After Reporting August 2016 as the hottest month in an 11-month streak of record-high temperatures, NASA reiterated what scientists have been stressing for years: climate change is upon us. While global climatic changes pose numerous environmental consequences, many scientists also claim these changes will greatly alter the distribution of diseases globally. A recent Lancet report even noted that the response to global climatic changes could be the greatest health opportunity of the 21st century.

There is a wide range of evidence linking climate change to disease distribution, but the unpredictability of the exact way that climate change will unfold makes disease burden difficult to forecast with absolute certainty. As a result, different types of predictive modeling using a wide range of climate change scenarios are employed to identify a variety of potential outcomes that can be used to inform decision making and the development of mitigation strategies.

For example, a study published in the Malaria Journal examined the distribution of two species of malaria-carrying mosquitos (A. gambiae and A. arabiensis) under both current environmental conditions (A) and under three probable but different climate change scenarios:

  1. A rise of 2°C Africa-wide temperature, with a 10% increase of summer rainfall and a 10% decrease in winter rainfall (B)
  2. A 0.1°C rise in summer and winter maximum and minimum temperatures, with a 10% increase in summer rainfall and a 10% decrease in winter (C)
  3. A rise of 4°C Africa-wide temperature, with a 20% increase in summer rainfall and a 20% decrease in winter rainfall (D)

As demonstrated in the graphic below, each scenario results in different levels of mosquito prevalence. Identifying regions where all three climate change scenarios indicate a high probability of mosquito presence allows researchers to be more confident that species will occur in these locations in the future. Decision makers in the health community can prioritize these areas in their strategies for addressing the shifting disease burden.


Another study published in the Medical and Veterinary Entomology journal demonstrates the wide variability that results when different types of predictive models are employed. This study applied two types of spatial models—statistical (A) and mechanistic (B)—to produce relative risk maps of the probability of disease presence in a specific mosquito species on the Bermuda Islands (see below).


Map A—a lighter shade overall—shows a higher probability of mosquito presence and, therefore, greater risk of disease contraction compared to map B. The variation between maps A and B illustrates the impact that different models can have on predictive modeling outcomes. It is critical for researchers to be cognizant of these differences. and to employ multiple models within a single study to present a more complete and accurate picture.

To effectively tackle today’s prominent global health issues, it is essential to continue refining the modeling process. This will involve continued research into how uncertainty in predictive modeling influences result outcomes, as well as increased information and best practice sharing across the research and health communities. At Nexight, we specialize in facilitating this critical communication among stakeholders to help industries, government agencies, and non-governmental organizations achieve their strategic goals. We look forward to helping our clients address the impacts of climate change on global health and resilience.