CDC Data Exercise

Dataset Overview

The dataset “Potentially Excess Deaths from the Five Leading Causes of Death” offers a detailed look at how many deaths in different areas of the U.S. Essentially, it highlights how many deaths occurred and what expected. In addition we are also provided with this following information:

Underlying cause of death is based on the International Classification of Diseases, and are as follows:

• Heart disease (I00-I09, I11, I13, and I20–I51)

• Cancer (C00–C97)

• Unintentional injury (V01–X59 and Y85–Y86)

• Chronic lower respiratory disease (J40–J47)

• Stroke (I60–I69)

• Locality (nonmetropolitan vs. metropolitan).

• CDC’s Potentially Excess Deaths webpage

• The dataset has 206k rows.

• 13 columns:

  • Year: The year of the data record - 2005 through 2015.
  • Cause of Death: Primary cause of death based on ICD-10 codes.
  • State: U.S. state where deaths occurred.
  • Locality: Classification as nonmetropolitan or metropolitan.
  • Observed Deaths: Actual number of deaths.
  • Expected Deaths: Number of deaths expected based on benchmark rates.
  • Potentially Excess Deaths: Deaths exceeding expected numbers.
  • Percent Potentially Excess Deaths: Percentage of deaths exceeding expectations.

Loading and Cleaning the Data

We first begin by loading and cleaning the dataset.

# Load necessary libraries
library(tidyverse)

Warning: package 'ggplot2' was built under R version 4.3.3

Warning: package 'tidyr' was built under R version 4.3.3

Warning: package 'dplyr' was built under R version 4.3.3

── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
✔ dplyr     1.1.4     ✔ readr     2.1.4
✔ forcats   1.0.0     ✔ stringr   1.5.0
✔ ggplot2   3.5.1     ✔ tibble    3.2.1
✔ lubridate 1.9.2     ✔ tidyr     1.3.1
✔ purrr     1.0.2     
── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag()    masks stats::lag()
ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors

library(readr)
library(RColorBrewer)
library(dplyr)
library(ggplot2)
library(tibble)

# Set the working directory to the project directory
data <- here::here("cdcdata-exercise","NCHS_-_Five_Leading_Causes_of_Death.csv")
# Define the file path and load the data
data <- read_csv("NCHS_-_Five_Leading_Causes_of_Death.csv")

Rows: 205920 Columns: 13
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (6): Cause of Death, State, State FIPS Code, Age Range, Benchmark, Locality
dbl (7): Year, HHS Region, Observed Deaths, Population, Expected Deaths, Pot...

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

# Tell us the name of the columns 
colnames(data)

 [1] "Year"                              "Cause of Death"                   
 [3] "State"                             "State FIPS Code"                  
 [5] "HHS Region"                        "Age Range"                        
 [7] "Benchmark"                         "Locality"                         
 [9] "Observed Deaths"                   "Population"                       
[11] "Expected Deaths"                   "Potentially Excess Deaths"        
[13] "Percent Potentially Excess Deaths"

#output two newline characters
cat("\n\n")

# Display the structure of the dataset
str(data)

spc_tbl_ [205,920 × 13] (S3: spec_tbl_df/tbl_df/tbl/data.frame)
 $ Year                             : num [1:205920] 2005 2005 2005 2005 2005 ...
 $ Cause of Death                   : chr [1:205920] "Cancer" "Cancer" "Cancer" "Cancer" ...
 $ State                            : chr [1:205920] "Alabama" "Alabama" "Alabama" "Alabama" ...
 $ State FIPS Code                  : chr [1:205920] "AL" "AL" "AL" "AL" ...
 $ HHS Region                       : num [1:205920] 4 4 4 4 4 4 4 4 4 4 ...
 $ Age Range                        : chr [1:205920] "0-49" "0-49" "0-49" "0-49" ...
 $ Benchmark                        : chr [1:205920] "2005 Fixed" "2005 Fixed" "2005 Fixed" "2010 Fixed" ...
 $ Locality                         : chr [1:205920] "All" "Metropolitan" "Nonmetropolitan" "All" ...
 $ Observed Deaths                  : num [1:205920] 756 556 200 756 556 ...
 $ Population                       : num [1:205920] 3148377 2379871 768506 3148377 2379871 ...
 $ Expected Deaths                  : num [1:205920] 451 341 111 421 318 103 451 341 111 784 ...
 $ Potentially Excess Deaths        : num [1:205920] 305 217 89 335 238 97 305 217 89 562 ...
 $ Percent Potentially Excess Deaths: num [1:205920] 40.3 39 44.5 44.3 42.8 48.5 40.3 39 44.5 41.8 ...
 - attr(*, "spec")=
  .. cols(
  ..   Year = col_double(),
  ..   `Cause of Death` = col_character(),
  ..   State = col_character(),
  ..   `State FIPS Code` = col_character(),
  ..   `HHS Region` = col_double(),
  ..   `Age Range` = col_character(),
  ..   Benchmark = col_character(),
  ..   Locality = col_character(),
  ..   `Observed Deaths` = col_double(),
  ..   Population = col_double(),
  ..   `Expected Deaths` = col_double(),
  ..   `Potentially Excess Deaths` = col_double(),
  ..   `Percent Potentially Excess Deaths` = col_double()
  .. )
 - attr(*, "problems")=<externalptr> 

cat("\n\n")

# Check for missing values
colSums(is.na(data))

                             Year                    Cause of Death 
                                0                                 0 
                            State                   State FIPS Code 
                                0                                 0 
                       HHS Region                         Age Range 
                                0                                 0 
                        Benchmark                          Locality 
                                0                                 0 
                  Observed Deaths                        Population 
                            10212                              5280 
                  Expected Deaths         Potentially Excess Deaths 
                            10212                             10212 
Percent Potentially Excess Deaths 
                            10212 

cat("\n\n")

# Remove rows with NA values
data_cleaned <- data %>%
  drop_na()

# Display the structure of the cleaned dataset
str(data_cleaned)

tibble [195,708 × 13] (S3: tbl_df/tbl/data.frame)
 $ Year                             : num [1:195708] 2005 2005 2005 2005 2005 ...
 $ Cause of Death                   : chr [1:195708] "Cancer" "Cancer" "Cancer" "Cancer" ...
 $ State                            : chr [1:195708] "Alabama" "Alabama" "Alabama" "Alabama" ...
 $ State FIPS Code                  : chr [1:195708] "AL" "AL" "AL" "AL" ...
 $ HHS Region                       : num [1:195708] 4 4 4 4 4 4 4 4 4 4 ...
 $ Age Range                        : chr [1:195708] "0-49" "0-49" "0-49" "0-49" ...
 $ Benchmark                        : chr [1:195708] "2005 Fixed" "2005 Fixed" "2005 Fixed" "2010 Fixed" ...
 $ Locality                         : chr [1:195708] "All" "Metropolitan" "Nonmetropolitan" "All" ...
 $ Observed Deaths                  : num [1:195708] 756 556 200 756 556 ...
 $ Population                       : num [1:195708] 3148377 2379871 768506 3148377 2379871 ...
 $ Expected Deaths                  : num [1:195708] 451 341 111 421 318 103 451 341 111 784 ...
 $ Potentially Excess Deaths        : num [1:195708] 305 217 89 335 238 97 305 217 89 562 ...
 $ Percent Potentially Excess Deaths: num [1:195708] 40.3 39 44.5 44.3 42.8 48.5 40.3 39 44.5 41.8 ...

Visualizations and Exploratory Data Analysis

1) Summary statistics for variables.

# Summary statistics for data 
summary(data)

      Year      Cause of Death        State           State FIPS Code   
 Min.   :2005   Length:205920      Length:205920      Length:205920     
 1st Qu.:2007   Class :character   Class :character   Class :character  
 Median :2010   Mode  :character   Mode  :character   Mode  :character  
 Mean   :2010                                                           
 3rd Qu.:2013                                                           
 Max.   :2015                                                           
                                                                        
   HHS Region      Age Range          Benchmark           Locality        
 Min.   : 0.000   Length:205920      Length:205920      Length:205920     
 1st Qu.: 3.000   Class :character   Class :character   Class :character  
 Median : 5.000   Mode  :character   Mode  :character   Mode  :character  
 Mean   : 5.231                                                           
 3rd Qu.: 8.000                                                           
 Max.   :10.000                                                           
                                                                          
 Observed Deaths    Population        Expected Deaths 
 Min.   :    10   Min.   :    55536   Min.   :     2  
 1st Qu.:   155   1st Qu.:   682826   1st Qu.:    92  
 Median :   508   Median :  1610936   Median :   298  
 Mean   :  2975   Mean   :  7008421   Mean   :  2113  
 3rd Qu.:  1586   3rd Qu.:  4681585   3rd Qu.:  1019  
 Max.   :493526   Max.   :315131659   Max.   :465126  
 NA's   :10212    NA's   :5280        NA's   :10212   
 Potentially Excess Deaths Percent Potentially Excess Deaths
 Min.   :     0.0          Min.   : 0.00                    
 1st Qu.:    41.0          1st Qu.:20.90                    
 Median :   159.0          Median :35.80                    
 Mean   :   875.1          Mean   :35.73                    
 3rd Qu.:   543.0          3rd Qu.:50.10                    
 Max.   :175703.0          Max.   :85.30                    
 NA's   :10212             NA's   :10212                    

cat("\n\n")

# Summary statistics for cleaned data
summary(data_cleaned)

      Year      Cause of Death        State           State FIPS Code   
 Min.   :2005   Length:195708      Length:195708      Length:195708     
 1st Qu.:2007   Class :character   Class :character   Class :character  
 Median :2010   Mode  :character   Mode  :character   Mode  :character  
 Mean   :2010                                                           
 3rd Qu.:2013                                                           
 Max.   :2015                                                           
   HHS Region    Age Range          Benchmark           Locality        
 Min.   : 0.0   Length:195708      Length:195708      Length:195708     
 1st Qu.: 3.0   Class :character   Class :character   Class :character  
 Median : 5.0   Mode  :character   Mode  :character   Mode  :character  
 Mean   : 5.3                                                           
 3rd Qu.: 8.0                                                           
 Max.   :10.0                                                           
 Observed Deaths    Population        Expected Deaths 
 Min.   :    10   Min.   :    55536   Min.   :     2  
 1st Qu.:   155   1st Qu.:   724049   1st Qu.:    92  
 Median :   508   Median :  1704462   Median :   298  
 Mean   :  2975   Mean   :  7177289   Mean   :  2113  
 3rd Qu.:  1586   3rd Qu.:  4783958   3rd Qu.:  1019  
 Max.   :493526   Max.   :315131659   Max.   :465126  
 Potentially Excess Deaths Percent Potentially Excess Deaths
 Min.   :     0.0          Min.   : 0.00                    
 1st Qu.:    41.0          1st Qu.:20.90                    
 Median :   159.0          Median :35.80                    
 Mean   :   875.1          Mean   :35.73                    
 3rd Qu.:   543.0          3rd Qu.:50.10                    
 Max.   :175703.0          Max.   :85.30                    

2) Visualization of trends over time for causes of death.

# Define a color palette
colors <- c("Heart Disease" = "red", 
            "Cancer" = "blue", 
            "Chronic Lower Respiratory Disease" = "darkgreen", 
            "Stroke" = "orange", 
            "Unintentional Injury" = "grey")

# Aggregate data by Year and Cause of Death, excluding "United States"
annual_totals <- data_cleaned %>%
  filter(State != "United States") %>%
  group_by(Year, `Cause of Death`) %>%
  summarize(Total_Observed_Deaths = sum(`Observed Deaths`, na.rm = TRUE), .groups = 'drop')

# Create the time series line plot with thicker lines and formatted year labels
ggplot(annual_totals, aes(x = Year, y = Total_Observed_Deaths, color = `Cause of Death`, group = `Cause of Death`)) +
  geom_line(linewidth = 1.2) +  # Use 'linewidth' for line thickness
  geom_point(size = 2) +  # Optionally, adjust point size
  scale_x_continuous(breaks = seq(min(annual_totals$Year), max(annual_totals$Year), by = 1),  # Ensure years are whole numbers
                     labels = scales::label_number(big.mark = "")) +  # Format year labels as whole numbers
  scale_color_manual(values = colors) +  # Apply custom colors
  labs(title = "Deaths Over Time by Cause of Death",
       x = "Year",
       y = "Total Observed Deaths") +  #  y-axis label
  theme_minimal() +
  theme(legend.position = "bottom",  # Adjust legend position if necessary
        legend.title = element_blank(),  # Remove legend title
        legend.key.width = unit(0.5, "cm"),  # Adjust legend key width
        legend.key.height = unit(0.5, "cm"),  # Adjust legend key height
        legend.text = element_text(size = 10))  # Adjust legend text size

Causes of Death:

• Cancer: This is the leading cause of death according to the graph, and the number of deaths has been slowly increasing over time. There appears to be a slight acceleration in the rate of increase in recent years.

• Heart Disease: Heart disease is the second leading cause of death. The number of deaths due to heart disease has fluctuated somewhat over the years, but there is a general downward trend from 2008 to 2013. Since then, there has been a slight uptick.

• Unintentional Injuries: Unintentional injuries are the third leading cause of death according to the graph. The number deaths due to unintentional injuries has been increasing steadily over time.

• Chronic Lower Respiratory Disease and Stroke: Chronic lower respiratory disease and stroke have the two lowest numbers of deaths depicted in the graph. The number of deaths due to these causes appears to be relatively stable.

3) Visualization: Causes of Death over time (Percentages).

# Aggregate total observed deaths by Year and Cause of Death, excluding "United States"
annual_totals <- data_cleaned %>%
  filter(State != "United States") %>%
  group_by(Year, `Cause of Death`) %>%
  summarize(Total_Observed_Deaths = sum(`Observed Deaths`, na.rm = TRUE), .groups = 'drop')

# Calculate total deaths by Year
total_deaths_by_year <- annual_totals %>%
  group_by(Year) %>%
  summarize(Total_Deaths = sum(Total_Observed_Deaths, na.rm = TRUE))

# Merge the total deaths with the annual totals to calculate percentages
annual_totals_with_percentages <- annual_totals %>%
  left_join(total_deaths_by_year, by = "Year") %>%
  mutate(Percentage_of_Deaths = (Total_Observed_Deaths / Total_Deaths) * 100)

# Reorder the Cause of Death factor levels based on total percentage
annual_totals_with_percentages <- annual_totals_with_percentages %>%
  mutate(`Cause of Death` = reorder(`Cause of Death`, Percentage_of_Deaths, FUN = sum))

# Create a bar plot with percentages and labels, with reordered bars
ggplot(annual_totals_with_percentages, aes(x = Year, y = Percentage_of_Deaths, fill = `Cause of Death`)) +
  geom_bar(stat = "identity", position = "stack") +  # Stacked bar plot
  geom_text(aes(label = sprintf("%.1f%%", Percentage_of_Deaths), y = Percentage_of_Deaths), 
            position = position_stack(vjust = 0.5), size = 3.5, color = "white") +  # Add percentage labels inside bars
  labs(title = "Percentage of Observed Deaths by Cause of Death Over Time",  # Update title
       x = "Year",
       y = "Percentage of Total Deaths") +
  scale_x_continuous(breaks = seq(min(annual_totals_with_percentages$Year), max(annual_totals_with_percentages$Year), by = 1), 
                     labels = scales::label_number(big.mark = "")) +  # Format year labels
  scale_fill_manual(values = colors) +  # Apply custom colors
  theme_minimal() +
  theme(legend.position = "bottom",  # Adjust legend position if necessary
        legend.title = element_blank(),  # Remove legend title
        legend.key.width = unit(0.5, "cm"),  # Adjust legend key width
        legend.key.height = unit(0.5, "cm"),  # Adjust legend key height
        legend.text = element_text(size = 10))  # Adjust legend text size

Percentage of Deaths Over Time:

• Cancer is the leading cause of death in the United States, representing around 40% of total deaths from the top 5 causes.

• Heart Disease is the second leading cause of death (after cancer). The graph shows a slight decrease in the percentage of deaths due to heart disease over time.

• Unintentional Injuries the graph shows a slight increase in the percentage of deaths due to unintentional injuries, meaning more deaths have occurred from unintentional causes over time.

• Chronic Lower Respiratory Disease the graph suggests a slight increase in deaths due to chronic lower respiratory disease.

• Stroke appears to have the lowest percentage of deaths among the top causes depicted in the graph. The percentage has remained relatively stable over time, with a possible slight decrease.

Overall, the graph shows the percentage of deaths by cause over time. The leading causes of death are cancer and heart disease, but the percentage of deaths due to these causes is slowly decreasing. Deaths due to unintentional injuries show a slight upward trend, while chronic lower respiratory disease and stroke show a stable or slightly increasing percentage of deaths.

4) Visualization: Causes of death by state.

# Exclude rows where State is "United States" and Total_Observed_Deaths is below 650,000
state_observed_deaths <- data_cleaned %>%
  filter(State != "United States") %>%
  group_by(State) %>%
  summarize(Total_Observed_Deaths = sum(`Observed Deaths`, na.rm = TRUE)) %>%
  filter(Total_Observed_Deaths >= 650000) %>%  # Filter out states with less than 650,000 observed deaths
  arrange(desc(Total_Observed_Deaths))

# Plotting the state-wise distribution of observed deaths excluding United States
ggplot(state_observed_deaths, aes(x = reorder(State, -Total_Observed_Deaths), y = Total_Observed_Deaths)) +
  geom_bar(stat = "identity", fill = "skyblue", width = 0.7) +  # Adjust bar width as needed
  labs(title = "Total Observed Deaths by State (Excluding United States)",
       x = "State",
       y = "Total Observed Deaths") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 90, hjust = 1, size = 7))  # Adjust angle and size

Here are the top ten states with the most deaths according to the graph:

1. California
2. Texas
3. Florida 4.New York
4. Pennsylvania
5. Ohio
6. Illinois
7. Michigan
8. North Carolina
9. Georgia

Intrigued by this, we will now dive deeper and explore the leading cause of death in these top 10 states. This will help provide and understanding mortality patterns across the country.

# Specify the states of interest
states_of_interest <- c("California", "Texas", "Florida", "New York", "Pennsylvania", 
                        "Ohio", "Illinois", "Michigan", "North Carolina", "Georgia")

# Filter data for specified states and calculate percentages by cause of death
state_cause_percentages <- data_cleaned %>%
  filter(State %in% states_of_interest) %>%
  group_by(State, `Cause of Death`) %>%
  summarize(Percentage = sum(`Observed Deaths`) / sum(data_cleaned$`Observed Deaths`) * 100, .groups = 'drop') %>%
  arrange(State, desc(Percentage))  # Arrange by State and descending Percentage

# Plotting separate bar plots for each state showing percentage distribution of causes of death
plots_list <- lapply(states_of_interest, function(state) {
  state_data <- state_cause_percentages %>%
    filter(State == state) %>%
    arrange(desc(Percentage))  # Sort within each state from highest to lowest percentage
  
  ggplot(state_data, aes(x = reorder(`Cause of Death`, -Percentage), y = Percentage, fill = `Cause of Death`)) +
    geom_bar(stat = "identity", width = 0.7) +  # Bar plot
    geom_text(aes(label = sprintf("%.1f%%", Percentage), y = Percentage), 
              position = position_stack(vjust = 0.5), size = 3.5, color = "white") +  # Add percentage labels inside bars
    labs(title = paste("Distribution of Causes of Death in", state),
         x = "Cause of Death",
         y = "Percentage") +
    scale_fill_manual(values = colors) +  # Apply custom colors
    theme_minimal() +
    theme(axis.text.x = element_text(angle = 45, hjust = 1),  # Rotate x-axis labels if needed
          plot.title = element_text(hjust = 0.5))  # Center title horizontally
})

# Print each plot separately
for (i in seq_along(plots_list)) {
  print(plots_list[[i]])
}

1. California

• Cancer: 2%
• Heart Disease: 1.4%
• Unintentional Injury: 0.7%
• Chronic Lower Respiratory Disease: 0.3%
• Stroke: 0.3%

2. Texas

• Cancer: 1.4%
• Heart Disease: 1.1%
• Unintentional Injury: 0.6%
• Chronic Lower Respiratory Disease: 0.2%
• Stroke:0.2%

3. Florida

• Cancer: 1.4%
• Heart Disease: 1.0%
• Unintentional Injury: 0.5%
• Chronic Lower Respiratory Disease:0.2%
• Stroke: 0.2%

4.New York:

• Cancer: 1.3%
• Heart Disease: 0.9%
• Unintentional Injury: 0.3%
• Chronic Lower Respiratory Disease: 0.2%
• Stroke: 0.1%

5. Pennsylvania

• Cancer: 1.0%
• Heart Disease: 0.7%
• Unintentional Injury: 0.4%
• Chronic Lower Respiratory Disease: 0.1%
• Stroke: 0.1%

6. Ohio

• Cancer: 0.9%
• Heart Disease: 0.7%
• Unintentional Injury: 0.3%
• Chronic Lower Respiratory Disease: 0.2%
• Stroke: 0.1%

7. Illinois

• Cancer: 0.9%
• Heart Disease: 0.6%
• Unintentional Injury: 0.3%
• Chronic Lower Respiratory Disease: 0.1%
• Stroke: 0.1%

8. Michigan

• Cancer: 0.7%
• Heart Disease: 0.6%
• Unintentional Injury: 0.2%
• Chronic Lower Respiratory Disease: 0.1%
• Stroke: 0.1%

9. North Carolina

• Cancer: 0.7%
• Heart Disease: 0.5%
• Unintentional Injury: 0.3%
• Chronic Lower Respiratory Disease: 0.1%
• Stroke: 0.1%

10. Georgia

• Cancer: 0.6%
• Heart Disease: 0.5%
• Unintentional Injury: 0.3%
• Chronic Lower Respiratory Disease: 0.1%
• Stroke: 0.1%

Conclusion:

The analysis of data from 2005 to 2015 reveals that Cancer is the leading cause of death, steadily rising over the years. Heart Disease follows, showing a slight decrease until 2013, and then a slight increase. Unintentional Injuries are continuously rising, presenting a growing public health concern. State-wise, California, Texas, and Florida report the highest number of deaths, with Cancer and Heart Disease consistently ranking at the top across all states analyzed.

Although deaths from Chronic Lower Respiratory Disease and Stroke remained stable from 2005 to 2015, any small increases in these areas should be watched closely to prevent them from becoming bigger health problems. The steady rise in deaths from Unintentional Injuries shows a need for targeted public health efforts and preventive measures to reduce these incidents. The state-by-state analysis reveals that Cancer and Heart Disease are the leading causes of death in all states, but the differences between states might be influenced by other factors. The visualizations and summary statistics highlight important trends and patterns and are valuable tools for policymakers.

Future research could look into underlying factors like demographics, policies, and access to affordable healthcare behind the trends seen in Cancer, Heart Disease, and Unintentional Injuries. A deeper understanding of these factors can lead to better health policies and interventions.

This section was contributed by Collin Real

Create synthetic dataset

library(tibble)
library(dplyr)

# Set seed for reproducibility
set.seed(123)

# Define the number of observations
n <- 195708

# Unique values lists (adjust as needed)
state_list <- c('Alabama', 'Alaska', 'Arizona', 'Arkansas', 'Delaware', 'California', 'Colorado', 'Connecticut', 
                 'New Jersey', 'District of Columbia', 'Florida', 'Georgia', 'Hawaii', 'Idaho', 'Illinois', 
                 'Indiana', 'Iowa', 'Rhode Island', 'Kansas', 'Kentucky', 'Louisiana', 'Maine', 'Maryland', 
                 'Massachusetts', 'Michigan', 'Minnesota', 'Mississippi', 'Missouri', 'Montana', 'Nebraska', 
                 'Nevada', 'New Hampshire', 'New Mexico', 'New York', 'North Carolina', 'North Dakota', 'Ohio', 
                 'Oklahoma', 'Oregon', 'Pennsylvania', 'South Carolina', 'South Dakota', 'Tennessee', 'Texas', 
                 'United States', 'Utah', 'Vermont', 'Virginia', 'Washington', 'West Virginia', 'Wisconsin', 'Wyoming')

state_fips_list <- 1:length(state_list)  # Example FIPS codes

hhs_list <- c('Region 1', 'Region 2', 'Region 3', 'Region 4', 'Region 5', 'Region 6', 'Region 7', 'Region 8', 
              'Region 9', 'Region 10')  # Adjust based on actual values

age_range_list <- c('0-4', '5-14', '15-24', '25-34', '35-44', '45-54', '55-64', '65-74', '75-84', '85+')  # Example ranges

benchmark_list <- c('Benchmark 1', 'Benchmark 2', 'Benchmark 3')  # Adjust as necessary
locality_list <- c('Urban', 'Rural')  # Adjust based on actual values

# Generate synthetic data
synth_df <- tibble(
  Year = sample(2005:2015, n, replace = TRUE),
  Cause.of.Death = sample(c('Cancer', 'Chronic Lower', 'Heart Disease', 'Stroke', 'Unintentional Injury'), n, replace = TRUE),
  State = sample(state_list, n, replace = TRUE),
  State.FIPS.Code = sample(state_fips_list, n, replace = TRUE),
  HHS.Region = sample(hhs_list, n, replace = TRUE),
  Age.Range = sample(age_range_list, n, replace = TRUE),
  Benchmark = sample(benchmark_list, n, replace = TRUE),
  Locality = sample(locality_list, n, replace = TRUE),
  Observed.Deaths = round(rnorm(n, mean = 2975, sd = 5000)),
  Population = round(rnorm(n, mean = 7177289, sd = 5000)),
  Expected.Deaths = round(rnorm(n, mean = 2113, sd = 5000)),
  Potentially.Excess.Deaths = round(rnorm(n, mean = 875.1, sd = 5000)),
  Percent.Potentially.Excess.Deaths = round(rnorm(n, mean = 35.73, sd = 5000))
)

# Display the synthetic data
print(synth_df)

# A tibble: 195,708 × 13
    Year Cause.of.Death     State State.FIPS.Code HHS.Region Age.Range Benchmark
   <int> <chr>              <chr>           <int> <chr>      <chr>     <chr>    
 1  2007 Heart Disease      Iowa               39 Region 9   25-34     Benchmar…
 2  2007 Unintentional Inj… Miss…               2 Region 7   15-24     Benchmar…
 3  2014 Heart Disease      Nort…              20 Region 4   65-74     Benchmar…
 4  2006 Stroke             Utah               26 Region 1   55-64     Benchmar…
 5  2010 Cancer             Okla…              42 Region 10  75-84     Benchmar…
 6  2015 Cancer             Virg…              23 Region 9   65-74     Benchmar…
 7  2009 Stroke             Colo…               7 Region 10  85+       Benchmar…
 8  2008 Chronic Lower      Arka…              33 Region 9   5-14      Benchmar…
 9  2010 Stroke             Wisc…              46 Region 7   55-64     Benchmar…
10  2013 Cancer             Virg…              27 Region 3   15-24     Benchmar…
# ℹ 195,698 more rows
# ℹ 6 more variables: Locality <chr>, Observed.Deaths <dbl>, Population <dbl>,
#   Expected.Deaths <dbl>, Potentially.Excess.Deaths <dbl>,
#   Percent.Potentially.Excess.Deaths <dbl>