New Zealand COVID-19 vaccine data (based on data released by Barry Young)

Analysis of New Zeland COVID-19 vaccine data leaked by Barry Young to a US journalist who then made it public.

The data directory has the data released by Barry Young (obfuscated to remove any PII and dates have been shifted slightly so no PII is released).

The code directory has the file to run to do the batch analysis. Just do python vax.py after you cd to the code directory.

The analysis directory has an excel file where we've used the output data in pivot tables to show the 1 year mortality rates for each batch for each age range.

Summary of vax.py

ChatGPT offers this helpful summary of vax.py:

1. Purpose: This code analyzes a dataset of vaccine records. It checks if people died within a year after receiving a vaccine and summarizes the results.

2. Steps in the Code:

   • Read the Data: The code reads a file containing vaccine records into a table (DataFrame).
   • Format Dates: It changes the date columns to a format that the code can work with easily.
   • Add a New Column: It creates a new column to show the month and year when the vaccine was given.
   • Calculate Time Difference: It figures out how many days passed between the vaccine date and the date of death.
   • Determine Deaths within a Year: It adds a column that indicates if someone died within a year of receiving the vaccine.
   • Group and Summarize Data: It groups the data by age, vaccine batch, dose number, and month-year of service. It counts the number of records and sums up the number of deaths within a year for each group.
   • Save the Results: Finally, it saves this summarized information into a new CSV file.

3. Outcome: The result is a new file that shows how many people received each batch of the vaccine, how many doses they got, and how many of those people died within a year of getting the vaccine, organized by age and month-year.

In essence, the code is taking a big list of NZ vaccine records, checking who died within a year after vaccination, summarizing the results, and then saving this summary into a new file.

Note

The data here is NOT the original NZ data, but it is a derivative work.

It contains no PII that matches any living or dead person anywhere in the world.

It does not violate any US law including copyright law.