC < 5)	769 (59.0)	<0.001	769 (59.0)	<0.001
Pleocytosis (Lymph ≥ 50%)	350 (26.8)	<0.001	350 (26.8)	<0.001
Pleocytosis (Lymph < 50%)	185 (14.2)	<0.001	185 (14.2)	<0.001

Table 3: WNV Testing by Guideline Adherence

# ── Table 3 variables ─────────────────────────────────────────────────────────
table3_vars <- c(
  "csfpcr_ordered", "csfpcr_rslt", "csfigm_ordered", "csfigm_rslt",
  "serumigm_ordered", "serumigm_rslt", "guideline"
)
factor_vars_t3 <- table3_vars

# ── Helper: build Table 3/4 style TableOne (all categorical) ──────────────────
build_tableone_cat <- function(data, vars, strata, fac_vars) {
  tobj <- tableone::CreateTableOne(
    vars = vars, strata = strata, data = data,
    factorVars = fac_vars, addOverall = TRUE
  )
  tmat <- print(tobj, showAllLevels = TRUE, printToggle = FALSE, noSpaces = TRUE)
  tdf <- as.data.frame(tmat, stringsAsFactors = FALSE)

  # Preserve unique rownames as row_id BEFORE any modifications
  tdf$row_id <- rownames(tdf)

  colnames(tdf)[1] <- "level"
  tdf$level <- trimws(tdf$level)

  fld <- lapply(fac_vars, function(x) levels(data[[x]]))
  names(fld) <- fac_vars
  tdf <- suppressMessages(copyPvalueToAllLevels(tdf, fld))

  # Relabel row headers
  row_label_dict <- c(
    "csfpcr_ordered...."   = "CSF PCR Ordered",
    "csfpcr_rslt...."      = "CSF PCR Result",
    "csfigm_ordered...."   = "CSF IgM Ordered",
    "csfigm_rslt...."      = "CSF IgM Result",
    "serumigm_ordered...." = "Serum IgM Ordered",
    "serumigm_rslt...."    = "Serum IgM Result",
    "guideline...."        = "Guideline Adherence"
  )
  for (old_nm in names(row_label_dict)) {
    new_nm <- row_label_dict[[old_nm]]
    tdf$level[rownames(tdf) == old_nm] <- new_nm
  }

  tdf <- tdf[-1, ]  # Remove n row
  if ("test" %in% colnames(tdf)) tdf <- dplyr::select(tdf, -test)
  return(tdf)
}

# ── Build Table 3 for OLD and NEW ────────────────────────────────────────────
t3_old <- build_tableone_cat(data_old, table3_vars, "guideline2", factor_vars_t3)

## [1] "--- Entering copyPvalueToAllLevels function ---"
## [1] "Processing variable: csfpcr_ordered"
## [1] "  Factor levels for csfpcr_ordered : CSF PCR Not Ordered, CSF PCR Ordered"
## [1] "  Rows in df_tableone for these levels: 2, 3"
## [1] "  Rows with p-value in factor levels of csfpcr_ordered : 2"
## [1] "  P-value found for csfpcr_ordered : <0.001"
## [1] "  P-value copied to rows: 2, 3"
## [1] "Processing variable: csfpcr_rslt"
## [1] "  Factor levels for csfpcr_rslt : Unresulted, Positive, Negative, Not Sent"
## [1] "  Rows in df_tableone for these levels: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "  Rows with p-value in factor levels of csfpcr_rslt : 4, 10, 16"
## [1] "  P-value found for csfpcr_rslt : <0.001"
## [1] "  P-value copied to rows: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "Processing variable: csfigm_ordered"
## [1] "  Factor levels for csfigm_ordered : CSF IgM Not Ordered, CSF IgM Ordered"
## [1] "  Rows in df_tableone for these levels: 8, 9"
## [1] "  Rows with p-value in factor levels of csfigm_ordered : 8"
## [1] "  P-value found for csfigm_ordered : <0.001"
## [1] "  P-value copied to rows: 8, 9"
## [1] "Processing variable: csfigm_rslt"
## [1] "  Factor levels for csfigm_rslt : Unresulted, Positive, Negative, Not Sent"
## [1] "  Rows in df_tableone for these levels: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "  Rows with p-value in factor levels of csfigm_rslt : 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "  P-value found for csfigm_rslt : <0.001"
## [1] "  P-value copied to rows: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "Processing variable: serumigm_ordered"
## [1] "  Factor levels for serumigm_ordered : Serum IgM Not Ordered, Serum IgM Ordered"
## [1] "  Rows in df_tableone for these levels: 14, 15"
## [1] "  Rows with p-value in factor levels of serumigm_ordered : 14"
## [1] "  P-value found for serumigm_ordered : <0.001"
## [1] "  P-value copied to rows: 14, 15"
## [1] "Processing variable: serumigm_rslt"
## [1] "  Factor levels for serumigm_rslt : Unresulted, Positive, Negative, Not Sent"
## [1] "  Rows in df_tableone for these levels: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "  Rows with p-value in factor levels of serumigm_rslt : 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "  P-value found for serumigm_rslt : <0.001"
## [1] "  P-value copied to rows: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "Processing variable: guideline"
## [1] "  Factor levels for guideline : Does Not Follow, Follows, Follows but Over-Tested"
## [1] "  Rows in df_tableone for these levels: 20, 21, 22"
## [1] "  Rows with p-value in factor levels of guideline : 20"
## [1] "  P-value found for guideline : <0.001"
## [1] "  P-value copied to rows: 20, 21, 22"
## [1] "--- Exiting copyPvalueToAllLevels function ---"
## [1] "--- End Debugging ---"

t3_new <- build_tableone_cat(data_new, table3_vars, "guideline2", factor_vars_t3)

## [1] "--- Entering copyPvalueToAllLevels function ---"
## [1] "Processing variable: csfpcr_ordered"
## [1] "  Factor levels for csfpcr_ordered : CSF PCR Not Ordered, CSF PCR Ordered"
## [1] "  Rows in df_tableone for these levels: 2, 3"
## [1] "  Rows with p-value in factor levels of csfpcr_ordered : 2"
## [1] "  P-value found for csfpcr_ordered : <0.001"
## [1] "  P-value copied to rows: 2, 3"
## [1] "Processing variable: csfpcr_rslt"
## [1] "  Factor levels for csfpcr_rslt : Unresulted, Positive, Negative, Not Sent"
## [1] "  Rows in df_tableone for these levels: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "  Rows with p-value in factor levels of csfpcr_rslt : 4, 10, 16"
## [1] "  P-value found for csfpcr_rslt : <0.001"
## [1] "  P-value copied to rows: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "Processing variable: csfigm_ordered"
## [1] "  Factor levels for csfigm_ordered : CSF IgM Not Ordered, CSF IgM Ordered"
## [1] "  Rows in df_tableone for these levels: 8, 9"
## [1] "  Rows with p-value in factor levels of csfigm_ordered : 8"
## [1] "  P-value found for csfigm_ordered : <0.001"
## [1] "  P-value copied to rows: 8, 9"
## [1] "Processing variable: csfigm_rslt"
## [1] "  Factor levels for csfigm_rslt : Unresulted, Positive, Negative, Not Sent"
## [1] "  Rows in df_tableone for these levels: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "  Rows with p-value in factor levels of csfigm_rslt : 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "  P-value found for csfigm_rslt : <0.001"
## [1] "  P-value copied to rows: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "Processing variable: serumigm_ordered"
## [1] "  Factor levels for serumigm_ordered : Serum IgM Not Ordered, Serum IgM Ordered"
## [1] "  Rows in df_tableone for these levels: 14, 15"
## [1] "  Rows with p-value in factor levels of serumigm_ordered : 14"
## [1] "  P-value found for serumigm_ordered : <0.001"
## [1] "  P-value copied to rows: 14, 15"
## [1] "Processing variable: serumigm_rslt"
## [1] "  Factor levels for serumigm_rslt : Unresulted, Positive, Negative, Not Sent"
## [1] "  Rows in df_tableone for these levels: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "  Rows with p-value in factor levels of serumigm_rslt : 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "  P-value found for serumigm_rslt : <0.001"
## [1] "  P-value copied to rows: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "Processing variable: guideline"
## [1] "  Factor levels for guideline : Does Not Follow, Follows, Follows but Over-Tested"
## [1] "  Rows in df_tableone for these levels: 20, 21, 22"
## [1] "  Rows with p-value in factor levels of guideline : 20"
## [1] "  P-value found for guideline : <0.001"
## [1] "  P-value copied to rows: 20, 21, 22"
## [1] "--- Exiting copyPvalueToAllLevels function ---"
## [1] "--- End Debugging ---"

cat("### Table 3 — OLD Dataset\n")

## ### Table 3 — OLD Dataset

knitr::kable(t3_old, caption = "Table 3 (OLD): WNV Testing by Guideline Adherence")

Table 3 (OLD): WNV Testing by Guideline Adherence

	level	Overall	Adherent	Nonadherent	p	row_id
csfpcr_ordered….	CSF PCR Ordered	171 (13.1)	153 (45.9)	18 (1.9)	<0.001	csfpcr_ordered….
X	CSF PCR Ordered	1133 (86.9)	180 (54.1)	953 (98.1)	<0.001	X
csfpcr_rslt….	CSF PCR Result	38 (2.9)	12 (3.6)	26 (2.7)	<0.001	csfpcr_rslt….
X.1	Positive	4 (0.3)	2 (0.6)	2 (0.2)	<0.001	X.1
X.2	Negative	1091 (83.7)	166 (49.8)	925 (95.3)	<0.001	X.2
X.3	Not Sent	171 (13.1)	153 (45.9)	18 (1.9)	<0.001	X.3
csfigm_ordered….	CSF IgM Ordered	953 (73.1)	0 (0.0)	953 (98.1)	<0.001	csfigm_ordered….
X.4	CSF IgM Ordered	351 (26.9)	333 (100.0)	18 (1.9)	<0.001	X.4
csfigm_rslt….	CSF IgM Result	63 (4.8)	60 (18.0)	3 (0.3)	<0.001	csfigm_rslt….
X.5	Positive	20 (1.5)	19 (5.7)	1 (0.1)	<0.001	X.5
X.6	Negative	268 (20.6)	254 (76.3)	14 (1.4)	<0.001	X.6
X.7	Not Sent	953 (73.1)	0 (0.0)	953 (98.1)	<0.001	X.7
serumigm_ordered….	Serum IgM Ordered	1169 (89.6)	257 (77.2)	912 (93.9)	<0.001	serumigm_ordered….
X.8	Serum IgM Ordered	135 (10.4)	76 (22.8)	59 (6.1)	<0.001	X.8
serumigm_rslt….	Serum IgM Result	14 (1.1)	7 (2.1)	7 (0.7)	<0.001	serumigm_rslt….
X.9	Positive	11 (0.8)	9 (2.7)	2 (0.2)	<0.001	X.9
X.10	Negative	110 (8.4)	60 (18.0)	50 (5.1)	<0.001	X.10
X.11	Not Sent	1169 (89.6)	257 (77.2)	912 (93.9)	<0.001	X.11
guideline….	Guideline Adherence	971 (74.5)	0 (0.0)	971 (100.0)	<0.001	guideline….
X.12	Follows	167 (12.8)	167 (50.2)	0 (0.0)	<0.001	X.12
X.13	Follows but Over-Tested	166 (12.7)	166 (49.8)	0 (0.0)	<0.001	X.13

cat("\n### Table 3 — NEW Dataset\n")

## 
## ### Table 3 — NEW Dataset

knitr::kable(t3_new, caption = "Table 3 (NEW): WNV Testing by Guideline Adherence")

Table 3 (NEW): WNV Testing by Guideline Adherence

	level	Overall	Adherent	Nonadherent	p	row_id
csfpcr_ordered….	CSF PCR Ordered	171 (13.1)	150 (45.5)	21 (2.2)	<0.001	csfpcr_ordered….
X	CSF PCR Ordered	1133 (86.9)	180 (54.5)	953 (97.8)	<0.001	X
csfpcr_rslt….	CSF PCR Result	38 (2.9)	12 (3.6)	26 (2.7)	<0.001	csfpcr_rslt….
X.1	Positive	4 (0.3)	2 (0.6)	2 (0.2)	<0.001	X.1
X.2	Negative	1091 (83.7)	166 (50.3)	925 (95.0)	<0.001	X.2
X.3	Not Sent	171 (13.1)	150 (45.5)	21 (2.2)	<0.001	X.3
csfigm_ordered….	CSF IgM Ordered	953 (73.1)	0 (0.0)	953 (97.8)	<0.001	csfigm_ordered….
X.4	CSF IgM Ordered	351 (26.9)	330 (100.0)	21 (2.2)	<0.001	X.4
csfigm_rslt….	CSF IgM Result	63 (4.8)	59 (17.9)	4 (0.4)	<0.001	csfigm_rslt….
X.5	Positive	20 (1.5)	19 (5.8)	1 (0.1)	<0.001	X.5
X.6	Negative	268 (20.6)	252 (76.4)	16 (1.6)	<0.001	X.6
X.7	Not Sent	953 (73.1)	0 (0.0)	953 (97.8)	<0.001	X.7
serumigm_ordered….	Serum IgM Ordered	1169 (89.6)	255 (77.3)	914 (93.8)	<0.001	serumigm_ordered….
X.8	Serum IgM Ordered	135 (10.4)	75 (22.7)	60 (6.2)	<0.001	X.8
serumigm_rslt….	Serum IgM Result	14 (1.1)	6 (1.8)	8 (0.8)	<0.001	serumigm_rslt….
X.9	Positive	11 (0.8)	9 (2.7)	2 (0.2)	<0.001	X.9
X.10	Negative	110 (8.4)	60 (18.2)	50 (5.1)	<0.001	X.10
X.11	Not Sent	1169 (89.6)	255 (77.3)	914 (93.8)	<0.001	X.11
guideline….	Guideline Adherence	974 (74.7)	0 (0.0)	974 (100.0)	<0.001	guideline….
X.12	Follows	164 (12.6)	164 (49.7)	0 (0.0)	<0.001	X.12
X.13	Follows but Over-Tested	166 (12.7)	166 (50.3)	0 (0.0)	<0.001	X.13

# ── Delta table with highlighting ─────────────────────────────────────────────
delta_t3_result <- build_delta_gt(t3_old, t3_new, "Table 3 Delta: WNV Testing by Guideline Adherence")
delta_t3 <- delta_t3_result$data
delta_t3_result$gt

Table 3 Delta: WNV Testing by Guideline Adherence
	OLD Dataset		NEW Dataset
	Overall (OLD)	P (OLD)	Overall (NEW)	P (NEW)
CSF PCR Ordered	171 (13.1)	<0.001	171 (13.1)	<0.001
CSF PCR Ordered	1133 (86.9)	<0.001	1133 (86.9)	<0.001
CSF PCR Result	38 (2.9)	<0.001	38 (2.9)	<0.001
Positive	4 (0.3)	<0.001	4 (0.3)	<0.001
Negative	1091 (83.7)	<0.001	1091 (83.7)	<0.001
Not Sent	171 (13.1)	<0.001	171 (13.1)	<0.001
CSF IgM Ordered	953 (73.1)	<0.001	953 (73.1)	<0.001
CSF IgM Ordered	351 (26.9)	<0.001	351 (26.9)	<0.001
CSF IgM Result	63 (4.8)	<0.001	63 (4.8)	<0.001
Positive	20 (1.5)	<0.001	20 (1.5)	<0.001
Negative	268 (20.6)	<0.001	268 (20.6)	<0.001
Not Sent	953 (73.1)	<0.001	953 (73.1)	<0.001
Serum IgM Ordered	1169 (89.6)	<0.001	1169 (89.6)	<0.001
Serum IgM Ordered	135 (10.4)	<0.001	135 (10.4)	<0.001
Serum IgM Result	14 (1.1)	<0.001	14 (1.1)	<0.001
Positive	11 (0.8)	<0.001	11 (0.8)	<0.001
Negative	110 (8.4)	<0.001	110 (8.4)	<0.001
Not Sent	1169 (89.6)	<0.001	1169 (89.6)	<0.001
Guideline Adherence	971 (74.5)	<0.001	974 (74.7)	<0.001
Follows	167 (12.8)	<0.001	164 (12.6)	<0.001
Follows but Over-Tested	166 (12.7)	<0.001	166 (12.7)	<0.001

Table 4: WNV Testing by Admitting Service

# ── Table 4 variables ─────────────────────────────────────────────────────────
table4_vars <- c(
  "csfpcr_ordered", "csfpcr_rslt", "csfigm_ordered", "csfigm_rslt",
  "serumigm_ordered", "serumigm_rslt"
)
factor_vars_t4 <- table4_vars

# ── Build Table 4 for OLD and NEW ────────────────────────────────────────────
build_tableone_cat_service <- function(data, vars, fac_vars) {
  tobj <- tableone::CreateTableOne(
    vars = vars, strata = "service", data = data,
    factorVars = fac_vars, addOverall = TRUE
  )
  tmat <- print(tobj, showAllLevels = TRUE, printToggle = FALSE, noSpaces = TRUE)
  tdf <- as.data.frame(tmat, stringsAsFactors = FALSE)

  # Preserve unique rownames as row_id BEFORE any modifications
  tdf$row_id <- rownames(tdf)

  colnames(tdf)[1] <- "level"
  tdf$level <- trimws(tdf$level)

  fld <- lapply(fac_vars, function(x) levels(data[[x]]))
  names(fld) <- fac_vars
  tdf <- suppressMessages(copyPvalueToAllLevels(tdf, fld))

  row_label_dict <- c(
    "csfpcr_ordered...."   = "CSF PCR Ordered",
    "csfpcr_rslt...."      = "CSF PCR Result",
    "csfigm_ordered...."   = "CSF IgM Ordered",
    "csfigm_rslt...."      = "CSF IgM Result",
    "serumigm_ordered...." = "Serum IgM Ordered",
    "serumigm_rslt...."    = "Serum IgM Result"
  )
  for (old_nm in names(row_label_dict)) {
    new_nm <- row_label_dict[[old_nm]]
    tdf$level[rownames(tdf) == old_nm] <- new_nm
  }

  tdf <- tdf[-1, ]
  if ("test" %in% colnames(tdf)) tdf <- dplyr::select(tdf, -test)
  return(tdf)
}

t4_old <- build_tableone_cat_service(data_old, table4_vars, factor_vars_t4)

## [1] "--- Entering copyPvalueToAllLevels function ---"
## [1] "Processing variable: csfpcr_ordered"
## [1] "  Factor levels for csfpcr_ordered : CSF PCR Not Ordered, CSF PCR Ordered"
## [1] "  Rows in df_tableone for these levels: 2, 3"
## [1] "  Rows with p-value in factor levels of csfpcr_ordered : 2"
## [1] "  P-value found for csfpcr_ordered : <0.001"
## [1] "  P-value copied to rows: 2, 3"
## [1] "Processing variable: csfpcr_rslt"
## [1] "  Factor levels for csfpcr_rslt : Unresulted, Positive, Negative, Not Sent"
## [1] "  Rows in df_tableone for these levels: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "  Rows with p-value in factor levels of csfpcr_rslt : 4, 10, 16"
## [1] "  P-value found for csfpcr_rslt : 0.004"
## [1] "  P-value copied to rows: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "Processing variable: csfigm_ordered"
## [1] "  Factor levels for csfigm_ordered : CSF IgM Not Ordered, CSF IgM Ordered"
## [1] "  Rows in df_tableone for these levels: 8, 9"
## [1] "  Rows with p-value in factor levels of csfigm_ordered : 8"
## [1] "  P-value found for csfigm_ordered : <0.001"
## [1] "  P-value copied to rows: 8, 9"
## [1] "Processing variable: csfigm_rslt"
## [1] "  Factor levels for csfigm_rslt : Unresulted, Positive, Negative, Not Sent"
## [1] "  Rows in df_tableone for these levels: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "  Rows with p-value in factor levels of csfigm_rslt : 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "  P-value found for csfigm_rslt : 0.004"
## [1] "  P-value copied to rows: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "Processing variable: serumigm_ordered"
## [1] "  Factor levels for serumigm_ordered : Serum IgM Not Ordered, Serum IgM Ordered"
## [1] "  Rows in df_tableone for these levels: 14, 15"
## [1] "  Rows with p-value in factor levels of serumigm_ordered : 14"
## [1] "  P-value found for serumigm_ordered : 0.016"
## [1] "  P-value copied to rows: 14, 15"
## [1] "Processing variable: serumigm_rslt"
## [1] "  Factor levels for serumigm_rslt : Unresulted, Positive, Negative, Not Sent"
## [1] "  Rows in df_tableone for these levels: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "  Rows with p-value in factor levels of serumigm_rslt : 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "  P-value found for serumigm_rslt : 0.004"
## [1] "  P-value copied to rows: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "--- Exiting copyPvalueToAllLevels function ---"
## [1] "--- End Debugging ---"

t4_new <- build_tableone_cat_service(data_new, table4_vars, factor_vars_t4)

## [1] "--- Entering copyPvalueToAllLevels function ---"
## [1] "Processing variable: csfpcr_ordered"
## [1] "  Factor levels for csfpcr_ordered : CSF PCR Not Ordered, CSF PCR Ordered"
## [1] "  Rows in df_tableone for these levels: 2, 3"
## [1] "  Rows with p-value in factor levels of csfpcr_ordered : 2"
## [1] "  P-value found for csfpcr_ordered : <0.001"
## [1] "  P-value copied to rows: 2, 3"
## [1] "Processing variable: csfpcr_rslt"
## [1] "  Factor levels for csfpcr_rslt : Unresulted, Positive, Negative, Not Sent"
## [1] "  Rows in df_tableone for these levels: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "  Rows with p-value in factor levels of csfpcr_rslt : 4, 10, 16"
## [1] "  P-value found for csfpcr_rslt : 0.004"
## [1] "  P-value copied to rows: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "Processing variable: csfigm_ordered"
## [1] "  Factor levels for csfigm_ordered : CSF IgM Not Ordered, CSF IgM Ordered"
## [1] "  Rows in df_tableone for these levels: 8, 9"
## [1] "  Rows with p-value in factor levels of csfigm_ordered : 8"
## [1] "  P-value found for csfigm_ordered : <0.001"
## [1] "  P-value copied to rows: 8, 9"
## [1] "Processing variable: csfigm_rslt"
## [1] "  Factor levels for csfigm_rslt : Unresulted, Positive, Negative, Not Sent"
## [1] "  Rows in df_tableone for these levels: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "  Rows with p-value in factor levels of csfigm_rslt : 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "  P-value found for csfigm_rslt : 0.004"
## [1] "  P-value copied to rows: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "Processing variable: serumigm_ordered"
## [1] "  Factor levels for serumigm_ordered : Serum IgM Not Ordered, Serum IgM Ordered"
## [1] "  Rows in df_tableone for these levels: 14, 15"
## [1] "  Rows with p-value in factor levels of serumigm_ordered : 14"
## [1] "  P-value found for serumigm_ordered : 0.016"
## [1] "  P-value copied to rows: 14, 15"
## [1] "Processing variable: serumigm_rslt"
## [1] "  Factor levels for serumigm_rslt : Unresulted, Positive, Negative, Not Sent"
## [1] "  Rows in df_tableone for these levels: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "  Rows with p-value in factor levels of serumigm_rslt : 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "  P-value found for serumigm_rslt : 0.004"
## [1] "  P-value copied to rows: 4, 5, 6, 7, 10, 11, 12, 13, 16, 17, 18, 19"
## [1] "--- Exiting copyPvalueToAllLevels function ---"
## [1] "--- End Debugging ---"

cat("### Table 4 — OLD Dataset\n")

## ### Table 4 — OLD Dataset

knitr::kable(t4_old, caption = "Table 4 (OLD): WNV Testing by Admitting Service")

Table 4 (OLD): WNV Testing by Admitting Service

	level	Overall	Neurology	Medicine/Subspecialties	Oncology	Other	p	row_id
csfpcr_ordered….	CSF PCR Ordered	171 (13.1)	101 (17.3)	52 (11.4)	10 (5.6)	8 (9.3)	<0.001	csfpcr_ordered….
X	CSF PCR Ordered	1133 (86.9)	483 (82.7)	404 (88.6)	168 (94.4)	78 (90.7)	<0.001	X
csfpcr_rslt….	CSF PCR Result	38 (2.9)	20 (3.4)	13 (2.9)	3 (1.7)	2 (2.3)	0.004	csfpcr_rslt….
X.1	Positive	4 (0.3)	3 (0.5)	1 (0.2)	0 (0.0)	0 (0.0)	0.004	X.1
X.2	Negative	1091 (83.7)	460 (78.8)	390 (85.5)	165 (92.7)	76 (88.4)	0.004	X.2
X.3	Not Sent	171 (13.1)	101 (17.3)	52 (11.4)	10 (5.6)	8 (9.3)	0.004	X.3
csfigm_ordered….	CSF IgM Ordered	953 (73.1)	369 (63.2)	363 (79.6)	150 (84.3)	71 (82.6)	<0.001	csfigm_ordered….
X.4	CSF IgM Ordered	351 (26.9)	215 (36.8)	93 (20.4)	28 (15.7)	15 (17.4)	<0.001	X.4
csfigm_rslt….	CSF IgM Result	63 (4.8)	37 (6.3)	15 (3.3)	6 (3.4)	5 (5.8)	0.004	csfigm_rslt….
X.5	Positive	20 (1.5)	5 (0.9)	13 (2.9)	2 (1.1)	0 (0.0)	0.004	X.5
X.6	Negative	268 (20.6)	173 (29.6)	65 (14.3)	20 (11.2)	10 (11.6)	0.004	X.6
X.7	Not Sent	953 (73.1)	369 (63.2)	363 (79.6)	150 (84.3)	71 (82.6)	0.004	X.7
serumigm_ordered….	Serum IgM Ordered	1169 (89.6)	506 (86.6)	419 (91.9)	165 (92.7)	79 (91.9)	0.016	serumigm_ordered….
X.8	Serum IgM Ordered	135 (10.4)	78 (13.4)	37 (8.1)	13 (7.3)	7 (8.1)	0.016	X.8
serumigm_rslt….	Serum IgM Result	14 (1.1)	3 (0.5)	9 (2.0)	0 (0.0)	2 (2.3)	0.004	serumigm_rslt….
X.9	Positive	11 (0.8)	4 (0.7)	6 (1.3)	1 (0.6)	0 (0.0)	0.004	X.9
X.10	Negative	110 (8.4)	71 (12.2)	22 (4.8)	12 (6.7)	5 (5.8)	0.004	X.10
X.11	Not Sent	1169 (89.6)	506 (86.6)	419 (91.9)	165 (92.7)	79 (91.9)	0.004	X.11

cat("\n### Table 4 — NEW Dataset\n")

## 
## ### Table 4 — NEW Dataset

knitr::kable(t4_new, caption = "Table 4 (NEW): WNV Testing by Admitting Service")

Table 4 (NEW): WNV Testing by Admitting Service

	level	Overall	Neurology	Medicine/Subspecialties	Oncology	Other	p	row_id
csfpcr_ordered….	CSF PCR Ordered	171 (13.1)	101 (17.3)	52 (11.4)	10 (5.6)	8 (9.3)	<0.001	csfpcr_ordered….
X	CSF PCR Ordered	1133 (86.9)	483 (82.7)	404 (88.6)	168 (94.4)	78 (90.7)	<0.001	X
csfpcr_rslt….	CSF PCR Result	38 (2.9)	20 (3.4)	13 (2.9)	3 (1.7)	2 (2.3)	0.004	csfpcr_rslt….
X.1	Positive	4 (0.3)	3 (0.5)	1 (0.2)	0 (0.0)	0 (0.0)	0.004	X.1
X.2	Negative	1091 (83.7)	460 (78.8)	390 (85.5)	165 (92.7)	76 (88.4)	0.004	X.2
X.3	Not Sent	171 (13.1)	101 (17.3)	52 (11.4)	10 (5.6)	8 (9.3)	0.004	X.3
csfigm_ordered….	CSF IgM Ordered	953 (73.1)	369 (63.2)	363 (79.6)	150 (84.3)	71 (82.6)	<0.001	csfigm_ordered….
X.4	CSF IgM Ordered	351 (26.9)	215 (36.8)	93 (20.4)	28 (15.7)	15 (17.4)	<0.001	X.4
csfigm_rslt….	CSF IgM Result	63 (4.8)	37 (6.3)	15 (3.3)	6 (3.4)	5 (5.8)	0.004	csfigm_rslt….
X.5	Positive	20 (1.5)	5 (0.9)	13 (2.9)	2 (1.1)	0 (0.0)	0.004	X.5
X.6	Negative	268 (20.6)	173 (29.6)	65 (14.3)	20 (11.2)	10 (11.6)	0.004	X.6
X.7	Not Sent	953 (73.1)	369 (63.2)	363 (79.6)	150 (84.3)	71 (82.6)	0.004	X.7
serumigm_ordered….	Serum IgM Ordered	1169 (89.6)	506 (86.6)	419 (91.9)	165 (92.7)	79 (91.9)	0.016	serumigm_ordered….
X.8	Serum IgM Ordered	135 (10.4)	78 (13.4)	37 (8.1)	13 (7.3)	7 (8.1)	0.016	X.8
serumigm_rslt….	Serum IgM Result	14 (1.1)	3 (0.5)	9 (2.0)	0 (0.0)	2 (2.3)	0.004	serumigm_rslt….
X.9	Positive	11 (0.8)	4 (0.7)	6 (1.3)	1 (0.6)	0 (0.0)	0.004	X.9
X.10	Negative	110 (8.4)	71 (12.2)	22 (4.8)	12 (6.7)	5 (5.8)	0.004	X.10
X.11	Not Sent	1169 (89.6)	506 (86.6)	419 (91.9)	165 (92.7)	79 (91.9)	0.004	X.11

# ── Delta table with highlighting ─────────────────────────────────────────────
delta_t4_result <- build_delta_gt(t4_old, t4_new, "Table 4 Delta: WNV Testing by Admitting Service")
delta_t4 <- delta_t4_result$data
delta_t4_result$gt

Table 4 Delta: WNV Testing by Admitting Service
	OLD Dataset		NEW Dataset
	Overall (OLD)	P (OLD)	Overall (NEW)	P (NEW)
CSF PCR Ordered	171 (13.1)	<0.001	171 (13.1)	<0.001
CSF PCR Ordered	1133 (86.9)	<0.001	1133 (86.9)	<0.001
CSF PCR Result	38 (2.9)	0.004	38 (2.9)	0.004
Positive	4 (0.3)	0.004	4 (0.3)	0.004
Negative	1091 (83.7)	0.004	1091 (83.7)	0.004
Not Sent	171 (13.1)	0.004	171 (13.1)	0.004
CSF IgM Ordered	953 (73.1)	<0.001	953 (73.1)	<0.001
CSF IgM Ordered	351 (26.9)	<0.001	351 (26.9)	<0.001
CSF IgM Result	63 (4.8)	0.004	63 (4.8)	0.004
Positive	20 (1.5)	0.004	20 (1.5)	0.004
Negative	268 (20.6)	0.004	268 (20.6)	0.004
Not Sent	953 (73.1)	0.004	953 (73.1)	0.004
Serum IgM Ordered	1169 (89.6)	0.016	1169 (89.6)	0.016
Serum IgM Ordered	135 (10.4)	0.016	135 (10.4)	0.016
Serum IgM Result	14 (1.1)	0.004	14 (1.1)	0.004
Positive	11 (0.8)	0.004	11 (0.8)	0.004
Negative	110 (8.4)	0.004	110 (8.4)	0.004
Not Sent	1169 (89.6)	0.004	1169 (89.6)	0.004

Figure 1: Yearly Adherence vs WNV Cases

# ── Helper: compute yearly summary ───────────────────────────────────────────
compute_yearly_summary <- function(data) {
  data %>%
    dplyr::filter(!is.na(csfdate)) %>%
    dplyr::mutate(
      year = format(as.Date(csfdate), "%Y"),
      is_adherent = (guideline2 == "Adherent"),
      is_wnv_positive = (wnvpos == "Positive WNV")
    ) %>%
    dplyr::group_by(year) %>%
    dplyr::summarise(
      adherence_rate   = mean(is_adherent, na.rm = TRUE) * 100,
      cohort_wnv_cases = sum(is_wnv_positive, na.rm = TRUE),
      .groups = "drop"
    ) %>%
    dplyr::filter(!is.na(year) & !is.na(adherence_rate) & !is.na(cohort_wnv_cases))
}

yearly_old <- compute_yearly_summary(data_old)
yearly_new <- compute_yearly_summary(data_new)

# ── Generate plots using helper function ──────────────────────────────────────
fig1_old <- plot_adherence_vs_cases(yearly_old) +
  ggplot2::ggtitle("OLD Dataset: Adherence vs WNV Cases")
fig1_new <- plot_adherence_vs_cases(yearly_new) +
  ggplot2::ggtitle("NEW Dataset: Adherence vs WNV Cases")

# ── Side-by-side with gridExtra ───────────────────────────────────────────────
fig1_combined <- gridExtra::arrangeGrob(
  fig1_old, fig1_new, ncol = 2,
  top = grid::textGrob("Figure 1 Comparison: Yearly Adherence vs WNV Cases",
                        gp = grid::gpar(fontsize = 16, fontface = "bold"))
)
grid::grid.draw(fig1_combined)

# ── Save ──────────────────────────────────────────────────────────────────────
ggplot2::ggsave(
  filename = fs::path(params$export_dir, "Figure1_Comparison_Adherence_vs_Cases.png"),
  plot = fig1_combined, width = 14, height = 6, dpi = 300, bg = "white"
)
message("Figure 1 comparison saved.")

# ── Yearly summary delta table ────────────────────────────────────────────────
yearly_delta <- dplyr::full_join(
  yearly_old %>% dplyr::rename(adherence_old = adherence_rate, cases_old = cohort_wnv_cases),
  yearly_new %>% dplyr::rename(adherence_new = adherence_rate, cases_new = cohort_wnv_cases),
  by = "year"
) %>% dplyr::mutate(
  delta_adherence = round(adherence_new - adherence_old, 2),
  delta_cases     = cases_new - cases_old
)
knitr::kable(yearly_delta, caption = "Yearly Summary Delta: OLD vs NEW")

Yearly Summary Delta: OLD vs NEW

year	adherence_old	cases_old	adherence_new	cases_new	delta_adherence	delta_cases
2016	33.33333	4	33.33333	4	0.00	0
2017	21.78218	2	21.28713	2	-0.50	0
2018	21.64948	11	21.64948	11	0.00	0
2019	22.63158	1	22.63158	1	0.00	0
2020	26.76056	0	26.05634	0	-0.70	0
2021	25.00000	3	25.00000	3	0.00	0
2022	24.26471	3	24.26471	3	0.00	0
2023	31.81818	2	31.25000	2	-0.57	0

Figure 2: Forest Plot (Logistic Regression)

# ── Helper: prepare model data ────────────────────────────────────────────────
prepare_model_data <- function(data) {
  data %>%
    dplyr::mutate(
      adherent_bin = ifelse(guideline2 == "Adherent", 1, 0),
      site = forcats::fct_relevel(site, "Brigham and Women's Hospital"),
      service2 = forcats::fct_collapse(service,
        "Neurology" = "Neurology",
        "Medicine/Other" = c("Medicine/Subspecialties", "Oncology", "Other")
      ) %>% forcats::fct_relevel("Medicine/Other"),
      immunosup = forcats::fct_relevel(immunosup, "Immunosuppressed"),
      csflymph_bin = forcats::fct_collapse(csflymph50,
        "Pleocytosis (Lymph >= 50%)" = "Pleocytosis (Lymph \u2265 50%)",
        "No pleocytosis / normal" = c("Normal CSF Cell Count (CSF TNC < 5)", "Pleocytosis (Lymph < 50%)")
      ) %>% forcats::fct_relevel("No pleocytosis / normal"),
      csftp60 = forcats::fct_relevel(csftp60, "CSF TP < 60")
    )
}

predictor_labels <- c(
  "Admission at MGH (vs BWH)",
  "Admission to Neurology (vs Medicine/Other)",
  "Immunocompetent (vs Immunosuppressed)",
  "Presence of CSF lymphocytic pleocytosis",
  "CSF protein elevated >60 mg/dL"
)

model_old <- prepare_model_data(data_old)
model_new <- prepare_model_data(data_new)

# ── Generate forest plots ─────────────────────────────────────────────────────
forest_old <- make_forest_plot_adherence(
  model_data = model_old,
  predictors = c("site", "service2", "immunosup", "csflymph_bin", "csftp60"),
  predictor_labels = predictor_labels
)
forest_new <- make_forest_plot_adherence(
  model_data = model_new,
  predictors = c("site", "service2", "immunosup", "csflymph_bin", "csftp60"),
  predictor_labels = predictor_labels
)

fig2_old <- forest_old$plot + ggplot2::ggtitle("OLD Dataset")
fig2_new <- forest_new$plot + ggplot2::ggtitle("NEW Dataset")

# ── Side-by-side with gridExtra ───────────────────────────────────────────────
fig2_combined <- gridExtra::arrangeGrob(
  fig2_old, fig2_new, ncol = 2,
  top = grid::textGrob("Figure 2 Comparison: Factors Associated with Guideline Adherence",
                        gp = grid::gpar(fontsize = 16, fontface = "bold"))
)
grid::grid.draw(fig2_combined)

# ── Save ──────────────────────────────────────────────────────────────────────
ggplot2::ggsave(
  filename = fs::path(params$export_dir, "Figure2_Comparison_Forest_Plot.png"),
  plot = fig2_combined, width = 14, height = 6, dpi = 300, bg = "white"
)
message("Figure 2 comparison saved.")

Regression Model Comparison

# ── Extract tidied OR tables from forest plot results ─────────────────────────
or_old <- forest_old$table %>%
  dplyr::select(predictor_display, estimate, conf.low, conf.high, p.value) %>%
  dplyr::rename(
    Predictor = predictor_display,
    OR_old = estimate, CI_low_old = conf.low, CI_high_old = conf.high, p_old = p.value
  )

or_new <- forest_new$table %>%
  dplyr::select(predictor_display, estimate, conf.low, conf.high, p.value) %>%
  dplyr::rename(
    Predictor = predictor_display,
    OR_new = estimate, CI_low_new = conf.low, CI_high_new = conf.high, p_new = p.value
  )

# ── Merge into comparison table ───────────────────────────────────────────────
or_compare <- dplyr::inner_join(or_old, or_new, by = "Predictor") %>%
  dplyr::mutate(
    OR_old_fmt  = sprintf("%.2f (%.2f - %.2f)", OR_old, CI_low_old, CI_high_old),
    OR_new_fmt  = sprintf("%.2f (%.2f - %.2f)", OR_new, CI_low_new, CI_high_new),
    Delta_OR    = round(OR_new - OR_old, 3),
    p_old_fmt   = ifelse(p_old < 0.001, "<0.001", sprintf("%.3f", p_old)),
    p_new_fmt   = ifelse(p_new < 0.001, "<0.001", sprintf("%.3f", p_new)),
    Sig_Change  = dplyr::case_when(
      p_old < 0.05 & p_new >= 0.05 ~ "Lost significance",
      p_old >= 0.05 & p_new < 0.05 ~ "Gained significance",
      TRUE ~ "No change"
    )
  ) %>%
  dplyr::select(Predictor, OR_old_fmt, p_old_fmt, OR_new_fmt, p_new_fmt, Delta_OR, Sig_Change)

# ── Display as gt table ──────────────────────────────────────────────────────
gt_regression <- or_compare %>%
  gt::gt() %>%
  gt::tab_header(
    title = gt::md("**Regression Model Comparison: OLD vs NEW**"),
    subtitle = "Logistic regression predicting guideline adherence"
  ) %>%
  gt::cols_label(
    Predictor   = "Predictor",
    OR_old_fmt  = "OR (95% CI) — OLD",
    p_old_fmt   = "P-value — OLD",
    OR_new_fmt  = "OR (95% CI) — NEW",
    p_new_fmt   = "P-value — NEW",
    Delta_OR    = gt::md("&Delta; OR"),
    Sig_Change  = "Significance Change"
  ) %>%
  gt::tab_style(
    style = gt::cell_fill(color = "lightyellow"),
    locations = gt::cells_body(rows = Sig_Change != "No change")
  ) %>%
  gt::cols_align(align = "center") %>%
  gt::cols_align(align = "left", columns = Predictor) %>%
  gt::tab_options(table.width = gt::pct(100), table.font.size = gt::px(12))

gt_regression

Regression Model Comparison: OLD vs NEW
Logistic regression predicting guideline adherence
Predictor	OR (95% CI) — OLD	P-value — OLD	OR (95% CI) — NEW	P-value — NEW	Δ OR	Significance Change
Admission at MGH (vs BWH)	1.92 (1.45 - 2.55)	<0.001	1.93 (1.46 - 2.56)	<0.001	0.010	No change
Admission to Neurology (vs Medicine/Other)	2.07 (1.58 - 2.73)	<0.001	2.02 (1.54 - 2.66)	<0.001	-0.057	No change
Immunocompetent (vs Immunosuppressed)	3.21 (1.81 - 6.17)	<0.001	3.98 (2.26 - 7.60)	<0.001	0.765	No change
Presence of CSF lymphocytic pleocytosis	2.66 (1.98 - 3.57)	<0.001	2.65 (1.97 - 3.57)	<0.001	-0.005	No change
CSF protein elevated >60 mg/dL	1.44 (1.08 - 1.91)	0.012	1.42 (1.06 - 1.89)	0.017	-0.022	No change

Executive Summary

# ── Auto-generated summary of key changes ─────────────────────────────────────
n_old <- nrow(data_old)
n_new <- nrow(data_new)

adherence_old <- mean(data_old$guideline2 == "Adherent", na.rm = TRUE) * 100
adherence_new <- mean(data_new$guideline2 == "Adherent", na.rm = TRUE) * 100

wnv_pos_old <- sum(data_old$wnvpos == "Positive WNV", na.rm = TRUE)
wnv_pos_new <- sum(data_new$wnvpos == "Positive WNV", na.rm = TRUE)

# Count significance changes in regression
sig_changes <- sum(or_compare$Sig_Change != "No change")

summary_df <- tibble(
  Metric = c(
    "Sample Size (N)",
    "Overall Adherence Rate (%)",
    "WNV Positive Cases (n)",
    "Regression: Predictors with Significance Change"
  ),
  OLD = c(
    as.character(n_old),
    sprintf("%.1f%%", adherence_old),
    as.character(wnv_pos_old),
    ""
  ),
  NEW = c(
    as.character(n_new),
    sprintf("%.1f%%", adherence_new),
    as.character(wnv_pos_new),
    ""
  ),
  Delta = c(
    as.character(n_new - n_old),
    sprintf("%+.1f pp", adherence_new - adherence_old),
    as.character(wnv_pos_new - wnv_pos_old),
    as.character(sig_changes)
  )
)

gt_summary <- summary_df %>%
  gt::gt() %>%
  gt::tab_header(title = gt::md("**Executive Summary: Key Changes Between Datasets**")) %>%
  gt::cols_align(align = "center") %>%
  gt::cols_align(align = "left", columns = Metric) %>%
  gt::tab_style(
    style = gt::cell_text(weight = "bold"),
    locations = gt::cells_column_labels()
  ) %>%
  gt::tab_options(table.width = gt::pct(80), table.font.size = gt::px(13))

gt_summary

Executive Summary: Key Changes Between Datasets
Metric	OLD	NEW	Delta
Sample Size (N)	1304	1304	0
Overall Adherence Rate (%)	25.5%	25.3%	-0.2 pp
WNV Positive Cases (n)	26	26	0
Regression: Predictors with Significance Change			0

Word Export

# ── Flextable helper ──────────────────────────────────────────────────────────
make_ft <- function(df, caption_text) {
  flextable::flextable(df) %>%
    flextable::bold(part = "header") %>%
    flextable::align(align = "center", part = "all") %>%
    flextable::align(align = "left", j = 1, part = "all") %>%
    flextable::border_remove() %>%
    flextable::border_outer(border = officer::fp_border(width = 1)) %>%
    flextable::border_inner_h(border = officer::fp_border(width = 1)) %>%
    flextable::border_inner_v(border = officer::fp_border(width = 1)) %>%
    flextable::autofit() %>%
    flextable::padding(padding = 2, part = "all") %>%
    flextable::set_caption(caption = caption_text)
}

# ── Build Word document ───────────────────────────────────────────────────────
doc <- officer::read_docx()

# Executive Summary
doc <- officer::body_add_par(doc, "Executive Summary: Key Changes Between Datasets", style = "heading 1")
doc <- flextable::body_add_flextable(doc, value = make_ft(summary_df, "Executive Summary"))
doc <- officer::body_add_break(doc)

# Helper: strip internal boolean columns before making flextable
strip_delta_cols <- function(df) {
  df %>% dplyr::select(-dplyr::any_of(c("overall_changed", "p_changed")))
}

# Table 1 Delta
doc <- officer::body_add_par(doc, "Table 1 Delta: Demographics by Guideline Adherence", style = "heading 1")
doc <- flextable::body_add_flextable(doc, value = make_ft(strip_delta_cols(delta_t1), "Table 1 Delta"))
doc <- officer::body_add_break(doc)

# Table New Delta
doc <- officer::body_add_par(doc, "Table New Delta: Demographics by Immune Status", style = "heading 1")
doc <- flextable::body_add_flextable(doc, value = make_ft(strip_delta_cols(delta_tn), "Table New Delta"))
doc <- officer::body_add_break(doc)

# Table 2 Delta
doc <- officer::body_add_par(doc, "Table 2 Delta: Demographics by WNV Positivity", style = "heading 1")
doc <- flextable::body_add_flextable(doc, value = make_ft(strip_delta_cols(delta_t2), "Table 2 Delta"))
doc <- officer::body_add_break(doc)

# Table 3 Delta
doc <- officer::body_add_par(doc, "Table 3 Delta: WNV Testing by Guideline Adherence", style = "heading 1")
doc <- flextable::body_add_flextable(doc, value = make_ft(strip_delta_cols(delta_t3), "Table 3 Delta"))
doc <- officer::body_add_break(doc)

# Table 4 Delta
doc <- officer::body_add_par(doc, "Table 4 Delta: WNV Testing by Admitting Service", style = "heading 1")
doc <- flextable::body_add_flextable(doc, value = make_ft(strip_delta_cols(delta_t4), "Table 4 Delta"))
doc <- officer::body_add_break(doc)

# Regression Comparison
doc <- officer::body_add_par(doc, "Regression Model Comparison", style = "heading 1")
doc <- flextable::body_add_flextable(doc, value = make_ft(or_compare, "Regression Comparison: OLD vs NEW"))
doc <- officer::body_add_break(doc)

# Figure 1
doc <- officer::body_add_par(doc, "Figure 1: Yearly Adherence vs WNV Cases", style = "heading 1")
fig1_path <- fs::path(params$export_dir, "Figure1_Comparison_Adherence_vs_Cases.png")
if (file.exists(fig1_path)) {
  doc <- officer::body_add_img(doc, src = fig1_path, width = 7, height = 3, style = "centered")
}
doc <- officer::body_add_break(doc)

# Figure 2
doc <- officer::body_add_par(doc, "Figure 2: Factors Associated with Guideline Adherence", style = "heading 1")
fig2_path <- fs::path(params$export_dir, "Figure2_Comparison_Forest_Plot.png")
if (file.exists(fig2_path)) {
  doc <- officer::body_add_img(doc, src = fig2_path, width = 7, height = 3, style = "centered")
}

# ── Save Word document ───────────────────────────────────────────────────────
report_path <- fs::path(params$export_dir, "Comparison_Report_Feb2026.docx")
print(doc, target = report_path)
message(glue("Word report saved to: {report_path}"))