Related Experiment Video
Updated: Apr 28, 2026
An In Vitro Model for Measuring Immune Responses to Malaria in the Context of HIV Co-infection
Published on: October 6, 2015
Testing a multi-malaria-model ensemble against 30 years of data in the Kenyan highlands
Daniel Ruiz1, Cyrille Brun, Stephen J Connor
1International Research Institute for Climate and Society, Lamont Doherty Earth Observatory, Columbia University in the City of New York, 61 Route 9 W, Palisades, PO Box 1000, New York 10964-8000, USA. pfcarlos@iri.columbia.edu.
Background:
Multi-model ensembles could overcome challenges resulting from uncertainties in models' initial conditions, parameterization and structural imperfections. They could also quantify in a probabilistic way uncertainties in future climatic conditions and their impacts.
Methods:
A four-malaria-model ensemble was implemented to assess the impact of long-term changes in climatic conditions on Plasmodium falciparum malaria morbidity observed in Kericho, in the highlands of Western Kenya, over the period 1979-2009. Input data included quality controlled temperature and rainfall records gathered at a nearby weather station over the historical periods 1979-2009 and 1980-2009, respectively. Simulations included models' sensitivities to changes in sets of parameters and analysis of non-linear changes in the mean duration of host's infectivity to vectors due to increased resistance to anti-malarial drugs.
Results:
The ensemble explained from 32 to 38% of the variance of the observed P. falciparum malaria incidence. Obtained R²-values were above the results achieved with individual model simulation outputs. Up to 18.6% of the variance of malaria incidence could be attributed to the +0.19 to +0.25°C per decade significant long-term linear trend in near-surface air temperatures. On top of this 18.6%, at least 6% of the variance of malaria incidence could be related to the increased resistance to anti-malarial drugs. Ensemble simulations also suggest that climatic conditions have likely been less favourable to malaria transmission in Kericho in recent years.
Conclusions:
Long-term changes in climatic conditions and non-linear changes in the mean duration of host's infectivity are synergistically driving the increasing incidence of P. falciparum malaria in the Kenyan highlands. User-friendly, online-downloadable, open source mathematical tools, such as the one presented here, could improve decision-making processes of local and regional health authorities.
