COVID-19 Deaths Analysis by Sex and Age¶

This notebook analyzes provisional COVID-19 deaths data by sex and age group across different states in the United States. The analysis aims to explore patterns, insights, and potential anomalies related to COVID-19 mortality based on demographic factors.

In [1]:

# Import Libraries
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import numpy as np

%matplotlib inline

Load Data¶

In [2]:

# Load CSV Data
try:
    df = pd.read_csv('provisional_covid-19_deaths_by_sex_and_age.csv')
    print("Data loaded successfully!")
except FileNotFoundError:
    print("Error: CSV file not found. Please ensure 'provisional_covid-19_deaths_by_sex_and_age.csv' is in the correct directory.")
    # To make it runnable without the file, we'll create a dummy DataFrame structure (remove in real use)
    df = pd.DataFrame(columns=[
        'Data As Of', 'Start Date', 'End Date', 'Group', 'Year', 'Month', 'State', 'Sex', 'Age Group',
        'COVID-19 Deaths', 'Total Deaths', 'Pneumonia Deaths', 'Pneumonia and COVID-19 Deaths', 'Influenza Deaths',
        'Pneumonia, Influenza, or COVID-19 Deaths', 'Footnote'
    ])
except Exception as e:
    print(f"An error occurred while loading the data: {e}")
    df = None # Handle cases where loading fails

Data loaded successfully!

Data Cleaning and Preprocessing¶

In [3]:

# Data Cleaning
if df is not None:
    # Replace suppressed values with NaN
    cols_to_numeric = ['COVID-19 Deaths', 'Total Deaths', 'Pneumonia Deaths', 'Pneumonia and COVID-19 Deaths', 'Influenza Deaths', 'Pneumonia, Influenza, or COVID-19 Deaths']
    for col in cols_to_numeric:
        df[col] = pd.to_numeric(df[col], errors='coerce')

    # Impute NaN values in numeric columns with 0 (or consider other imputation methods if appropriate)
    df[cols_to_numeric] = df[cols_to_numeric].fillna(0)

    # Remove rows with Footnote (indicating data suppression issues in other columns if needed, or handle based on analysis goal)
    df = df[df['Footnote'].isnull()]
    df = df.drop(columns=['Footnote']) # Drop Footnote column after filtering

    print("Data cleaning and preprocessing complete.")
else:
    print("Data cleaning skipped due to loading errors.")

Data cleaning and preprocessing complete.

Data Overview¶

In [4]:

if df is not None:
    print("DataFrame Head:")
    print(df.head())
    print("\nDataFrame Information:")
    print(df.info())
    print("\nDataFrame Description:")
    print(df.describe())
else:
    print("Data overview skipped due to loading errors.")

DataFrame Head:
   Data As Of  Start Date    End Date     Group  Year  Month          State  \
0  09/27/2023  01/01/2020  09/23/2023  By Total   NaN    NaN  United States   
1  09/27/2023  01/01/2020  09/23/2023  By Total   NaN    NaN  United States   
2  09/27/2023  01/01/2020  09/23/2023  By Total   NaN    NaN  United States   
3  09/27/2023  01/01/2020  09/23/2023  By Total   NaN    NaN  United States   
4  09/27/2023  01/01/2020  09/23/2023  By Total   NaN    NaN  United States   

         Sex     Age Group  COVID-19 Deaths  Total Deaths  Pneumonia Deaths  \
0  All Sexes      All Ages        1146774.0    12303399.0         1162844.0   
1  All Sexes  Under 1 year            519.0       73213.0            1056.0   
2  All Sexes    0-17 years           1696.0      130970.0            2961.0   
3  All Sexes     1-4 years            285.0       14299.0             692.0   
4  All Sexes    5-14 years            509.0       22008.0             818.0   

   Pneumonia and COVID-19 Deaths  Influenza Deaths  \
0                       569264.0           22229.0   
1                           95.0              64.0   
2                          424.0             509.0   
3                           66.0             177.0   
4                          143.0             219.0   

   Pneumonia, Influenza, or COVID-19 Deaths  
0                                 1760095.0  
1                                    1541.0  
2                                    4716.0  
3                                    1079.0  
4                                    1390.0  

DataFrame Information:
<class 'pandas.core.frame.DataFrame'>
Index: 39804 entries, 0 to 137696
Data columns (total 15 columns):
 #   Column                                    Non-Null Count  Dtype  
---  ------                                    --------------  -----  
 0   Data As Of                                39804 non-null  object 
 1   Start Date                                39804 non-null  object 
 2   End Date                                  39804 non-null  object 
 3   Group                                     39804 non-null  object 
 4   Year                                      38499 non-null  float64
 5   Month                                     34947 non-null  float64
 6   State                                     39804 non-null  object 
 7   Sex                                       39804 non-null  object 
 8   Age Group                                 39804 non-null  object 
 9   COVID-19 Deaths                           39804 non-null  float64
 10  Total Deaths                              39804 non-null  float64
 11  Pneumonia Deaths                          39804 non-null  float64
 12  Pneumonia and COVID-19 Deaths             39804 non-null  float64
 13  Influenza Deaths                          39804 non-null  float64
 14  Pneumonia, Influenza, or COVID-19 Deaths  39804 non-null  float64
dtypes: float64(8), object(7)
memory usage: 4.9+ MB
None

DataFrame Description:
               Year         Month  COVID-19 Deaths  Total Deaths  \
count  38499.000000  34947.000000     3.980400e+04  3.980400e+04   
mean    2021.225694      6.484419     6.735513e+02  7.215437e+03   
std        1.029564      3.412469     9.396709e+03  9.666126e+04   
min     2020.000000      1.000000     0.000000e+00  0.000000e+00   
25%     2020.000000      4.000000     0.000000e+00  2.200000e+01   
50%     2021.000000      7.000000     2.800000e+01  3.090000e+02   
75%     2022.000000      9.000000     1.090000e+02  1.357000e+03   
max     2023.000000     12.000000     1.146774e+06  1.230340e+07   

       Pneumonia Deaths  Pneumonia and COVID-19 Deaths  Influenza Deaths  \
count      3.980400e+04                   39804.000000      39804.000000   
mean       6.814865e+02                     336.652924         13.745905   
std        9.340308e+03                    4733.412587        184.363712   
min        0.000000e+00                       0.000000          0.000000   
25%        0.000000e+00                       0.000000          0.000000   
50%        3.100000e+01                      13.000000          0.000000   
75%        1.190000e+02                      52.000000          0.000000   
max        1.162844e+06                  569264.000000      22229.000000   

       Pneumonia, Influenza, or COVID-19 Deaths  
count                              3.980400e+04  
mean                               1.030559e+03  
std                                1.416063e+04  
min                                0.000000e+00  
25%                                0.000000e+00  
50%                                4.700000e+01  
75%                                1.800000e+02  
max                                1.760095e+06

COVID-19 Deaths by Age Group (United States - All Sexes)¶

In [5]:

if df is not None:
    us_all_sexes_age = df[(df['State'] == 'United States') & (df['Sex'] == 'All Sexes') & (df['Age Group'] != 'All Ages') & (df['Age Group'] != '0-17 years') & (df['Age Group'] != '18-29 years')] #Exclude aggregated age groups for clarity
    plt.figure(figsize=(12, 6))
    sns.barplot(x='Age Group', y='COVID-19 Deaths', data=us_all_sexes_age, order= ['Under 1 year', '1-4 years', '5-14 years', '15-24 years', '25-34 years', '30-39 years', '35-44 years', '40-49 years', '45-54 years', '55-64 years', '65-74 years', '75-84 years', '85 years and over'])
    plt.title('COVID-19 Deaths by Age Group in the United States (All Sexes)')
    plt.xlabel('Age Group')
    plt.ylabel('COVID-19 Deaths')
    plt.xticks(rotation=45, ha='right')
    plt.tight_layout()
    plt.show()
else:
    print("Visualization skipped due to loading errors.")

No description has been provided for this image

Insight: This bar plot shows the distribution of COVID-19 deaths across different age groups in the United States, considering all sexes. We can observe the trend of deaths increasing with age, particularly prominent in older age groups.

COVID-19 Deaths by Sex (United States - All Ages)¶

In [6]:

if df is not None:
    us_sex_all_ages = df[(df['State'] == 'United States') & (df['Age Group'] == 'All Ages') & (df['Sex'] != 'All Sexes')]
    plt.figure(figsize=(8, 8))
    plt.pie(us_sex_all_ages['COVID-19 Deaths'], labels=us_sex_all_ages['Sex'], autopct='%1.1f%%', startangle=90, colors=sns.color_palette('pastel')[0:2])
    plt.title('COVID-19 Deaths Distribution by Sex in the United States (All Ages)')
    plt.ylabel('COVID-19 Deaths')
    plt.tight_layout()
    plt.show()
else:
    print("Visualization skipped due to loading errors.")

Insight: This pie chart illustrates the proportion of COVID-19 deaths by sex for all age groups in the United States. It provides a quick visual comparison of mortality rates between males and females.

Top States by COVID-19 Deaths (All Sexes, All Ages)¶

In [7]:

if df is not None:
    state_deaths_all_sexes_all_ages = df[(df['Group'] == 'By Total') & (df['Sex'] == 'All Sexes') & (df['Age Group'] == 'All Ages') & (df['State'] != 'United States')]
    top_10_states = state_deaths_all_sexes_all_ages.nlargest(10, 'COVID-19 Deaths')

    plt.figure(figsize=(12, 6))
    sns.barplot(x='State', y='COVID-19 Deaths', data=top_10_states, palette='viridis')
    plt.title('Top 10 States by COVID-19 Deaths (All Sexes, All Ages)')
    plt.xlabel('State')
    plt.ylabel('COVID-19 Deaths')
    plt.xticks(rotation=45, ha='right')
    plt.tight_layout()
    plt.show()
else:
    print("Visualization skipped due to loading errors.")

/tmp/ipykernel_159/2996980020.py:6: FutureWarning: 

Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the same effect.

  sns.barplot(x='State', y='COVID-19 Deaths', data=top_10_states, palette='viridis')

Insight: This bar plot displays the top 10 states with the highest number of COVID-19 deaths across all sexes and age groups. This visualization helps identify the states most severely impacted by the pandemic in terms of mortality count.

Correlation Analysis¶

In [8]:

if df is not None:
    correlation_matrix = df[cols_to_numeric].corr()
    plt.figure(figsize=(8, 6))
    sns.heatmap(correlation_matrix, annot=True, cmap='coolwarm', fmt=".2f")
    plt.title('Correlation Matrix of Death Causes')
    plt.tight_layout()
    plt.show()
else:
    print("Correlation analysis skipped due to loading errors.")

Insight: The heatmap presents the correlation matrix between different causes of death. It helps understand the relationships between COVID-19 deaths, pneumonia deaths, influenza deaths, and total deaths. High positive correlations are expected between 'Pneumonia Deaths', 'COVID-19 Deaths', and 'Pneumonia, Influenza, or COVID-19 Deaths' as they are related and sometimes overlapping categories.