From Methodology Development to Policy Applications
Lancaster University
Common risk factors:
- Poor sanitation and lack of clean water
- Limited healthcare access
- Poverty and overcrowding
- Exposure to disease vectors (e.g., mosquitoes, flies)
- Lack of education and awareness
where \(Y=(Y_1, \ldots, Y_n)\) and \(S = (S(x_1), \ldots, S(x_n))\)
Giorgi, E. and Fronterre, C. (expected August 2025) Model-based geostatistics for global public health using R. Chapman and Hall/CRC. The R Series.
Question: Which spatio-temporal correlation structures are suitable for modelling disease prevalence data?
Giorgi, E., Diggle, P. J., Snow, R. W., Noor, A. M. (2018). Geostatistical methods for disease mapping and visualization using data from spatio-temporally referenced prevalence surveys. International Statistical Review. https://doi.org/10.1111/insr.12268
Questions: 1) How we predict a gold-standard diagnostic using cheap biased diagnostics? 2) How do we model the bivariate relationship between two complementary gold-standards?
Amoah, B., Diggle, P. J., Giorgi, E. (2019). A geostatistical framework for combining data from multiple diagnostic tests. Biometrics. doi:10.1111/biom.13142
Question: How can we effectively model the bivariate relationship between disease suitability and disease prevalence?
Diggle, P. J., Giorgi, E. (2016). Model-based geostatistics for prevalence mapping in low-resource settings (with discussion). Journal of the American Statistical Association. 111:1096-1120
Using multivariate geostatistical models to analyse serological data from multiple diseases
Incorporating mathematical models for the impact of interventions into geostatistical models
Developing more robust methods for the validation of geostatistical models for counts data.
Dealing with spatial and temporal misalignment in repeated cross-sectional survey data
Sasanami, M., et al. (2023). Using model-based geostatistics for assessing the elimination of trachoma. PLoS Neglected Tropical Diseases, 17(7): e0011476. https://doi.org/10.1371/journal.pntd.0011476
Complexity of methods – Advanced models are difficult to interpret for non-specialists.
Data uncertainty – Policymakers may struggle to incorporate probabilistic estimates into decision-making.
Lack of user-friendly tools – Limited access to tools that allow interactive exploration of spatial predictions.
Lack of local geostatistical expertise – Few or no local experts support to support control programmes.
Technical meeting on geostatistical methods for trachoma elimination
Lancaster University, 4-5 March 2024
Let’s Connect
🔗 giorgistat.github.io
📧 e.giorgi@lancaster.ac.uk
📍 CHICAS, Lancaster Medical School, Lancaster University, UK