Data Source

The dataset was provided by the Mexican government. This dataset contains an enormous number of anonymized patient-related information including pre-conditions. The raw dataset consists of 21 unique features and 1,048,576 unique patients.

The covid_sub.csv contains data from 20% of the population, and we will provide data Covid Data_full.csv for the full population to check the reproducibility of code.

Project Description

The Covid-19 report has four analysis parts.

1. code/01_demography.R from L9otus is Demographic Analysis

• Analyze the distribution of different demographic characteristics, such as age and gender.
  • Pie chart output/pie_chart_sex.png : Displaying the proportion of two genders.
  • Bar chart output/bar_chart_age.png and box plot output/box_plot_age.png : Displaying the distribution of different ages.
• Examine the distribution of pregnant and immunosuppressed patients, focusing on their proportions among hospitalized or not.
  • Classification table output/classification_table_immu.png : Displaying the proportion of immunosuppressed/non-immunosuppressed patients in each patient_type.
  • Classification table output/classification_table_pregnancy.png: Displaying the proportion of pregnant/non-pregnant in each patient_type.

2. code/02_health_outcome.R from kmg1024 is Case Severity and Health Outcomes Analysis

• Distribution of COVID-19 Cases by Classification Levels output/classification_bar_chart.png
  • Generates a bar chart showing the number of cases categorized as mild, moderate, and severe.
• Relationship Between Pre-existing Health Conditions and Case Severity output/diabetes_severity_scatter_plot.png
  • Creates a scatter plot illustrating the association between conditions like diabetes and hypertension with the severity of COVID-19 cases.
• Classification Distribution Table output/classification_table.csv
  • Summarizes the number and percentage of cases in each classification level.
• Patient Type Outcomes Table output/patient_outcomes_table.csv
  • Compares mortality rates, ICU admissions, and intubation rates between different patient types (e.g., hospitalized vs. returned home).

3. code/03_association.R from zye229zye is Association Analysis

• severity_results.csv
  • Contains logistic regression results for ICU admission factors (e.g., diabetes, renal chronic disease, hypertension).
• severity_odds_ratios.png
  • Visualizes the odds ratios from the logistic regression analysis of ICU admissions.
• mortality_results.csv
  • Summarizes logistic regression results for mortality factors (e.g., age, hypertension, diabetes, obesity).
• mortality_odds_ratios.png
  • Displays odds ratios for factors influencing mortality.
• tobacco_obesity_results.csv
  • Contains logistic regression results for the impact of tobacco use and obesity on ICU admissions.
• tobacco_obesity_odds_ratios.png
  • Visualizes odds ratios for tobacco use and obesity in predicting ICU admissions.
• tobacco_obesity_stacked_bar.png
  • Shows a stacked bar chart of ICU admissions categorized by tobacco use and obesity status.

4. code/04_temporal_trend.R from Shicy621 is Temporal Trends and Outcome Analysis

• Case Trends Over Time output/daily_cases.png output/weekly_cases.png output/monthly_cases.png
  • Visuals and summary statistics showing case counts across intervals (daily, weekly, monthly), with insights on significant fluctuations during key periods.
• Mortality and Hospitalization Trends output/mortality_hospitalization_rates.png
  • Charts showing mortality and hospitalization rate changes over time, highlighting any significant increases or decreases across different periods.
• ICU Admission and Intubation Rates output/icu_intubation_rates.png
  • Analysis of ICU and intubation rates during seasonal and pandemic peaks, with comparisons to low-demand periods to assess resource strain.
• Date-of-Death and Severe Case Prediction output/death_trends.png
  • Death trends over time, with potential predictive insights for severe cases, aimed at aiding resource planning for peak periods.
• Summary output/summary_statistics.rds
  • Key findings and recommendations on managing resources effectively during high-demand intervals.

The analysis parts are combined in covid_report.Rmd.

code/00_render_report.R is used for rendering the combined report to the root directory from the command line.

Report Building Instruction

To compile the report using make, follow these steps:

1. To generate the default report (COVID-19 cases):

Run the following command:

make

This will automatically analyze COVID-19 cases, as WHICH_CONFIG=default is the default setting.

2. To generate the non-COVID-19 cases report:

Set the WHICH_CONFIG parameter to healthy and run:

export WHICH_CONFIG=healthy
make

If you do not set the WHICH_CONFIG parameter, it will default to analyzing COVID-19 cases. Note: There's only Part III has non-Covid version.

3. To clean the output files in the output/ directory and other unnecessary by-products, run:

make clean

Synchronize Package Repository

When you clone this repository for the first time, use the following command to restore the required packages:

make install

If you add, remove, or update any packages during development, use the following command to update the project library and synchronize the renv.lock file:

renv::snapshot()