There is a small, but growing number of applications of geographic information systems towards health-related issues. This review provides a synthesis and bibliographic resource on emerging applications within this realm (see also Albert, Gesler, and Wittie 1995; Siebert 1994; Barnes and Peck 1994; Davenhall 1993).

Disciplines

Medical geography is "the application of geographical concepts and techniques to health-related problems" (Hunter 1974, 3). Two divisions within the field include the spatial patterning of health services and the spatial patterning of disease (Overhead 1). Geographic information systems integrate spatial data to support decisions (Cowen 1990). Star and Estes (1990) coined the term the "four M's" in reference to the measurement, mapping, monitoring, and modeling capabilities of geographic information systems (Overhead 2). This review encompasses the intersection between medical geography (MG) and geographic information systems (GIS).

Medical Geography and GIS

The literature encompassing the intersection between MG and GIS represents four distinct groups of articles - potential, cautionary, preliminary, and application (Albert, Gesler, and Wittie 1995) (Overhead 3). Connecting these orientations is a dialogue, an academic and professional exchange through workshops, conferences, and journals that facilitate communications between these groups.

The potential group (Verhasselt 1993; Gould 1992; Fisherden 1991; Nicol 1991; Scholten et al. 1991; Twigg 1990; Wrigley 1991; Mohan and Maguire 1985) suggests the benefits of geographic information systems for hypothesis generation, crossing geopolitical boundaries, and small area analysis. Quoting Verhasselt (1993, 121), GIS "open up new possibilities in ecological associative analysis."

The cautionary group argues for a critical approach to using GIS (Matthews 1990) for the following reasons. First, GIS suffers from definitional vagueness; there is no definition having universal acceptance. Second, techniques often overshadow research questions. Third, data suffer from the four (in' s): intensive, inaccessible, incompatible, and inaccurate data (Taylor 1991; Heywood 1990; Twigg 1990).

The preliminary group suggests potential disease/health applications of geographic information systems. One suggestion involves assessing danger to lead poisoning through neighborhood risk scores attributable to socioeconomic status (0 = high; 3 = low), soil contamination (0 = clean; 3 maximum), and air contamination (0 = no effect; 3 maximum). Exposure rates (number of cases /number of blocks) are found for each risk score (Wartenberg 1992). The second suggestion (Stallones et al. 1992) involves producing a composite map of water wells and associated residences using a contaminated aquifer. The third suggestion is to use geographic information systems to analyze patterns of stomach and esophageal cancer for ecological association because of the enormous environmental data requirements (Lam 1986).

The application group uses geographic information systems to examine spatial patterns of health care services and disease. These have been set apart into the following subdivisions: distributions; location/allocation; catchment areas; monitoring and prevention; surveillance, modeling, and simulations; and cluster analysis (Overhead 4).

Albert (1995) and Albert and Gesler (1997) provide analyses of physicians' primary, secondary, and tertiary medical practice locations using a geographic information system (MapInfo®). The physician database (18,253 records with 22 fields) and attribute data on settlement size and county-level designations on metropolitan and health professional shortage status were input within GIS. Some findings include: secondary and tertiary medical locations have lower population thresholds than primary medical locations; and, higher-order physicians (more specialized) travel greater distances between locations than do lower-order (less specialized) physicians.

Van Creveld (1991) describes an ambulance service application. The "Command and Control Program" presents a status report on the nearest 24 rescue vehicles to a monitor. Planned upgrades include a radio beacon system to dispatch and track the optimum vehicle to the accident location. Another example of an emergency response application is the use of a geographic information system to conduct search and rescue activities during the collapse of a section the Nimitz Freeway in October 1987 (Tyler 1990).

Zwarenstein et al. (1991) demonstrate the use of Thiessen polygons to assess the effect of removing race restrictions on hospitals in Natal/KwaZulu, South Africa in 1985. The results indicate that the population/bed ratios for blacks were still high even after removing race restrictions on hospital admissions.

Other applications include a geographic information system to monitor measles epidemics (Solarsh and Dummann 1992), HIV/AIDS (Fost 1990), and access to health care resources for the homeless (Lee 1996).

Some applications present modeling and simulations of health hazards. For example, isopleths were drawn around an industrial plant producing toxic emissions to indicate areas having an excess risk for cancer (Moore 1991). Another project found geographic information systems useful in comparing observed versus expected patterns of lead exposure among children (Guthe et al. 1992).

Several projects examine the incidence of cancer using geographic information systems (Tobias et al. 1996). Fitspatrick-Lins et al. (1990) found that potential radon exposure is highest in areas with metamorphic rocks and that sandstone and shale have a mediating influence on potential radon levels. Oppenshaw et al. (1987, 1988) found age-sex adjusted incidence rates for various cancers corresponding to circles drawn around points of a superimposed grid. Those circles with significant incidence rates were drawn to indicate the location of major cancer clusters.

Classroom Linkage

The instructor can use some of the research articles cited in this review to illustrate certain concepts and functions of geographic information systems. For example the following GIS capabilities can be referenced with specific research (Figure 1): database attribute queries and report generation (Tyler 1990), database queries (Fost 1990), overlay operations (Fitspatrick-Lins et al. 1990; Guthe et al. 1992), exploratory data analysis (Fitspatrick-Lins et al. 1990), Thiessen polygons (Twigg 1990; Zwarenstein et al. 1991), isopleths (Moore 1991), radii searches and cluster significance (Openshaw et al. 1987), network routing (Van Creveld 1991), scale (Solarsh and Dummann 1992), and modeling (Moore 1991; Guthe et al. 1992). Further, since most GIS courses have a term project of some sort, this review provides students with ideas and direction that can be used to develop a project more in line with improving on previous research or identifying new avenues of research.

Conclusions

This review represents a resource and reference guide to current health and disease applications of geographic information systems. Similar reviews exist documenting applications of geographic information systems for other geographic subdisciplines and other disciplines.

Acknowledgment

National Council for Geographic Education for permission to adapt "Geographic Information Systems and Health: An Educational Resource (Albert et al. 1995).

References

Albert, D. 1995. Is there a doctor near the house? MapInfo analyzes health care access in North Carolina. GlobalNews (Summer): 7.

Albert, D. and W. Gesler. 1997. Multiple locations of medical practice in North Carolina: Findings and health care policy implications. Carolina Health Services and Policy Review, 4: 55-75.

Albert, D., W. Gesler, and P. Wittie. 1995. Geographic information systems and health: An educational resource. Journal of Geography 94:350-356.

Barnes, S. and A. Peck. 1994. Mapping the future of health care. Geo Info Systems (April): 31-39.

Cowen, D. 1990. GIS versus CAD versus DBMS: What are the differences? In Introductory Readings in Geographic Information Systems, ed. D. Peuquet and D. F. Marble, 52-61. London: Taylor & Francis.

Davenhall, W. 1993. Health care reform puzzle: 1,000 not-so-easy pieces. GIS World (November/December): 36-38.

Fishenden, J. 1991. Towards more healthy living. In Geographic Information 1991: The Yearbook of the Association of Geographic Information, ed. J. Cadoux-Hudson and I. Heywood, 126-127. London: Taylor & Francis.

Fitspatrick-Lins, K., T. Johnson, and J. Otton. 1990. Radon potential defined by exploratory data analysis and geographic information systems. Selected Papers in the Applied Computer Sciences 1990. U. S. Geological Survey Bulletin 1908: E1-E10.

Fost, D. 1990. Main St. marketing: Using maps to tackle AIDS. American Demographics 12(4):22.

Gould, M. 1992. The use of GIS and CAC by health authorities: Results from a postal questionnaire. Area 24(4):391-401.

Guthe, W., R. Tucker, E. Murphy, R. England, E. Stevenson, and J. Luckhardt. 1992. Reassessment of lead exposure in New Jersey using GIS technology. Environmental Research 59:318-325.

Heywood, I. 1990. Geographic information systems in the social sciences. Environment and Planning A 22:849-854.

Hunter, J. 1974. The challenge of medical geography. In The Geography of Health and Disease, ed. J. Hunter. Chapel Hill: Department of Geography, University of North Carolina.

Lam, N. 1986. Geographical patterns of cancer mortality in China. Social Science and Medicine 23 (3):241-247.

Lee, H. 1996. Health care for San Francisco's homeless. Geo Info Systems (June): 46-47.

Matthews, S. 1990. Epidemiology using a GIS: The need for caution. Computers, Environment and Urban Systems 17(14):213-221.

Mohan, J. and D. Maguire. 1985. Harnessing a breakthrough to meet the needs of health care. Health and Social Service Journal, (May 9): 580-581.

Moore, T. 1991. Application of GIS technology to air toxics risk assessment: meeting the demands of the California air toxics "Hot Spots" Act of 1987. GIS/LIS '91 Proceedings 2:694-714.

Nicol, J. 1991 Geographical information systems within the National Health Service: the scope for implementation. Planning Outlook 34(1):37-42.

Openshaw, S., M. Charlton, C. Wymer, and A. Craft. 1987. A mark I geographical analysis machine for the automated analysis of point data sets. International Journal of Geographical Information Systems 1(4):335-358.

Openshaw, S., A. Craft, M. Charlton, and J. Birch. 1988. Investigation of leukaemia clusters by use of a geographical analysis machine. The Lancet, (February) 6: 272-273.

Scholten, H., and J. de Lepper Marion. 1991. The benefits of the application of geographical information systems in public and environmental health. World Health Statistics Quarterly 44: 160-170.

Siebert, D. 1994. The geography of healthcare reform. GIS World (March/April): 36.

Stallones, L., J. Nuckols, and B. Berry. 1992. Surveillance around hazardous waste sites: Geographic information systems and reproductive outcomes. Environmental Research 59:81-92.

Star, J. and J. Estes. 1990. Geographic Information Systems: An Introduction. Englewood Cliffs, New Jersey: Prentice Hall.

Solarsh, G. and D. Dammann. 1992. A community paediatric information system ñ a tool for measles surveillance in a fragmented health ward. South Africa Medical Journal 82:114-118.

Taylor, D. 1991. Geographic Information Systems: The Microcomputer and Modern Cartography. Oxford, UK: Pergamon Press.

Tobias, R., R. Roy, C. Alo, and H. Howe. 1996. Tracking human health statistics in "Radium City" Geo Info Systems (July): 50-53.

Twigg, L. 1990. Health based geographical information systems: Their potential examined in the light of existing data sources. Social Science and Medicine 30(1):143-155.

Tyler, S. 1990. Computer assistance for California earthquake rescue effort. The Police Chief 57(3):42-43.

Van Creveld, I. 1991. Geographic information systems for ambulance services. In Geographic Information 1991: The Yearbook of the Association of Geographic Information, ed. J. Cadoux-Hudson and I. Heywood, 128-130. London: Taylor & Francis.

Verhasselt, Y. 1993. Geography of health: Some trends and perspectives. Social Science and Medicine 36(2): 119-123.

Wartenberg, D. 1992. Screening for lead exposure using a geographic information system. Environmental Research 59:310-317.

Wrigley, N. 1991. Market-based systems of health-care provision, the NHS Bill, and geographical information systems. Environment and Planning A, 23 (1): 5-8.

Zwarenstein, M., D. Krige, and B. Wolff. 1991. The use of a geographical information system for hospital catchment area research in Natal/KwaZulu. South African Medical Journal 80: 497-500.

Figure 1. Applications of Geographic Information Systems to Health Research
Author(s)	Application	Geography	Variables	Concept Illustrated	Software
Tyler 1990	Search and rescue	Micro scale San Francisco and Alameda County	Vehicle registration Power lines	Database query reports	Unspecified
Van Creveld 1991	Automated emergency response	West Midlands 345 square miles	Real-time tracking of 24 rescue vehicles	Network routing	West Midland's Command and Control Program
Fost 1990	AIDS surveillance	San Francisco	Street network Questionaire data	Database queries	MapInfo
Twigg 1990	Assessment of GPs and clinics relative to school catchment districts	Enumeration district approximation using Thiessen polygons	Households without WC and bath GP surgeries and clinics School districts	Thiessen polygons	ARC/INFO
Zwarenstein et al. 1991	Measuring accessibility	Defining hospital catchment areas using Thiessen polygons	Persons/bed ratios	Thiessen polygons	ARC/INFO
Moore 1991	Degree of public exposure Assess health risks from emissions	Isopleths of cancer risk	Carcinogenic and noncarcinogenic health risks	Isopleths Modeling	PC ARC/INFO
Guthe et al. 1992	Prediction of high levels of lead in blood of children	Identification of high-risk census tracts	Blood screening records Local sources of industrial and hazardous wastes Traffic volume	Overlay operation Modeling	ARC/INFO
Fitspatrick-Lins et al. 1990	Regional variations in radon potential	Within county	Underlying geology soils Indoor radon data	Overlay operation Exploratory data analysis	ARC/INFO
Openshaw et al. 1987	Significance of cancer clusters	Point locations	childhood leukemia incidence	Neighborhood operation- search using radii	GAM
Solarsh and Dummann 1992	Measles surveillance	Subdistrict (Edenal Health Ward, southern Natal)	Hospital inpatient data	Scale	dBase III PLUS Epi Info (v. 3 CDC)Harvard Graphics