Estimating the burden of Chagas disease and impact of interventions in Latin America

Modelling the transmission dynamics of infectious diseases relies not only on complex mathematical and statistical models, but equally on reliable data. In the context of Chagas disease, we argue that such balance has yet to be met. This project will build a general framework that will allow the use of routine sero-prevalence studies in Latin America to improve estimates of the past, current and potential future burden of Chagas disease. To achieve this, results of age-stratified sero-prevalence studies will be used to:

  1. Estimate temporal and spatial heterogeneities in incidence trends - formally estimating the Force-of-Infection (FoI), or rate at which individuals susceptible to infection become infected.
  2. Estimate the basic progression rates from the different infection/disease stages and the excess morbidity and mortality associated with each of those stages.
  3. Estimate the expected number of cases stratified by ages and infection stages (e.g. cardiomyopathy) and associated excess deaths due to Chagas.

The platform allows the whole process – from input data to model outputs – to take place in standardised, open-source, clear, reliable and reproducible way. Ultimately, the model results depend to a great extent on the coverage and data quality of the age-stratified sero-prevalence studies available throughout Latin America. This step involves an intense process of systematic review and data management to ensure the input used is reliable and representative of the different populations.

At present, and relying mostly on publicly accessible information, data have been collated and Force-of-Infection estimates have been obtained for seven countries, with which initial estimates of burden have been calculated for one country. Future steps in this research will involve:

  • Collating non-publicly available data from National and/or Subnational Chagas disease control programmes.
  • Introducing Machine Learning techniques to improve predictions in locations where sero-prevalence data are not available.
  • Performing systematic external validations on predictions.
  • Engaging with countries and stakeholders to improve the use and impact of the proposed platform.

Scientific Approach

The general approach is to build on a proven functional modelling framework combining five modelling components:

  1. Modelling and estimating the Force-of-Infection of Chagas Disease
  2. Modelling and estimating the Burden of Chagas Disease
  3. Predicting future trends and the impact of potential interventions
  4. Providing guidelines for data collection to inform the current epidemiological situation
  5. Guiding diagnostics strategies, and procedures for the evaluation of the impact of control.

These components rely on systematically reviewing the published literature, as well as sharing unpublished data with endemic countries. All data will be centralised through a data secured platform called DICTUM (Decreasing the Impact of Chagas disease Through Modelling), hosted at Imperial College London in collaboration with PAHO.

The ultimate aim of this project will be to help Latin American countries to monitor and optimise their progress toward the control of Chagas disease.

Primary LCNTDR organisation

External partners

  • University of Sussex, UK 
  • University of Princeton, USA 
  • Pan American Health Organization, USA