Total Deaths/Mortality by Selected UCD Chapters [USA, Western Census Region]

chart_mortality_ucd_chapter_usa.r

library(data.table)
library(ggplot2)

sf <- 2
width <- 900 * sf
height <- 450 * sf
options(vsc.dev.args = list(width = width, height = height, res = 72 * sf))

# Load data
s <- fread(
  "https://s3.mortality.watch/data/population/usa/usa_states_iso3c.csv"
)
p <- fread("https://s3.mortality.watch/data/population/usa/20y.csv")[
  age_group == "all"
]
d1 <- fread(
  "https://apify.mortality.watch/cdc-wonder/year-icd_chapter.txt",
  select = c("State", "Year Code", "UCD - ICD Chapter Code", "Deaths")
)
setnames(d1, c("jurisdiction", "year", "ucd-chapter", "deaths"))
d2 <- fread(
  "https://apify.mortality.watch/cdc-wonder/provisional-year-icd_chapter.txt",
  select = c("Residence State", "Year Code", "UCD - ICD Chapter Code", "Deaths")
)
setnames(d2, c("jurisdiction", "year", "ucd-chapter", "deaths"))

df <- p[d2[s, on = .(jurisdiction)], on = .(iso3c, year), nomatch = NULL]
df <- df[year < 2025]

western_states <- c(
  # Mountain Division
  "Arizona", "Colorado", "Idaho", "Montana",
  "Nevada", "New Mexico", "Utah", "Wyoming",

  # Pacific Division
  "Alaska", "California", "Hawaii", "Oregon", "Washington"
)

# Handle suppressed values and convert to numeric
df[deaths == "Suppressed", deaths := "5"]
df[, deaths := as.numeric(deaths)]

# Filter for the specified ICD chapters and Western states
comparison_data <- df[jurisdiction %in% western_states &
  `ucd-chapter` %in% c("U00-U99", "C00-D48", "I00-I99", "J00-J98")]

# Calculate mortality rates and deaths - U00-U99 vs sum of the other three
mortality_comparison <- comparison_data[, .(
  total_deaths = sum(as.numeric(deaths), na.rm = TRUE),
  total_population = sum(population, na.rm = TRUE)
), by = .(year, `ucd-chapter`)][
  ,
  mortality_rate := (total_deaths / total_population) * 100000
]

# Create the comparison for both mortality rates and deaths
final_comparison <- rbind(
  # U00-U99 rates and deaths
  mortality_comparison[`ucd-chapter` == "U00-U99", .(
    year,
    category = "U00-U99",
    mortality_rate,
    deaths = total_deaths
  )],

  # Sum of the other three categories - CORRECTED
  mortality_comparison[`ucd-chapter` %in% c("C00-D48", "I00-I99", "J00-J98"), .(
    total_deaths = sum(total_deaths, na.rm = TRUE),
    total_population = sum(total_population, na.rm = TRUE)
  ), by = year][, .(
    year,
    category = "C00-D48 + I00-I99 + J00-J98",
    mortality_rate = (total_deaths / total_population) * 100000,
    deaths = total_deaths
  )]
)

# Create mortality rate chart
chart_rates <- ggplot(final_comparison, aes(x = year, y = mortality_rate)) +
  geom_line(linewidth = 1) +
  geom_point() +
  labs(
    title = "Crude Mortality Rate by Selected UCD Chapters [USA, Western Census Region]",
    subtitle = "Source: CDC Wonder, U00-U99 vs Combined C00-D48 + I00-I99 + J00-J98",
    x = "Year",
    y = "Deaths per 100,000 population"
  ) +
  facet_wrap(~category, scales = "free") +
  theme_minimal()

# Create deaths chart
chart_deaths <- ggplot(final_comparison, aes(x = year, y = deaths)) +
  geom_line(linewidth = 1) +
  geom_point() +
  labs(
    title = "Total Deaths by Selected UCD Chapters [USA, Western Census Region]",
    subtitle = "Source: CDC Wonder, U00-U99 vs Combined C00-D48 + I00-I99 + J00-J98",
    x = "Year",
    y = "Total Deaths"
  ) +
  facet_wrap(~category, scales = "free") +
  theme_minimal()

# Combine both charts into a 2x2 grid
library(gridExtra)
combined_chart <- grid.arrange(chart_deaths, chart_rates, ncol = 1)

ggsave(
  filename = "chart1.png",
  plot = combined_chart,
  width = width,
  height = height * 2, # Double height for 2 charts
  units = "px",
  dpi = 72 * sf,
  device = grDevices::png,
  type = "cairo"
)