State-Specific Prevalence of Selected Health Behaviors, by Race and Ethnicity -- Behavioral Risk Factor Surveillance System, 1997

Julie C. Bolen, Ph.D.
Luann Rhodes, M.P.A., M.P.H.
Eve E. Powell-Griner, Ph.D.
Shayne D. Bland, M.Sc.
Deborah Holtzman, Ph.D.
Division of Adult and Community Health
National Center for Chronic Disease Prevention and Health Promotion

Problem/Condition: In the United States, disparities in risks for chronic disease (e.g., diabetes, cardiovascular disease, and cancer) and injury exist among racial and ethnic groups. This report summarizes findings from the 1997 Behavioral Risk Factor Surveillance System (BRFSS) of the distribution of access to health care, health-status indicators, health-risk behaviors, and use of clinical preventive services across five racial and ethnic groups (i.e., whites, blacks, Hispanics, American Indians or Alaska Natives, and Asians or Pacific Islanders) and by state.

Reporting Period Covered: 1997.

Description of System: The BRFSS is a state-based telephone survey of the civilian, noninstitutionalized, adult (i.e., persons aged greater than or equal to 18 years) population. In 1997, all 50 states, the District of Columbia, and Puerto Rico participated in the BRFSS.

Results: Variations in risk for chronic disease and injury among racial and ethnic groups exist both within states and across states. For example, in Arizona, 11.0% of whites, 26.2% of Hispanics, and 50.5% of American Indians or Alaska Natives reported having no health insurance. Across states, the median percentage of adults who reported not having this insurance ranged from 10.8% for whites to 24.5% for American Indians or Alaska Natives. Other findings are as follows. Blacks, Hispanics, American Indians or Alaska Natives, and Asians or Pacific Islanders were more likely than whites to report poor access to health care (i.e., no health-care coverage and cost as a barrier to obtaining health care). Blacks, Hispanics, and American Indians or Alaska Natives were more likely than whites and Asians or Pacific Islanders to report fair or poor health status, obesity, diabetes, and no leisure-time physical activity. Blacks were substantially more likely than other racial or ethnic groups to report high blood pressure. Among all groups, American Indians or Alaska Natives were the most likely to report cigarette smoking. Except for Asians or Pacific Islanders, the median percentage of adults who reported not always wearing a safety belt while driving or riding in a car was greater than or equal to 30%. The Papanicolaou test was the most commonly reported screening measure: greater than or equal to 81% of white, black, and Hispanic women with an intact uterine cervix reported having had one in the past 3 years. Among white, black, and Hispanic women aged greater than or equal to 50 years, greater than or equal to 63% reported having had a mammogram in the past 2 years. Approximately two thirds of white, black, and Hispanic women aged greater than or equal to 50 years reported having had both a mammogram and a clinical breast examination in the past 2 years; this behavior was least common among Hispanics and most common among blacks. Screening for colorectal cancer was low among whites, blacks, and Hispanics aged greater than or equal to 50 years: in each racial or ethnic group, less than or equal to 20% reported having used a home-kit blood stool test in the past year, and less than or equal to 30% reported having had a sigmoidoscopy within the last 5 years.

Interpretation: Differences in median percentages between racial and ethnic groups, as well as between states within each racial and ethnic group, are likely mediated by various factors. According to published literature, socioeconomic factors (e.g., age distribution, educational attainment, employment status, and poverty), lifestyle behaviors (e.g., lack of physical activity, alcohol intake, and cigarette smoking), aspects of the social environment (e.g., educational and economic opportunities, neighborhood and work conditions, and state and local laws enacted to discourage high-risk behaviors), and factors affecting the health-care system (e.g., access to health care, and cost and availability of screening for diseases and health-risk factors) may be associated with these differences.

Action Taken: States will continue to use the BRFSS to collect information about health-risk behaviors among various racial and ethnic groups. Analysis of these data will enable states to monitor factors that may affect the rate of chronic disease- and injury-related morbidity and mortality and to develop public health programs and policies to address these problems.

Behaviors (e.g., physical inactivity, excessive alcohol consumption, and cigarette smoking) can contribute to chronic disease- and injury-related morbidity and mortality in the United States (1). Preventive health practices (e.g., cholesterol screening, mammography, the blood stool test, and sigmoidoscopy) can help identify early stages of chronic disease (e.g., heart disease, breast cancer, and colorectal cancer), thereby reducing the morbidity and mortality rates from these leading causes of death among the U.S. adult (i.e., persons aged greater than or equal to 18 years) population (2). Increasing the use of screening for chronic disease and reducing high-risk behaviors are among the year 2000 national health objectives (3).

In the United States, disparities in risk for chronic disease and injury exist between non-Hispanic whites and persons belonging to other racial or ethnic groups (4). For example, in 1996 blacks had the highest death rates for seven causes of deaths, and American Indians or Alaska Natives had the highest death rates for two of the causes (Table 1). Conversely, Asians or Pacific Islanders had the lowest death rates for six of the causes of death, and Hispanics had the lowest death rates for four causes. In 1970, persons in racial and ethnic minority groups accounted for 16% of the population; by 1998, this proportion had increased to 27%. The proportion is expected to continue increasing, to nearly 50% by 2050 (5).

Although persons in racial and ethnic minority groups are accounting for increasingly larger proportions of the U.S. population, information about the health behaviors of persons in minority groups is insufficient, especially at the state and local levels (6). Ongoing state-specific information is important in identifying or monitoring the prevalence of access to health care, health-status indicators, health-risk behaviors, and use of clinical preventive services among various racial and ethnic groups. These data can be used to identify disparities among racial and ethnic groups and to plan, implement, and evaluate culturally appropriate prevention programs at the state and local levels.

State-specific data about modifiable risk factors for chronic diseases and other leading causes of death among adults are collected each year through the Behavioral Risk Factor Surveillance System (BRFSS), an ongoing surveillance system. The BRFSS is administered by state health departments each month in all 50 states, the District of Columbia, and Puerto Rico.* This system can be used to measure achievement toward national health objectives (3) and specific state objectives. The history, rationale and use of the BRFSS have been previously described (7-10). In this report, 1997 state-specific BRFSS estimates of access to health care, health-status indicators, health-risk behaviors, and use of clinical preventive services are presented for five racial and ethnic groups.

Each state health department used random digit dialing to select samples of adults in households with telephone. The samples represent each state's civilian, noninstitutionalized, adult population (7). States used the three-stage cluster sampling based on the Waksberg method (11) or simple random sampling, stratified random sampling, or other sampling designs.

The BRFSS includes a core set of questions asked in all participating states each year and a rotating set of core questions asked every other year. Rotating core questions asked in 1997 were about awareness of high blood pressure and high blood cholesterol, alcohol use, safety belt use, testing for cholesterol, and colorectal cancer screening. Because the rotating core question regarding leisure-time physical activity was not asked in 1997, the 1996 data for this health behavior are used in this report.

If a substantial portion of a state's population does not speak English proficiently, the state can use a Spanish version of the core questionnaire provided by CDC. If the interviewer determines that the respondent is not proficient in the available languages, the interviewer does not administer the survey and notes language barrier as the reason for ending the telephone call.

In each state, during the interview period in each month, BRFSS data are collected from randomly selected adults. The data are sent to CDC after the monthly interviewing cycle ends. CDC edits and checks the reliability of the data collected.

A computer-assisted telephone interviewing (CATI) system, which permits direct entry of data into a computer file during an interview, was used in 49 states in 1997. CATI helps reduce errors in data collection by facilitating data coding and entry and by enabling monitoring of the interviewers.

CDC aggregates the edited monthly data files to create a yearly sample for each state. Each state's yearly data file is weighted to the respondent's probability of selection and to the age- and sex-specific or race-, age-, sex-specific population from the most current census data (or intercensal estimates) for each state (9,12). CDC uses these weighted data to estimate the prevalence of risk factors for each state's population. SUDAAN, a software package for analyzing complex survey data, is used to calculate the standard errors and the 95% confidence intervals (CI) around the prevalence estimates (13).

In 1997, the number of interviews completed in each state ranged from 1,505 to 4,923 (median: 2,340). Response rates were calculated according to the methodology of the Council of American Survey Research Organization (CASRO) (14) and ranged from 36.7% to 88.9% (median: 62.1%).

Each BRFSS respondent reports his or her race and ethnicity. These self-reports were the basis of the five categories used in this report: white, black, Hispanic, American Indian or Alaska Native, and Asian or Pacific Islander. Any respondent reporting Hispanic ethnicity was categorized as Hispanic, regardless of race. Survey respondents who reported they were of another race and not Hispanic (0.6% of respondents) and those who replied "don't know" or refused to answer the questions on both race and ethnicity (less than 0.5% of respondents) were excluded from this analysis. In this report, state-level estimates for any race- and ethnicity-specific group that had less than 50 respondents are considered unstable and are not shown. No statistical testing was performed for comparisons presented in this report; therefore, these findings should be considered descriptive.

In the 1997 BRFSS, whites comprised 75.4% of the respondent group (Table 2). Blacks represented 9.7% of the cohort, Hispanics represented 11.1%, American Indians or Alaska Natives accounted for 1.0%, and Asians or Pacific Islanders accounted for 2.8%. The number of respondents was sufficient (i.e., greater than or equal to 50) in 51 states for whites, 35 states for blacks, 36 states for Hispanics, 11 states for American Indians or Alaska Natives, and 10 states for Asians or Pacific Islanders. Survey questions that were restricted to adults of a certain age (e.g., colorectal cancer screening among persons aged greater than or equal to 50 years) or sex (e.g., breast cancer screening among women) further reduced the number of states that had a sufficient number of respondents for analysis.

The factors related to access to health care, health-status indicators, and health-risk behaviors (described in Tables 3-16) are those that place adults at high risk for chronic disease and injury. In contrast, use of clinical preventive services (described in Tables 17-23) reduces the risk of chronic disease, because screening allows early detection and treatment.

The median percentage of adults who reported having less than a high school education was 9.8% for whites (range: 0.7%-22.2%), 17.2% for blacks (range: 2.8%-31.0%), 20.9% for Hispanics (range: 11.3%-47.4%), 17.0% for American Indians or Alaska Natives (range: 3.4%-32.0%), and 9.5% for Asians or Pacific Islanders (range: 3.5%-23.9%) (Table 3).

The median percentage of adults who reported that they did not have any kind of health-care coverage, including prepaid plans (e.g., health maintenance organizations) and government plans (e.g., Medicare), varied considerably between states and between racial and ethnic groups (Table 4). Hispanics and American Indians or Alaska Natives were more than twice as likely as whites to lack health-care coverage. The median percentage of respondents who were uninsured was 10.8% for whites (range: 6.4%-18.4%), 16.4% for blacks (range: 5.8%-34.9%), 22.6% for Hispanics (range: 1.3%-44.9%), 24.5% for American Indians or Alaska Natives (range: 12.5%-50.5%), and 14.3% for Asians or Pacific Islanders (range: 4.8%-31.1%).

The median percentage of adults who reported they needed to see a doctor in the last 12 months but could not because of the cost, irrespective of insurance status, ranged from 9.4% to 16.2% (Table 5). Whites were the least likely racial or ethnic group to report that cost was a barrier to obtaining health care. The median percentage was 9.4% for whites (range: 5.4%-24.3%), 13.2% for blacks (range: 6.6%-27.7%), 16.2% for Hispanics (range: 7.9%-30.1%), 12.6% for American Indians or Alaska Natives (range: 9.2%-26.7%), and 11.6% for Asians or Pacific Islanders (range: 4.7%-16.3%).

Blacks were the least likely racial or ethnic group to report having had no routine physical examination within the past 2 years (Table 6). The median percentage of BRFSS participants who reported not having had a recent, routine physical examination was 18.0% for whites (range: 11.6%-28.4%), 8.7% for blacks (range: 4.7%-19.3%), 18.2% for Hispanics (range: 7.7%-33.9%), 14.5% for American Indians or Alaska Natives (range: 8.8%-30.0%), and 17.1% for Asians or Pacific Islanders (range: 5.9%-25.2%).

The race- and ethnicity-specific median percentage of adults who reported fair or poor health status ranged from 8.8% to 19.4% (Table 7). The median percentage was 11.6% for whites (range: 3.4%-21.7%), 19.4% for blacks (range: 10.3%-30.5%), 16.1% for Hispanics (range: 2.1%-35.2%), 17.5% for American Indians or Alaska Natives (range: 10.1%-35.4%), and 8.8% for Asians or Pacific Islanders (range: 5.6%-17.2%).

A U.S. man or woman having a body mass index of greater than or equal to 30 kg/m2 is considered obese (15). This criterion conforms to World Health Organization guidelines (16). In the 1997 BRFSS, respondents were asked their weight and height, and body mass index was calculated. More than one-fourth of blacks and American Indians or Alaska Natives were determined to be obese (Table 8). The median percentage of respondents who were obese was 15.6% for whites (range: 5.8%-21.2%), 26.4% for blacks (range: 7.6%-33.2%), 18.2% for Hispanics (range: 8.1%-34.7%), 30.1% for American Indians or Alaska Natives (range: 13.0%-32.3%), and 4.8% for Asians or Pacific Islanders (range: 1.7%-16.9%).

High Blood Pressure. The median percentage of adults who reported having ever been told by a health professional that their blood pressure was high was 23.0% for whites (range: 11.0%-29.6%), 30.9% for blacks (range: 21.9%-45.4%), 18.6% for Hispanics (range: 11.4%-28.4%), 20.7% for American Indians or Alaska Natives (range: 16.6%-30.7%), and 16.3% for Asians or Pacific Islanders (range: 4.5%-27.9%) (Table 9).

Diabetes. The median percentage of adults who reported having ever been told by a health professional that they had diabetes was less than 8% in each racial and ethnic group (Table 10). The median percentage was 4.4% for whites (range: 0.7%-6.2%), 7.6% for blacks (range: 2.6%-17.9%), 5.5% for Hispanics (range: 0.4%-14.4%), 7.6% for American Indians or Alaska Natives (range: 3.3%-14.0%), and 4.6% for Asians or Pacific Islanders (range: 0.0%-8.2%).

High Blood Cholesterol. The racial- and ethnic-specific median percentages of adults who had had their blood cholesterol checked and reported having ever been told by a health professional that they had high blood cholesterol were very similar, from 25.6% for Hispanics to 29.7% for whites (Table 11). The median percentage was 29.7% for whites (range: 14.3%-35.2%), 26.0% for blacks (range: 7.6%-36.7%), 25.6% for Hispanics (range: 12.5%-41.0%), and 27.3% for Asians or Pacific Islanders (range: 16.3%-35.8%). In the states that had a sufficient number of American Indian or Alaska Native respondents for this question, the percentage was 26.0% (95% CI: 17.7%-34.2%) for Alaska, 28.6% (95% CI: 14.2%-42.9%) for Oklahoma, and 26.5% (95% CI: 9.3%-43.8%) for Washington.

The BRFSS rotating core question on leisure-time physical activity was asked in 1996 but not in 1997. Leisure-time physical activity is exercise, recreation, or physical activities (e.g., running, calisthenics, golfing, gardening, or walking) that are not performed as part of regular job duties. The median percentage of persons who reported no leisure-time physical activity was greater than or equal to 25% in each racial and ethnic group (Table 12). For each group, the percentage varied by greater than or equal to 17 percentage points across states. The percentage ranged from 13.0% to 50.6% for whites (median: 25.1%), from 17.8% to 54.6% for blacks (median: 38.2%), from 15.0% to 51.9% for Hispanics (median: 34.2%), from 26.7% to 43.2% for American Indians or Alaska Natives (median: 37.2%), and from 21.5% to 40.3% for Asians or Pacific Islanders (median: 28.9%).

Current Drinking. For each racial and ethnic group, the median percentage of adults who reported current drinking (i.e., consumption of greater than or equal to 1 drink of beer, wine, wine cooler, or liquor in the past month) varied across states by greater than or equal to 33 percentage points (Table 13). The range was 27.0%-71.3% for whites (median: 55.4%), 25.1%-73.2% for blacks (median: 40.4%), 26.5%-70.0% for Hispanics (median: 50.8%), 30.8%-64.2% for American Indians or Alaska Natives (median: 50.5%), and 10.2%-58.5% for Asians or Pacific Islanders (median: 38.2%).

Binge Drinking. Asians or Pacific Islanders were the group least likely to report binge drinking (i.e., consumption of greater than or equal to 5 drinks on at least one occasion in the past month); American Indians or Alaska Natives were the most likely to report this behavior (Table 14). The median percentage of adults who reported binge drinking was 14.3% for whites (range: 7.2%-23.6%), 8.7% for blacks (range: 3.0%-23.6%), 16.2% for Hispanics (range: 4.3%-28.0%), 18.9% for American Indians or Alaska Natives (range: 11.4%-30.2%), and 6.7% for Asians or Pacific Islanders (range: 0.0%-31.9%).

The median percentage of adults who reported having ever smoked greater than or equal to 100 cigarettes and currently smoked varied almost fourfold across the racial and ethnic groups (Table 15). The median percentage was 23.6% for whites (range: 13.3%-30.8%), 22.8% for blacks (range: 14.8%-37.0%), 23.1% for Hispanics (range: 10.5%-43.6%), 41.3% for American Indians or Alaska Natives (range: 3.1%-48.6%), and 10.7% for Asians or Pacific Islanders (range: 4.7%-36.1%).

The median percentage of adults who reported not always wearing a safety belt while driving or riding in a car was greater than or equal to 30% for each racial and ethnic group except Asians or Pacific Islanders (Table 16). The median percentage of adults who reported this behavior was 30.0% for whites (range: 13.0%-60.6%), 37.6% for blacks (range: 11.7%-63.0%), 30.3% for Hispanics (range: 10.9%-57.5%), 40.9% for American Indians or Alaska Natives (range: 20.6%-75.1%), and 18.6% for Asians or Pacific Islanders (range: 9.4%-27.4%).

The median percentage of adults who reported having had their blood cholesterol checked within the last 5 years was greater than or equal to 55% in each racial and ethnic group (Table 17). The median percentage was 71.2% for whites (range: 56.1%-82.8%), 67.4% for blacks (range: 45.3%-81.7%), 59.3% for Hispanics (range: 45.7%-81.7%), 54.7% for American Indians or Alaska Natives (range: 49.9%-75.4%), and 67.8% for Asians or Pacific Islanders (range: 36.7%-74.1%).

The median percentage of white, black, or Hispanic women with an intact uterine cervix who reported having had a Pap test in the past 3 years was greater than or equal to 81% (Table 18). The median percentage was 84.7% for whites (range: 77.7%-93.4%), 91.1% for blacks (range: 83.5%-97.0%), and 80.9% for Hispanics (range: 69.2%-92.9%). In Alaska, the only state that had greater than or equal to 50 Native American or Alaska Native respondents for this question, the percentage of women with an intact uterine cervix who reported having had a recent Pap test was 95.5% (95% CI: 91.1%-100.0%). In the states that had a sufficient number of Asian or Pacific Islander respondents for this question, the percentage was 84.2% for Hawaii (95% CI: 80.5%-88.0%), 75.9% for New York (95% CI: 63.5%-88.3%), and 84.1% for Washington (95% CI: 73.6%-94.5%).

Mammogram. The percentage of women aged greater than or equal to 50 years who reported having had a mammogram in the past 2 years was 73.7% for whites (range: 58.7%-89.7%), 76.1% for blacks (range: 44.3%-85.5%), and 63.5% for Hispanics (range: 59.7%-79.6%) (Table 19). Alaska was the only state that had greater than or equal to 50 Native American or Alaska Native respondents to this question; the percentage who reported having had a recent mammogram was 93.5% (95% CI: 87.7%-99.2%). Hawaii was the only state that had greater than or equal to 50 Asian or Pacific Islander respondents; the percentage was 80.7% (95% CI: 75.2%-86.2%).

Clinical Breast Examination. The median percentage of white, black, or Hispanic women aged greater than or equal to 50 years who reported having had a clinical breast exam in the past 2 years was greater than 75% (Table 20). The median percentage was 77.5% for whites (range: 64.7%-88.4%), 78.2% for blacks (range: 57.6%-90.0%), and 75.5% for Hispanics (range: 58.3%-78.7%). In Alaska, the only state that had greater than or equal to 50 American Indians or Alaska Native respondents for this question, the percentage of women who reported having had a recent clinical breast exam was 93.4% (95% CI: 87.1%-99.7%). In Hawaii, the only state with a sufficient number of Asian or Pacific Islander respondents, the percentage was 77.8% (95% CI: 71.3%-84.3%).

Mammogram plus Clinical Breast Exam. The median percentage of women aged greater than or equal to 50 years who reported having had both a mammogram and a clinical breast exam in the past 2 years was 67.6% for whites (range: 51.3%-81.3%), 67.8% for blacks (range: 35.7%-79.5%), and 57.8% for Hispanics (range: 53.7-72.3) (Table 21). Alaska was the only state that had greater than or equal to 50 American Indian or Alaska Native respondents for this question; 88.5% reported having had both a mammogram and a clinical breast exam in the past 2 years (95% CI: 80.1%-96.8%). Hawaii was the only state that had greater than or equal to 50 Asian or Pacific Islander respondents; 71.2% reported having had both screening procedures recently (95% CI: 64.3%-78.2%).

Home-Kit Blood Stool Test. A blood stool test (also called fecal occult blood test) is a test to determine whether the stool contains blood. The test may be done at home with a special kit. The median percentage of adults aged greater than or equal to 50 years who reported having used a home-kit blood stool test in the past year was 18.2% for whites (range: 9.1%-31.2%), 20.3% for blacks (range: 3.0%-43.3%), and 14.2% for Hispanics (range: 5.8%-28.6%) (Table 22). In Alaska, the only state that had greater than or equal to 50 Native American or Alaska Native respondents, the percentage who reported having recently used a home-kit blood stool test was 12.3% (95% CI: 4.3%-20.4%). Two states had greater than or equal to 50 Asian or Pacific Islander respondents for this question; the percentage was 2.6% in California (95% CI: 0.0%-6.2%) and 23.8% in Hawaii (95% CI: 18.9%-28.7%).

Sigmoidoscopy. Sigmoidoscopy (also called proctoscopy) is a procedure involving a thin tube being inserted into the rectum to view the bowel for signs of cancer and other health problems. The median percentage of adults aged greater than or equal to 50 years who reported having had a sigmoidoscopy within the last 5 years was 30.4% for whites (range: 15.6%-49.1%), 28.2% for blacks (range: 10.9%-38.3%), and 22.4% for Hispanics (range: 18.1%-40.7%) (Table 23). In Alaska, the only state that had greater than or equal to 50 American Indian or Alaska Native respondents for this question, the percentage who reported having had a recent sigmoidoscopy was 27.6% (95% CI: 16.7%-38.5%). Two states had greater than or equal to 50 Asian or Pacific Islander respondents to this question; the percentage was 24.3% in California (95% CI: 11.4%-37.2%) and 40.7% in Hawaii (95% CI: 34.9%-46.5%).

Adults in the United States continue to engage in behaviors that increase their risk of chronic disease and injury (1). Since 1984, BRFSS data have documented substantial state-to-state variation in the prevalence of these behaviors (8,17-24). This report presents 1997 BRFSS data by state and documents disparities in the prevalence of health-status indicators, health-risk behaviors, and receipt of clinical preventive services across five racial and ethnic groups (a summary is provided in Table 24).

The differences in median percentages between racial and ethnic groups, as well as between states within each racial and ethnic group, are likely mediated by various factors. Socioeconomic factors (e.g., age distribution, educational attainment, employment status, and poverty), lifestyle behaviors (e.g., lack of physical activity, alcohol intake, and cigarette smoking), aspects of the social environment (e.g., educational and economic opportunities, neighborhood and work conditions, and state and local laws enacted to discourage high-risk behaviors), and factors affecting the health-care system (e.g., access to health care, and cost and availability of screening for diseases and health-risk factors) are thought to be major reasons for such differences (25-28). Level of education is highly correlated with the prevalence of many health risks (e.g., obesity, lack of physical activity, and cigarette smoking) (27,29-31). In this report, the percentage of adults who did not have at least a high school education in 1997 varied more than twofold across the five racial and ethnic groups. Persons in the United States who have more education usually have higher incomes, are more likely to have health-care coverage, and perhaps are better able to access and use the health-care system (4,31). In addition, having health insurance is closely associated with employment status. If a person is unemployed or is employed in a part-time or low-wage job, employee health benefits may not be available or the portion of the cost the employee must bear may be prohibitive (32). Persons in the United States who do not have health insurance are less likely to receive clinical preventive services (33,34). Thus, education, employment, and health insurance all affect the health and health behavior of U.S. adults.

The literature indicates that recent immigration and lack of fluency in English may affect the prevalence of risks for chronic disease and injury among certain racial and ethnic groups. In 1997, 61% of the Asian or Pacific Islander portion and 38% of the Hispanic portion of the U.S. population were foreign born; in contrast, only 8% of the white portion, 6% of the black portion, and 6% of the American Indian or Alaska Native portion were foreign born (5). Although some immigrants are highly educated and have high incomes (5), lack of familiarity with the U.S. public and private health systems, different cultural attitudes about the use of traditional and U.S. conventional medicine (6), and lack of fluency in English may pose barriers to obtaining appropriate health care (35).

Conditions associated with very poor urban areas (e.g., high rates of circulatory diseases, accidents, homicide, crime, infection with the human immunodeficiency virus, and exposure to environmental hazards) can negatively affect health (36). On the other hand, living outside metropolitan areas can also be a risk factor for poor health, because fewer opportunities may exist for health-care and clinical preventive services (37). Racial and ethnic groups tend to distribute differently in urban and rural settings. For example, American Indians or Alaska Natives are the most likely group to live outside metropolitan areas (5). More than one half of Hispanics and blacks and nearly one half of Asians or Pacific Islanders, but only one fourth of whites, live in central cities (5).

Of the five racial and ethnic groups in this report, Hispanics reported the highest median prevalence of three of the four factors associated with limited access to health care (i.e., less than a high school education, cost as a barrier to obtaining health care, and no routine physical examination). Hispanics also reported the second- highest prevalence of no health-care coverage. The latter finding is consistent with observations from the 1997 Current Population Survey (38), the 1998 National Health Interview Survey (4), and the 1996 Medical Expenditure Panel Survey (39) that Hispanics and blacks were the groups most likely to report having no health insurance. These three surveys did not present data for American Indians or Alaska Natives or for Asians or Pacific Islanders. All four national surveys found that whites were the least likely racial and ethnic group to report that they had no health-care coverage. The BRFSS and Medical Expenditure Panel Survey also found that whites were the least likely to report that cost was a barrier to obtaining health care. In the BRFSS, blacks were the least likely to report not having a recent physical examination, and Asians or Pacific Islanders were the least likely to report having less than a high school education.

The National Cholesterol Education Program recommends that all adults in the United States have their total blood cholesterol checked every 5 years (40), but in this report, the median percentage of adults who reported having had this screening procedure within the last 5 years ranged from only 54.7% among American Indians or Alaska Natives to 71.2% among whites. Only seven states had achieved the year 2000 national health objective that greater than or equal to 75% of adults have had their blood cholesterol checked within the last 5 years (3). In this report, a timely Pap test was the most frequently reported clinical preventive service: the median percentage of white, black, or Hispanic women with an intact uterine cervix who reported having had a Pap test in the past 3 years was greater than or equal to 81%. For any racial or ethnic group, only three states had reached the national health objective that greater than or equal to 95% of women have had a timely Pap test, but most states had reached the objective that greater than or equal to 60% of women aged greater than or equal to 50 years have had both a mammogram and a clinical breast exam within the past 2 years (3). The median percentage of white, black, or Hispanic adults who reported having been screened for colorectal cancer was low. The data from this report and a previous report indicate that less than or equal to 20% had used the home-kit blood stool test in the past year and less than or equal to 30% had had a sigmoidoscopy within the last 5 years (41). The American Cancer Society recommends that all persons aged greater than or equal to 50 years have a blood stool test each year and a sigmoidoscopy every 5 years (42).

Caution must be used in interpreting these BRFSS data. In 1997, only 11 states had a sufficient number of (i.e., greater than or equal to 50) American Indian or Alaska Native respondents to produce stable state-level estimates, and only 10 states had a sufficient number of Asian or Pacific Islander respondents. When responses to BRFSS questions were stratified by age and sex, as few as one state had greater than or equal to 50 respondents in a racial or ethnic group. Thus, the prevalence reported may not be representative of the nation for these groups.

In addition, combining several populations into a broad category may mask important differences between the populations. BRFSS surveys conducted among Chinese, Korean, and Vietnamese persons in California have demonstrated that each subpopulation has some distinct characteristics in educational attainment, income, prevalence of health-risk behaviors, and use of clinical preventive services (6,35,43). Thus, a median prevalence for the broader category of Asian or Pacific Islander may not accurately reflect the prevalence among the subpopulations in this group.

The racial and ethnic categories used in this report may be not only too broad, but also imprecise. The proportion of persons in the United States who identify their race as American Indian or Alaska Native has increased since the 1960s (44). The increase is larger than can be accounted for by deaths, births, immigrations, and improvements in census coverage. This disproportionate increase suggests that persons are now more likely to identify themselves on censuses and surveys as American Indian or Alaska Native.

The BRFSS-estimated prevalences of access to health care, health-status indicators, health-risk behaviors, and clinical preventive services pertain only to the U.S. adult population living in households with telephones. Overall, 95% of U.S. households have telephones; coverage ranges from 87% to 98% across states and varies for subpopulations as well (45). In 1996 an estimated 5.0% of white; 14.6% of black; 14.6% of Hispanic; 16.8% of American Indian, Aleut, or Eskimo; and 4.7% of Asian or Pacific Islander households had no telephone (46). Approximately 8% of households in the South, 5.5% in both the Northeast and West, and 5.3% in the West had no telephone. Persons in lower socioeconomic groups also typically had lower telephone coverage. The BRFSS uses no direct method to compensate for no telephone in the home, but poststratification weights are used and may partially correct for any bias caused by non-telephone coverage. BRFSS weights adjust for differences in probability of selection, nonresponse, and non-telephone coverage, and they must be used to derive representative population-based prevalence estimates.

Most of the risks behaviors associated with chronic disease and injury and described in this report are modifiable. Lack of health insurance, cost as a barrier to obtaining health care, obesity, high blood pressure, high blood cholesterol, lack of leisure-time physical activity, binge drinking, cigarette smoking, not always wearing a safety belt, and screening for cancer could be improved through more effective state and local public health programs. Health-care practitioners could do more to encourage their patients to reduce harmful behaviors and adopt healthier behaviors (47). In addition, health-care delivery systems could offer patients more counseling on preventive measures (e.g., losing weight, stopping smoking, wearing safety belts, or getting a timely mammogram) and facilitate patients' access to clinical preventive services (e.g., by coordinating mammography vans that visit worksites or neighborhoods, or hosting health fairs at shopping centers that offer blood pressure and cholesterol screening) (48). Rapid improvements in modifiable risks and particularly in preventive services are possible, as evidenced by the doubling of timely screening for breast cancer in the United States from 1987 through 1992 (49). In the early 1980s, use of mammography had been underused by women, especially among black women, but the gap between whites and blacks disappeared by 1992.

State-specific data and racial- and ethnic-specific data from the BRFSS can provide a sound basis for developing and evaluating public health programs to reduce racial and ethnic disparities in health risks. In Georgia, analysis of BRFSS data has prompted implementation of a parallel, local survey in a large urban county to further investigate such disparities (personal communication, Linda Martin, BRFSS coordinator, Georgia Department of Human Resources, Division of Public Health). In Alaska, according to BRFSS data, the median percentage of adults who smoke is higher than that for the nation, and in this state American Indians or Alaska Natives have the highest percentage of smokers of all racial or ethnic groups. Alaska is using BRFSS data to support state tobacco control efforts to reduce the prevalence of smoking and its harmful health effects (50).

This report is a first step toward understanding some of the reasons for racial and ethnic disparities in health and health behaviors. More analytic work is needed to shed light on the reasons for these differences. A better understanding of these differences should help states develop effective, culturally sensitive public health prevention programs to decrease the prevalence of high-risk behaviors and increase the use of preventive services. The BRFSS is also a valuable tool in monitoring progress toward national year 2000 and state health objectives for racial and ethnic populations.

References

1. McGinnis JM, Foege WT. Actual causes of death in the United States. JAMA 1993;270:2207-12.
2. Hahn RA, Teutsch SM, Rothenberg RB, Marks JS. Excess deaths from nine chronic diseases in the United States, 1986. JAMA 1990;264:2654-9.
3. Public Health Service. Healthy people 2000: national health promotion and disease prevention objectives -- full report, with commentary. Washington, DC: US Department of Health and Human Services, Public Health Service, 1991; DHHS publication no. (PHS)91-50212.
4. National Center for Health Statistics. Health, United States, 1998 with socioeconomic status and health chartbook. Hyattsville, MD: US Department of Health and Human Services, CDC, National Center for Health Statistics, 1998; DHHS publication no. (PHS)98-1232.
5. Council of Economic Advisors for the President's Initiative on Race. Changing America: indicators of social and economic well-being by race and Hispanic origin. Washington, DC: Council of Economic Advisors for the President's Initiative on Race, 1998.
6. CDC. Behavioral risk factor survey of Vietnamese -- California, 1991. MMWR 1992;41:69-72.
7. Nelson DE, Condon K. Objectives and design of the Behavioral Risk Factor Surveillance System. In: American Statistical Association 1998 proceedings of the Section on Survey Research Methods. Alexandria, VA: American Statistical Association, 1999:214-8.
8. Powell-Griner E, Anderson JE, Murphy W. State- and sex-specific prevalence of selected characteristics -- Behavioral Risk Factor Surveillance System, 1994 and 1995. In: CDC surveillance summaries (August 1). MMWR 1997;46(No. SS-3).
9. Frazier EL, Franks AL, Sanderson LM. Behavioral risk factor data. In: Using chronic disease data: a handbook for public health practitioners. Atlanta: US Department of Health and Human Services, Public Health Service, CDC, 1992:4-1-4-17.
10. Remington PL, Smith MY, Williamson DF, Anda RF, Gentry EM, Hogelin GC. Design, characteristics, and usefulness of state-based behavioral risk factor surveillance: 1981-87. Public Health Rep 1988;103:366-75.
11. Waksberg J. Sampling methods for random digit dialing. J Am Stat Assoc 1978;73:40-6.
12. Gentry EM, Kalsbeek WD, Hogelin GC, et al. The behavioral risk factor surveys: II. Design, methods, and estimates from combined state data. Am J Prev Med 1985;1:9-14.
13. Shah BV, Barnwell BG, Bieler GS. SUDAAN: software for the statistical analysis of correlated data. User's manual, release 7.0 [Software documentation]. Research Triangle Park, NC: Research Triangle Institute, 1996.
14. White AA. Response rate calculation in RDD telephone health surveys: current practices. In: American Statistical Association 1983 proceedings of the Section on Survey Research Methods. Washington, DC: American Statistical Association, 1984:277-82.
15. National Heart, Lung, and Blood Institute. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults. The evidence report. Bethesda, MD: US Department of Health and Human Services, Public Health Service, National Institutes of Health, 1998; NIH publication no. 98-4083.
16. World Health Organization. Obesity: preventing and managing the global epidemic. Report of a WHO consultation of obesity, Geneva, 3-5 June 1997. Geneva: World Health Organization, 1998.
17. Frazier EL, Okoro CA, Smith C, McQueen DV. State- and sex-specific prevalence of selected characteristics -- Behavioral Risk Factor Surveillance System, 1992 and 1993. In: CDC surveillance summaries (December 27). MMWR 1996;45(SS-6).
18. Siegel PZ, Frazer EL, Mariolis P, Brackbill RM, Smith C. Behavioral risk factor surveillance, 1991: monitoring progress toward the nation's year 2000 health objectives. In: CDC surveillance summaries (August 27). MMWR 1993;42(SS-4).
19. Siegel PZ, Brackbill RM, Frazer EL, Mariolis P, Sanderson LM, Waller MN. Behavioral risk factor surveillance, 1986-1990. In: CDC surveillance summaries (December). MMWR 1991;40(SS-4):1-23.
20. Anda RF, Waller MN, Wooten KG, Mast EE, Escobedo LG, Sanderson LM. Behavioral risk factor surveillance, 1988. In: CDC surveillance summaries (June 1). MMWR 1990;39(SS-2).
21. CDC. Behavioral risk factor surveillance -- selected states, 1986. MMWR 1987;36:252-4.
22. CDC. Behavioral risk factor surveillance in selected states -- 1985. MMWR 1986;35:441-4.
23. CDC. Behavioral risk factor surveillance -- selected states, 1984. MMWR 1986;35:253-4.
24. CDC. Behavioral risk factor surveillance system: reprints from the MMWR, 1990-1998. Atlanta: CDC, 1999.
25. Lillie-Blanton M, Laveist T. Race/ethnicity, the social environment, and health. Soc Sci Med 1996;43:83-91.
26. Lillie-Blanton M, Parsons PE, Gayle H, Dievler A. Racial differences in health: not just black and white, but shades of gray. Annu Rev Public Health 1996;17:411-48.
27. CDC. Prevalence of selected risk factors for chronic disease by education level in racial/ethnic populations -- United States, 1991-1992. MMWR 1994;43:894-9.
28. Otten MW Jr, Teutsch SM, Williamson DF, Marks JS. The effect of known risk factors on the excess mortality of black adults in the United States. JAMA 1990;263:845-50.
29. Guralnik JM, Leveille SG. Annotation: race, ethnicity, and health outcomes -- unraveling the mediating role of socioeconomic status. Am J Public Health 1997;87:728-30.
30. Shea S, Stein AD, Basch CE, et al. Independent associations of educational attainment and ethnicity with behavioral risk factors for cardiovascular disease. Am J Epidemiol 1991;134:567-82.
31. Winkleby MA, Fortmann SP, Barrett DC. Social class disparities in risk factors for disease: eight-year prevalence patterns by level of education. Prev Med 1990;19:1-12.
32. Kuttner R. The American health care system: employer-sponsored health coverage. N Engl J Med 1999;340:248-52.
33. CDC. Health insurance coverage and receipt of preventive health services -- United States, 1993. MMWR 1995;44:219-25.
34. Makuc DM, Freid VM, Parsons PE. Health insurance and cancer screening among women. Advance data from vital and health statistics; no. 254. Hyattsville, MD: US Department of Health and Human Services, Public Health Service, CDC, National Center for Health Statistics, 1994.
35. CDC. Behavioral risk factor survey of Korean Americans -- Alameda County, California, 1994. MMWR 1997;46:774-7.
36. Geronimus A, Bound J, Waidmann TA, Hillemeier MM, Burns PB. Excess mortality among blacks and whites in the United States. N Engl J Med 1996;335:1552-8.
37. Mueller KJ, Patil K, Boilesen E. The role of uninsurance and race in healthcare utilization by rural minorities. Health Serv Res 1998;33(pt 1):597-610.
38. Bennefield RL. Health insurance coverage: 1997. The haves and have-nots. Washington, DC: US Department of Commerce, Economics and Statistics Administration, Bureau of the Census, 1998. (Current population report P60-202).
39. Kass BL, Weinick RM, Monheit AC. Racial and ethnic differences in health, 1996. Rockville, MD: Agency for Health Care Policy and Research, 1999; AHCPR publication no. 99-0001. (MEPS chartbook no. 2).
40. Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Summary of the second report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel II). JAMA 1993;269:3015-23.
41. CDC. Screening for colorectal cancer -- United States, 1997. MMWR 1999;48:116-21.
42. Byers T, Levin B, Rothenberger D, Dodd GD, Smith RA. American Cancer Society guidelines for screening and surveillance for early detection of colorectal polyps and cancer: update 1997. CA Cancer J Clin 1997;47:154-60.
43. CDC. Behavioral risk factor survey of Chinese -- California, 1989. MMWR 1992;41:266-70.
44. Passel JS. The growing American Indian population, 1960-1990: beyond demography. In: Sandefur GD, Rindfuss RR, Cohen B, eds. Changing numbers, changing needs: American Indian demography and public health. Washington, DC: National Academy Press, 1996:79-102.
45. US Bureau of the Census. Phoneless in America. Washington, DC: US Department of Commerce, Economics and Statistics Administration, Bureau of the Census, 1994. (Statistical brief 94-16).
46. Giesbrecht LH, Kulp DW, Starer AW. Estimating coverage bias in RDD samples with Current Population Survey (CPS) data. In: American Statistical Association 1996 proceedings of the Section on Survey Research Methods. Alexandria, VA: American Statistical Association, 1997:503-8.
47. US Preventive Services Task Force. Guide to clinical preventive services. 2nd ed. Baltimore: Williams & Wilkins, 1996.
48. Walsh JM, McPhee SJ. A systems model of clinical preventive care: an analysis of factors influencing patient and physician. Health Educ Q 1992;19:157-75.
49. Anderson LM, May DS. Has the use of cervical, breast, and colorectal cancer screening increased in the United States? Am J Public Health 1995;85:840-2.
50. Owen P, Ingle DE, Schumacher C. The prevalence of tobacco use among Alaska adults. Alaska Med 1996;38:21-51.

* For simplicity in this report, the term state hereafter includes the District of Columbia and Puerto Rico.