Statistical methods for linking geostatistical maps and transmission models: application to lymphatic filariasis in East Africa

Research output: Contribution to journalArticlepeer-review

Authors

Colleges, School and Institutes

Abstract

Infectious diseases remain one of the major causes of human mortality and suffering. Mathematical models have been established as an important tool for capturing the features that drive the spread of the disease, predicting the progression of an epidemic and hence guiding the development of strategies to control it. Another important area of epidemiological interest is the development of geostatistical methods for the analysis of data from spatially referenced prevalence surveys. Maps of prevalence are useful, not only for enabling a more precise disease risk stratification, but also for guiding the planning of more reliable spatial control programmes by identifying affected areas. Despite the methodological advances that have been made in each area independently, efforts to link transmission models and geostatistical maps have been limited. Motivated by this fact, we developed a Bayesian approach that combines fine-scale geostatistical maps of disease prevalence with transmission models to provide quantitative, spatially-explicit projections of the current and future impact of control programs against a disease. These estimates can then be used at a local level to identify the effectiveness of suggested intervention schemes and allow investigation of alternative strategies. The methodology has been applied to lymphatic filariasis in East Africa to provide estimates of the impact of different intervention strategies against the disease.

Bibliographic note

Acknowledgements: The authors gratefully acknowledge funding of the NTD Modeling Consortium by the Bill and Melinda Gates Foundation [OPP1152057, OPP1053230, OPP1156227, OPP1186851]. The views, opinions, assumptions or any other information set out in this article should not be attributed to the Bill and Melinda Gates Foundation or any person connected with the Bill and Melinda Gates Foundation. The authors thank Rachel L. Pullan and Jorge Cano for sharing the geostatistical maps of LF prevalence and Michael A. Irvine and Paul Brown for improving the model code. We also thank Dr. Nick Golding for providing helpful comments on our manuscript.

Details

Original languageEnglish
Article number100391
JournalSpatial and Spatio-temporal Epidemiology
Volume2020
Issue number00
Early online date21 Nov 2020
Publication statusE-pub ahead of print - 21 Nov 2020

Keywords

  • Bayesian methods, Fine-scale spatial predictions, Linking maps with models, Lymphatic filariasis, Projections, Uncertainty

Sustainable Development Goals