Overview

This work aims to systematically compare three individual-based malaria transmission models that have been used to inform global, national, and product-development decision-making.

Mathematical models are increasingly being used by countries and by global partners to predict the potential impact of interventions. Many models have been independently developed, and in malaria, three individual-based models (EMOD, malariasimulation, and OpenMalaria) have been used to inform decision-making at the country or global level.

However, the extent to which the three models give similar or different results, under what conditions, and why, is not known. Furthermore, while best practice is to use multiple models to inform decision-making, often a single model is used. Ensemble modeling with individual-based malaria models has been challenging due to the expertise required to run each model correctly, the extent to which harmonization across models is needed to enable comparison, and the lack of a common software platform on which to use the models.

AHADI is collaborating with multiple partners to systematically characterize the behavior of these three models (EMOD, malariasimulation, and OpenMalaria) so that countries and the global community can better interpret model results, understand drivers of differences in model outputs, and identify common knowledge gaps where new data is needed to inform the models. We are building a software framework to enable anyone to run one or all of these models under harmonized conditions.

Partners

1. The Kids Research Institute Australia, Australia
2. University of Western Australia, Australia
3. University of Basel, Switzerland
4. Swiss Tropical and Public Health Institute, Switzerland
5. Imperial College London, UK
6. Institute for Disease Modeling, USA
7. Northwestern University, USA

Outputs

Articles

How can modeling responsibly inform decision-making in malaria?

When models are used to inform decision-making, both their strengths and limitations must be considered. Using malaria as an example, we explain how and why models are limited and offer guidance for ensuring a model is well-suited for its intended purpose.

MultiMalModPy - A novel software framework for running three individual-based malaria transmission models

The multi-malaria-modeling Python framework (MultiMalModPy) allows three individual-based malaria models (EMOD, OpenMalaria, and malariasimulation) to run simultaneously, alone or in combination. This article presents the software structure of MultiMalModPy with example inputs and outputs.

Harmonization and comparison of baseline behavior of three individual-based malaria models

This article reports behaviors of three malaria models (EMOD, malariasimulation, and OpenMalaria) at steady state under perennial transmission, seasonal transmission, and in response to a step-change in carrying capacity. The models were aligned as much as possible on critical input parameters, such as transmission intensity, case management, and diagnostic performance while maintaining structural model differences.

Software

MultiMalModPy: a software framework for running multiple individual-based malaria models

‍Documentation on MultiMalModPy

Funding

AHADI’s contribution to this work is funded by the Gates Foundation.