Contact FutureLearn for Support
Skip main navigation
We use cookies to give you a better experience, if that’s ok you can close this message and carry on browsing. For more info read our cookies policy.
We use cookies to give you a better experience. Carry on browsing if you're happy with this, or read our cookies policy for more information.

Spatial data analysis

The purpose of analysis is always to gain a deeper understanding about a topic. Analysis helps to reveal information that cannot be seen directly from the original data presented in a map. For example, you may see many points located relatively close together on a map but to know if these points represent a spatial cluster, you will need further analysis. When we are analyzing GeoHealth data our aim is to find information that is essential for planning interventions and ultimately perhaps disease prevention.

The two groups of applications within the GeoHealth domain were defined in the first step of this course: “health care geography” and “spatial epidemiology”. Both of these domains have different but partially overlapping sets of analytical methods.

Important for spatial epidemiology are:

  1. Cluster Analysis – this group of techniques provides information on clustering and is often used to find possible associations between a disease and the environment. Does a cluster coincide with environmental exposure?
  2. Geographic correlation studies – this group studies the variation in exposure to environmental or social variables.
  3. Diffusion studies - these techniques are not trying to identify clusters or a correlation with the environment but assume that a disease re-locates to new locations. Diffusion analysis focus on the speed and direction of spread over time. This type of study is often conducted for infectious diseases.

It is not so that each of the groups listed above leads to one type of analysis. We have many techniques and approaches available for testing for clusters, or disease diffusion. Certain techniques are also applied to many different types of disease or health conditions. The type of data you have available (and the quality and completeness) also has an influence on the analytical techniques that can/should be applied. A small summary for techniques in the field of spatial epidemiology is provided below but this is far from complete.

Aim Analysis: Type technique: Type Disease:
Cluster Analysis Analytical techniques; Statistical methods Cancer; hodgkin disease
Correlation studies Statistical methods; Machine learning; Simulation Cancer; cardiovascular disease; vector born disease (vector to environmental features water to mosquitoes, or ticks to vegetation types)
Diffusion studies Statistical; Machine learning; Simulation Infectious diseases; Influenza, measles etc.

Health Care Geography

Analytical techniques for health care geography are concerned with assessing the spatial distribution and accessibility of health services. Good access to health care is often regarded as a right of each human being. But how do you measure good access? And how do you plan new facilities to improve this accessibility? This group of analytical methods provides you with the tools to do this.

For health care analysis:

  1. Accessibility analysis
  2. Planning of health facilities
Aim Analysis: Type technique: Type Disease:
Accessibility analysis Minimum distance analysis; Isochrone measures; Potential measures Calculate the distance from a household to the nearest health facility; Draw lines around a health facility that represent 15 minutes drivetime; Methods to evaluate potential access to care
Planning health facilities Location-Allocation modelling From a set of possible new locations for a health facility, pick the location with the highest coverage

Any of the analysis listed above can be carried out in two different ways: in Space and in Space and time (spatio-temporal analysis). For example, I can identify a cluster in space based on a dataset that just represents one moment in time, for example, the time of the survey that you conducted. This will reveal if there is a cluster at precisely this moment but it cannot tell you if this is a permanent cluster or not. The temporal element is especially important for diffusion studies and surveillance of diseases including early warning systems as they detect, and potentially forecast, changes in disease occurrence in space and time.

Share this article:

This article is from the free online course:

Geohealth: Improving Public Health through Geographic Information

University of Twente

Course highlights Get a taste of this course before you join: