Understanding Uncertainty in Temperature Effects on Vector-borne Disease: a Bayesian Approach

Authors

Leah R. Johnson, University of South FloridaFollow
Tal Ben-Horin, University of California
Kevin D. Lafferty, University of California
Amy McNally, University of California
Erin Mordecai, University of California
Krijn P. Paaijmans, Universitat de Barcelona
Samraat Pawar, University of Chicago
Sadie J. Ryan, State University of New York

Document Type

Article

Publication Date

2015

Keywords

Anopheles gambiae, basic reproductive number, Bayesian statistics, climate envelope, malaria, Plasmodium falciparum, sensitivity analysis, thermal physiology, uncertainty analysis

Digital Object Identifier (DOI)

https://doi.org/10.1890/13-1964.1

Abstract

Extrinsic environmental factors influence the distribution and population dynamics of many organisms, including insects that are of concern for human health and agriculture. This is particularly true for vector-borne infectious diseases like malaria, which is a major source of morbidity and mortality in humans. Understanding the mechanistic links between environment and population processes for these diseases is key to predicting the consequences of climate change on transmission and for developing effective interventions. An important measure of the intensity of disease transmission is the reproductive number R₀. However, understanding the mechanisms linking R₀ and temperature, an environmental factor driving disease risk, can be challenging because the data available for parameterization are often poor. To address this, we show how a Bayesian approach can help identify critical uncertainties in components of R₀ and how this uncertainty is propagated into the estimate of R₀. Most notably, we find that different parameters dominate the uncertainty at different temperature regimes: bite rate from 15°C to 25°C; fecundity across all temperatures, but especially ~25–32°C; mortality from 20°C to 30°C; parasite development rate at ~15–16°C and again at ~33–35°C. Focusing empirical studies on these parameters and corresponding temperature ranges would be the most efficient way to improve estimates of R₀. While we focus on malaria, our methods apply to improving process-based models more generally, including epidemiological, physiological niche, and species distribution models.

Was this content written or created while at USF?

Yes

Citation / Publisher Attribution

Ecology, v. 96, issue 1, p. 203-213

Scholar Commons Citation

Johnson, Leah R.; Ben-Horin, Tal; Lafferty, Kevin D.; McNally, Amy; Mordecai, Erin; Paaijmans, Krijn P.; Pawar, Samraat; and Ryan, Sadie J., "Understanding Uncertainty in Temperature Effects on Vector-borne Disease: a Bayesian Approach" (2015). Integrative Biology Faculty and Staff Publications. 548.
https://digitalcommons.usf.edu/bin_facpub/548