Race, Ethnicity, and Language Data: Standardization for Health Care Quality Improvement

Chapter 2: Evidence of Disparities Among Ethnicity Groups (continued)

Black or African American Groups

In Census 2000, 12.9 percent of the U.S. population (36.2 million people) self-identified with the Black or African American category.8 The Office of Management and Budget (OMB) and Census definition for the Black or African American race category is "a person having origins in any of the Black racial groups of Africa" (OMB, 1997; U.S. Census Bureau, 2000).

Demographic Characteristics 

The Black population, like the AIAN and White populations, is more likely than other groups to be born in the United States (nearly 94 percent vs. 89 percent for the total U.S. population, as compared with 59.8 percent of Hispanics, 31.1 percent of Asians, and 80.1 percent of NHOPI). The origins of foreign-born Blacks are as follows: approximately 59 percent from the Caribbean, 24 percent from Africa, and 13 percent from Central and South America (McKinnon and Bennett, 2005). While English is the primary language of 94 percent of Blacks, nearly one-third of those over age 5 who speak a language other than English at home speak English less than "very well"; additional detail is provided in Table 2-3 on groups who speak a language other than English at home. One in four Blacks live in poverty; 14 percent over age 25 have a bachelor's degree, while 19.6 percent have not graduated from high school.

Health-Related Differences Among Black or African American Groups

For the most part, few studies subdivide the Black population for study; when they are, the literature has generally subdivided this category into U.S.-born Blacks, Caribbean-born Blacks, and African-born Blacks although some have distinguished other groups by using additional countries of birth which may not necessarily represent ethnicity (e.g., born in Europe to African parents). The available literature has examined health and health care differences among these groups by overall self-rated health, mental health, cancer, low birthweight, and cardiovascular health.

Overall Self-Rated Health

In a study comparing U.S.-born, European-born, African-born, and West-Indian-born Black ethnic groups aged 18 and older (utilizing merged 2000-2001 NHIS data), groups were examined for differences in self-rated health status, any self-assessed activity limitation in general and then specifically due to hypertension (Read et al., 2005b). Multivariate regression analyses adjusted for demographic characteristics and socioeconomic status including educational attainment, insurance status and income. The study does not distinguish between Blacks of different ethnicities born in the United States. U.S.- and European-born Blacks had worse ratings on all the measures compared with those born in Africa or Whites born in the United States. West Indian-born Blacks had poorer self-rated health status, more activity limitation, and more hypertension-related activity limitation compared with those born in Africa. European-born Blacks had the worst results of all categories; those who are African born had the best values. These findings lead the authors to conclude that the health advantage ascribed to Black immigrants in other studies can be due to the influence of data on African-born groups.

Mental Health

Williams and colleagues (2007) studied mental health among Caribbean Black groups of different ethnicities as well as African Americans with no Caribbean roots by using data derived from the National Survey of American Life. The Caribbean groups included persons born in the United States as well as those who immigrated to this country. Caribbean Black women had significantly lower odds than African-American women of suffering from any mental disorder in terms of either lifetime prevalence or occurrence in the last 12 months. Caribbean Black men were significantly more likely to suffer from any disorder in the past 12 months but not for lifetime prevalence compared with U.S. African American men. Among the Caribbean ethnicities, those whose ethnic origins were in Spanish-speaking countries had higher odds of lifetime prevalence of any disorder than those from English speaking countries. Using first-generation Blacks as the reference group, third-generation immigrants had greater odds of lifetime prevalence of any disorder. The authors note the importance of understanding associations between ethnicity and other factors in order to better describe heterogeneous populations, concluding "that the mental health risk profile of Caribbean Blacks differs from that of other African-Americans. Moreover, the Black Caribbean immigrant category itself masks considerable heterogeneity" (p. 57) as is illustrated by the differences exhibited for Spanish- and English-speaking countries of origin.

Rates of Cancer Mortality

Data on differences in cancer mortality rates among Blacks at more granular ethnicity levels are limited. One study, based on New York City death certificates dating from 1988-1992 linked with U.S. Census data, found that Caribbean-born non-Hispanic Blacks had lower rates than U.S.-born non-Hispanic Blacks for the types of cancer studied with the exception of prostate cancer. For that, the Caribbean-born group rate was significantly higher than that of any other group (Fang et al., 1997). The authors posit that differences in the Caribbean diet may be protective for certain types of cancers such as colon, rectum, and breast. Descriptive statistics indicated that more Caribbean Blacks graduated from high school, but cancer rates were not adjusted for educational attainment.

Low Birthweight

Pallotto and colleagues (2000) used vital records from Illinois (1985-1990) to assess the low- birthweight distributions for infants born to U.S.-born Black women, Caribbean-born Black women, and U.S.-born White women. They classified women into low and high risk categories based on reproductive risk factors (e.g., trimester, parity) and socioeconomic risk factors (e.g., age, education). Even for the lowest risk mothers, there were differences in relative risks for moderately low birthweight infants (1,500-2,499 g); compared with infants of non-Hispanic White mothers, the risk for infants delivered of U.S.-born non-Hispanic Black mothers was 2.7 (95 percent CI 2.1-3.4) and for infants delivered of Caribbean-born Black mothers 1.2 (95 percent CI 0.4-3.1). This mirrored the relative risk profile for delivery of moderately low birthweight infants among all mothers in these ethnic groups regardless of whether they themselves were assessed as high or low risk for low birthweight outcomes. The relative risk for very low birthweight infants (less than 1,500 g) was elevated for both groups of Black mothers compared with non-Hispanic White mothers, but the Black groups were not significantly different from each other. A similar study of deliveries in Illinois found lower relative risk of low birthweight for infants whose mothers were born in Africa; in fact, for women classified as low risk on demographics and reproductive factors, the relative risk was similar for women born in Africa and for U.S. born White women, yet the risk remained high for U.S. born African-American women (David and Collins, 1997).

Cardiovascular Health

A study by Lancaster and colleagues (2006) used data from the National Health and Nutrition Examination Survey (NHANES) III to assess differences in dietary intake, coronary heart disease (CHD) risk factors, and predicted 10-year risk of CHD for subgroups of Black adults (non-Hispanic Blacks born in the United States and both non-Hispanic and Hispanic Blacks born outside of the United States). Multivariate analyses controlled for education as a socioeconomic marker as well as for age, sex, and body mass index. The study found that non- Hispanic Black, U.S.-born participants had a higher intake of calories and fat; a lower intake of fruits, fiber and micronutrients; and a higher predicted 10-year risk of developing CHD (5.8 percent) than both immigrant groups (non-Hispanic Black 3.7 percent, p <0.001; Hispanic Black 4.7 percent, p = 0.017). However, it is notable that there are differences between the two immigrant groups in terms of their 10-year risk as well. In addition, proportionally more non-Hispanic Black immigrants had elevated fasting glucose, while more Hispanic Blacks had elevated serum triglycerides and low HDL cholesterol. The authors conclude that there is a need to study dietary and health differences within the Black population and tailor dietary interventions to subgroups of Blacks.


In the Black category, U.S.-born Blacks disproportionately suffered worse mental health and cardiovascular outcomes and were at greater risk for having low-birth-weight infants than Blacks born in the Caribbean or Africa. A few notable exceptions were found, such as a significantly elevated incidence of prostate cancer among Caribbean men. The authors of these studies and other studies describe heterogeneity within the Black population in health and cultural factors such as diet, and the need to continue to examine the Black population in greater detail (Kington and Nickens, 2001). Differentials have been primarily explored by distinguishing Black populations born in the United States and elsewhere. Heterogeneity, however, was also described among various immigrant ethnicities.

Asian Groups

While Census 1990 collected data for the single category "Asian or Pacific Islander," Census 2000 split the categories into "Asian" and "Native Hawaiian or Other Pacific Islander," as required by the 1997 OMB standards. The Census and OMB definition for who fits into the Asian category includes "people having origins in any of the original peoples of the Far East, Southeast Asia, or the Indian subcontinent" (OMB, 1997; U.S. Census Bureau, 2000).

Demographic Characteristics 

In Census 2000, 4.2 percent of the U.S. population (11.8 million people) self-identified with the Asian category.9 The largest Asian ethnic groupings are listed in Table 2-4. Each group has different characteristics pertaining to amount of time in the United States, English proficiency, educational attainment, and risk of poverty. Many of Chinese and Japanese ethnicity have lived in the United States for generations, while Vietnamese and Hmong populations are more often recent immigrants. The median age for Japanese is almost 43 years compared with the Hmong population, which has a median age of 16 years (Reeves and Bennett, 2004).

Health-Related Differences Among Asian Groups

For the most part, the Asian categories have been subdivided according to country of ancestry, although some authors also include more general categories (e.g., South Asian). The available literature has examined access to and utilization of health care, cancer and cancer screening, low birthweight, and asthma outcomes.

Access to and Utilization of Health Care Services

Huang and Carrasquillo (2008) used cross-sectional analyses of CPS data to examine differences in insurance coverage across the six largest distinct groups of Asian populations in the United States: Chinese, Filipino, Indian, Korean, Vietnamese, and Japanese. Differences among Asian subgroups in coverage can be greater than the difference between all Asians and non-Hispanic Whites. Persons of Korean heritage had the highest overall proportion lacking coverage (29.8 percent), followed by 21.5 percent in the Vietnamese group and 16.8 percent in the Chinese group (Huang and Carrasquillo, 2008). U.S. born-Koreans reported about twice the rate of uninsurance of other Asian subgroups born in the United States. People of Asian Indian, Filipino, and Japanese heritage had insurance rates similar or better than those of non-Hispanic Whites. The authors note that insurance expansions based solely on income may not resolve the higher rates of uninsurance for Koreans who may have incomes too high to qualify for public programs, but as small business owners find affordability of insurance an issue.

A lack of health coverage can lead to problems in having a usual source of health care. A recent study released by the Kaiser Family Foundation and the Asian and Pacific Islander American Health Forum found that uninsured Asians are more than four times as likely to lack a usual source of care compared with insured Asians (Kaiser Family Foundation and APIAHF, 2008). The same study found that 20 percent of Asian Indians and 21 percent of those falling into the Other Asian category lack a usual source of care, while Chinese and Filipino Americans have rates similar to those for non-Hispanic Whites. The percentage of uninsured persons having a doctor's visit in the past year also varies among subgroup ethnicities. The differential among Asian American groups—for example, fewer insured Filipinos lack a usual source of care (7 percent) compared with insured Asian Indians (13 percent)—is greater than a comparison of the broad Asian category (11 percent) with insured non-Hispanic Whites (9 percent).

Analyses of NHIS survey data from 2004-2006 reveal differences among Asian subgroups in access and utilization (CDC, 2008). For example, 25 percent of Korean adults are without a usual source of car, about twice the rate of other Asian subgroups. Vietnamese adults are more likely to identify a clinic or a health center as their usual source of care (23 percent) compared with other groups (13-14 percent for Japanese, Asian Indians, and Filipinos). The Japanese group was more likely than others to receive influenza and pneumonia vaccinations; Asian Indians are more likely to have received hepatitis B vaccines. A study of Asian and Pacific Islander children using NHIS data from 1997-2000 showed that children identified as falling into the heterogeneous "other Asian and Pacific Islander" category were the most likely to lack a usual source of care (6.3 percent) compared with Chinese (3.8), Filipino children (3.6), Asian Indian (1.9) and non-Hispanic Whites (1.7) (Yu et al., 2004). The poverty rate was greatest among these categories for the Asian Indian and other Asian and Pacific Islander families, but their rates of access differed.

Health Status

Asian Americans tend to rate their health status more highly than do other groups, just 11 percent of Asian Americans rate their health status as fair or poor, compared with 13 percent of non-Hispanic Whites, 18 percent of Hispanics, 22 percent of African Americans, and 23 percent of American Indians/Alaska Natives (Kaiser Family Foundation and APIAHF, 2008). Among Asian American ethnic groups, the proportion rating their health status as fair or poor ranges from 8 percent among those of Japanese ethnicity to 15 percent among those of Vietnamese or South East Asian extraction. Thus, the difference among some Asian groups is greater than the difference revealed by simply comparing the rates for all Asian Americans to non-Hispanic Whites.

Cancer and Cancer Screening among Asian Ethnicities

Breast and cervical cancer screening rates are lower for Asian American women than for any other ethnic group in California. To better understand Asian intragroup differences, Kagawa-Singer and colleagues (2007) used the 2001 California Health Interview Survey to evaluate Pap smear and mammography screening rates for a representative sample of 2,239 Asian American women. Reported rates of Pap test use for those aged 18 and older ranged from 81 percent (Filipina) to 61 percent (Vietnamese). Reported mammography rates for women aged 40 and older ranged from 78 percent (Japanese) to 53 percent (Korean). Somewhat surprisingly, Korean and Japanese immigrants with more than 10 years of U.S. residency had higher rates of Pap screening than their U.S. born counterparts, but this pattern did not hold up for Korean immigrants on mammogram screening. While trends suggested increased used of screening with increasing income, the difference was only found to be significant for Chinese Americans utilizing Pap tests. For women whose income was less than 200 percent of the federal poverty limit (FPL), the Pap screening rates still varied by ethnicity from 53 percent for Chinese Americans to 78 percent for Filipina Americans. Similarly among insured women, the range was 64 percent for Vietnamese and Cambodian Americans to 82 percent for Filipina Americans. Utilization of mammography among women below 200 percent of FPL also varied by ethnicity, from 53 percent for Korean American women to 86 percent among Asian Indian women. Rates for insured women also varied from 59 percent among Korean Americans to 78 percent among Japanese American women. The authors stress that "different factors were independently associated with lower screening rates for each group" (p. 706), and thus it is important to tailor interventions to specific ethnic subgroups.

Asian groups differ with respect to not only screening rates but also mortality. Using data from the California Cancer Registry, which collects approximately 140,000 new cancer case reports annually, the Kaplan-Meier method was applied to calculate 5- and 10-year survival probabilities for cervical cancer by Asian subgroup, and the Cox proportional hazard method was applied for calculating adjusted survival rates (Bates et al., 2008). Among the California women, once adjusted for age, socioeconomic status, stage, and treatment, the risk of mortality was found to be significantly lower for all groups compared to non-Hispanic Whites except Chinese and Japanese women. Of the six groups studied (Chinese, Filipino, Japanese, Korean, South Asian, and Vietnamese), South Asian women were found to have the highest unadjusted survival rates at both 5 and 10 years (85.8 percent for both), followed by Korean (85.7 and 82.5 percent), and Vietnamese (82.1 and 79.7 percent) groups, compared with non-Hispanic Whites (77.5 and 75.4 percent) and Japanese (72.3 and 69.5 percent). Incidence rates are highest among Vietnamese, Filipino, and Korean ethnic groups and lowest among Chinese, Japanese, and South Asian groups; the authors note that incidence rates tend to mirror rates found in international surveillance reports for distinct ethnicities with a few exceptions.

The Centers for Disease Control and Prevention (CDC, 2008, p. 2) states that "although cancer mortality rates for Asian Americans are low compared with other U.S. populations, Asian Americans have the highest incidence rates of liver and stomach cancer for both sexes compared with Hispanic, non-Hispanic Whites, or non-Hispanic Blacks" (CDC, 2008). Furthermore, subgroup differences can be illustrated by differential mortality rates for liver cancer for different Asian ethnicities in California data: specifically 54.3 per 100,000 males for Vietnamese, 33.9 for Korean, 23.3 for Chinese, 16.8 for Filipino, and 9.3 for Japanese compared with a rate of 6.8 for non-Hispanic White males in the state. Disproportionate risks exist for some Asian subgroups, but not all, for a variety of other conditions, including chronic obstructive pulmonary disease, chronic hepatitis B, tuberculosis, and diabetes.

Low Birthweight

Comparisons of the birthweight outcomes for two Asian subgroups (Asian Indian and Chinese) were derived from analysis of the National Center for Health Statistics Natality File for 293,211 singleton births during 1998-2003. Even when the mothers were themselves born in the United States there were ethnic differences in outcomes. Infants born to Asian Indian mothers were more likely to have a lower mean birthweight as well as higher proportions of very low birthweight (VLBW) and moderately low birthweight (MLBW) compared with Chinese mothers, once data were adjusted for age, education, marital status, and parity. Infants born of U.S.-born Asian Indian mothers were 1.87 times as likely to be VLBW and 1.59 times more likely to be MLBW than infants born to U.S.-born Chinese mothers. The likelihood of VLBW and MLBW infants was even higher for non-U.S.-born Asian Indian mothers compared with non-U.S.-born Chinese mothers (Hayes et al., 2008).


A large study published by Davis and colleagues (2006) compared asthma prevalence among various Asian American and Pacific Islander ethnic groups using data from the California Healthy Kids Survey on 462,147 public school students in the state from school years spanning 2001-2002 and 2002-2003. While the analyses could not adjust for sociodemographic characteristics beyond grade and gender, the existence of distinct rates among the groups is clear. Pacific Islander and Filipino groups had higher lifetime prevalence rates for asthma (21.0 and 23.8 percent, respectively) than eight other subgroups (e.g., Korean [10.9 percent], Vietnamese [13.6 percent], Chinese [14.4 percent], and Asian Indians [16.3 percent]). The authors note that prevalence studies can be influenced substantially by the composition of the population under study, and recommend more precise categorization by subgroups for utilization in such studies.


In the Asian category, differences exist across ethnic groups, and disparities differ on health care and health measures. For example, Japanese persons appear less likely to experience cancer screening disparities compared with the reference population, while differences were often noted for Korean, and Vietnamese subgroups. Yet Japanese women have high cervical cancer mortality. Each study stresses the importance of distinct reporting by subgroup to illuminate differences in order to tailor responses accordingly. Pooling of data over several years is often necessary to have a substantial sample to distinguish among subgroups.

Native Hawaiian or Other Pacific Islander Groups

In Census 2000, 0.31 percent of the U.S. population (860,965 people) self-identified with the Native Hawaiian or Other Pacific islander (NHOPI) category.10 This grouping is defined by the Census and OMB as "people having origins in any of the original peoples of Hawaii, Guam, Samoa, or other Pacific Islands" (OMB, 1997; U.S. Census Bureau, 2000).

Demographic Characteristics 

The largest NHOPI groups are listed in Table 2-5. Native Hawaiians, Samoans, and Guamanians make up 74 percent of the Pacific Islander population. Fully 80 percent of NHOPIs are native to the United States since persons born in American Samoa, Guam, or Hawaii are included in the native-born population. Native Hawaiians fare better on ability to speak English, graduation from high school, and having a lower poverty rate relative to most of the other Pacific Islander subgroups (Harris and Jones, 2005).

Health-Related Differences Among Native Hawaiian or Other Pacific Islander Groups

The NHOPI group has been subdivided for analysis according to islands of ancestry. Few studies are available for examining this population in greater detail. Philippine Islanders are classified under the Asian category according the OMB convention; however, some studies examine their health and health care profile along with groups classified as Pacific Islanders by OMB definition.

Access to and Utilization of Health Care Services

A study using a three-year average from CPS data found that the most recent estimate of uninsurance for the NHOPI population was 20.5 percent (DeNavas-Walt et al., 2008). Another study found that the level of uninsurance to be 24 percent (Kaiser Family Foundation and APIAHF, 2008). Several studies also note the low health care utilization rates of Native Hawaiian women compared with other Asian populations (Blaisdell-Brennan and Goebert, 2001; Van Ta and Chen, 2008).

Cancer among Pacific Islander Groups

Among Pacific Islander groups living in all 17 SEER registry sites of the United States, significant health disparities have been found for Native Hawaiian and Samoan groups (Goggins and Wong, 2007; Miller et al., 2008). A study by Goggins and Wong (2007) showed that Samoans were significantly more likely to present with advanced cancer and had the poorest cause-specific survival of all groups studied, including Native Hawaiians, other Pacific Islanders, African Americans, Native Americans, and Whites. While all Pacific Islander ethnicities had poorer cause-specific survival than Whites, Samoan women had an especially elevated risk of mortality from breast cancer (relative risk [RR] = 3.05, 95 percent CI 2.31-4.02) and Samoan men had an especially elevated risk of mortality from prostate cancer (RR = 4.82, 95 percent CI 3.38-6.88). Similar findings are presented in a study by Miller and colleagues (2008), where overall cancer incidence rates were lower for Asians and Pacific Islanders in the sample (i.e., Asian Indians, Chinese, Filipinos, Guamanians, Japanese, Koreans, Native Hawaiians, Samoans, and Vietnamese) compared with White non-Hispanics in the United States; the one exception was Native Hawaiian women. The age-adjusted all cancer mortality rate among Asian and Pacific Islander men was highest for Native Hawaiians (263.7 per 100,000) and Samoans (293.9) in contrast to Guamanians (147.0) and Asian populations such as Japanese (173.7) or Vietnamese (159.9). The pattern of mortality rates among women was the same. The authors suggest that the higher risk for poor outcomes among Samoans may be due to failure to target interventions to small groups whose outcomes are masked when their data are combined with all Pacific Islander and Asian data.


Pacific Islander groups are little studied in comparative research, but among those studied, Samoans appear to suffer disproportionate rates of poor cancer outcomes. Additional data sources indicate that NHOPIs experience high levels of health disparities compared with other groups in the United States as well. For example, Native Hawaiians aged 36-65 are nearly 1.5 times as likely to experience heart disease as other racial groups in the United States (Asian & Pacific Islander American Health Forum, 2006). In California, NHOPI and Filipino adults have higher rates of obesity and being overweight (70 and 46 percent, respectively) compared with the state average (34 percent) (Ponce et al., 2009). Native Hawaiians also have the second highest rate of Type II diabetes among racial groups in the United States (Mau et al., 2001). However, sparse information on Pacific Islander subgroups may be related to the fact their numbers are proportionately small nationally and thus are not reflected in sufficient numbers for analysis in national surveys.

White Groups

In Census 2000, 77 percent of the U.S. population (216.9 million people) self-identified with the White race (Grieco, 2001b).11 Because this is the largest racial group in the United States, it heavily influences reported levels of quality of health and health care achieved in the nation, as well as national rates of indicators, such as poverty. The OMB definition for the White race is "a person having origins in any of the original peoples of Europe, the Middle East, or North Africa," (OMB, 1997) and the Census Bureau definition further elaborates with examples including Irish, German, Italian, Lebanese, Near Easterner, Arab, or Polish (U.S. Census Bureau, 2000).

Demographic Characteristics 

The poverty rate among those of White race alone in 2007 was 10.5 percent, nearly the same as the overall average rate for Asian and Pacific Islanders but half the rate among Blacks and Hispanics. The national poverty rate for the total U.S. population as of 2007 was 12.5 percent (DeNavas-Walt et al., 2008). With respect to the number of persons in poverty, however, there are more Whites (25.1 million) in poverty than Blacks (9.2 million) and Hispanics (9.9 million) combined. Similarly, as of 2000, White non-Hispanics included a lower percentage of persons aged 25 and older who did not graduate from high school (14.5 percent) compared with Blacks (27.7 percent) and Hispanics of any race (47.6 percent) (U.S. Census Bureau, 2006a)—a rate that still translates into 19.4 million White non-Hispanics over age 25 without a high school diploma (U.S. Census Bureau, 2003a). The White population, like the AIAN and Black populations, is more likely to be born in the United States than other racial groups (Malone et al., 2003). (Go to Table 2-6.)

Comparative information on different ethnicities within the White population is limited for both demographics and health and health care differences. The Census has published only one in depth analysis of an ancestry grouping that falls within the White category, and that is of the U.S. Arab population. Three-fifths of the Arab population is of Lebanese, Syrian, and Egyptian ancestry (de la Cruz and Brittingham, 2003), but Lebanese are the largest group, consisting of more than a quarter (28.8 percent) of the U.S. Arab population (Brittingham and de la Cruz, 2005). About half of all Arabs in the country were born here (46.4 percent) (Brittingham and de la Cruz, 2005). Of those who speak Arabic at home, approximately one in four speak English less than very well. Sixteen percent of Arabs here over age 25 have not graduated from high school. The overall poverty rate for U.S. Arab groups (16.7 percent) is somewhat higher than the national rate (12.5 percent) (Brittingham and de la Cruz, 2005); some Arab ancestry groups (e.g., Palestinian, Moroccan, Iraqi) have higher poverty rates. About half of the Arab population resides in only five states: California, Florida, Michigan, New Jersey, and New York (de la Cruz and Brittingham, 2003).

Health-Related Differences Among Select White Groups

While recent research is limited in this area, differences in health care and health outcomes among ethnicities who categorize themselves as White among the OMB categories have been documented. The sections that follow review more recent evidence on this topic, with an emphasis on differences found between groups of Arab and European descent. Reliable data on differences among other ethnic groups within the broad White category could not be identified, representing an area that could benefit from more study that would be informed by granular ethnicity data collection.

Self-Reported Health

Naturalized Middle Eastern immigrants reported worse health compared with their non-naturalized Middle Eastern counterparts in a study based on data from the NHIS. Overall, however, Arab Americans were less likely to report health-related limitations than U.S.-born Whites of European descent (Read et al., 2005a).

Cancer Screening

Lower rates of mammography have been found among Middle Eastern women than in the population as a whole. One telephone survey of 365 Arab American women in metropolitan Detroit found that only 70 percent reported ever having had a mammogram, compared with the overall rate for Michigan of 92.6 percent (Schwartz et al., 2008). This 70 percent rate is lower than the rate for other racial and ethnic groups nationally for mammograms as well. One group, Lebanese women, was considerably more likely than other groups of Arab women to have ever had a mammogram. Other predictors of screening among Middle Eastern women in this sample included being married, having health insurance, and having resided in the United States for 10 or more years (Schwartz et al., 2008).

Cultural beliefs pertaining to cancer among Middle Eastern immigrants in New York appear to be significantly different from those of their White peers of European descent and can affect their access to optimal care. In a qualitative study of focus groups designed to explore barriers to cancer care for Arab immigrants, barriers that emerged included experiences of discrimination, fears of immigration enforcement, and differences in beliefs surrounding causes of cancer (Shah et al., 2008).

However, another study that examined participation in breast cancer genetic counseling found no correlation between ethnicity of the participants in the study, which included European American women and women of Ashkenazi Jewish ancestry, and willingness to accept such counseling (Culver et al., 2001). This study did not control for socioeconomic factors except for level of education attained, because the genetic counseling was being offered at no charge in order to remove cost and access barriers for the participants.

Preterm Birth

A study found lower rates of preterm birth among mothers of Middle Eastern nativity than among those who were U.S.-born of Middle Eastern descent and U.S.-born non-Hispanic Whites (El Reda et al., 2007).


Disparities in health for non-Hispanic Whites compared with other racial groups include high levels of mortality from melanoma, chronic lower respiratory deaths, and prostate cancer, each of which is potentially responsive to health care interventions (Keppel, 2007). While the data on differences among White subgroups is very limited, significant differences can be found among persons of Middle Eastern and European descent. International statistics provide some insight into the differences among European nations, which make up the ancestry of significant portions of the U.S. citizenry as well as the recent immigrant population (Brittenham and de la Cruz, 2004). For example, life expectancy in Eastern European countries and Russia is lower than in Western Europe (Ginther, 2009; WHO, 2009). Foodways, the eating practices and customs of a group of people (e.g., lack of vitamin C intake among Russian men), and high rates of smoking and alcohol consumption all contribute. A high incidence of more lethal cancers, particularly of lung cancer, is common in Eastern Europe (Bray et al., 2002). Men and women in these countries also have the highest mortality rates from ischemic heart disease of all the Organisation for Economic Co-operation and Development (OECD) countries (OECD, 2007). Breast cancer incidence and mortality differs across Europe, being higher in Denmark than other northern European nations (Althuis et al., 2005). These findings represent very preliminary evidence in favor of the collection and reporting of more granular ethnicity data separately for White subgroups. It remains to be seen which other White subgroups experience considerable differences in care or health outcomes, and collecting granular ethnicity data will make the picture clearer.

Page last reviewed October 2014
Page originally created September 2012
Internet Citation: Chapter 2: Evidence of Disparities Among Ethnicity Groups (continued). Content last reviewed October 2014. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/research/findings/final-reports/iomracereport/reldata2a.html