Hidden Heterogeneity: How the White Racial Category Masks Interethnic Health Inequality

Abstract

Inequality research has often used non-Hispanic Whites as the reference category in measuring U.S. racial and ethnic health disparities, with less attention paid to diversity among Whites. Immigration patterns over the last several decades have led to greater ethnic heterogeneity among Whites, which could be hidden by the aggregate category. Using data from the National Health Interview Survey (2000–2018), we disaggregate non-Hispanic Whites by nativity status (U.S.- and foreign-born) and foreign-born region of birth (Europe, Former Soviet Union, and the Middle East) to examine diversity in health among adults aged 30+ (n = 290,361). We find that foreign-born Whites do not have a consistent immigrant health advantage over U.S.-born Whites, and the presence of an advantage further varies by birth region. Immigrants from the Former Soviet Union (FSU) are particularly disadvantaged, reporting worse self-rated health and higher rates of hypertension (high blood pressure) than U.S.-born and European-born Whites. Middle Eastern immigrants also fare worse than U.S.-born Whites but have health outcomes more similar to European immigrants than to immigrants from the FSU. These findings highlight considerable diversity in health among White subgroups that is masked by the aggregate White category. Future research must continue to monitor growing heterogeneity among Whites and consider more carefully their use as an aggregate category for gauging racial inequality.

Introduction

Whites have been the numerically, politically, and socially dominant racial group throughout U.S. history. Congress first invoked the White racial category in 1790 to legally exclude non-Whites from gaining U.S. citizenship (Haney-López 2006). Progressively racist policies throughout the nineteenth and early twentieth centuries further limited U.S. citizenship rights to lighter skinned Whites from the Northern and Western Hemispheres (i.e., Western Europe). Immigration legislation until the 1960s used a restrictive national-origin quota system to control emigration from the Southern and Eastern Hemispheres, ensuring that Whites were a relatively homogeneous population (Budiman et al. 2020; Cohn 2015). But far from being an objective and consistent category, who was classified as White changed over time as new groups arrived in the United States and the racial order was reorganized to maintain a status hierarchy (Haney-López 2006; Hochschild and Powell 2008; McDermott and Ferguson 2022).

The 1965 Immigration and Nationality Act was particularly pivotal in ushering in new racial and ethnic immigrant groups, and the 1964 Civil Rights Act helped pave the way for more active assessments of racial inequities that were based on comparisons with non-Hispanic Whites (hereafter, Whites). Subsequent federal policies enacted by the U.S. Census Bureau solidified a racial classification system that relied on six broad categories to assess disparities: White or Caucasian, Black or African American, Latino or Hispanic, Asian American, Native Hawaiian and Pacific Islander, and American Indian and Alaska Native (Revisions to the Standards 1997). Whites, specifically U.S.-born Whites, became the default reference category in inequalities research—a practice that remains widely used today (Kauh et al. 2021; Read et al. 2021). Population-level evidence on racial and ethnic health disparities, for example, is based mainly on comparisons between the U.S.-born White population and other groups, such as Black and Hispanic Americans (U.S. Department of Health and Human Services 2022).

Several shifts in the demographic composition of Whites over the past several decades render their use as a reference group problematic. First, Whites in the United States are projected to be a numerical minority by 2040 (Vespa et al. 2020), which begs the question of the appropriate reference category in studies of inequality, particularly for policies driven by numerically larger groups. Second, the ethnic composition of Whites has changed dramatically since the 1965 Immigration and Nationality Act and is far less homogeneous than typically assumed (Iceland 2014; Ro et al. 2015). Whites include persons who trace their ancestries to Europe, the Middle East, and North Africa, and all federally collected data¹ over the past 45 years have classified them as such (U.S. Department of Commerce 1978:37–38). When the U.S. Census first introduced the ancestry question in the 1980 decennial census, Western European² Whites made up more than 70% of the U.S. population and nearly 60% of the White immigrant population (Farley 1991; Read 2024). By 2023, Whites from these European regions had dropped to approximately one third of the U.S. population and roughly the same (33%) of the White immigrant population (authors’ tabulations of U.S. Census data 2023; Read et al. 2021). In contrast, the proportion of Whites of Eastern European and Middle Eastern/North African descent (hereafter, Middle Eastern) has climbed steadily (Read et al. 2020). In 2023, immigrants from the two regions made up 58% of the foreign-born White population, and for the first time, Eastern Europeans surpassed Western Europeans as the single largest White immigrant group (authors’ tabulations of U.S. Census data 2023).

Changes in the ethnic origins of Whites are tied to global events that have destabilized many world regions and altered the characteristics of newer immigrant arrivals. Relative newcomers from Eastern Europe and the Middle East are more linguistically, phenotypically, and culturally diverse than their Western European predecessors, and many have migrated under challenging circumstances owing to political and civil unrest in their countries of origin (Engelman et al. 2017; Ro et al. 2015). A large swath of Eastern Europeans come from the Ukraine, Russia, and other countries that belonged to the Former Soviet Union (FSU),³ and most Middle Eastern immigrants originate in Iraq, Syria, and Iran (Figures A3 and A4, online appendix; authors’ tabulations from U.S. Census data 2023; Read et al. 2019). Such diversity suggests that the White category might obscure within-group inequalities between White subgroups based on birth region. In studies that focus exclusively on U.S.-born Whites, interethnic inequalities among foreign-born Whites are overlooked altogether (Bakhtiari 2018; Kindratt et al. 2022).

To date, few studies have systematically examined whether and how the aggregate White category masks heterogeneity between White subgroups from different world regions. The current study aims to address this gap by disaggregating Whites by nativity status (U.S.- and foreign-born) and birth region (Europe, Middle East, the FSU) and examining differences in self-rated health and diagnosed hypertension (high blood pressure). These dimensions of health capture subjective and objective assessments of well-being and are valid predictors of comorbidities and mortality. The analysis uses nationally representative data from the National Health and Interview Survey (2000–2018) to examine whether health differs across foreign-born Whites and U.S.-born Whites. Put differently, we ask whether White immigrants experience an immigrant health advantage over their U.S.-born White peers. We then assess whether and how birth region differentiates the health of foreign-born Whites. A key distinction we make throughout the analyses is between White immigrants from Europe, the FSU, and the Middle East. These three groups represent the major birth regions for White immigrants, and each region has distinctive sociodemographic characteristics that are likely associated with health differences. The analysis considers this possibility by examining the extent to which socioeconomic, demographic, and healthcare characteristics explain observed health disparities between White subgroups based on nativity and birth region.

Background

Compositional Changes in the Origins of Whites

U.S. immigration policies have played a key role in shaping the composition of racial and ethnic groups and are a driving force behind growing ethnic diversity among Whites (Bolter 2022). The 1790 Naturalization Act passed by Congress ruled that only “free white persons” could be naturalized as U.S. citizens. In the face of challenges to citizenship laws in the nineteenth century, federal courts had to decide who was classified as White, and they did so discriminately and narrowly based on ancestry in “Caucasian” countries, skin tone, and cultural and religious background (Haney-López 2006). Legislation through the early part of the twentieth century upheld these mandates and limited U.S. citizenship rights to Whites, broadly understood as immigrants arriving from Northern and Western Europe (Massey 1995). Immigrants from these regions were overwhelmingly lighter skinned Whites of Christian origin who fit into the Eurocentric immigration model, and discriminatory policies passed in the 1920s continued to favor them over immigrants from all other world regions (Cohn 2015). Nearly 70% of all entry visas up until the mid-1960s were allotted to White immigrants from Germany, Ireland, and Great Britain (Budiman et al. 2020). As a result, Western European immigrants and their offspring made up the lion's share of the total U.S. population through the 1980s (Farley 1991).

Several watershed events in the 1960s altered the composition of White immigrants arriving in subsequent decades. The 1965 Immigration and Nationality Act abolished the national origin quota system and, for the first time, placed an annual cap on arrivals from Western and Northern Europe (Iceland 2014). The act also established a new preference category for family reunification that, together with numerical restrictions on the Western Hemisphere, began a slow and steady decline in immigration from most European regions. In 1960, Europeans of non–Eastern European origin made up more than one half (58%) of the White immigrant population; by 2023, the proportion had dropped to one third (Figure 1). Immigrants from Eastern Europe initially declined from 28% of White immigrants in 1960 to 21% in 1990 (i.e., the last decade of the Cold War) before jumping to 31% in the years following the dissolution of the FSU. In 2023, Eastern Europeans composed one third (33%) of the White immigrant population, and nearly one half (16%) originated from countries belonging to the FSU (authors’ tabulations from American Community Survey [ACS] data). Middle Eastern immigrants likewise grew exponentially from less than 2% of the White immigrant population in 1960 to 25% in 2023.

Today, White immigrants represent 18% of the total U.S. foreign-born population and are projected to make up 20% by 2060 (Read 2024; Vespa et al. 2020). The majority are immigrants from the FSU and Middle East who arrived in the United States after 1990 (Figure 2; authors’ tabulations from the Yearbook of Immigration Statistics).⁴ This period is critical for understanding diversity among Whites: global events during the late 1980s and early 1990s transformed the size and characteristics of subsequent immigrant arrivals. The establishment of the European Union in 1993 provided greater economic security in Europe that led to further declines in annual admissions, whereas the fall of the FSU (1991), the Iran–Iraq War (1980–1989), and the first Gulf War (1991) generated economic and political instabilities that had the opposite effect (Abuelezam et al. 2018; Read et al. 2021). Emigration from the FSU skyrocketed in the immediate aftermath of its collapse (Figure 2).

Between 1992 and 1995, more than 220,000 immigrants arrived from the FSU—a fourfold increase over the prior decade (authors’ tabulations from the Yearbook of Immigration Statistics). Most FSU immigrants arriving in 1992–1997 were refugees from Russia and Ukraine, with smaller numbers entering through family, diversity, and employment pathways (Figure A1, online appendix). The proportion of FSU immigrants arriving as refugees peaked again in 2000–2001 and 2005–2006, with all other years dominated by admissions based on family reunification. Emigration from the Middle East likewise grew, with a spike in admissions from Iran in 1989 and 1990 and growth from Iraq beginning around the time of the U.S.-led invasion of Iraq (1991). The chief pathway of entry for Middle Eastern immigrants during the 1990s was through family reunification (Figure A2, online appendix), whereas the decade of 2009–2018 saw substantial growth in the number arriving as refugees.

New global conflicts emerged at the beginning of the twenty-first century, exacerbating tensions between the United States and several countries in the FSU and Middle East. The terrorist attacks on September 11, 2001, and the rise of ISIS throughout the Middle East led to more restrictive immigration policies and the creation of the U.S. Department of Homeland Security (2002). Annual admissions dropped substantially for all immigrant groups between 2002 and 2004 before rebounding to pre–9/11 levels in 2005. The U.S. military's drawdown in Iraq (2008), uprisings during the Arab Spring (2011), and the Syrian civil war (2011 to the present) displaced hundreds of thousands of individuals and led to an uptick in the number of refugees arriving from the Middle East that lasted until the first Trump administration (2017–2021) (Figure A2, online appendix). Between 2010 and 2017, immigrants from Iraq (45%) and Syria (75%) were among the fastest-growing groups, far outpacing the average growth for all immigrants (11%) (Zong and Batalova 2019). A majority had been deprived of opportunities in their origin countries and arrived with lower levels of educational attainment and English language proficiency and higher poverty rates (Harjanto and Batalova 2022).

Emigration from the FSU was likewise affected by regional disputes emanating from the Russian annexation of Crimea (2014) and, more recently, Russia's invasion of Ukraine (2021 to the present). Annual admissions from Ukraine began climbing in 2015 and remained steady during the first Trump administration, even increasing slightly in 2018 and 2019 (Figure 2 and Figure A3, online appendix). In contrast, the ban on many Muslim-majority countries resulted in a marked decline in emigration from the Middle East: between 2018 and 2020, annual admissions from Iraq dropped by 82%, and those from Syria dropped by 78% (Figure A4, online appendix).

The United States’ involvement in many of the aforementioned conflicts and its toll on the U.S. military and economy have exacerbated negative views on immigrants from these regions and created a more hostile context of U.S. reception (Ewing 2008; Jamal and Naber 2008). Anti-immigration policies and rhetoric under the first Trump administration (2017–2021) popularized the belief in the “right kind of immigrant”—one who was White, Christian, and of European origin (Gjelten 2018). Trump's equivocal stance toward Russia's invasion of Ukraine, validation of President Putin's aggressive policies, and the Muslim travel ban further heightened awareness of and discrimination against emigrants from the Middle East and FSU (Awad et al. 2019; Smith 2022). Unlike their European predecessors, many are phenotypically darker skinned, speak a native language other than English (e.g., Arabic), and arrived as refugees or through family reunification (Figures A1 and A2, online appendix).

Health Implications of Ethnic Diversity Among Whites

Despite growing ethnic diversity among Whites, analyses of inequality tend to use predefined federal racial and ethnic categories (Monk 2022). As a result, most research on Whites examines them in the aggregate and frequently as the reference category to measure racial and ethnic health disparities in the United States (Kauh et al. 2021). For example, the annual National Healthcare Disparities Report uses non-Hispanic U.S.-born Whites as the comparison group in evaluating health disparities (U.S. Department of Health and Human Services 2022). This practice stems from Whites’ privileged status that was codified by the federal government in the wake of the Civil Rights Movement (Read et al. 2021). Federal agencies were tasked with enforcing civil rights laws and needed consistent and comparable data on race and ethnicity to ensure the equitable distribution of resources. In 1977, the White House OMB issued federal standards to guide the collection and analysis of all data on race and ethnicity. Critical to the classification of Whites is that the standards defined them as “person(s) having origins in any of the original peoples of Europe, North Africa, or the Middle East” (U.S. Department of Commerce 1978:37). However, when the standards were created, most Whites were lighter skinned, U.S.-born citizens of Northern and Western European descent, with relatively few from other world regions.

Since then, the proportion of Whites of Western European origin has been declining across all segments of the U.S. population. Lower fertility and higher mortality rates among U.S.-born Whites, combined with increased emigration from Eastern Europe and the Middle East, have contributed to these fluctuations (Vespa et al. 2020). Moreover, the classification of certain groups as “White” has come under intense scrutiny, especially in the case of Middle Eastern immigrants (Abboud et al. 2019). In line with long-standing criticisms of U.S. Census categories (Prewitt 2013, 2018), many have argued that the legal and statistical definitions of racial categories do not always reflect the diversity of lived experiences and the social outcomes of individuals classified within them (Monk 2022). Further, a tendency to focus on the U.S.-born population has resulted in much less attention to a growing proportion of White immigrants who fit this description (Read et al. 2020).

A few studies have differentiated foreign- and U.S.-born Whites in analyses of health disparities (e.g., Antecol and Bedard 2006; Dupre et al. 2012). Studies that further disaggregated Whites by ancestry or birth country/region are scarce relative to research on diversity within other racial groups (Abuelezam et al. 2018; Read et al. 2021). The dearth of research on White immigrants is due largely to a lack of data sources with questions on ethnicity and sufficiently large samples for analyses of White subgroups. The National Health Interview Survey (NHIS) and ACS are exceptions to this pattern, and scholars have used questions on nativity, ancestry, and birth region to expand knowledge about Whites. Others have relied on state- and community-level data to parse out White ethnic groups in areas where they are highly concentrated (e.g., Arab communities in Dearborn and Detroit, Michigan).

Studies examining ethnic diversity in health among Whites broadly fall into one of two categories: (1) those focusing on health outcomes for specific White ethnic groups, such as Arab Americans or Eastern Europeans (e.g., Bulut and Brewster 2021; Dallo and Kindratt 2016; Mehta and Elo 2012); and (2) those including Whites as part of a larger analysis on ethnic diversity in health among other racial and ethnic groups, such as Hispanics or Asians (e.g., Antecol and Bedard 2006; Brown 2018; Read et al. 2021). This literature has produced mixed results regarding the presence of an immigrant health advantage among Whites. Some studies found that foreign-born Whites have little to no health advantage over their U.S.-born counterparts across multiple indicators of health, including mortality (Dupre et al. 2012), cognitive limitations (Kindratt et al. 2022), and disability (Read et al. 2019). Others found a slight health advantage among foreign-born Whites on some conditions (e.g., depression, anxiety, and hypertension) but not on others (e.g., diabetes and chronic lung disease) (Kindratt et al. 2022; Read 2024). The lack of a consistent immigrant health advantage among Whites might partly reflect U.S. Whites’ overall privileged status, which sets a high bar for identifying an immigrant “advantage.”

Inconsistent evidence of an immigrant health advantage among Whites might also reflect diversity in health among the foreign-born based on their birth region. For example, Read and colleagues (2020) found that lower disability levels among foreign-born Whites were due almost entirely to the better health of Western European immigrants, who made up a sizable portion of the population and contributed to overall average health. These averages, in turn, masked higher physical disability rates among ethnic subgroups from Eastern Europe and the Middle East. Other studies have similarly found higher rates of hypertension, disability, and poorer self-rated health among FSU immigrants relative to U.S.-born Whites (Mehta and Elo 2012; Reynolds et al. 2016; Yi et al. 2014), even after controlling for differences in educational attainment, smoking, and heavy alcohol use (i.e., standard explanations for health disparities). Emigrants from the Middle East likewise have higher rates of diabetes (Abuelezam et al. 2021), heart disease (Dallo and Kindratt 2016), poor self-rated health (Read et al. 2005), and cognitive limitations (Kindratt et al. 2022). Country-specific findings (i.e., Iraq, Syria, Ukraine) are generally consistent with overall patterns of poor health, again finding little support for a healthy immigrant effect among Whites (Read et al. 2019; Szaflarski 2023).

Explanations for the poorer health of FSU and Middle Eastern immigrants highlight historical conditions in the sending and destination countries that have led to more negative health selection. Violent conflicts throughout the FSU and Middle East have stagnated economic development and resulted in immigrants arriving with fewer socioeconomic resources, greater exposure to war-related trauma, and more accumulated stressors that harm health (Haas and Ramirez 2022; Reynolds et al. 2016). New legal modes of U.S. entry made possible through family reunification and refugee status have likewise reduced health selectivity relative to those entering through employment and diversity pathways (Ro et al. 2016). The U.S. reception context has also been more hostile to immigrants from these regions, especially toward those who look and sound different from Western European Whites (Awad et al. 2019; Pew Research Center 2015). As others have argued, not all groups classified as White share the same experience of Whiteness (Maghbouleh et al. 2022). Darker skinned, non-English-speaking immigrants have been most affected by anti-immigrant sentiment and xenophobia, reporting greater discrimination and poorer mental and physical well-being than U.S.-born Whites (Abdulrahim et al. 2012; Bulut and Brewster 2021). A recent study on maternal and infant health, for example, found that the 2017 Muslim travel ban increased preterm births among women born in the banned countries but had no impact on preterm births among U.S.-born White women (Samari et al. 2020).

Taken together, the literature suggests growing health disparities among Whites that deserve further attention. Specifically, we aim to address several unknowns in the current study. First, few studies have moved beyond comparing foreign- and U.S.-born Whites to consider how birth region shapes health diversity among White immigrants from different regions. Studies on immigrants from the FSU and the Middle East found poorer health outcomes for these groups relative to U.S.-born Whites, but they did not clarify how immigrants from these regions fare relative to each other or to White immigrants born in Europe. European immigrants have long dominated the ethnic origins of Whites but are largely missing from analyses of health disparities among Whites. A comparison of the three major origin regions (Europe, the FSU, and the Middle East) could provide useful insight into health patterns among White immigrants, particularly for a growing segment from the FSU and the Middle East and North Africa (MENA). Finally, to our knowledge, no study has included information on the demographic and migration characteristics of White immigrants by region of origin. Doing so in the same analytic framework as health outcomes could provide a more complete picture of health diversity among Whites.

Current Study

This study makes several contributions to the literature on racial and ethnic health disparities and immigration and health. By focusing on non-Hispanic Whites, we extend recent work underscoring growing ethnic heterogeneity within racial groups, most of which has concentrated on racial and ethnic minorities (for a review, see Kauh et al. 2021). We also challenge the common practice of using an aggregated non-Hispanic White category as a reference group to measure racial gaps in health (Read et al. 2021). We capitalize on data from several nationally representative sources to create a more robust account of the demographic and immigration characteristics of Whites by birth region. We then analyze data from multiple waves of the NHIS to examine the extent to which nativity and birth region affect health outcomes among White adults. Our primary objective is to assess diversity in health between Whites born in Europe, the FSU, and the Middle East. We also evaluate broad patterns by nativity to determine whether a health advantage exists among foreign-born Whites relative to U.S.-born Whites. In both cases, we test whether differences in sociodemographic, immigration, and healthcare characteristics impact observed health disparities. More formally stated, our analyses focus on the following research questions:

Data and Methods

NHIS Data

To answer these questions, we use merged data from the 2000–2018 NHIS, an annual multipurpose health survey conducted by the National Center for Health Statistics and the Centers for Disease Control and Prevention and administered by the U.S. Census Bureau. The NHIS is one of the few nationally representative datasets that contain sufficient information and sample sizes to disaggregate non-Hispanic Whites by nativity and birth region. The NHIS releases data on birth region, rather than country, to protect respondents’ anonymity and confidentiality. Data were obtained from IPUMS, which integrates person and sample adult data files to allow for complete information on each individual (Blewett et al. 2022).

The sample is based on data from U.S.-born (n = 280,265) and foreign-born (n = 10,096) non-Hispanic White adults aged 30 or older. Foreign-born adults exclude individuals born in a U.S. territory or born abroad to U.S. parents. We further disaggregate the foreign-born into the three largest birth regions: Europe (n = 7,415), the Middle East (n = 1,467), and the FSU (n = 1,214).⁵ The NHIS category for Europe includes Northern, Western, Southern, and some Eastern European countries (e.g., Poland and Romania); it also separates other Eastern European countries that belonged to the FSU (e.g., Russia, Ukraine). For consistency with the NHIS definitions, we use the terms “Europe” and “FSU” throughout the analyses. Finally, we exclude other birth regions because of small sample sizes for non-Hispanic Whites (e.g., Asia, Africa, South America). Notably, the survey questionnaire was administered in English and Spanish but not in other languages common among White ethnic subgroups, such as Arabic and Russian. Thus, our results are likely conservative estimates of health disparities owing to the underrepresentation of groups with low English language proficiency. Language barriers are known to restrict health service access and utilization, decreasing the likelihood of receiving adequate healthcare (Read and Smith 2018).

Region of Origin Characteristics

We supplement NHIS data with information from several other data sources to provide more demographic details on the three regions. First, we use data from the 1986–2022 Yearbook of Immigration Statistics and from the 1960–2023 decennial census and the ACS to isolate the top sending countries in each region (Figures A3–A5, online appendix). For Europe, the largest origin countries are England, Germany, and Ireland; for the Middle East, the countries are Iran, Iraq, and Lebanon, with a growing number from Syria; and for the FSU, Russia and Ukraine are, by far, the largest countries of origin. We also use these data to (1) estimate the ethnic composition of White immigrants over time (Figure 1), (2) calculate annual admissions by region of origin (Figure 2), and (3) identify the major admissions classes (i.e., visa categories) for FSU and Middle Eastern immigrants (Figures A1 and A2, online appendix).

Measures

Our dependent variables include two indicators of health status, and both are coded to capture poor health. First, self-rated health gauges respondents’ subjective assessments of their well-being and is known to be a valid predictor of morbidity and mortality (Idler and Benyamini 1997). The self-reported health responses are excellent, very good, good, fair, and poor. We dichotomize responses into fair/poor health and excellent/very good/good health (reference category). Because self-rated health might be affected by cultural differences in responses, we include a second, objective indicator of health status that requires a healthcare professional's diagnosis: hypertension (i.e., high blood pressure). Hypertension affects almost half of all U.S. adults and is a precursor for several leading causes of death, including heart attack and stroke (Adams and Wright 2020). It is largely asymptomatic and requires a doctor's diagnosis for detection and treatment (Centers for Disease Control and Prevention 2024). The question reads, “Have you ever been told by a doctor or other healthcare professional that you had hypertension, also called high blood pressure?” We code responses as “yes, has been diagnosed” and “no, has not been diagnosed” (reference category). In ancillary analyses, we included heart disease as an additional dependent measure and found patterns that were substantively similar to those of the other health indicators but with sample sizes too small for meaningful interpretation (results available on request).

The key independent variables are nativity and birth region. Nativity is coded as U.S.-born or foreign-born, and birth region classifies the foreign-born as having been born in Europe, the Middle East, or the FSU. We include an indicator for arrival cohort to account for potential differences among immigrants based on the period of U.S. arrival (reference = before 1990; 1 = 1991–2000; 2 = 2001–2018). We also include several sets of independent variables known to influence health. We are mainly interested in the roles of socioeconomic status and healthcare access, given that they are interrelated mechanisms with policy implications for understanding health diversity among Whites. Socioeconomic status is captured with a categorical measure of educational attainment (reference = less than high school; 1 = high school graduate or equivalent; 2 = some college; 3 = college degree or more) and a measure of poverty status (reference = above poverty threshold; 1 = below poverty threshold). We include a “missing” category for poverty to account for the high number of missing values (n = 38,442; 13.2%).

Healthcare access is captured with three variables: health insurance coverage (reference = no coverage), usual place of care (reference = no), and having seen a doctor or other healthcare professional in the past 12 months (reference = no). Finally, the analysis controls for several demographic and health behaviors: age and age squared (continuous in years), sex (reference = male), marital status (reference = currently married), children in the home (reference = no), parents in the home (reference = no), U.S. citizenship (reference = noncitizen), region of U.S. residence (reference = Northeast), body mass index (reference = normal), drinking status (reference = never), and smoking status (reference = never).

Analytic Strategy

The analysis begins with descriptive comparisons of our dependent and independent variables, separately by nativity and birth region (Table 1). We use Pearson chi-square and Wald tests with Rao and Scott's second-order correction and Wilcox rank-sum test for complex surveys to identify significant differences between U.S.- and foreign-born Whites (RQ1), as well as between U.S.-born Whites and Whites born in Europe, the Middle East, and the FSU (RQ2). We also compare respondents from the Middle East and the FSU with those from Europe to examine the extent of variation among foreign-born Whites (RQ3). These descriptive analyses provide initial evidence for our research questions.

Next, we run a series of logistic regression models estimating self-rated health (Table 2) and diagnosed hypertension (Table 3) by birth region. As a robustness check, we also ran the analyses using alternative specifications of our health measures: once with a combined measure (i.e., fair/poor health and hypertension) and once with a measure of either condition (i.e., fair/poor health or hypertension). Because the substantive results were similar in both cases and were harder to interpret, we kept the measures separate. Tables 2 and 3 first use U.S.-born Whites as the reference group to determine whether birth region differentiates health for White immigrants relative to their U.S.-born peers (RQ2). We then limit the sample to foreign-born respondents and use European-born Whites as the reference category (Table 4, RQ3) to look more closely at variation among White immigrants, beyond variation relative to U.S.-born Whites.

Each regression table uses a model-building sequence that identifies the utility of several sets of independent variables (RQ4) in explaining observed variation in predicted health (e.g., Reynolds et al. 2016). We evaluated the Bayesian information criterion for each model to aid in model selection. Model 1 examines the impact of birth region, adjusting for age and gender; Model 2 adds socioeconomic and immigration characteristics; and Model 3 adds health behaviors and healthcare access. All models control for the survey year. Tables 2, 3, and 4 display odds ratios (ORs) with 95% confidence intervals (CIs). Figures 3 and 4 plot the ORs for self-rated health (Table 2) and hypertension (Table 3) to ease the interpretation of the findings in the tables. The figures also plot health patterns for U.S.-born and foreign-born Whites (RQ1) before disaggregating foreign-born Whites by birth region. All analyses were conducted using R version 4.1.1 (R Core Team 2021) and the survey package (Lumley 2004, 2021) to incorporate NHIS sample weights.

Results

Descriptive Statistics

Table 1 presents descriptive statistics for U.S.- and foreign-born Whites in the aggregate and separately by birth region. Foreign-born Whites do not differ significantly from U.S.-born Whites in their self-rated health (14.0% and 13.4%, respectively, report fair or poor health), but they report lower levels of diagnosed hypertension (31.3% vs. 35.2%). When the results are disaggregated by birth region, FSU-born Whites have significantly higher rates of fair/poor self-rated health (20.3%) than U.S.-born Whites (13.4%), European-born Whites (12.3%), and Middle Eastern–born Whites (16.1%). FSU immigrants also have the highest rate of diagnosed hypertension (33.6%), which is significantly higher than that of Middle Eastern immigrants (27.2%) but similar to that of European-born (32.0%) or U.S.-born (35.2%) Whites.

European-born Whites have a more established history in the United States. Among these individuals, 27.5% arrived in the earliest cohort (before 1990), compared with 16.1% and 9.6% of Middle Eastern and FSU immigrants, respectively. A smaller proportion (20%) of European-born Whites arrived in the most recent cohort (2001–2018) relative to Middle Eastern (33.5%) and FSU (30.5%) immigrants. FSU immigrants are less likely to be U.S. citizens (68.7%) than European (73.4%) and Middle Eastern (71.0%) immigrants. Turning to socioeconomic status, FSU (61.8%) and Middle Eastern (51.8%) immigrants are significantly more likely than European (38.8%) immigrants to have obtained a college degree or more. However, both groups have higher poverty rates (13.0% and 14.9%) than European immigrants (5.1%), who have rates similar to those of U.S.-born Whites (5.4%).

In terms of healthcare access, foreign-born Whites are significantly more likely than U.S.-born Whites to lack health insurance coverage (10.1% vs. 7.6%) and are more likely to rely on public (10.0% vs. 5.8%) rather than private (67.3% vs. 74.2%) health insurance programs. Relatedly, foreign-born Whites are less likely to have a usual place of care and to have seen a healthcare professional in the prior year. Among the foreign-born, however, European Whites have similar rates of health insurance coverage as U.S.-born Whites and are only slightly less likely to have a usual place of care or to have seen a doctor in the prior year. The gap between U.S.-born Whites and Middle Eastern and FSU immigrants is significantly larger, with rates of being uninsured nearly twice as high for these immigrant groups (13.2% and 13.1%, respectively). Middle Eastern and FSU immigrants are likewise more disadvantaged in terms of having a usual place of care and having seen a healthcare professional in the past year. Overall, the findings in Table 1 suggest that European immigrants track closely with U.S.-born Whites on most characteristics, whereas Middle Eastern and, especially, FSU immigrants appear more disadvantaged.

Diversity in Immigrant Health Relative to U.S.-born Whites

Table 2 provides results from logistic regression models predicting self-rated health by birth region (RQ2).⁶ Results for Model 1 show that relative to U.S.-born Whites, Middle Eastern and FSU immigrants have significantly higher odds of reporting fair/poor self-rated health, whereas immigrants from Europe have significantly lower odds. The inclusion of socioeconomic and immigrant characteristics (Model 2) reduces the European advantage to nonsignificance (OR = 0.97, 95% CI = 0.85, 1.10), does little to affect the poorer health of Middle Eastern immigrants (OR = 1.43, 95% CI = 1.09, 1.87), and considerably amplifies the poorer health of FSU immigrants (OR = 2.71, CI = 2.13, 3.44) relative to U.S.-born Whites. In the fully adjusted model that includes health behaviors and healthcare characteristics (Model 3), European and Middle Eastern Whites no longer differ statistically from U.S.-born Whites in self-reported health. However, FSU immigrants remain more than twice as likely as U.S.-born Whites to report poor health (OR = 2.52, 95% CI = 1.95, 3.24).

A similar story emerges for FSU immigrants with diagnosed hypertension (Table 3). European Whites again exhibit better health than U.S.-born Whites (OR = 0.78, 95% CI = 0.73, 0.84), as do Middle Eastern immigrants (OR = 0.76, 95% CI = 0.64, 0.90). Accounting for differences in education, poverty, arrival cohort, and citizenship status (Model 2) and health behaviors and healthcare (Model 3) decreases the European and Middle Eastern advantage slightly, but both immigrant groups maintain significantly better health than U.S.-born Whites. In contrast, the inclusion of these factors increases the gap between FSU immigrants and U.S.-born Whites when moving from Model 1 (OR = 1.04, 95% CI = 0.88, 1.22) to Model 3 (OR = 1.29, 95% CI = 1.04, 1.59). In sum, Tables 2 and 3 indicate that the descriptive disadvantage for FSU immigrants seen in Table 1 is amplified in the multivariate context, even when socioeconomic, immigrant, health behavior, and healthcare characteristics are considered. In contrast, Middle Eastern and European immigrants are significantly less likely than U.S.-born Whites to report diagnosed hypertension in the baseline model (Model 1), and the advantage holds for both groups in the fully adjusted model.

Figures 3 and 4 plot the ORs for self-rated health and hypertension, respectively, by nativity and birth region. The black triangles represent the baseline model, Model 1, which is adjusted for survey year, age, age squared, and gender; the blue circles represent Model 3, which is the fully adjusted model. All ORs are depicted with 95% CIs. In the baseline model of Figure 3, foreign-born Whites do not differ significantly from U.S.-born Whites in terms of self-rated health. However, after we disaggregate the foreign-born by birth region, European-born Whites have significantly lower odds of fair/poor self-rated health, and Middle Eastern– and FSU-born Whites have significantly higher odds of fair/poor self-rated health. In the fully adjusted model (Model 3), foreign-born Whites have significantly higher odds of fair/poor self-rated health, which is primarily driven by the poorer health of FSU-born Whites, the only group that remains significantly different from U.S.-born Whites.

Figure 4 displays the same structure of results for diagnosed hypertension. In the baseline model, foreign-born Whites have significantly lower odds of hypertension than U.S.-born Whites. Both European- and Middle Eastern–born Whites have lower odds of hypertension than U.S.-born Whites, whereas FSU-born Whites do not differ significantly from U.S.-born Whites. For the fully adjusted model (Model 3), foreign-born Whites, European-born Whites, and Middle Eastern–born Whites maintain their statistically lower odds of hypertension. However, the odds of hypertension among FSU-born Whites increase and are significantly higher than that among U.S.-born Whites.

Diversity in Immigrant Health Relative to European-born Immigrants

Table 4 more closely examines variations in health among the foreign-born using European-born Whites as the reference category (RQ3). As seen in panel A for self-rated health, Middle Eastern (OR = 1.93, 95% CI = 1.47, 2.55) and FSU immigrants (OR = 2.43, 95% CI = 1.97, 3.00) have higher odds of reporting fair/poor self-rated health than European-born Whites. Controlling for socioeconomic and healthcare characteristics (RQ3) diminishes the effect for Middle Easterners to nonsignificance (OR = 1.15, 95% CI = 0.85, 1.55) but does little to modify the disadvantaged health status of FSU immigrants (OR = 2.03, 95% CI = 1.59, 2.60). A similar story emerges for diagnosed hypertension in panel B: the poorer health of FSU immigrants relative to European-born Whites is amplified slightly moving across the models. FSU immigrants have 1.39 times the odds of diagnosed hypertension in Model 1 (95% CI = 1.15, 1.69) but 1.50 times the odds in Model 3 (95% CI = 1.22, 1.84). Middle Eastern immigrants, on the other hand, do not differ significantly from European-born immigrants across model specifications.

Discussion and Conclusion

This study aimed to examine ethnic diversity in health among non-Hispanic Whites, a group typically treated in the aggregate and as the reference category for measuring U.S. health disparities. Our findings indicate considerable heterogeneity in health among White subgroups—heterogeneity that is hidden by the aggregate White category. When differences are examined by nativity, White immigrants do not have a consistent health advantage over U.S.-born Whites (i.e., healthy immigrant effect; RQ1). White immigrants fare worse on self-rated health but better on diagnosed hypertension (high blood pressure). These patterns are further complicated when foreign-born Whites are disaggregated by birth region. European immigrants most closely resemble U.S.-born Whites across sociodemographic and health indicators. Similarities between European immigrants and U.S.-born Whites are not altogether surprising, given that previous cohorts of European immigrants largely assimilated to a cultural Whiteness that increased their social standing in the U.S. racialized social system (Barrett and Roediger 1997; Bonilla-Silva 1997).

In contrast, FSU immigrants have significantly worse health than U.S.-born Whites (RQ2) and their European-born peers (RQ3). Middle Eastern immigrant health is even more variable, depending on the health outcome and reference category in question. The poorer health of FSU immigrants is particularly troubling in light of Russia's invasion of Ukraine in 2021 and the ensuing war that continues today. The Ukrainian population in the United States recently surpassed 1 million (Szaflarski 2023), and immigrant admissions from Ukraine and other FSU countries have been relatively high for the past decade. Consequently, a growing number of FSU immigrants in the United States will likely have been affected by the conflict, either directly through displacement or indirectly through family and social ties. They will have arrived through pathways that are less selective on health status (e.g., refugee and family reunification), which could both contribute to and explain their poorer health (Ro et al. 2016).

The health of Middle Eastern immigrants is also noteworthy, not only in establishing diversity in health among Whites but also for illustrating how the legal and statistical definition of Whites can encompass groups that are not necessarily considered White in their everyday lives (Maghbouleh et al. 2022). The case of Middle Eastern immigrants is consistent with broader criticisms of using taken-for-granted racial categories in analyses of inequality (e.g., Monk 2022). For example, the addition of a MENA response category on the race and ethnicity question in the U.S. Census—separate from their historical classification as White—reflects years of community advocacy and mounting research evidence of racialized and discriminatory experiences faced by MENA groups that distinguish them from Whites (Abboud et al. 2019). Scholars have effectively argued that the failure to classify MENA separately from Whites undermines the identification of vulnerable groups that are collapsed in the broad White category (American Medical Association 2021; Awad et al. 2022).

The addition of a new MENA response category in the U.S. Census might amplify knowledge about a heretofore hidden White subgroup (Revisions to OMB 2024). At the same time, the 2024 OMB revisions to the classification of U.S. racial and ethnic populations—including a combined race and Hispanic ethnicity question—must be viewed with caution. Whether respondents choose to check the MENA option in combination with or isolation from another racial category (e.g., White, Black) or perhaps skip the question altogether remains to be seen. Understanding the characteristics of respondents in the new census categories compared with those responding to the older census formats (e.g., ancestry questions and write-in responses to race) will be critical for establishing confidence in the reliability and comparability of data over time. For example, if we begin to see changes in racial gaps in health, we would want to know whether they reflect actual changes in health or if they are due to compositional changes in racial groups that result from the altered measurement of race and ethnicity.

Our findings also highlight several changes in the ethnic composition of Whites that have broader implications for knowledge about the White racial category. First, Western European Whites make up a declining proportion of the White population across the life course. They have lower fertility rates than non–Western European Whites, make up a smaller percentage of the White workforce, and represent a dwindling segment of the White elderly population (Vespa et al. 2020; authors’ tabulations of ACS data)—all of which have significant implications for U.S. population health and the healthcare system. These declines are partly due to a growing segment of second- and third-generation Whites who identify as American rather than European (Read et al. 2021), but they are much more attributable to population aging and immigration. By 2030, international migration will surpass natural population increase (via fertility) as the main driver of U.S. population growth (Vespa et al. 2020). Second, and relatedly, change among Whites will continue to be shaped by growth in the Eastern European and MENA populations. Thus, understanding diversity within the White population will be essential for accurately identifying vulnerable subgroups and for disentangling more established Western Europeans from newer immigrant arrivals.

This study has limitations, some of which offer fruitful opportunities for future research. First, we could not disaggregate White immigrants by their birth country or visa admission type owing to the lack of data available in the NHIS. We attempted to address the limitation by drawing on external data sources, such as the Yearbook of Immigration Statistics, to provide contextual information on characteristics not included in the NHIS. The external data are not linked to the NHIS, and our tabulations and analyses should be interpreted as supplementary to the regression results. Second, the NHIS does not collect ancestry information on U.S.-born Whites, precluding comparisons with second- and third-generation Whites. Third, responses to self-rated health could vary across ethnic subgroups owing to cultural differences in meaning; we addressed this possibility by including an objective measure of hypertension (high blood pressure) and conducting robustness checks using alternative specifications of health (e.g., heart disease). We found substantively similar patterns, giving us additional confidence in the findings. Finally, our analyses focused on non-Hispanic Whites and excluded a growing number of Whites who identify as Hispanic (Lopez et al. 2024). Our sample was selected to achieve the primary goal of debunking homogeneous representations of non-Hispanic Whites in inequalities research.

Overall, this study highlights growing ethnic heterogeneity among Whites that complicates approaches that treat them in the aggregate. Continuing to use the aggregate category might produce biased or misleading estimates of racial inequality between Whites and other populations, especially when comparing trends over time. Trends assume relative stability in the composition of populations, which, as we have shown, is not the case for Whites. Beyond implications for interethnic inequality, the aggregate White category can also mask the unique needs of immigrant subgroups classified as White, such as those from the Former Soviet Union. Future projections indicate that immigrants and their descendants will make up 88% of U.S. population growth through 2065. Therefore, understanding ethnic heterogeneity within racial populations will remain a pressing concern for years to come. Moving forward, we must continue to monitor changes in the ethnic origins of Whites and examine more closely the standard practice of using them in the aggregate to measure race-based health disparities.

Acknowledgments

This study was supported by a pilot grant from the Center for Population Health and Aging and received support from the Centers on the Demography and Economics of Aging Program award (P30 AG034424) by the Division of Behavioral and Social Research at the National Institute on Aging.

Notes

References