GIS in Human Health Studies

Abstract

Databases used in the field of medical geology are generally comprised of geospatial and/or temporal elements. Although these are not requirements for all medical geology research projects, much of the discussion in this chapter will be focused on databases incorporated into geographic information systems (GIS). GIS are computer-based (or manual) methods that allow a user to input, store, retrieve, manipulate, analyze, and output spatial data (Aronoff 1989). There are four major systems of GIS: engineering mapping systems (computer-aided design/computer-assisted mapping; CAD/CAM), geographic base file systems, image processing systems, and generalized thematic mapping systems. Various software packages are available that perform one or more of these systems, and the relative ability to move data back and forth between them can be critical to the needs and success of a particular GIS. Relational databases are the most commonly used types of databases in GIS (Cromley and McLafferty 2002). Relational database management models are convenient for linking formerly disparate databases together in a GIS. The databases to be joined must share one common attribute, usually an identifier such as coded patient number, sample site, or latitude/longitude. Other database management structures, such as hierarchical and network systems, are not as well suited to health GIS applications, although they may be useful for extremely large databases.

The views expressed by the author are his own and do not represent the views of the United States Geological Survey or the United States.

Other Sources

A. Sources of Earth Science/Geospatial Information

FGDC (http://www.fgdc.gov/)

The FGDC coordinates the sharing of geographic data, maps, and online services, that searches metadata held within the NSDI Clearinghouse Network. The National Geospatial Program (http://www.usgs.gov/ngpo/) organizes data through various portals which are listed here:

http://nationalmap.gov – base data, map products and geospatial web services

http://geo.data.gov – geopatial information discovery and access, communities, and partnership marketplace

http://nationalatlas.gov – maps of America for everyone’s use

http://geonames.usgs.gov/domestic – geographic names authority and gazetteer for the nation

http://cegis.usgs.gov – leadership for geospatial information research

Seamless Data Warehouse (http://seamless.usgs.gov/)

The Seamless Data Warehouse is the ideal location to explore and retrieve data. U.S. Geological Survey (USGS) and the Earth Resources Observation and Science (EROS) are committed to providing access to geospatial data through The National Map. An approach is to provide free downloads of national base layers, as well as other geospatial data layers. These layers are divided into framework categories:

Places, Structures, Transportation, Boundaries, Hydrography, Orthoimagery, Land Cover, and Elevation

Along with providing access to the data, this site contains:

Tutorial to help with downloads

Information about the downloadable products

Frequently Asked Questions (FAQs)

Links to product homepages and information pages

The Seamless Data Warehouse is always growing with new data, tools, features, and much more.

For selected online earth science/geospatial journals see Appendix B. For selected biomedical/health information see Appendix B. URLs were valid at the time of original publication (March 2002).

B. Libraries (for Further Research)

U.S. Geological Survey Library

950 National Center 12201 Sunrise Valley Drive

Reston, VA 20192

e-mail: library@usgs.gov

U.S. Library of Congress

101 Independence Avenue,

S. E. Washington, DC 20540

e-mail: lcweb@loc.gov

U.S. National Library of Medicine

National Institutes of Health

8600 Rockville Pike

Bethesda, MD 20894

e-mail: NIHInfo@OD.NIH.GOV

C. Where Does One Start a Search for Relevant Databases?

The FGDC Web site http://www.fgdc.gov provides access to over 38 simultaneously searchable data clearinghouses in the United States and internationally. These include databases related to Earth sciences, geography, landform information, ecosystem health, biological resources, and satellite imagery (see Other Sources, part A). The focus of this chapter has been on electronic sources, but don’t forget to check your library’s reference section, trade journals, or other specialized periodicals and books. To find databases on the Internet, you might use a search engine. Be aware that different search engines work in different ways, and that what may be overlooked by one search engine might be found by another one. There are also metasearch engines that use several search engines simultaneously (Hock 2001). Of course, another very efficient way to find out what databases are used by experts in a given field is to simply ask them. Contact information for university professors is often listed on their institution’s Web site, which can be found on any search engine. Even if the principal investigator is hard to reach in person, his or her postdoctoral fellows, graduate students, and technicians may be willing to help.

An efficient strategy when starting out on medical geology research projects is to seek out relevant database clearinghouses. Using clearinghouses also offers some protection from rapidly changing unique or uniform resource locators (URLs), which are often referred to as Web site addresses. One example is the U.S. Geological Survey (USGS), which is a major clearinghouse for Earth science data. The URL for some particular databases contained therein may change, but the URL for a clearinghouse such as USGS generally remains stable over time. The primary clearinghouse organization will maintain proper internal links and keep access to all of their individual databases current.

The geographic and temporal range of the data needed must be ascertained at the outset of any medical geology GIS. It is better to err on the side of obtaining more information then deleting unnecessary elements, because it can be difficult to add data later if it is decided to examine additional parameters. But the initial cost of the data and the cost in resources to store data must also be taken into account. Because different databases will likely contain data archived in a variety of formats, it is advisable to store the initial downloaded data as is and make copies of it before any subsequent manipulation. The text-only ASCII file format is a “common denominator” useful for merging data from different sources into a single data set.

The use of Internet-derived databases can be made frustrating and difficult by two realities of this medium. One reality is that URLs change quickly, and so the Web site address that worked in the past may not take you to the same page today. This potential pitfall can be avoided by using the “gatekeeper” URL to a database clearinghouse as mentioned above, rather than by using direct URLs to individual databases. For example, one is advised to use a main clearinghouse Web site rather than a more specific URL for some individual database, such as for water table depths in India. The other major problem is the so-called Invisible Web (Sherman and Price 2001). There are a great many databases accessible via the Internet with no easy way to find them or to find out about them. Many databases can only be accessed after registering and entering a password. Search engines will miss these and other relevant sites, and they will often come up with totally irrelevant sites. Investigators must be mindful, too, of the reliability of database sources accessible via the Internet. If associated metadata are not available, that database should not be used.

Once a database of interest has been identified and the legitimacy of the organization that maintains it is verified, you are ready to download. Appendix B of this volume lists a number of earth science/geospatial and biomedical/human health databases as examples. Make sure you have the minimum requirements and sufficient memory space on your computer before proceeding. As always when downloading any software or data to a personal computer, remember to have some tool in place for screening computer viruses. It is critical to ensure that once downloaded, the data were not corrupted in the process. You must examine the source data carefully and confirm that they match the data in the form that has been downloaded. Problems can arise, for example, if the source data are tab delimited and your default download is space delimited. As soon as you have downloaded the source data, you should make a backup copy before doing anything with the data. It is generally convenient to keep such files on a compact disc (CD). Now you are ready to open up your data with your spreadsheet software package and import it to your GIS application or to a statistical analysis package. In a GIS environment, you can easily query the data. That is, by clicking with a mouse on a location visibly displayed on a map, you can extract attributes of that point.

You will need to join data from different databases for use in a GIS project. Get to know the raw data well as you must always maintain quality assurance/quality control (QA/QC). It is easy to mix up or somehow corrupt data when manipulating it. For instance, if you sort the data for some reason, make sure you keep a copy of the original unsorted data. Also be careful not to sort only one field, but rather keep your unique identifiers tied to the data in the proper order. If your ultimate aim is to do some statistical analysis of the data, you should work closely with a statistician right from the start. The statistician will help you determine the appropriate data you need to answer the questions you are asking.