/bmm-opioid

A Bayesian multistate model to understand the transitions between opioid use, misuse, and fatal overdoses

Primary LanguageR

Updated: 2023-07-07.

Bayesian Multistate Modelling of the Opioid Crisis

The project aims to provide a theoretical for estimating the transition probabilities between different stages of opioid use.

In this repository, you will find all codes related to the project and short summaries and plots. To reproduce the analysis, run the following make commands in the project directory:

Simulation

  • make gen-sim: Generate data for simulation.
  • make fit-sim: Fit simulated data both using the ratio estimates and without using the ratio estimates.
  • make plot-sim: Generate diagnostic and exploratory plots of the fitted estimates.

US case study

  • make download-us: Download data files containing multiple causes of death reports and National Survey on Drug Use and Health.
  • make parse-us: Parse the downloaded data files to prepare for analysis.
  • make fit-us: Fit the parsed data for both non-Hispanic black and non-Hispanic white US populations.
  • make plot-us: Generate diagnostic and exploratory plots of the fitted estimates.

Publications and presentations

  • make plot-paa: Generate plots used for the PAA 2023 poster.

You can find the resulting plots in the following directories:


Authors

  • Michael J. Moon, University of Toronto
  • Monica Alexander, University of Toronto