Combining two sciences – geography and statistics – geostatistics is a field of statistics that deals with analyzing and modeling data distributed in space or time. This discipline originated from the prestigious Paris School of Mines in the 1960s’ where engineers and mathematicians were looking for new ways to improve the evaluation of recoverable reserves in mineral deposits.
The solution?
Drilling on a large and more or less regular grid before extrapolating data from sampled cores (i.e., ore grade, thickness of deposit) to the entire field to estimate recoverable resources. Given that history, the reader might be left to wonder how these techniques apply to the health science field.
Geostatistics and its success
The success of geostatistics resides in its ability to capitalize on the first law of geography (Tobler, 1970), stating that “Everything is related to everything else, but near things are more related than distant things.” Although this statement refers to proximity in the geographic space (distance between observations), it also applies to time.
In other words, observations recorded a few days apart, such as pollutant concentrations, are more likely to be related than data collected one month apart.
One of the main characteristics of data collected in fields like earth science or public health is their structured distribution in space and time. This reflects the impact of various factors (e.g., geology, weather, hydrodynamic processes, human activities, socio-economic status) operating at different spatial and temporal scales. Geostatistics starts by modeling this space-time structure, which allows one to predict locations and instants that are not being monitored.
Geostatistics and health today
Recent years have witnessed the application of geostatistics in medical geography and spatial epidemiology, including the study of spatial patterns of disease incidence and mortality and the identification of potential “causes” of disease, such as environmental exposures or socio-demographic factors.
One application of geostatistics in public health is the study of air pollution and its health effects.
Geostatistical methods can model the spatial distribution of air pollution to estimate the exposure of individuals to pollutants. As well as investigate the association between exposure to air pollution and health outcomes such as respiratory diseases, cardiovascular diseases, and cancer. Another example is the study of infectious diseases and their spatial patterns.
Geostatistical methods can also model the spatial distribution of disease occurrence, identify clusters of cases, and investigate the spatial risk factors associated with disease transmission.
Improving our health outlook
Overall, geostatistics has massive potential for improving our understanding of the complex relationships between environmental exposures, social determinants, and health outcomes. By identifying spatial patterns and risk factors, geostatistical methods can help inform public health interventions and policies aimed at reducing health disparities and improving the health of populations.
BioMedware’s upcoming release of Vesta Version 3.0 includes a major innovation in geostatistics, which accounts for relationships among variables (e.g. “multivariate”) and across space and time. This innovation is the first of its kind and will greatly enhance the ability to model multi-variable datasets through space and time. Subscribe to our newsletter to get first access to Vesta product updates.
Tobler W., 1970. A computer movie simulating urban growth in the Detroit region. Economic Geography 46(2): 234-240.