Paper:

Friedman, J. & Akre, S. COVID-19 and the Drug Overdose Crisis: Uncovering the Deadliest Months in the United States, January‒July 2020. Am J Public Health e1–e8 (2021). doi:10.2105/ajph.2021.306256

Updated 07-17-2021

• Overdose data by state for analysis purposes in: Useable_Overdose_Data_Through_Oct_2020.csv

• src/: Contains python scripts used in analysis
  • defilter_data.py: Disaggregates rolling sum data
  • estimate_error.py: Calculates error in disaggregation process from historical data
  • load_data.py: Cleans and preprocesses input datasets
  • tests/: Unit tests to ensure desired functions work as intended
• input/: Input datafiles used for analysis
• output/: Generated intermediate files
• visuals/: Generated figures and tables
  • state_timeseries_2021-05-16.pdf: State by state data
• CDC_overdose_monthly_recovery.ipynb: Python notebook used to disaggregate CDC overdose death data and calculate errors in that process
• CDC_monthly_imputation.R: Imputes missing data from monthly "ground truth" overdose data
• CDC_monthly_recovery_analysis.R: Run analysis and visualizations of monthly overdose data. Used to generate figures in manuscript.
• Useable_Overdose_Data_Through_Sep_2020.csv: Cleaned monthly overdose death data by state. Intended for use by others.

Python 3.7.6

• pandas==1.2.0
• numpy==1.19.4
• scipy==1.5.4
• matplotlib==3.3.3
• seaborn==0.11.1
• jupyterlab==3.0.0 .

R 4.0.3

1. Set up R and python environments
2. Create an empty folder output and visuals inside of repository
   • Intermediate files and manuscript figures will be placed here
3. Run CDC_monthly_imputation.R
   • Make sure to change root variable to point to location of this repository on your local system
4. Run the full CDC_overdose_monthly_recovery.ipynb python notebook
5. Run CDC_monthly_recovery_analysis.R
   • Make sure to change root variable to point to location of this repository on your local system