BACKGROUND

A variety of GIS projects and technologies exist today that involve health care. Here is a sampling of these projects, technologies, and research.

GIS

A geographic information system (GIS) is a “computer-based system for Text Box: A “geographic information system (GIS) is a computer-based system for the integration and analysis of geographic data.” - GIS and Public Health  the integration and analysis of geographic data.” GIS can be distinguished as a technology by the following software functions: “the ability to store or compute and display spatial relationships between objects, the ability to store many attributes of objects, the ability to analyze spatial and attribute data in addition to simply managing and retrieving data, and the ability to integrate spatial data from different sources.”

GIS is a part of a “larger constellation of computer technologies for processing geographical data.” Other technologies in this “constellation” include GPS, remote sensing, scanning, and digitizing. However, these technologies are used for spatial data collection by persons using GIS, and they are not considered GIS. The data structure and software functions of GIS technology are specifically designed to integrate and analyze data based on location. The GIS software stores locations of features on the earth’s surface to allow identification of different features on the same location.

GIS technology has a wide range of applicability. Possible applications include scientific investigations, resource management, asset management, urban planning, criminology, history, route planning, environmental impact assessment, route planning, and natural disaster emergency response planning.

Geographic information science is the science underlying geographic information systems and their applications.

Risk Mapping

GIS has proven to be a powerful tool in spatial modeling of environmental risk factors. The technology can be used to model the spread of contaminants, vector habitats, host habitats, and any other geographical patterns of risk factors. With this ability to combine the factors of importance in specific desired locations, health investigators are able to gage the level of risk of specific diseases in various areas.

ArcGIS

ArcGIS is an integrated collection of GIS software products produced by ESRI . The geodatabase of all ArcGIS products allow the user to handle rich data types, apply sophisticates rules and relationships, and access large volumes of geographic data stored in files and data bases. ArcGIS provides several levels of GIS products: Desktop GIS, Server GIS, Mobile GIS, Hosted GIS, and ArcGIS Online.

The primary platform, Desktop GIS, is designed for professionals to compile, author, and use geographic information and knowledge. The software allows one to view spatial data, create maps, perform basic spatial analysis, and manipulate shapefiles and geodatabases. Server GIS centralizes GIS servers in application servers to deliver GIS capabilities to large numbers of users over networks. Mobile GIS allows field-based personnel to utilized GIS technologies. Hosted GIS delivers a cost-effective way to include mapping services in Web-enabled applications. ArcGIS Online provides online map services that are ready to use with ArcGIS.

A significant amount of training is required to effectively be able to use the ArcGIS software. A certificate can be obtained after the completion of web courses, instructor-led courses, and instructor-led virtual classroom courses. Training requires an average of 3 days to complete.

Notable Records of Disease Mapping:

1690-1692 Plague: Bari, Naples

When the plague returned in the 1600s to attack much of Europe, nobody knew what caused the disease or how it was transmitted. Two maps by Filippo Arrieta from this period have survived. Arrieta mapped the areas most affected and the boundaries of military quarantine imposed to prevent its spread to neighboring towns and other provinces. Land was labeled to denote areas where the plague was active and areas where the plague had recently been active. “Inherent in the map is a theory of the disease, one based on its past activity, contemporary pattern of occurrence, and known potential for diffusion.”

1798 Yellow Fever: New York.

Surgeon Valentine Seaman argued that the origin of New York City’s disease outbreak was the smell that arose from the city’s garbage and sewage that accumulated in the harbor area. His arguments were supported with maps published in his article in the Medical Rpository. These maps of the city’s docklands mapped slips, puddles, filth, and garbage (marked “S”), areas of open sewers with human waste (marked “x”), and fatal cases. Although his theory on smallpox and odors was not accurate, the waste in the docklands did hold critical epidemiological relation to the incidence of yellow fever in the city. Seaman’s map “admirably advanced a thoroughly ecological argument.”

1854 Cholera: Britain

In 1849, Doctor John Snow theorized that cholera was transmitted by contaminated food or water. Against the widely-held belief that cholera is contracted by breathing noxious vapors, Snow agued that cholera could not have been an airborne disease since it does not affect the lungs. In 1854, an outbreak allowed Snow to prove his theory. After mapping cholera deaths, he noticed that about 500 people died in ten days near the intersection of Cambridge and Broad Streets. All the people in this area shared a single water supply. Convinced that the water supply was the source of cholera transmission, Snow persuaded city officials to remove the pump handle from the water source. New incidences of cholera in the area reduced significantly to near extinct levels.

RAGFIL

RAGFIL, or Research on Rapid Assessment of Geographical Assessment of Bancroftian Filariasis, is a disease control program aimed at eradicating lymphatic filariasis (link to Parasite tab). The International Task Force identified a need to understand the geographic distribution of lymphatic filariasis before they could know where to target mass treatment. Because the procedures for diagnosing lymphatic filariasis are time-consuming, cumbersome, expensive, and intrusive, the presence of clinical filarial disease is used as a proxy to measure the levels of endemecity of filariasis. Once a method for rapid assessment had been field-tested and proven effective in Ghana, Tanzania, Myanmar and India, it was proposed as the cornerstone for a standardized, rapid-mapping system that could be applied throughout Africa. Figure 5 illustrates the three-dimensional model of the prevalence of filariasis across various longitudinal locations.

NASA Applied Sciences Program

The NASA Applied Sciences Program is an initiative aimed to enhance the organization’s decision-support capabilities by “enabling their expanded use of Earth science research results, information, and technology to serve their management and policy responsibilities to society,” explained John Haynes, Program Manager of NASA’s Aviation and Public Health Applications. NASA has a lot of Earth science information to provide and has identified twelve areas of national priority through which to use this information for public good.

The Public Health Program Element, one of the twelve areas, focuses on building relationships between NASA Earth observation systems, modeling systems, and partner-led decision support systems for epidemiologic surveillance in the areas of infectious disease, environmental health, and emergency response and preparedness. NASA partners with many organizations that have public health responsibilities, including the CDC, the National Institutes of Health, the U.S. Environmental Protection Agency, the Department of Health and Human Services, the National Oceanic and Atmospheric Administration, the Department of Defense (DOD), and the Department of Energy.

One of the Public Health Program Element projects involving prevention of parasitic disease spread is the Malaria Modeling and Surveillance (MMS) project. This project obtains malaria epidemiological records from the Thai Army’s Pramongkul Hospital and the U.S. Armed Forces Research Institute for Medical Sciences in Thailand. Then, the team uses the data to remotely sense environmental observations to create a nonparametric model for predicting malaria cases. The team has been very successful in pinpointing potential cases of Mekong malaria and filariasis and is in the process of increasing the accuracy.

U.S. Geological Survey: Mapping Malaria

Text Box: “Understanding environmental and ecological health is a prerequisite to protecting public health.”           USGS Health Forum U.S. Geological Survey is one of the world’s leading experts on mapping. We interviewed Eliot Christian, former manager of Data Information Systems at USGS, on his recent involvement with parasite-related projects through USGS’ health sector. “Public health problems caused by environmental contamination and emerging infectious diseases are a growing concern worldwide.” Pressures of development, including population growth, make it difficult to effectively enforce public health prevention strategies against parasitic diseases, including vector-borne and zoonotic diseases, water contamination, airborne contaminants, bioaccumulative contaminants in the food chain, and environmental threats to public health. Understanding environmental and ecological health is a prerequisite to protecting public health. USGS, the country’s natural science agency, has begun to play a significant role in providing scientific knowledge necessary to improve our understanding of environmental contributions to disease.

“Mapping has been a core tool for epidemiology since it was first developed as a discipline,” Christian remarked. Christian made it clear that the issue of correlating mapping data such as remote sensing to disease progression is a matter for epidemiologists to work out. However, tracking of disease incidence and prevalence can be done through USGS’s risk mapping. It is often difficult to strictly map incidence of disease due to vast underreporting at the clinical level in the developing world. In other words, it is possible to map the precursor conditions that correlate with disease occurrence. This data is then useful to the WHO, the United Nations, or non-governmental organizations such as Doctors Without Borders.

Google Maps Mashups

Google Maps, one of the many products that Google offers, provides directions, interactive maps, and satellite/aerial imagery of the United States. One can also search by keyword, such as type of business, within the map. A mashup [http://en.wikipedia.org/wiki/Mashup_(web_application_hybrid)] is a website or application that combines content from more than one source into an integrated experience. Google Maps has released an application programming interface (API) version of their product, allowing users to create mashups of Google Maps.

Jessica Lee, Google Maps Product Manager at the Google headquarters in Mountain View, CA, displayed a variety of mashups that are already in existence. “Google maps by itself is really pushing the limits of what JavaScript can do already,” she remarked. Google Maps relies on AJAX technology, or technology that allows the map to move within the browser itself without refreshing. The satellite imagery is acquired from a multitude of sources. “It takes time to process the satellite data, and the satellites take a long time to orbit” Lee remarked. In any given location, it takes anywhere from six months to five years to update the Google maps imagery.

Figure 6 illustrates EpiSpider (epispider.org), a Google Maps mashup that uses RDF and RSS to aggregate global surveillance information on epidemics and world health problems. EpiSpider.org tracks news feeds and ProMED reports created by the International Society for Infectious Diseases and then visually locates them by mining the article contents. The goal is to alert health researchers around the world of where the hotspots are. Some speculate that the EpiSpider technology may have broad utility for drug safety surveillance as well.

Google Earth
Janies, et al. (2007) created an innovative compilation on Google Earth using genomic data from samples of the avian strain of influenza (H5N1) collected across the globe. Gathering and mapping the host taxa and the H5N1 genomic data allowed for display of this information in Google Earth according to an interactive timescale and geographic format. The data file is available to anyone that downloads Google’s free desktop software, enabling the general public to visually explore the spread and evolution of H5N1 influenza across the globe up until Dec 31, 2005. The data collection also allowed for the identification of the key genotype for successful transmission of H5N1, Lys-627 in polymerase basic protein 2, PB2, as well as geographically distinct influenza characteristics.

WHO

The World Health Organization (WHO) launched a Public Health Mapping and GIS Program in 1993 to collect detailed health information from around the world, compile detailed geographic models, and disseminate this information quickly and easily. The WHO gathers its information largely from remote field data collection tools, especially hand-held GPS and computer units, to record and geo-code data. All information is stored in the Public Health Mapping and GIS program database with political maps, community locations and infrastructure, population data, health conditions, and environmental features. The Public Health Mapping and GIS Program developed HealthMapper, a computer program that accesses the geographic, demographic, and health information at the WHO, to allow users to use the software’s data management and mapping interface to examine data on their own computers without the background and training necessary for more complex commercial programs. The Global Atlas is similar to HealthMapper by allowing anyone with an internet connection to look at similar information, view maps, and create custom maps in real time online using the geo-coded information collected by the WHO relating to infectious disease. These free computer and online available programs broaden the availability of interacting with WHO data to compile detailed information and analysis in a visual format.

Stochastic Modeling

Professor Shripad Tuljapurkar’s Group at Stanford University conducts scientific research on stochastic modeling. Stochastic model is a tool for estimating probability distributions of potential outcomes by allowing for random variation in one or more inputs over time. Figure 9 shows the population pyramid of the U.S. in 2045. To the naked eye it is a simple graph, but there is a lot of data depicted. There is a middle bulge because of the baby boomers, and a second bulge because of the baby boom echo. At the top, the graph is skinny because of the 4 million people who die each year.

Just like digital mapping, stochastic modeling is another way to synthesize large amounts of data into a graphic depiction. Professor Tuljapurkar explained that the key feature of stochastic models is the ability to build uncertainty into the projections. With parasitic disease, there is a high degree of uncertainty. How many of the exposed will get the disease? What percent of people respond to drug treatment? Through stochastic modeling, one can convey information about the death rate, survivorship, life expectancy at birth, or life expectancy at age X.

CDC

The Center for Disease Control and Prevention (CDC) launched BioSense in 2003 to collect, track, and disseminate secure, real-time health care data from hospitals, laboratories, and public health organizations in a variety of formats, including through maps. The program is designed to monitor anomalies in local health status and make an array of current clinical information available to professionals. The overarching goal is to improve disease detection and monitoring. Currently there are over 350 hospitals and major labs reporting to the BioSense network and information is uploaded every 15 minutes for near real-time data availability. The secure network to make the information available to professionals also includes the ability to display collected data in a map and allows users to explore health activity and reports using color gradients across defined geographic areas. Currently the network is not available to the public and past criticism of the program has centered on its effectiveness and circumvention of local public health offices as data collection centers.

MAP

The Malaria Atlas Project (MAP) was launched in May 2006 with the goal of collecting and verifying data for a new, detailed model of P. falciparum and P. vivax malarial infection incidence throughout the globe. It is funded by the Wellcome Trust and is a joint project between the Malaria Public Health & Epidemiology Group, Centre of Geographic Medicine in Kenya, and the Spatial Ecology & Epidemiology Group at Oxford University. As of May 2007, it has 3670 parasite rate surveys from 79 countries and aims to make its complete database available to the public and for research by June 2009. Sources for the project include journal articles, reports, and theses with each data set verified for exact geo-positioning information. Over half of all the studies collected for the project are from the last five to six years, allowing for a sweeping update to malaria data that may be as old as 1960.

Schistosomiasis in China

Terracing and irrigation canals, especially in the Sichuan province, offer a prime location for snail growth in close proximity to humans and an environment conducive to schistosomiasis transmission. The Chinese government has collected schistosomiasis epidemiology data in the past, but only for very low-resolution mapping. Higher resolution or more detailed data collection using GPS units offers a much more accurate understanding of the disease and how to develop the most effective public health initiatives to combat the disease. One of many schistosomiasis studies in China includes Spear, et al (2004). The study examined what kind of agriculture was most likely to encourage schistosomiasis and where the parasite was most prevalent using GPS ditch maps, water-contact questionnaires, and crop type data. Satellite data allowed for analysis of vegetation and crops through a normalized difference vegetation index (NDVI) in an area, but all remote sensing still required field validation since the technology is still considered new by many researchers. As described in an interview with Edmund Seto, a researcher in the study and a lecturer and assistant researcher in the School of Public Health at the University of California at Berkeley, the GIS findings revealed in ways not possible with conventional analysis that schistosomiasis is much more likely to be spread by raw manure in dry agriculture such as with vegetables or tobacco, rather than in rice paddies. Most importantly for public health, Chinese officials welcomed the GIS data and programs are ongoing in villages calculated to be at-risk to encourage longer fermentation of manure through household biogas systems that dramatically reduces parasite load in the manure and in the community.

Cromley E. K., & McLafferty S. L. (2002) GIS and Public Health (pp. 1-20). New York: Guilford Press.

ESRI. “Arc GIS- the complete geographic information system.” ESRI GIS and Mapping Software. http://esri.com/software/arcgis/index.html. 5/20/07.

Koch T. (2005) Cartographies of Disease- Maps, Mapping, and Medicine, (19-23). Redlands, CA: ESRI Press.

Koch T. (2005) Cartographies of Disease- Maps, Mapping, and Medicine, (26-34). Redlands, CA: ESRI Press

Guynup S. (2004) Cholera: Tracking the First Truly Global Disease. National Geographic Channel.

“Research on Rapid Geographical Assessment of Bancroftian Filariasis” James Cook University, Townsville, Australia. 22-25 July 1997.

Gyapong, J.O et al, “The use of spatial analysis in mapping the distribution of bancroftian filariasis in four West African countries.” Annals of Tropical Medicine & Parasitology, Vol. 96, No. 7, 695-705 (2002)

Haynes, John. Program Manager, Aviation Applications/Public Health Applications. National Aeronautics and Space Administration. 4 May 2007.

Venezia, Robert & Haynes, John. “NASA Space Systems Enhance Public Health Science for Society.” http://www.eomonline.com/EOM_Aug05/article.php?Article=feature02

USGS Forum on Health, http://health.usgs.gov/

Ibid.

Christian, Eliot. U.S. Geological Survey former Data Information Systems Manager. Interview 26 April 2007.

Ibid.

Lee, Jessica. Google Product Manager, Mountain View, CA. Interview 2 May 2007.

Ibid.

Epispider: Exploring the Limits. Fifth International Semantic Web Conference, November 2006. Available at: http://blog.epispider.net/

Janies, Daniel, Andrew Hill, Robert Guralnick, Farhat Habib, Eric Waltari, and Ward Wheeler. “Genomic Analysis and Geographic Visualization of the Spread of Avian Influenza (H5N1).” Syst Biol 56 (2007): 321-329.

“WHO’s Public Health Mapping and GIS Programme.” World Health Organization < http://www.who.int/health_mapping/en/> May 2007.

Tuljapurkar, Shripad. Professor of Population Studies and Biological Sciences. Interview 10 May 2007.

“BioSense.” Center for Disease Control and Prevention. < http://www.cdc.gov/biosense/> 12 Oct. 2006.

“Welcome to the MAP Website.” Malaria Atlas Project < http://www.map.ox.ac.uk/> 1 May 2006.

Spear, Robert, et al. “Factors Influencing the Transmission of Schistosoma Japonicum in the Mountains of Sichuan Province of China.” Am J Trop Med Hyg 70 (2004): 48-56.

Seto, Edmund. Telephone interview. 10 May 2007.

 


© 2007 The Board of Trustees of the Leland Stanford Junior University. All rights reserved.