The long term objectives of this proposal are to precisely characterize and to elucidate determinants of the trends and distribution of stroke in the United States during 1900-1990.
Specific aims (all five years of the award) are to (1) determine the geographic distribution of stroke mortality in the United States for sequential periods during 1900-1990; (2) determine the regional variation in the decline of stroke mortality during 1900-1990; (3) determine whether migration between geographic areas of high and low risk influences stroke mortality rates; (4) relate the geographic and secular changes in stroke mortality and morbidity to changes in potential explanatory variables; (5) relate the geographic and secular changes in stroke mortality to changes in total mortality and mortality due to ischemic heart disease, diabetes, and hypertension. Data sets with broad temporal and spatial coverage will be used to address these issues. Initial phases of the study will utilize United States annual mortality data and decennial census data. Average annual age-adjusted, race- and gender-specific rates will be calculated for the period of each revision of the International List of Causes of Death. Mapping by state, and by state-economic areas, will facilitate interpretation and presentation of these large amounts of data. Quantitative analysis will include determination of statistically significant deviations of age-adjusted rates from the national rate, tests for geographic clustering, and centrographic methods. Later phases of the study will involve merging morbidity, mortality, demographic, and environmental exposure data sets from several sources and testing of explanatory hypotheses generated in the earlier phases of this project. Available morbidity data with geographic information will be examined and related to potential explanatory variables. Data sets for these studies will include applicable subsets of the National Health Interview Survey, the National Health and Nutrition Examination Survey, and data on health resources utilization from the National Hospital Discharge Survey, the National Ambulatory Medical Care Survey, National Nursing Home Survey, National Medical Care Utilization and Expenditure Survey, Medicare Data files. Analytic methods for the later phases will include multivariate weighted regression methods which will provide additional information not visually evident from the maps, and will help to refine and narrow the hypotheses suggested by the spatial and temporal patterns of stroke mortality and morbidity.
Showing the most recent 10 out of 32 publications
