ABSTRACT
Many diseases have risk factors that are distributed unevenly in the environment. Examples for cancers include bladder cancer and arsenic (Silverman et al., 2006), leukemia and benzene (Linet et al., 2006), and lung cancer and radon (Spitz et al., 2006). It is reasonable, therefore, to expect spatial pattern in disease risk, which may be explained by the uneven distribution of risk factors that are known or unknown. When risk factors are unknown, studying spatial-temporal patterns in disease events may reveal clues about disease etiology. There is a long history of research analyzing geographic patterns in cancer incidence and mortality with the objective of discovering environmental determinants of disease (Devesa, Grauman and Fraumeni, 1998; Devesa et al., 1999; Fraumeni and Blot, 1977). Examples of risk factors revealed by analytic epidemiologic studies that followed upon observations of geographic patterns of cancer include chronic use of snuff as a risk factor for oral and pharyngeal cancer among women in the southern United States (Winn et al., 1981) and exposure to asbestos from shipyards as a risk factor for lung cancer among men along the southeastern United States seaboard (Blot et al., 1979). Other examples of investigating spatial patterns in cancer risk include studies on childhood leukemia (Alexander, 1993; Bithell and Vincent, 2000; Wheeler, 2007) and bladder cancer (Jacquez et al., 2006).
