© 2004 by Society
© 2004 by the Society forNeuro-Oncology
Brain cancer mortality in the United States, 1986 to 1995: A geographicanalysis
Department of Human Genetics, University of CaliforniaLos Angeles School of Medicine, Los Angeles, CA 90095-7088 (Z.F.); Department of Ambulatory Care and Prevention, HarvardMedical School and Harvard Pilgrim Health Care, Boston, MA 02215 (M.K.); Department of Community Medicine and Health Care,University of Connecticut School of Medicine, Farmington, CT 06030-6325(D.I.G.); USA
2 Address correspondence to Martin Kulldorff, Department of Ambulatory Care andPrevention, Harvard Medical School and Harvard Pilgrim Health Care, 133Brookline Avenue, 6th Floor, Boston, MA 02215, USA(Martin_Kulldorff{at}hms.harvard.edu).
 |
Abstract |
|---|
The Atlas of Cancer Mortality in the United States, 1950-94(Devesa et al.) published in 1999 by the National Institutes of Healthsuggests that there are elevated rates of brain and other nervous systemcancer in the northwestern, north central, and southeastern parts of thecountry. Being descriptive in nature, the atlas does not evaluate whetherobserved patterns are simply due to random variation or if they are reflectiveof true geographical differences in disease risk or treatment practices. Toformally test for geographical clustering of disease, we analyzed U.S. braincancer mortality data from 1986 to 1995 with Tango's Excess Events test, theCuzick-Edwards k-Nearest-Neighbors test, and the spatial scan statistic. Alltests revealed statistically significant geographical clustering for bothadult men and women. The spatial scan statistic indicated that the most likelycluster of high mortality was in parts of Arkansas, Mississippi, and Oklahoma(relative risk [RR] = 1.22, P < 0.0001) for women and in parts ofTennessee and Kentucky (RR = 1.15, P < 0.0001) for men. Severalsecondary clusters were detected, but there were no statistically significantclusters of a very localized nature and a high RR. For childhood brain cancer,there were no statistically significant geographical clusters. It isreassuring that no local brain cancer mortality "hot spots" withvery high RRs were found. While the causes of the large geographical clusterswith modest RRs are unclear, the geographical pattern of brain cancermortality provides valuable information that can help in formulatingetiological hypotheses and in targeting high-risk populations for furtherepidemiological and health services research.
Received August 6, 2003; Accepted January 2, 2004