The Geography of Care: USA’s Health Maps

Nine health indicators, one view—this faceted map visualizes standardized z-scores for each metric, highlighting geographic trends and disparities across U.S. states. Built using {ggplot2}, {sf}, and data wrangling with {dplyr}, it helps reveal patterns not visible in raw numbers alone.

About the Data

The dataset for this week’s #TidyTuesday explores state-level measurements of timely and effective care in hospitals across the United States, sourced from Medicare.gov. Collected and published by the Centers for Medicare and Medicaid Services (CMS), the data captures how hospitals perform across various metrics such as emergency room wait times and treatment timelines for different conditions. With 22 unique measure IDs covering six condition categories, each record includes the state, the measured outcome, and the relevant time window. This dataset provides an opportunity to analyze geographic disparities in care quality and timeliness, potentially uncovering how factors like state population, staffing levels, or hospital capacity affect patient experiences. It was curated by Jon Harmon from the Data Science Learning Community, with inspiration from a Visual Capitalist map by Kayla Zhu and Christina Kostandi, highlighting how emergency room wait times vary dramatically across states.

Figure 1: This graphic displays nine choropleth maps comparing U.S. states across major healthcare indicators such as ER visits, immunization rates, preventable hospitalizations, and mental health service use. Each map shows standardized z-scores, enabling meaningful comparisons despite differing units and scales. States with higher-than-average values are shaded in darker hues, revealing regional patterns and disparities in healthcare access and utilization.

How I made this graphic?

To create this faceted map visualization of U.S. state-level healthcare indicators, I primarily used the ggplot2 framework from the tidyverse suite for plotting, along with several powerful supporting packages. The spatial geometries were handled using sf, while data wrangling was done with dplyr, tidyr, and forcats. Each of the nine maps represents a different performance indicator (selected using their measure_id) and is scaled using standardized z-scores for comparability. The base maps came from usmapdata, and I used geom_sf() to draw state boundaries and geom_sf_text() to overlay actual indicator values. Color gradients were applied with paletteer and refined using colorspace, while plot text was enhanced via showtext and ggtext. The maps were arranged using facet_wrap(), revealing regional patterns—such as better emergency care metrics in the Midwest but lower vaccination scores. Final composition tweaks used patchwork, and social media icons were integrated with fontawesome for a polished and branded appearance. Overall, this approach combined geospatial visualization, normalization, and typography to craft an insightful and visually cohesive narrative.

Loading required libraries

# Data Import and Wrangling Tools
library(tidyverse)            # All things tidy

# Final plot tools
library(scales)               # Nice Scales for ggplot2
library(fontawesome)          # Icons display in ggplot2
library(ggtext)               # Markdown text support for ggplot2
library(showtext)             # Display fonts in ggplot2
library(colorspace)           # Lighten and Darken colours

library(magick)               # Download images and edit them
library(ggimage)              # Display images in ggplot2
library(patchwork)            # Composing Plots

library(sf)                   # For maps and area computation

care_state <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/main/data/2025/2025-04-08/care_state.csv')

Visualization Parameters

# Font for titles
font_add_google("Saira",
  family = "title_font"
) 

# Font for the caption
font_add_google("Saira Extra Condensed",
  family = "caption_font"
) 

# Font for plot text
font_add_google("Saira Condensed",
  family = "body_font"
) 

showtext_auto()

mypal <- c("yellow", "blue", "grey30")
# cols4all::c4a_gui()

# A base Colour
bg_col <- "white"
seecolor::print_color(bg_col)

# Colour for highlighted text
text_hil <- mypal[3]
seecolor::print_color(text_hil)

# Colour for the text
text_col <- mypal[3]
seecolor::print_color(text_col)

line_col <- "grey30"

# Define Base Text Size
bts <- 120

# Caption stuff for the plot
sysfonts::font_add(
  family = "Font Awesome 6 Brands",
  regular = here::here("docs", "Font Awesome 6 Brands-Regular-400.otf")
)
github <- "&#xf09b"
github_username <- "aditya-dahiya"
xtwitter <- "&#xe61b"
xtwitter_username <- "@adityadahiyaias"
social_caption_1 <- glue::glue("<span style='font-family:\"Font Awesome 6 Brands\";'>{github};</span> <span style='color: {text_hil}'>{github_username}  </span>")
social_caption_2 <- glue::glue("<span style='font-family:\"Font Awesome 6 Brands\";'>{xtwitter};</span> <span style='color: {text_hil}'>{xtwitter_username}</span>")
plot_caption <- paste0(
  "**Data:** medicare.gov, CMS & Jon Harmon", 
  " |  **Code:** ", 
  social_caption_1, 
  " |  **Graphics:** ", 
  social_caption_2
  )
rm(github, github_username, xtwitter, 
   xtwitter_username, social_caption_1, 
   social_caption_2)

# Add text to plot-------------------------------------------------
plot_title <- "The Geography of Care: USA’s Health Maps"
plot_subtitle <- "Nine health indicators, one view—this faceted map visualizes standardized z-scores for each metric, highlighting geographic trends and disparities across U.S. states. Built using {ggplot2}, {sf}, and data wrangling with {dplyr}, it helps reveal patterns not visible in raw numbers alone."

Exploratory Data Analysis and Wrangling

# Overall exploration of the data
# library(summarytools)
# care_state |> 
#   dfSummary() |> 
#   view()

# Lets look at the variables - names - meaning
# care_state |> 
#   distinct(measure_id, measure_name) |> 
#   print(n = Inf)

# Lets look at correlation between the various variables
# care_state |>
#   select(state, measure_id, score) |>
#   pivot_wider(
#     id_cols = state,
#     names_from = measure_id,
#     values_from = score
#   ) |>
#   select(-state) |>
#   GGally::ggpairs()
# Pair wise scatterplot is too heav to make sense.

# Using ggcorrplot instead
care_state |>
  select(state, measure_id, score) |>
  pivot_wider(
    id_cols = state,
    names_from = measure_id,
    values_from = score
  ) |>
  select(-state) |>
  cor(use = "pairwise.complete.obs") |> 
  ggcorrplot::ggcorrplot() +
  scale_fill_gradient2(
    low = "red", high = "blue", mid = "white"
  )

# Select out the relevant variables that make more sense
care_state |> 
  filter(
    measure_id %in% selected_vars
  ) |>
  distinct(measure_id, measure_name)

selected_vars <- c(
  "OP_18b", "OP_18c", "OP_22", 
  "OP_29", "SAFE_USE_OF_OPIOIDS", "SEP_1",
  "OP_23", "HCP_COVID_19", "IMM_3"
  )

# A correlation plot between the variables
care_state |>
  filter(
    measure_id %in% selected_vars
  ) |>
  select(state, measure_id, score) |>
  pivot_wider(
    id_cols = state,
    names_from = measure_id,
    values_from = score
  ) |>
  select(-state) |>
  cor(use = "pairwise.complete.obs") |>
  ggcorrplot::ggcorrplot() +
  scale_fill_gradient2(
    low = "red", high = "blue", mid = "white"
  )

# A facet labeller vector
df_temp <- care_state |> 
  filter(measure_id %in% selected_vars) |> 
  distinct(measure_id, measure_name)
facet_names <- df_temp$measure_name
# Improve names (remove unwanted words)
facet_names <- str_wrap(str_replace(df_temp$measure_name, "\\b(Higher|A lower|Lower)\\b.*", ""), 50)
names(facet_names) <- df_temp$measure_id
# Reverse the variable for which lower value is better
rev_vars = c("OP_18b", "OP_18c", "OP_22")
rm(df_temp)

# Create a final usable tibble for faceted map of USA
df1 <- care_state |> 
  filter(
    measure_id %in% selected_vars
  ) |> 
  select(
    state, 
    measure_id, 
    # measure_name,
    score
  ) |> 
  group_by(measure_id) |> 
  mutate(
    score = if_else(
      measure_id %in% rev_vars,
      -score,
      score
    ),
    # Using standardized z-score for Fill Scale
    score_scaled = (score - mean(score, na.rm = TRUE)) / 
                      sd(score, na.rm = TRUE),
    
    
    score_display = if_else(
      score == min(score, na.rm = T) | score == max(score, na.rm = T),
      score,
      NA
    )
  ) |> 
  ungroup() |> 
  mutate(measure_id  = fct(measure_id, levels = selected_vars)) |> 
  rename(abbr = state)

# Check if values are propery distributed.
# df1 |> 
#   ggplot(aes(x = score_scaled)) +
#   geom_boxplot() +
#   facet_wrap(~measure_id, ncol = 1)


df_plot <- usmapdata::us_map() |> 
  select(abbr, full, geom) |> 
  left_join(df1) |> 
  mutate(area_var = as.numeric(st_area(geom)))

The Plot

g <- ggplot(
  data = df_plot,
  mapping = aes(
    fill = score_scaled
  )
) +
  geom_sf(
    colour = bg_col,
    linewidth = 0.2
  ) +
  geom_sf_text(
    mapping = aes(
      label = round(abs(score), 1),
      size = area_var
    ),
    family = "caption_font",
    colour = text_col
  ) +
  scale_size(range = c(bts/40, bts/8)) +
  facet_wrap(
    ~ measure_id,
    ncol = 3,
    labeller = labeller(
      measure_id = facet_names
    )
  ) +
  paletteer::scale_fill_paletteer_c(
    "grDevices::Tropic",
    direction = -1,
    limits = c(-1, 1),
    oob = scales::squish,
    breaks = c(-0.6, 0.6),
    labels = c("Worse than National Average",
               "Better than National Average")
  ) +
  guides(size = "none") +
  coord_sf(clip = "off", expand = FALSE) +
  labs(
    title = plot_title,
    subtitle = str_wrap(plot_subtitle, 100),
    caption = plot_caption,
    fill = NULL
  ) +
  ggthemes::theme_map(
    base_family = "body_font",
    base_size = bts
  ) +
  theme(
    
    # Overall
    plot.margin = margin(5,-5,5,-15, "mm"),
    plot.title.position = "plot",
    text = element_text(
      colour = text_col,
      lineheight = 0.3,
      hjust = 0.5
    ),
    
    # Labels and Strip Text
    plot.title = element_text(
      colour = text_hil,
      margin = margin(15,0,5,0, "mm"),
      size = bts * 1.8,
      lineheight = 0.3,
      hjust = 0.5
    ),
    plot.subtitle = element_text(
      colour = text_hil,
      margin = margin(5,0,15,0, "mm"),
      size = 0.8 * bts,
      lineheight = 0.3,
      hjust = 0.5
    ),
    plot.caption = element_textbox(
      margin = margin(5,0,0,0, "mm"),
      hjust = 0.5,
      colour = text_hil,
      size = 0.5 * bts
    ),
    plot.caption.position = "plot",
    
    # Legend
    legend.position = "bottom",
    legend.margin = margin(-20,0,0,0, "mm"),
    legend.box.margin = margin(-20,0,0,0, "mm"),
    legend.title = element_text(
      margin = margin(0,0,0,0, "mm")
    ),
    legend.text = element_text(
      margin = margin(5,0,0,0, "mm"),
      hjust = 0.5
    ),
    legend.key.height = unit(10, "mm"),
    legend.key.width = unit(60, "mm"),
    legend.justification = c(0.5, 1),
    
    # Strip Text
    strip.text = element_text(
      margin = margin(0,0,0,0, "mm"),
      lineheight = 0.3,
      hjust = 0.5,
      vjust = 0.5,
      family = "caption_font",
      size = bts / 2
    ),
    strip.background = element_rect(
      fill = NA, colour = NA
    ),
    panel.spacing.y = unit(5, "mm"),
    panel.spacing.x = unit(-15, "mm"),
    
    # Axes
    axis.ticks = element_blank(),
    axis.ticks.length = unit(0, "mm"),
    axis.title = element_blank(),
    axis.text = element_blank()
  )


ggsave(
  filename = here::here(
    "data_vizs",
    "tidy_us_states_care.png"
  ),
  plot = g,
  width = 400,
  height = 500,
  units = "mm",
  bg = bg_col
)

Savings the thumbnail for the webpage

# Saving a thumbnail

library(magick)

# Saving a thumbnail for the webpage
image_read(here::here("data_vizs", 
                      "tidy_us_states_care.png")) |> 
  image_resize(geometry = "x400") |> 
  image_write(
    here::here(
      "data_vizs", 
      "thumbnails", 
      "tidy_us_states_care.png"
    )
  )

Session Info

# Data Import and Wrangling Tools
library(tidyverse)            # All things tidy

# Final plot tools
library(scales)               # Nice Scales for ggplot2
library(fontawesome)          # Icons display in ggplot2
library(ggtext)               # Markdown text support for ggplot2
library(showtext)             # Display fonts in ggplot2
library(colorspace)           # Lighten and Darken colours

library(magick)               # Download images and edit them
library(ggimage)              # Display images in ggplot2
library(patchwork)            # composing Plots


sessioninfo::session_info()$packages |> 
  as_tibble() |> 
  select(package, 
         version = loadedversion, 
         date, source) |> 
  arrange(package) |> 
  janitor::clean_names(
    case = "title"
  ) |> 
  gt::gt() |> 
  gt::opt_interactive(
    use_search = TRUE
  ) |> 
  gtExtras::gt_theme_espn()

Table 1: R Packages and their versions used in the creation of this page and graphics