Will the Health Status of the Changing Hispanic Population Remain ‘Paradoxical’?

The term “Hispanic paradox” was originally associated with an “epidemiologic paradox” observed between socioeconomic status and health status in the mid-1980s.¹ Specifically, Markides and Coreil noted that health outcomes of the Hispanic/Latino population, mostly Mexican Americans, in the southwestern region of the United States, was more similar to non-Hispanic White people than non-Hispanic Black or African American people.¹ Notably, Hispanic/Latino people then and now shared more similarities with non-Hispanic Black or African American people with regard to socioeconomic status than with non-Hispanic White people. Better outcomes for infant mortality, overall life expectancy, adult mortality for all causes and cardiovascular diseases and cancers specifically, and functional health served as the foundation for what is commonly referred as the “Hispanic paradox” and sometimes the “Latino health paradox.”

Several explanations have been proposed for this paradoxical observation, including cultural practices, family support and ties, and a healthy or selective immigrant effect.¹ Other explanations have emerged over the years, including healthy dietary habits, acculturation, genetic factors, ethnic enclaves, data reliability, and death misclassification.² While the health advantage continues to be observed among Mexican Americans (the largest Hispanic/Latino subgroup) and for foreign-born and recent immigrants, the Hispanic/Latino population is not only very heterogeneous but also is experiencing demographic changes such as aging, country of origin, and nativity status over the years, which may have implications for the paradox.³ In this editorial, we discuss the implications of these changes together with socioeconomic position (SEP), access to health care, immigration patterns, and racial dynamics for the Hispanic paradox, and for racial/ethnic inequities in the United States for years to come.

GROWTH AND CHANGES IN THE HISPANIC POPULATION

The Hispanic/Latino population represents the only ethnic group recognized in the United States and the largest minoritized subgroup of the US population (hereafter referred to as Hispanic). As of 2022, Hispanic people constituted approximately 63.7 million or 19% of the US population representing a 77% growth rate since 1980 (14.5 million).⁴ This growth comes with an increase in the heterogeneity of a population considered to be monolithic or homogenous under the Hispanic ethnicity self-identification US Census category. This heterogeneity includes country of origin, nativity status, immigration patterns, language proficiency, educational attainment, and racial self-identification.

While Mexican Americans (37.4 million) continue to be the largest subgroup within the Hispanic population (63.7 million in 2022),⁴ Hispanics self-identified as coming from or having ties with at least 19 countries in Latin America and the Caribbean in the 2020 US Census. The next subgroups with more than 2 million people are Puerto Ricans, Salvadorians, Cubans, and Dominicans. However, populations originating in Venezuela (169%), Dominican Republic (60%), Guatemala (60%), Honduras (54%), and Colombia (46%) grew at least twice as much as the Hispanic population as a whole (23%) from 2010 to 2021,⁵ with the Mexican-origin population (13%) experiencing the smallest increase among all Hispanic subgroups. In 2021, the Hispanic population constituted 43.9% of the total US foreign-born population (44.8 million).⁶ However, the share of foreign-born people among the Hispanic population decreased between 2010 (37%) and 2021 (32%), with people from Mexico showing the greatest decrease (from 36% to 29%).⁵

While immigration to the United States from Latin America and the Caribbean was first linked to political, economic, and better life opportunities between 1960 and 1990, the latest waves of immigrants represented a mix of traditional immigrants and increasing numbers of refugees and asylum seekers.⁷ The latter groups may be the result of the sociopolitical conditions of the home countries (e.g., Venezuela).⁷ Notably, early waves of immigrants were mostly younger people and men who tended to be health-selected, whereas recent waves are composed of immigrants ranging in age across the lifespan and people with poor health status.

Because of their history of colonization and ancestry, the Hispanic population can be of any race and self-identify with multiple US Census racial categories.³ In the 2020 US Census, most Hispanics self-identified with “Some Other Race” (42.2%) followed by White (20.3%), American Indian and Alaska Native (2.4%), and Black (1.9%). It is worth noting that racial self-identification varies depending upon country of origin: people from Argentina (32%) and Chile, Costa Rica, Cuba, and Mexico (20% for each country) are more likely to self-identify as White whereas those from Panama (17%), Dominican Republic (6%), and Puerto Rico (6%) are more likely to self-identify as Black.⁸

In general, Hispanic people are younger, have lower educational attainment levels, and are less likely to have health insurance than both non-Hispanic Black and White people. However, while Hispanics have lower median household income and a higher proportion of people living below the poverty level than non-Hispanic White people, their SEP indicators (i.e., median household income and poverty level) are better than for non-Hispanic Black people. Within the US Hispanic population, Mexican Americans are younger, less educated, and less likely to have health insurance than Puerto Ricans and Cubans.⁵ The finding of low SEP and access to health insurance for Mexican Americans is consistent with the “Hispanic paradox “observation from almost 40 years ago.

HEALTH OUTCOMES ASSOCIATED WITH THE HISPANIC PARADOX

Consistent with the Hispanic paradox,¹ Hispanics had age-adjusted death rates for all causes (523.8), heart diseases (111.3), and cancers (105.6) that are lower than those of non-Hispanic Black people (884.0, 208.6, and 173.1, respectively) and White people (739.9, 166.4, and 152.0 respectively) in 2019.⁹ Moreover, Hispanic people’s life expectancy (81.9 years) is higher than that of non-Hispanic Black people (74.8) and White people (78.8).¹⁰ Similarly, the rate of dying for infant babies of Hispanic women (4.9/1000 live births) is closer to that of non-Hispanic White (4.6) women but much lower than that for non-Hispanic Black (10.7) women.¹¹ Unfortunately, infant mortality data were only available for Mexican American (4.9), Puerto Rican (5.6), and Cuban (3.9) infants.¹¹

Even during the COVID-19 pandemic, when Hispanic and non-Hispanic Black populations were disproportionally affected,¹² Hispanic men (915.6) and women (599.8) had lower age-adjusted mortality rates than non-Hispanic White and Black men (1380.2 and 1055.3, respectively) and women (750.6 and 921.9, respectively) in 2021.⁹ The same pattern was observed for life expectancy in 2020 with Hispanics (78.8 years) outliving non-Hispanic White (77.8 years) and Black (72.8 years) people.¹² These data support the “Hispanic paradox” with the caveat that estimates are not available for all Hispanic subgroups based on country of origin. Therefore, we cannot rule out that the aggregate estimates may be driven by the largest subgroup within Hispanics, Mexican Americans, masking inequities within the ethnic category.

THE FUTURE OF THE HISPANIC PARADOX

Potential explanations for the Hispanic paradox include cultural practices, healthy dietary habits, healthy or selective immigration effect, acculturation, family supports and ties, genetic factors, ethnic enclaves, data reliability, and death misclassification.^1,2,13 These factors may act independently or jointly to support the paradox. For example, most Hispanic immigrants arrive in better health than the native population but may adopt unhealthy behaviors leading to an increase in obesity prevelance.¹⁴ This convergence to host country culture has been criticized for ignoring other factors such as the stresses accompanying the immigration experience, including exposure to xenophobia and racial discrimination, often harsh socioeconomic conditions,¹⁵ and a lifetime of substandard medical care.¹⁶ Considering the diversity and heterogeneity within the Hispanic population, it is crucial to examine how demographic changes and social dynamics may affect the “Hispanic paradox” in the future.

First, the cultural diversity associated with Hispanic people from countries other than Mexico may have implications for the traditions and practices, behaviors, and beliefs associated with the health advantage of Mexican Americans. For example, Puerto Rican women are more likely to smoke, and their infants tend to have lower birth weights and die at a higher rate than infants of Mexican American and Central and South American women. Similarly, it is important to consider the high influx of immigrants from Venezuela, Dominican Republic, Honduras, and Guatemala as they bring a great variation in cultural traditions and beliefs, SEP, and legal status.⁷ Related to the latter, the “Hispanic paradox” has been found among recent immigrants, an observation associated with a health selection effect. However, recent waves of immigration comprise a mix of individuals of all ages and, as such, with different health statuses. For example, evidence suggests a loss of health advantage in Mexican immigrants with time in the United States with a greater prevalence of disability, and overall poor health in both immigrants and US-born older adults of Mexican origin.¹⁷ Similarly, there is evidence that disability and cognitive impairment among those aged 75 years or older have increased from the early 1990s to the middle of the 2000s because of significant increases in the prevalence of diabetes among older Mexican adults.¹⁷ The latter may call attention to another Hispanic paradox of living longer but accompanied by poor health.

Another ignored contribution to the Hispanic population diversity is the role of indigeneity. As of 2010, there were approximately 45 million Indigenous people in Central and South America, representing more than 800 Indigenous groups, including those countries of origin for immigrants from Mexico (15.1%), Guatemala (41%), Honduras (7%), Colombia (3.4%), and Venezuela (2.7%). Like ethnicity, indigeneity or the culture, norms, language, and traditional knowledge of people before colonization may afford health protective benefits and contribute to the paradox. However, information on indigeneity is lost when Hispanic Indigenous people are asked to self-identify as only Hispanic in most surveys. Finally, the contribution to the Hispanic population growth has shifted over the past 40 years from immigration before 2000 to newborns after 2000.¹⁸ Therefore, the healthy migrant effect may disappear and, together with it, the health advantages of recent immigrants and older Hispanic adults.

Second, acculturation, the process by which individuals assimilate into the host country’s culture and norms, has been associated with both positive and negative health outcomes, depending on how individuals use its features for their health and well-being. In addition, acculturation may erode some Hispanic traditions associated with better health, such as respect for the elderly and familism. The latter serves as a pillar of social support and ties among Hispanic families. Losing these traditions and practices will have implications for the health status of infants and older adults. Similarly, the diversity of the Hispanic population translates into a variety of SEPs, access to health insurance, and language proficiency, which are important for navigating and accessing the health care system. For example, Panamanian adults have the highest English proficiency (87%) and one of the lowest proportions of people without health insurance (8%), while Honduran adults have the lowest English proficiency and highest proportion of people without health insurance (40% for both) among the Hispanic population.⁸

Third, the Hispanic population is not evenly distributed across the United States, and some of the states where they reside have policies that disproportionally and unequally affect this population. For example, the increase in xenophobia over the past decade has and will continue to have implications for the Hispanic immigrant population. Specifically, states with structural xenophobia because of anti-immigrant laws and policies (i.e., Georgia and Alabama) do not provide a welcome context for immigrants.¹⁹ Residing in such states may have implications for the health status of the Hispanic immigrant population. In contrast, residing in states with more inclusive laws and policies, such as California, Oregon, and New York, may have health benefits for this population.

Evidence suggests that anti-immigrant rhetoric, policies, and climate have been associated with high emergency department use in Californian adults,²⁰ mental health and chronic pain in US children,²¹ and satisfaction with health care in US Hispanic adults.²² Similarly, living in a sanctuary city may offer a buffer against immigration policies for Hispanic people, regardless of their legal status. In fact, evidence suggests that Hispanic adolescents and adults residing in sanctuary cities may have better mental health relative to their counterparts who reside in nonsanctuary cities in California.²³ Moreover, Hispanic people tend to live in neighborhoods with high proportions of minoritized populations (68%), mostly composed of Hispanic populations (47%), which are sometimes referred to as ethnic enclaves. Such neighborhoods may mitigate the stress related to xenophobia and racial discrimination, with evidence suggesting that ethnic enclaves provide advantages leading to better health outcomes for older Hispanics.¹⁶

Fourth, accurate and reliable data at times of birth and death are important not only for Hispanics but also for other US racial/ethnic groups. It is imperative to have mechanisms in place to link birth and death certificates to avoid misclassifications at the time of death. Moreover, agreements in countries from which the Hispanic population originates should be established to avoid underreporting or missed death events occurring outside of the United States (i.e., salmon bias). Finally, the value and importance of collecting accurate and disaggregated data for the Hispanic population (e.g., country of origin, nativity status, length of stay in the United States) cannot be stressed enough. Such data could serve to further our understanding of the Hispanic paradox and identify health inequities within the Hispanic population.

Fifth, because Hispanics can be of any race, they can self-identify with various US Census racial categories. Self-identification with either the Black or White category could have implications for experiences of racial discrimination that may affect individuals’ everyday lives as well as their life chances and opportunities. Furthermore, because of the social construction of race, racial self-identification among the Hispanic population could lead to the same Black‒White health gap observed for non-Hispanic Black and White people. Evidence suggests that Hispanic people with darker skin are more likely to self-identify as Black or to report experiences of discrimination than their lighter skinned or White counterparts (64% vs 54%). Moreover, darker skinned Hispanic people are more likely to be treated unfairly by someone who is Hispanic than lighter skinned Hispanic people (41% vs 25%).²⁴ The latter resulted in a double jeopardy when it comes to racial discrimination of Hispanic Black people: they could experience discrimination from non-Hispanic Black and White people, and from Hispanic White people.²⁵

Moreover, social and environmental exposures related to discrimination may lead to gene‒environment interactions and affect the health of the Hispanic population. However, the Hispanic population is underrepresented in genomics research relative to their share of the US population (< 0.5% vs 19%). As we move forward with precision medicine and genomics research, the inclusion of a diverse population such as the Hispanic population could help our understanding of socio-epigenetics. The All of Us Research Program offers a unique opportunity to address the lack of representation of the Hispanic population in biomedical, genomic, or genetic research.

CONCLUSIONS

As the Hispanic population grows, it is imperative to acknowledge its diversity, which may affect the Hispanic paradox. This includes examining changes in sociodemographic characteristics, culture, norms and practices, immigration patterns, and racial identity. The protective effect of these factors is beginning to diminish, and the Hispanic health advantage observed over the past 40 years is at risk for disappearing. The Mexican American population is shrinking and acculturating, while the demographics of newcomers are changing, threatening to erase this advantage. We call attention to collecting granular data using a multidimensional lens when examining the Hispanic population, allowing for a disaggregated and in-depth examination of its diversity. This approach could also be applied to other racial/ethnic groups considered homogenous, such as Asian Americans, to provide a clearer picture of racial/ethnic health inequities in the United States and identify factors related to the Hispanic paradox that could benefit the health of the entire US population.