The construction of disease maps has been a central part of descriptive epidemiology throughout its history. The map is a useful tool for disease surveillance but is a tool that is often used incorrectly.

There are many ways to lie with maps. At their best, maps are useful when examined by people who are knowledgeable about the characteristics of local populations, the local environment, and the factors that contribute to ill-health in such environments.

Maps rarely definitively answer questions; but they often help to generate interesting questions and guide the search for causes of ill-health in specific populations by assisting in the generation of hypotheses about the causes of diseases in specific populations. They are useful for monitoring changes in the health of specific populations and to monitor the effects of efforts to improve health in specific areas.

Unfortunately, maps that show disease rates for small areas are often unreliable because of the small numbers of people with any specific disease in a small area. This problem is illustrated and discussed in Lecture #3 Empirical Bayes Estimation, in the Lecture Hall section of this Web Project.