Goal 2: Eliminate Health Disparities

The second goal of Healthy People 2010 stems from the observation that there are substantial disparities among populations in specific measures of health, life expectancy, and quality of life. The second goal is to eliminate health disparities that occur by race and ethnicity, gender, education, income, geographic location, disability status, or sexual orientation.¹ As discussed in the section on Healthy People objectives and as shown in Figure 5 and Table 3, there has been widespread improvement in objectives for nearly all of the populations associated with these characteristics. However, progress toward the target for individual populations and progress toward the goal to eliminate disparities are independent of each other.¹⁰ Improvements for individual populations—even improvements for all of the populations for a characteristic—do not necessarily ensure the elimination of disparities. This section focuses specifically on relative disparities between populations and changes in these relative disparities over time, regardless of whether the rates for specific populations are moving toward or away from the targets for each objective (see the Technical Appendix for further discussion of the measurement of disparities and changes in disparities).

Disparities between populations and the persistence of disparities over time have been well documented.^{1, 11, 12, 13, 14, 15, 16, 17, 18}  Unlike previous Healthy People initiatives, Healthy People 2010 calls for monitoring objectives for an extensive array of specific population characteristics. All population-based objectives and subobjectives were monitored by race and ethnicity, by income or education, and by gender (if applicable). Monitoring for other characteristics (that is, geographic location and disability status) was optional. Healthy People 2010, therefore, provides the basis for a broad examination of disparities among populations and changes in disparities over time. Findings for specific objectives and populations are presented in 27 of the 28 focus area chapters. None of the objectives in Public Health Infrastructure (Focus Area 23) call for data according to population characteristics. The findings concerning disparities among populations are summarized below. The following conclusions are based on this summary:

• Substantial disparities between populations were evident for many Healthy People 2010 objectives.
• Both increases and decreases in relative disparities were evident for individual populations for specific objectives and subobjectives; however, there was no change in disparity for most of the objectives and subobjectives with data for any group.

For specific population characteristics:

• Among 195 objectives and subobjectives with trend data for racial and ethnic groups, disparities decreased for 24 and increased for 14.
• Among 238 objectives and subobjectives with trend data for males and females, disparities decreased for 25 and increased for 15. Females more often had the best group rate, and reductions in disparity were more frequent among males.
• Among education groups, disparities decreased for 3 objectives and subobjectives and increased for 14.
• Among income groups, among geographic groups, and between persons with disabilities and persons without disabilities, there were few changes in disparities.

Measuring Disparities

There are 498 population-based objectives and subobjectives for which disparities among populations could be measured. The second figure in each focus area chapter provides detailed information about disparities among populations for the objectives and subobjectives in that focus area. These figures provide information about the availability of data for each population, the size of relative disparities between populations for each characteristic, and the size of changes in these disparities between the Healthy People 2010 baseline and the most recent data point for each objective and subobjective. Data are not available for all populations for each objective and subobjective, and tracking data are not always available to assess changes in disparity from the baseline. Data are not available by sexual orientation for any of the Healthy People 2010 objectives.

In this midcourse review, disparities are measured using the “best” or most favorable group rate as the reference point.¹⁰  Best is used to identify the group with the most favorable rate among the groups associated with a particular characteristic. Best does not imply that no further improvement is called for. Disparities by race and ethnicity, for example, are measured using the rate for the racial and ethnic population with the best rate as the reference point. Disparities are measured in relative terms as the percent difference between the rate for each of the other populations and the rate for the best group for each characteristic. In the measurement of disparities, objectives are generally expressed in terms of adverse events, such as death rates, to facilitate comparisons among them (except for a small number of objectives that cannot be expressed in adverse terms). Changes in disparity are measured by subtracting the percent difference from the best group rate at the baseline from the percent difference from the best group rate at the most recent data point. As a result, changes in disparity are measured in percentage points. In addition, when more than two groups are associated with a characteristic (race and ethnicity, education, or income), a summary index is used to describe the average percent difference from the best group rate for all of the other group rates.^{10, 18} The summary index provides a basis for conclusions about changes in the size of the disparities associated with these characteristics. A detailed description of the methods used to measure and evaluate disparities is provided in the Technical Appendix.

Finding Concerning Disparities

Race and Ethnicity
Information about disparities among racial and ethnic populations at the most recent data point based on the disparity tables for each focus area is summarized in Figure 9. The measurement of disparities depends on the availability of data for each group. The number of objectives and subobjectives with the data needed to measure disparities varies from 43 for the Native Hawaiian or other Pacific Islander population to 356 for the black non-Hispanic population.

Figure 9. Percent Distribution of Healthy People 2010 Objectives and Subobjectives
by Size of Disparity for Racial and Ethnic Groups at the Most Recent Data Point

View text version of Figure 9.

American Indian or Alaska Native Population
The data needed to assess disparities for the American Indian or Alaska Native population are available for 155 objectives and subobjectives (Figure 9). This group had the best rate for 13 percent of these objectives and subobjectives, including the least exposure to ozone (8-1a), the lowest stroke death rate (12-7), and the lowest case rate for hepatitis A (14-6). The American Indian or Alaska Native population had a larger proportion of disparities in the greater than or equal to 100 percent category than any of the other racial and ethnic populations. The American Indian or Alaska Native population had rates at least 100 percent worse than (or at least twice as high as) the best group rate for 24 percent of these objectives and subobjectives, including health insurance coverage among persons under age 65 (1-1), new AIDS cases (13-1), infant deaths (16-1c), and deaths of persons aged 15 to 24 years (16-3b and c). Disparities increased for the American Indian or Alaska Native population for 13 objectives and subobjectives, including increases of 100 percentage points or more for hepatitis B among persons aged 40 years and older (14-3c) and women smoking during pregnancy (16-17c). Disparities from the best group rate decreased for 12 objectives and subobjectives, including reductions of 100 percentage points or more for end-stage renal disease due to diabetes (4-7), HIV testing of tuberculosis patients aged 25 to 44 years (13-11), new cases of hepatitis A (14-6), physical assault (15-37), gonorrhea among females aged 15 to 24 years (25-2b), and primary and secondary syphilis (25-3). Reductions of 100 percentage points do not mean that the disparity has been eliminated. For example, the percent difference from the best group rate for the American Indian or Alaska Native population for the physical assault objective (15-37) declined by 254 percentage points, from 626 percent at baseline in 1998 to 372 percent in 2001.

Asian and Native Hawaiian or Other Pacific Islander Populations
The data needed to assess disparities for the Asian population (excluding the Native Hawaiian or other Pacific Islander population) are available for 107 objectives and subobjectives (Figure 9). The Asian population had the best group rate for 36 percent of these objectives and subobjectives, including preterm birth (16-11a, b, and c) and fetal and infant mortality (16-1a through f). This population had rates at least 100 percent worse than the best group rate for 7 percent of the objectives and subobjectives with data for this population, including source of ongoing care for persons aged 18 years and under (1-4b), Pap test ever received (3-11a), and tuberculosis cases (14-11). For the Asian population, disparities increased for four objectives and subobjectives, and there were none for which disparities decreased.

Data for the Native Hawaiian or other Pacific Islander population are available for 43 objectives and subobjectives (Figure 9). This population had a smaller percentage of best group rates (16 percent) and a greater percentage of large disparities (21 percent) than the Asian population. Disparities increased for this population for five objectives and subobjectives, including increases of 100 percentage points or more for cigarette smoking during pregnancy (16-17c), and disparities decreased for three objectives and subobjectives.

Data are available for the combined Asian or Pacific Islander population for 76 objectives and subobjectives (Figure 9). This combined population had the best group rate for 55 percent of these objectives and subobjectives, including many causes of death in Cancer (Focus Area 3); Injury and Violence Prevention (Focus Area 15); and Maternal, Infant, and Child Health (Focus Area 16). The Asian or Pacific Islander population had rates at least 100 percent worse than the best group rate for two objectives—HIV testing among tuberculosis patients aged 25 to 44 years (13-11) and congenital syphilis (25-9). Disparities increased for the Asian or Pacific Islander population for six objectives, including an increase of 100 percentage points or more for congenital syphilis (25-9). Disparities decreased for this population for six objectives and subobjectives, including decreases of 100 percentage points or more for cases of hepatitis B (14-3a, b, and c).

Hispanic Population
The data needed to assess disparities for the Hispanic population are available for 330 objectives and subobjectives (Figure 9). The Hispanic population had the best group rate for 18 percent of these objectives and subobjectives, including objectives and subobjectives related to injury and violence (Focus Area 15), nutrition (Focus Area 19), and tobacco use (Focus Area 27). This population had rates at least 100 percent worse than (or at least twice as high as) the best group rate for 12 percent of these objectives and subobjectives, including health insurance coverage among persons under age 65 years (1-1), source of ongoing care (1-4a, b, and c), new AIDS cases and HIV infection deaths (13-1 and 13-14), and objectives related to violence (Focus Area 15). Between the baseline and the most recent data point, disparities increased for 22 objectives and subobjectives, including increases of 100 percentage points or more for exposure to particulate matter (8-1b), new cases of tuberculosis (14-11), physical assault (15-37), and congenital syphilis (25-9), and decreased for 25 objectives and subobjectives, including decreases of 100 percentage points or more for new AIDS cases (13-1), hepatitis A and B (14-6 and 14-3c), nonfatal firearm-related injuries (15-5), and primary and secondary syphilis (25-3).

Black Non-Hispanic Population
Data needed to assess disparities for the black non-Hispanic population (or, in some cases, for blacks, including persons of Hispanic origin) are available for 356 objectives and subobjectives (Figure 9). This population had the best rate for 21 percent of these objectives and subobjectives, including counseling about health risks (Focus Areas 1 and 2) and quality of their health providers’ communication skills (Focus Area 11). This population had rates at least 100 percent worse than the best group for 20 percent of these objectives and subobjectives, including most causes of death in many focus areas. Between the baseline and the most recent data point, disparities increased for 32 objectives and subobjectives, including increases of 100 percentage points or more for prostate cancer deaths (3-7), invasive early onset group B streptococcal disease (14-16), firearm-related deaths (15-3), smoking during pregnancy (16-17c), and genital herpes in persons aged 20 to 29 years (25-4). Disparities decreased for 26 objectives and subobjectives, including decreases of 100 percentage points or more for new AIDS cases (13-1), invasive pneumococcal infections in persons aged 5 years and under (14-5a), residential fire deaths (15-25), asthma deaths among persons aged 15 to 34 years (24-1c), new cases of gonorrhea (25-2a), total cases of gonorrhea in women aged 15 to 44 years (25-2b), total cases of primary and secondary syphilis (25-3), and cases of congenital syphilis (25-9). To reiterate, reductions of 100 percentage points do not mean that the disparity has been eliminated.

White Non-Hispanic Population
Data needed to assess disparities for the white non-Hispanic population (or, in some cases, for whites, including persons of Hispanic origin) are available for 355 objectives and subobjectives (Figure 9). This population had the best rate for 49 percent of these objectives and subobjectives and disparities greater than or equal to 100 percent for 5 percent. The large disparities included deaths from lung cancer (3-2), female breast cancer (3-3), prostate cancer (3-7), melanoma (3-8), firearms (15-3), poisoning (15-8), unintentional injury (15-13), suicide (18-1), and chronic obstructive pulmonary disease (24-10). Between the baseline and the most recent data point, disparities increased for 25 objectives and subobjectives, including increases of 100 percentage points or more for poisoning deaths (15-8), smoking during pregnancy (16-17c), and drug-induced deaths (26-3). Between the baseline and the most recent data point, disparities declined for 16 objectives and subobjectives. All decreases in disparity were less than 50 percentage points.

Two or More Races
Data for individuals who identified with more than one race are available for 82 objectives and subobjectives (Figure 9). This population had the best rate for 23 percent of these objectives and subobjectives, including physical activities that enhance muscular strength (22-4) and flexibility (22-5). This population had rates 100 percent worse than the best group rate for 10 percent of these objectives and subobjectives, including activity limitations due to chronic lung disease (24-9), activity limitations due to chronic back conditions (2-11), and delay or difficulty in getting emergency care (1-10). Disparities increased for this group for two air pollution subobjectives: exposure to particulate matter and carbon monoxide (8-1b and c, respectively).

Changes in Disparity Among Racial and Ethnic Groups
In addition to the findings for individual racial and ethnic groups, a summary index allows the evaluation of changes in disparity over time among all racial and ethnic populations. There was no change in disparities among racial and ethnic populations for 157 objectives and subobjectives or 81 percent of the 195 objectives and subobjectives with the data needed to calculate the index and assess change over time. The average percent difference from the best group rate decreased for 24 objectives and subobjectives and increased for 14 objectives and subobjectives (Figure 10). Disparity increased by at least 100 percentage points for smoking during pregnancy (16-17c). Disparities decreased by at least 100 percentage points for new AIDS cases (13-1), hepatitis A (14-6), nonfatal firearm-related injuries (15-5), gonorrhea (25-2a and b), primary and secondary syphilis (25-3), and congenital syphilis (25-9).

Figure 10. Number of Objectives and Subojectives With Changes in Disparity
From the Healthy People 2010 Baseline to the Most Recent Data Point for
Each Population Characteristic

View text version of Figure 10.

Gender
Data by gender are available for 304 objectives and subobjectives (Figure 11). For some objectives and subobjectives only baseline data are available. As noted below, trends in disparity can only be measured for 238 objectives and subobjectives. Disparities by gender are not relevant to objectives and subobjectives that apply only to females or males, including those in Family Planning (Focus Area 9) and a number of objectives and subobjectives in other focus areas. Results by gender are summarized in Figure 11.

Figure 11. Percent Distribution of Healthy People 2010 Objectives and Subojectives
by Size of Disparity for Females and Males at the Most Recent Data Point

View text version of Figure 11.

Females had the better group rate for 64 percent of these objectives and subobjectives, compared with 40 percent for males. There were a number of cases in which males and females had the same rate; therefore, both were counted as the better rate. Females had a smaller percentage of objectives and subobjectives that were greater than or equal to 100 percent worse than males. Cases of genital herpes (25-4) and physical assault by an intimate partner (15-34) were more than twice as common among females.

Males had the better group rate for 40 percent of these objectives and subobjectives. Males had a larger percentage of objectives that were greater than or equal to 100 percent worse than females, including less favorable rates for oropharyngeal cancer deaths (3-6), melanoma deaths (3-8), and lower extremity amputations in persons with diabetes (5-10) and nine objectives related to injury and violence (Focus Area 15).

Changes in Disparity Between Gender Groups
The data needed to evaluate changes in gender disparities over time are available for 238 objectives and subobjectives. There was no change in disparity for 198 objectives and subobjectives, or 83 percent of the total with data. Disparity decreased for 25 objectives and subobjectives and increased for 15 (Figure 10). The number of objectives and subobjectives with decreases in disparity, therefore, outnumbered those with increases. Changes in disparity were more common among males, in part because females more often had the better group rate. Among males, disparities decreased for 21 objectives and subobjectives and increased for 13. In a few instances, reductions in the difference between males and females were associated with increases in death rates—deaths from poisoning (15-8), falls (15-27), and drugs (26-3) (data not shown)—of both males and females.

Education Level
The data needed to assess disparities among populations by education level are available for 158 to 161 objectives and subobjectives in 23 focus areas (Figure 12). Education level was not included as a characteristic in all focus areas. The population with at least some college education had the best rate for 82 percent of the objectives and subobjectives with data by education. The population with less than a high school education and high school graduates had the best group rate for 9 percent and 11 percent of the objectives and subobjectives with data by education, respectively. For example, the least educated population had the best or most favorable rates for three subobjectives concerning communication with health care providers (11-6a, c, and d). There were no objectives or subobjectives for which the difference between the population with at least some college education and the best group differed by at least 100 percent. The population that completed high school had rates at least 100 percent higher than the best group rate for 17 percent of these objectives and subobjectives, and the population with less than a high school education had rates at least 100 percent higher than the best group rate for 25 percent of these objectives. Among persons with less than a high school education, disparity increased for nine objectives and subobjectives and decreased for nine. Among persons with a high school education, disparity increased for 24 objectives and subobjectives and decreased for 2. Among persons with at least some college, there was one objective with an increase in disparity. There were no increases or decreases in disparity of 100 percentage points or more for any of the three populations by education level.

Figure 12. Percent Distribution of Healthy People 2010 Objectives and Subobjectives
by Size of Disparity for Education Groups at the Most Recent Data Point

View text version of Figure 12.

Changes in Disparity Among Education Groups
In addition to the findings for individual populations by education level, the summary index permits the evaluation of changes in disparities over time among all three populations by level of education. There was no change in disparity among populations by education level for 92 objectives and subobjectives or 84 percent of the 109 objectives and subobjectives with the data needed to calculate the index and assess change over time. The average percent difference from the best group rate decreased for 3 objectives and subobjectives and increased for 14; therefore, increases in disparity by level of education outnumbered decreases in disparity (Figure 10). There were no increases or decreases of 100 percentage points or more.

Income Level
Income level was not included as a characteristic in all focus areas. All of the objectives and subobjectives in Nutrition and Overweight (Focus Area 19) and six subobjectives in Immunization and Infectious Diseases (Focus Area 14) are excluded from this summary because data by income were available for only two populations. This summary is based on 99 to 104 objectives and subobjectives with data for populations by income level (Figure 13). The population with middle/high income had the best rate for 76 percent of the objectives and subobjectives with data by income level. The poor and near-poor income populations each had the best rate for 16 percent of their objectives and subobjectives. The poor population had the most favorable rates for counseling about diet and nutrition (1-3b) and counseling about reduced alcohol consumption (1-3d) and for eight objectives and subobjectives in Substance Abuse (Focus Area 26). In almost all cases, disparities for the other populations by income level were less than 10 percent.

Figure 13. Percent Distribution of Healthy People 2010 Objectives and Subobjectives
by Size of Disparity for Income Groups
at the Most Recent Data Point

View text version of Figure 13.

There were no objectives or subobjectives for which the difference between persons with middle/high income and the best group differed by 100 percent or more. The near-poor income population had rates at least 100 percent higher than the best group rate for 6 percent of the objectives and subobjectives with data for this population. The poor or lowest income population had rates at least 100 percent higher than the best group rate for 12 percent of the objectives and subobjectives with data.

Changes in Disparity Among Income Groups
The summary index enables the evaluation of changes in disparity over time among populations by income level. The data needed to evaluate changes in disparity among the populations by income level are available for 53 objectives and subobjectives (Figure 10). There was little evidence of any change in disparity among populations by income level. The average percent difference from the best group rate did not increase for any objectives or subobjectives and decreased for one objective related to births within 24 months of a previous birth (9-2).

Geographic Location
Geographic location is defined in different ways in Healthy People 2010. For some objectives, the distinction is between urban and rural areas, while for others, the distinction is between metropolitan and nonmetropolitan areas. Findings for disparities by geographic location for 52 objectives and subobjectives in 13 focus areas are summarized in Figure 14.

Figure 14. Percent Distribution of Healthy People 2010 Objectives and Subobjectives
by Size of Disparity for Geographic Groups
at the Most Recent Data Point

View text version of Figure 14.

Urban or metropolitan areas had the better rate for 65 percent of the objectives and subobjectives with data. Urban or metropolitan areas also had more objectives and subobjectives with larger disparities than rural or nonmetropolitan areas. There was at least a 100 percent difference from nonmetropolitan areas for four of the six air pollution subobjectives (8-1a, b, c, and e).

Rural or nonmetropolitan areas had the best rate for 42 percent of these objectives and subobjectives. Urban and rural areas had the same rate—therefore also the better rate—for four objectives. There were no objectives or subobjectives for which rural or nonmetropolitan areas had rates at least twice as high as rates in urban or metropolitan areas.

Changes in Disparity Among Geographic Groups
The data needed to evaluate changes in disparities between geographic areas are available for 38 objectives and subobjectives. Disparities from the best geographic group rate declined for three subobjectives, and there were no increases (Figure 10). Between the baseline in 1997 and the data point in 2001, disparities declined by at least 100 percentage points for two air pollution subobjectives—exposure to particulate matter and carbon monoxide (8-1b and 8-1c)—and declined by a smaller margin for exposure to ozone (8-1a). Reductions of 100 percentage points do not mean that the disparity has been eliminated.

Disability Status
Data for persons with disabilities and persons without disabilities for 63 objectives and subobjectives in 17 focus areas are summarized in Figure 15. Persons with disabilities had the better group rate for 46 percent of these objectives and subobjectives, and persons without disabilities had the better group rate for 54 percent. The similarity in the percentage with the better group rate between persons with disabilities and persons without disabilities may be a reflection of more frequent contacts with health care providers among persons with disabilities. Persons with disabilities had rates at least 100 percent higher than persons without disabilities for 6 percent of these objectives and subobjectives, including delay or difficulty getting emergency care (1-10), new cases of diabetes (5-2), overall cases of diabetes (5-3), and blindness and visual impairment (28-4).

Figure 15. Percent Distribution of Healthy People 2010 Objectives and Subobjectives
by Size of Disparity for Persons With Disabilities and Persons Without Disabilities
at the Most Recent Data Point

View text version of Figure 15.

Changes in Disparity Between Persons With Disabilities and Persons Without Disabilities
The data needed to evaluate changes in disparities between disability groups are available for 48 objectives and subobjectives (Figure 10). There were few changes in disparity by disability status. Disparities between these populations declined for one objective and one subobjective. Between the baseline in 1997 and the data point in 2003, disparity declined by at least 100 percentage points for blindness and visual impairment (28-4). Between 2000 and 2003, the disparity declined by a smaller margin for sun exposure (3-9b). Disparity from the population with the better rate (persons with disabilities) increased for persons without disabilities for source of ongoing care (1-4a).

Data Limitations
Several factors limit the number of objectives for which changes in disparity can be assessed. First, the number of years on which this assessment is based varies greatly. Among 328 population-based objectives with trend data, 16 percent are based on an interval of 2 years or less between the baseline and the most recent data point. Second, this assessment is limited by a lack of data for specific populations. There are no racial and ethnic data for about 15 percent of the population-based objectives, with larger proportions without data among American Indian or Alaska Native, Asian, Hispanic, and Native Hawaiian or other Pacific Islander populations. Data by gender are not available for about 15 percent of the relevant population-based objectives. Data by education level and income level are not available for about 30 percent and 55 percent, respectively, of the population-based objectives and subobjectives for which education level and income level were selected as characteristics to be tracked.

These findings are also subject to the limitations of the data on which they are based. This assessment is based only on data at the baseline and at the most recent data point; intervening data values are not considered. The findings presented here are also limited by the reliability and validity of information about the persons on which the data are based. The reporting of race and income from some data systems is particularly problematic.^{2, 19} Assessments of the probability that disparities or changes in disparity are due to random fluctuations in the data are limited by the lack of estimates of variability (that is, standard errors) for some of the data on which Healthy People 2010 objectives are based (see the Technical Appendix).

<<  Previous—Goal 1   |   Table of Contents   |   Next—Summary and Future Directions  >>