Heterogeneity or behaviour modification - Can we tell?

Overview

This project aims to use simulations to explore the influence of heterogeneity and behavioural modification on the transmission dynamics of infectious diseases. The group will scale transmission rates by prevalence, a reflection of heterogeneity, mortality, a potential signal of behaviour modification, or both. The group may also investigate various delays from infection to death and compare the impact these have on the dynamics when combined with heterogeneity and behaviour modification.

Extensions to this project may involve participants fitting models to their simulated data to identify which scenarios allow for accurate identification of the underlying mechanism and which do not. Another possible extensions is to explore more complex methods for accounting for heterogeneity and behaviour modification. Finally, participants may wish to explore the impact of different levels of heterogeneity and behaviour modification on the performance of short-term forecasts.

Things to Consider

• When deciding on your interest in this project, please consider the following:

• This group is recommended for:
  • Participants working on outbreak and epidemic response.
  • Participants with an interest in infectious disease dynamics, particularly heterogeneity and behaviour modification.
  • Participants interested in designing scenarios to test the impact of different transmission dynamics.
  • Those interested in learning about software options for fitting dynamic models to time series data or who have experience with these methods.
  • Participants interested in short-term forecasting and forecast evaluation.
• This group will have the opportunity to engage in any of the following:
  • Design and run simulations based on different disease transmission dynamics.
  • Analyse simulated data to compare dynamics across different model assumptions.
  • Literature review of heterogeneity and behaviour modification in infectious disease modelling.
  • Learn about software options for fitting dynamic models to time series data, potentially including Bayesian methods and/or POMP.
• Sessions that may be useful for this project from week 1:
  • Introduction to models and data: HIV in Harare
  • HIV tutorial summary
  • Consequences of heterogeneity and modelling options
  • Distributed Delay Models of Survival

Data

Data for this project will be simulated by the group, potentially informed by the HIV Tutorial.

Resources

References

• Declining HIV prevalence and incidence in perinatal women in Harare, Zimbabwe
• Forecasting Ebola with a regression transmission model

Tutorials

• Getting started with scoringutils and forecast evaluation
• Bayesian workflow for disease transmission modeling in Stan
• POMP: Statistical inference for partially observed Markov processes