Kuller, L H. (Johns Hopkins Univ. School of Hygiene and Public Health, Baltimore, Md. 21205), A. Bolker, M. S. Saslaw, B. L Paegel, C Sisk, N. Borhani, J. A. Wray, H. Anderson, D. Peterson, W. Winkelstein, Jr., J. Cassel, P. Spiers, A. G. Robinson, H. Curry, A. M. LJIienfefd and R. Seltser. Nationwide cerebrovascular disease mortality study. I. Methods and analysis of death certificates. Amer. J. Epid., 1969, 90; 536–544.—Large differences in cerebrovascular disease (CVD) mortality among the geographic areas of the United States have been reported. In order to determine whether these geographic differences might be due to differences in certification practices or accuracy of the diagnosis of stroke, a study of death certificates in 9 areas of the United States—3 with high, 3 intermediate and 3 with low cerebrovascular disease death rates for white males ages 35–74—was completed. A stratified sample of 6, 314 death certificates was included in the study. The information on the death certificate was then compared with clinical data from hospital records, physician‘s reports and medical examiner’s records. Cerebrovascular disease had been listed as underlying cause of death on 1, 232 death certificates in the original sample and 1, 310 after adjusting for sampling. Stroke was listed as underlying cause for 1, 310 (66.8%) of the 1, 960 certificates listing stroke as either underlying or contributing cause of death. There were no substantial differences among the areas. The age and sex distribution, place of death, type of stroke listed on the certificate and the frequency with which stroke appeared on the same certificate with hypertension, arteriosclerotic heart disease and diabetes were similar in the high, low and intermediate cerebrovascular death rate areas. The variations in death rates could not be explained by differences in certification practices such as choice of the underlying cause of death among all death certificates listing stroke.