UNLABELLED: PROBLEM/CONDITIONS: Cardiovascular disease (CVD), including coronary heart disease (CHD) and stroke, is the leading cause of death in the United States, and state rates of CVD vary by state and by region of the country. Several behavioral risk factors (i.e., overweight, physical inactivity, smoking, hypertension, and diabetes mellitus) and preventive practices (i.e., weight loss and smoking cessation) are associated with the development of CVD and also vary geographically. This summary displays and analyzes geographic variation in the prevalences of selected CVD risk factors. REPORTING PERIOD: 1994 (1992 for prevalence of hypertension). DESCRIPTION OF SYSTEM: The Behavioral Risk Factor Surveillance System (BRFSS) is a state-based random-digit-dialing telephone survey of noninstitutionalized adults aged > or =18 years; 50 states and the District of Columbia participated in BRFSS in 1994, and 48 states and the District of Columbia participated in 1992. METHODS: Several different analyses were conducted: a) analysis of state risk factor and preventive practice prevalences by sex and race (i.e., black and white); b) mapping; c) cluster analysis; d) correlations of state prevalence rates by sex and race; and e) regression of state risk factor prevalences on state CHD and stroke mortality rates. RESULTS: Mapping the prevalence of selected CVD risk factors and preventive health practices indicates substantial geographic variation for black and white men and women, as confirmed by cluster analysis. Data for blacks are limited by small sample size, especially in western states. Geographic clustering is found for physical inactivity, smoking, and risk factor combinations. Risk factor prevalences are generally lower in the West and higher in the East. White men and white women are more similar in state risk factor rates than other race-sex pairs; white women and black women ranked second in similarity. State prevalences of physical inactivity and hypertension are strongly associated with state mortality rates of CVD. INTERPRETATION: Geographic patterns of risk factor prevalence suggest the presence (or absence) of sociocultural environments that promote (or inhibit) the given risk factor or preventive behavior. Because the risk factors examined in this summary are associated with CVD, further exploration of the reasons underlying observed geographic patterns might be useful. The BRFSS will continue to provide geographic data about cardiovascular health behaviors with a possible emphasis on more data-based small- area analyses and mapping. This will permit states to more adequately monitor trends that affect the burden of CVD in their regions and the United States. Mapping also facilitates the exploration of patterns of morbidity, health-care use, and mortality, as well as the epidemiology of risk factors. Finally, by identifying those segments of the population with high levels of these risk factors and lower levels of the preventive health practices, public health personnel can better allocate resources and target intervention efforts for the prevention of CVD.