Abstract
This chapter identifies the need for GIS and geospatial applications in health research. Spatial epidemiology (or geo-epidemiology) and geo-medicine are two rapidly growing fields in medical research and epidemiology. They could offer tremendous potential to the healthcare industry, health researchers and students, as well as the community in general. Their main advantage is the wide range of methodologies that they utilize in order to study health outcomes and incidences in three or four dimensions (2 or 3D space and 1D time). Therefore, this chapter also introduces selected GIS applications with a special focus on epidemiology and public health. Furthermore, in this chapter we will critically appraise the most frequently used tests and spatio-temporal methodologies in the literature and discuss the future emerging trends and research opportunities that arising.
TopIntroduction
Medicine, public health and epidemiology, as well as other social and health sciences, have always been considering “place” as a significant factor for individual particularities, population variations and inequalities. The concept of individual health and holistic healthcare and health management was first introduced by Hippocratus and was later expanded and further studied in the 1990s (Macintyre & Ellaway, 2003). The idea that place, individual characteristics and attitudes along with the external environment affects health has nowadays become widely accepted. “Place” is not only referring to the geographical location (e.g. location of residence), it is a wider concept in the context of space (including each location an individual has visited, environmental conditions and exposures, societal framework, culture, religion, economy, habits, human interactions etc). The terms of “place” or “space” refer to all the phenomena that occur on earth or any sub-entity, such as human. These phenomena actually occur in the so called “real world” or more correctly the “spacetime”. Spacetime refers to any mathematical model that combines space and time into a single interwoven continuum, while in our universe it is usually interpreted from an Euclidean space perspective (3 dimensions of space and 1 dimension of time). These concepts are extensively used in data analysis in physics, geography and mathematics contrary to health research that employs conventional methods of analysis.
These concepts refer to a quantitative and qualitative approach that is met in the philosophy of the Geographical Information Systems (GIS). GIS have a long history in research with a range of applications in various scientific fields. Roger Tomlinson inserted the term “Geographic Information System” in 1968, while Tomlinson was acknowledged as the “father” of GIS and John Snow as the “father” of spatial epidemiology (Coppock & Rhind, 1991). Spatial epidemiology (or geo-epidemiology) and geo-medicine are two rapidly growing fields in health research and epidemiology. GIS offer tremendous potential to the healthcare industry, health researchers, students and the community, but are still considered as a new tool. Thus, further efforts are required in order to explore GIS in health and incorporate them in the already established medical practice.
The main purpose of this chapter is to present a range of GIS applications in selected fields of health research, with a special focus on epidemiology and public health. More specifically, it serves the following objectives:
- 1.
Identifies and discusses the existing applications of GIS on epidemiological research, public health, clinical practice, hospital and healthcare management and community health
- 2.
Presents selected methods of geospatial and spatio-temporal analysis
- 3.
Critically appraises these methods in terms of validity, effectiveness and fitness in major health topics and
- 4.
Highlights impact of GIS, space and time on health research.
Furthermore, the examples of geospatial and spatio-temporal methods of analysis that are discussed involve descriptive, exploratory and explanatory statistics. For instance, major analytical tests that are appropriate for analyzing disease patterns or health outcomes, mapping clusters, assessing autocorrelation, grouping common trends, identifying risk regions and hot spots, modeling spatial relationships, predicting spatial trends or future dispersion of disease and health-related events [e.g. Getis-ord Gi*, Anselin Local Moran's I, Geary’s c, Ripley's K Function, Interpolation Kriging, Geographically Weighted Regression (GWR), Ordinary Least Squares (OLS), Knox spatio-temporal test and other].
Key Terms in this Chapter
Geographical Information Systems (GIS): GIS is a system designed to capture, store, manipulate, analyze, manage, and present all types of spatial or geographical data in all dimensions of space and time. It refers to all the following: the science and theory of geography and its analytics, the software and hardware, the analyst/expert and the simple users. It is often used to refer to the academic discipline or career of working with GIS and is a large domain within the broader academic discipline of Geoinformatics.
Exposure: The dose of a substance reaching an individual (e.g. environmental exposures like particular matters).
Spatial Analysis/Statistics: Any formal techniques that study entities using their topological, geometric, or geographic properties. It includes a variety of techniques based on mathematics and statistics, while many of these techniques are still developing.
Personalized Medicine: Personalized medicine refers to a form of medicine that takes under consideration the needs and wishes of each patient in order to provide custom-tailored prevention, diagnosis or treatment services. The aim of personalized medicine can be summarized in one phrase: “to prescribe the right treatment for the right person at the right time”. The core principle of personalized medicine is the person-centered approach that gives emphasis on treating the patient as a person and provide care suited to the patient's condition.
Primary Health Care (PHC): The Primary Health Care (PHC) is the first point of contact within a country’s health care system. The PHC services are fundamental for the population’s health and welfare. PHC aims to promote the overall health status of a community, prevent diseases or early diagnose them and manage patients with chronic diseases. Among its core principles involvement of community and equality of access are included. Spatial Epidemiology: It is considered to be a subfield of health geography that focuses on the study of spatial distribution of several health outcomes. It is closely involves the description and examination of disease and its geographic and temporal variations, while in considers all types of factors that may affect health outcomes in real life (e.g. demographic, environmental, behavioral, socioeconomic, genetic, and infections risk factors).
Cluster: A disease cluster is the occurrence of a greater than the expected number of cases of a particular disease or health outcome within a group of individuals, a geographically defined area or a period of time.
Geocoding: The process by which GIS matches each record in an attribute database with the geographic files. The geocoding process assigns each record in the attribute database to a point on a map (e.g. by using the address information in an attribute database and comparing this with the address information in a stored spatial database).
Healthcare Management: The term healthcare management (or healthcare administration) is defined as supervising the functions of a healthcare organization. Healthcare managers tasks include providing leadership, management and direction to healthcare units (such as hospitals or other health care systems) in order to ensure the best delivery of the available healthcare services.
Spatial Autocorrelation: Spatial dependency is the co-variation of properties within geographic space. In other words, characteristics at proximal locations appear to be correlated, either positively or negatively. It leads to the spatial autocorrelation problem in statistics because this “violates” standard statistical techniques that assume independence among observations. For instance, regression analyses that don’t compensate for spatial dependency can yield unreliable significance tests due to several unstable parameter estimates. Spatial regression models capture these relationships and don’t suffer from these weaknesses.