Untitled Document

Case Studies: GIS and the Spatial Epidemiology of Tuberculosis.

"Molecular and Geographic Patterns of Tuberculosis Transmission After
15 Years of Directly Observed Therapy."
Bishai, W. R. et al.
Journal of the American Medical Association
1998, 280(18): 1679-1684

A cooperative study involving the US Departments of International Health, Molecular Microbiology and Immunology, and Epidemiology; the Johns Hopkins University School of Medicine and Hygiene, and the School of Public Health; the Baltimore City Health Department, and the Maryland Department of Health employed GIS as a means of discerning patterns of tuberculosis transmission in Baltimore. Research was aimed at differentiating whether TB was transmitted largely as a means of recent contact between individuals with active disease and non-infected individuals, or whether certain genetic strains of disease are concentrated in particular areas in the city and thus TB becomes active in long-time residents of TB-endemic regions as a function of latent disease.

A concentric-zone method was used to identify individuals most likely to have been exposed to infection through contact with culture-positive persons. Tuberculosis risk factors, compiled from profiles of and interviews conducted with the 182 culture-positive patients, included intravenous and non-injection drug abuse, incidence of unemployment within 2 years prior to diagnosis, and homelessness at any time within one year's time prior to diagnosis.

Positive-culture patients were classified on the basis of genetic (DNA strain) and epidemiological (close contact with a positive-TB patient prior to diagnosis) variables. Three groups were identified: persons sharing a matching genetic and an epidemiological link to another positive case (Group 1), cases with matching genetic fingerprints but no epidemiological link (Group 2), and individuals with unique genetic fingerprints (Group 3). Spatial clustering revealed that 4 out of 26 clusters, with an average of 4.5 individuals per cluster, were strongly geographically aggregated. Furthermore, spatial analysis also revealed that Group 1 clusters showed a much greater degree of geographic aggregation than individuals in Group 2 clusters. Results indicated not only that the majority of patients diagnosed contracted TB through recent transmission, but also that patients belonging to TB clusters resided in geographical proximity to each other relative to non-clustered cases. Communities with high incidences of TB were characterized by high drug abuse and crime rates, as well as a lack of adequate housing and membership in low socioeconomic brackets.

"The Use of a Geographical Information System (GIS) to Evaluate the
Distribution of Tuberculosis in a High Risk Community"
Beyers, N., Gie, R., Zietsman, H., Kunneke, M., Hauman, J., Tatley, M,
and Donald, P.
South African Medical Journal
1996, 86(1): 40-44

With an average 718 new tuberculosis cases per 100,000 population each year, the Western Cape province has the highest TB incidence rate in all of South Africa. As tuberculosis rates were increasing rapidly amongst the colored population, GIS was utlised to identify the geographical distribution of TB cases in Uitsig and Ravensmead, the two suburbs with the highest rates of TB infection in the province. Data was obtained for all individuals identified as culture-positive within a ten-year period from 1985 through 1994. Digital cadastral boundaries were used as a base map, onto which were added socioeconomic, demographic, and clinical data. Property data was overlaid with enumerator sub-district (ESD) boundaries in order to compare incidence rates at the end of the study period with 1991 incidence rates organised by ESD. The use of GIS yielded a graphic display of spatial TB concentrations occurring within specific areas of both suburbs.

"TB or not TB? Increasing door-to-door response for Screening."
Cegielski, J.P, Clark, P.A.,and Hassell, W.
Public Health Nursing
1997, 14(5): 268-271

As Tuberculosis has made a dramatic resurgence in the United States within the last twenty years, identification of high-risk communities is essential in the development and implementation of TB prevention and treatment programs. Dr. Peter Cegielski's preventive treatment strategy developed at the University of Texas Health Center employed GIS as a means of spatially locating high-risk neighborhoods in Tyler, Texas where resources should be allocated for the early detection of active TB cases. High-risk neighborhoods were identified on the basis of geographical clustering of previous tuberculosis incidences. Spatial analysis isolated two communities in central Tyler that were particularly prone to experiencing above average rates of TB incidence. This allowed for the implementation of a community-based screening program providing free skin tests to residents of both high risk-neighborhoods.

Background:
Biology | Current Situation | Case Studies | Glossary