who-programme-costing-utilities

A command-line utility for producing modular programme costs.

Context

When building models, we tend to consider costs at the individual-level. That is, costs that are incurred when an individual, or group of individuals do a particular thing. However, there are often costs that are incurred by virtue of a programme being run. For example, a law being passed on sugar-sweetened beverages still has a cost to pass the law. For example, a breast-cancer screening intervention has costs to run the programme, train staff, etc. For these costs, we have built a separate utility, which allows us to consider the costs of a programme, rather than the costs of an individual. The hope is that this utility can be incorporated into specific modelling, once a simple method has been established. Until then, it's more clear to have another utility which does this. This is that utility.

Use

• You must have the prerequisite databases (sqlite databases)
  • ./data/who_choice_price_database.db
• Clone the repository using git clone
• Install the requirements using pip install -r requirements.txt
• You can check the tests work by running python -m unittest discover tests
• Run the program by using python main.py -i X -o Y
  • where X is the input file and Y is the output file
  • Note, the input file must be a .JSON, utf-8 encoded file
  • Note, the output file will be a .CSV, utf-8 encoded file

Input file

The input file must be a .JSON, utf-8 encoded file. These are the keys that should be included in the input file:

• start_year: int (DEFAULT: 2023) the year the programme starts
• end_year: int (DEFAULT: 2030) the year the programme ends
• discount_rate: float (DEFAULT: 1) the discount rate to use for the programme
• currency: str (DEFAULT: USD) The currency to use for the programme
• currency_year: int (DEFAULT: 2018) The year to use for the currency
• modules: list (DEFAULT: Programme costs template) The list of modules relevant to the programme