Allostatic Load Patterns by U.S. Citizenship Status and Length of U.S. Residency Among Adults, 2009–2018

HIGHLIGHTS

• •Citizenship and length of U.S. residency have different impacts for chronic stress.
• •Longer residing noncitizens and naturalized citizens have greater stress over time.
• •Citizenship status is an important factor for immigrant health inequities.

Abstract

Introduction

Emerging research has documented that having a non-citizen status (e.g., temporary visa, undocumented) induces stress and, as a result, has adverse impacts on health. It is unclear whether chronic stress differs by citizenship status and length of U.S. residency using clinical biomarkers. The objective of this study was to examine allostatic load (or cumulative stress) by citizenship status and length of U.S. residency among U.S. adults.

Methods

The study sample included 27,705 adult respondents (aged ≥20 years) from the 2009–2018 National Health and Nutrition Examination Surveys. Multivariable Poisson regression models were estimated with U.S. citizenship status (U.S.-born citizens, naturalized citizens, noncitizens) and U.S. residency (shorter: <15 years, longer: ≥15 years) on allostatic load. Allostatic load was defined with summative scores of 10 biomarkers (systolic blood pressure, diastolic blood pressure, high-density lipoprotein, total cholesterol, HbA1c, BMI, albumin, estimated glomerular filtration rate, white blood cell count, and asthma).

Results

Naturalized citizens with shorter U.S. residency (females only) and noncitizens with shorter U.S. residency had lower allostatic load than U.S.-born citizens. Naturalized citizens with longer U.S. residency had greater allostatic load than naturalized citizens with shorter U.S. residency (females only). Noncitizens with longer U.S. residency had higher allostatic load than both noncitizens with shorter U.S. residency and naturalized citizens with shorter U.S. residency (females only).

Conclusions

This study demonstrates nuanced impacts on allostatic load by citizenship status and length of U.S. residency, with differences by sex. The findings infer that citizenship status contributes to health inequities among immigrants, with greater attention needed to unpack citizenship-stress mechanisms.

INTRODUCTION

In 2018, foreign-born residents accounted for 13.7% (44.8 million) of the total population in the U.S. The majority are noncitizens, including lawful permanent residents or green card holders (27%), undocumented visa holders (23%), and temporary visa holders (5%). The extant literature has found that immigrants tend to be healthier than their U.S.-born counterparts.^1,2 These findings have been attributed to behavioral factors (e.g., diet) and strong community support.3, 4, 5 Furthermore, evidence demonstrates that healthier immigrants are more likely to migrate than their nonmigrant counterparts.^6,7 In considering beyond individual level factors, legal/citizenship status has been considered to be a social determinant of health contributing to health disparities.8, 9, 10, 11 Moreover, legal/citizenship status is enacted and enforced through policies, practices, and laws that contribute to and reproduce inequities through access to social and economic resources and opportunities.^8,12 Noncitizens are more likely to live in poverty than U.S.- born citizens or naturalized citizens.¹³ Despite eligibility and service availability, healthcare and other social services utilization may be underutilized among noncitizens owing to increased anti-immigrant sentiment resulting in more restrictive policies (e.g., changes in residency rules, visa, or work requirements) and policy changes leading to consequential levels of immigration enforcement.^14,15 Hence, noncitizens may be experiencing greater cumulative stress over time than their U.S.-born counterparts.

Empirical scholarship has demonstrated differences in a constellation of health outcomes across citizenship status categories. Noncitizens are less likely to have health insurance^16,17 and are more likely to report worse self-rated health^18,19 and greater distress¹⁸ than U.S. citizens. Noncitizens are less likely to be diagnosed and treated for hypertension and diabetes than U.S.-born citizens, with health insurance as a key factor for obtaining treatment and diagnoses for noncitizens.^20,21 Furthermore, noncitizens are less likely to have access to health care than U.S.-born citizens.²² However, limited research has examined differences in health by legal/citizenship status and length of U.S. residency. In addition, empirical research is underdeveloped in investigating the relationship between legal/citizenship status and physical health outcomes²³ and especially for allostatic load or physiologic dysregulation, which captures how chronic stress get under the skin. Allostatic load is a summative multisystem composite measure of stress-induced risk,^24,25 with a wealth of research demonstrating the relationship between allostatic load and cardiovascular disease, worse physical health, greater cognitive decline, and mortality.26, 27, 28, 29

Emerging research has documented that having a precarious immigration status, specifically a noncitizen status, has deleterious consequences on health over time because of the stress and stress processes associated with having this status. A recent study found that noncitizens and undocumented adults had higher allostatic load than U.S.-born adults.³⁰ Another study observed high levels of stress among temporary visa migrants.³¹ The backlog of visa-processing times and waiting in limbo (e.g., visa status changes or legal residency status) were key drivers of their stress and negatively impacting their health (e.g., sleep, mental health). Indeed, wait times for obtaining legal residency status for visa migrants may span well over a decade.^31,32 Recent studies have found that stress proliferation processes with citizenship and other stressors are associated with poorer health. Specifically, having a noncitizen status in combination with another stressor (e.g., discrimination) amplifies the effect of these stressors on health.³³ In summary, having a precarious immigration status, including a noncitizen status, may be attributed to increased stress34, 35, 36; however, previous studies have utilized perceived stress indicators and community-based samples.³⁷ More empirical investigations are needed to examine the role of legal/citizenship status for cumulative physiologic stress and health among representative samples of U.S. adults.

Moreover, additional research is needed to investigate the role of legal/citizenship status in combination with length of U.S. residency for cumulative stress. A previous study did not observe significant differences between legal status and health (self-rated health, blood pressure, hypertension); however, when parsed out by documentation status and length of U.S. residency, the authors found that documented immigrants with shorter U.S. residency had lower blood pressure than documented immigrants with longer U.S. residency and also U.S.-born citizens.³⁷ On the basis of previous empirical research, the authors hypothesize that naturalized citizens and noncitizens with shorter U.S. residency will have significantly lower allostatic load than naturalized and noncitizens with longer residency and U.S.-born citizens. In addition, the authors anticipate that noncitizens with a longer U.S. residency will have greater chronic stress than their U.S. citizen counterparts (naturalized citizens with shorter and longer residency, U.S-born citizens) and also compared with noncitizens with shorter U.S. residency. To extend the literature on citizenship status and chronic stress, this study examines the differences in allostatic load by citizenship status and length of U.S. residency among a representative sample of U.S. adults.

METHODS

Study Sample

Data were extracted from the 2009–2018 National Health and Nutrition Examination Surveys (NHANES). NHANES is a cross-sectional survey administered by the National Center for Health Statistics using a multistage probability sampling procedure to obtain nationally representative estimates of the U.S. noninstitutionalized population. Information is collected on demographic and health information through self-reported interviews, physical examinations, and laboratory testing.³⁸ The analytic sample was restricted to adults aged 20–80 years (N=27,705). The 2009–2018 survey cycles were used owing to the sociopolitical climate and exponential increases in immigration enforcement from 2010 onward.³⁹ Data are publicly available and deidentified without any identifiable information and, hence, did not require IRB approval.

Measures

Allostatic load is a count measure that included 10 biomarkers representing cardiovascular, metabolic, and immune systems. These measures were collected through physical examination, apart from asthma. Cardiovascular risk measures included diastolic blood pressure, systolic blood pressure, high-density lipoprotein, and total cholesterol. Metabolic biomarkers included HbA1c, BMI, albumin, and estimated glomerular filtration rate. White blood cell count and self-reported asthma diagnosis were included to measure inflammation. Asthma is an inflammatory condition impacting the airways and is often aggravated by other infections, including an increase in cytokine IL-6 production.⁴⁰ These indicators have been utilized for measuring allostatic load in previous empirical studies.^41,42

Prior research demonstrates that sex-specific scores provide greater nuance for allostatic load.⁴³ In this study, high-risk thresholds were calculated for each biomarker across females and males. For each biomarker, scores were determined on the basis of whether the readings were in the high-risk quartile (assigned a value of 1 or, alternatively, 0 for those who were either above or below the high-risk quartile), following previously recommended guidelines and previous studies that have utilized similar calculations for allostatic load.^30,44, 45, 46 Biomarker readings that fell above the 75th percentile or high risk included systolic blood pressure, diastolic blood pressure, total cholesterol, BMI, HbA1c, and white blood cell count. For high-density lipoprotein, estimated glomerular filtration rate, and albumin, readings that fell below the 25th percentile were categorized as high risk. For asthma, participants with a self-reported current asthma diagnosis were high risk (1), and those with a past diagnosis were at moderate risk (0.5). After dichotomizing each biomarker, a summative index was created in combining each of the biomarkers, with values ranging from 0 to 10. Table 1 presents percentages, means, and high-risk cutoff scores by sex across all biomarkers. Race-specific adjustments were not used in calculating allostatic load scores because race is a social construct, and using these adjustments would be based on notions that race differences are biological.⁴⁷

Table 1.

Weighted Percentage, Means, and High-Risk Cutoff for Allostatic Load Biomarkers by Sex, NHANES 2009–2018

	Female n=10,860		Male n=10,330
Biomarker	Mean (SD) or n (weighted %)	High-risk cutoff	Mean (SD) or n (weighted %)	High-risk cutoff
Systolic blood pressure	121.89 (19.21)	132	125.55 (16.88)	134
Diastolic blood pressure	68.72 (12.65)	76	71.49 (13.16)	79
HDL cholesterol	57.77 (16.14)	46	48.26 (14.56)	38
Total cholesterol	194.38 (40.81)	218	188.19 (41.79)	213
BMI	29.74 (7.70)	33.8	28.83 (6.15)	31.8
HbA1C	5.74 (1.06)	5.9	5.82 (1.12)	5.9
eGFR	59.76 (30.62)	32.19	73.38 (37.53)	39.48
Albumin	35.10 (248.80)	3.9	53.71 (333.21)	4.3
White blood cell	7.31 (4.38)	8.5	7.17 (2.73)	8.2
Asthma (n, weighted %)
No asthma	9,031 (82.69%)	—	9,045 (87.70%)	—
Past diagnosis	631 (6.19%)	—	648 (6.30%)	—
Current diagnosis	1,198 (11.11%)	—	637 (5.99%)	—
Allostatic load	2.46 (1.58)	—	2.44 (1.52)	—

eGFR, estimated glomerular filtration rate; HDL, high-density lipoprotein; NHANES, National Health and Nutrition Examination Surveys.

The authors created a measure combining citizenship status and length of U.S. residency. Citizenship status was a self-reported measure indicating whether participants were either U.S.-born citizen, naturalized citizen, or noncitizen. Previous studies have demonstrated pronounced differences in health after 15 years^1,37,48,49 and especially considering that the manifestations of stress take longer; hence, 15-year cut points were utilized. In addition, the NHANES data set comprises majority U.S.-born citizens (83.4%), with limited immigrant samples for naturalized and noncitizen samples. To ensure robust estimations with adequate sample sizes for population estimates, the authors utilized 15-year cut points (additional sensitivity analyses using 10-year cut points were conducted, with further details provided in subsequent sections). The authors categorized respondents as U.S.-born citizens, naturalized citizens and noncitizens with shorter U.S. residency (<15 years), and naturalized citizens and noncitizens with longer U.S. residency (≥15 years).

In line with previous research on allostatic load and citizenship status, sociodemographic and health covariables were included as potential confounders in the analyses.^20,41,43,50, 51, 52, 53 Sociodemographic factors included family poverty level, age (20–44, 45–64, ≥65 years), marital status (married/living with partner, widowed/divorced/separated, never married), educational attainment (less than high school, high-school graduate/GED, some college, bachelor’s degree more), and race/ethnicity (non-Hispanic White, Latinx, Black, other). The authors also included smoking status (never smoked, former smoker, current smoker) and health insurance status. Physical activity was constructed using a series of questions of respondents’ reported levels of activity (moderate or vigorous) over the past 30 days (work or recreational). No responses to either physical activity in work or recreational activity questions were categorized as none. Responses of yes to both work and recreational activities were coded as active, and a yes response to either vigorous or moderate levels of activity at work or recreational was coded as some activity.

Analyses included 27,705 of U.S. adults. Allostatic load (n=4,090) and poverty level (n=2,735) had the highest levels of missing data, and other covariables in the study had no missing data or up to 2% missing data. In the final Poisson regression models, a total of 6,515 respondents had missing data (n=21,190). To ensure precision of the results, the authors employed multiple imputation by chained equations algorithms to replace missing data. Multiple imputation by chained equations is a robust and rigorous method because each variable is imputed on the basis of its distribution and imputation model.⁵⁴

Statistical Analysis

Data were analyzed using Stata 14.0 (StataCorp, College Station, TX). Survey weights were incorporated to account for the different sampling probabilities and participation rates to obtain representative estimates of the U.S. population. The authors computed descriptive statistics for sample sociodemographic and health characteristics stratified by citizenship status and length of U.S. residency. Bivariate analyses were conducted to compare proportions and mean differences across groups (i.e., chi-square test, t-tests).

The authors conducted imputed sex-stratified multivariable Poisson regression models. First, the authors estimated the relationship between citizenship status and length of U.S. residency (crude model), followed by models adjusted for sociodemographics, health behaviors, and health insurance separately by sex (Model 1). Additional Poisson models were estimated to assess differences in allostatic load across each of the citizenship status and length of U.S. residency categories.

Previous empirical research has noted high levels of stress and stressors among Latinx immigrants.^55,56 For noncitizen White and Black adults, the sample sizes were small, and hence, the authors could not conduct estimations for these groups. NHANES did not include Asian as a separate category in their sampling for race/ethnicity prior to 2011 (in other words, for the 2009–2010 interview data). Therefore, the authors conducted additional sensitivity analyses to examine citizenship status and U.S. residency on allostatic load only among the Latinx sample. The authors also conducted additional analyses using 10-year cut points to compare and determine whether similar relationships between citizenship status and allostatic load were observed as in the main models. Statistical tests were 2 sided, and statistical significance was considered for p<0.05.

RESULTS

Table 2 presents weighted descriptive statistics by citizenship status and length of U.S. residency. Of 21,190 adults, 83.4% were U.S.-born citizens. Noncitizens were more likely to be Latinx than U.S.-born citizens (<15 years: 50.5%, ≥15 years: 71.0%). Noncitizens were more likely to be male than U.S.-born citizens (<15 years: 52.2%, ≥15 years: 52.0% vs U.S.-born citizens: 48.5%). Noncitizens (<15 years: 59.4%, ≥15 years: 64.2%) and naturalized citizens (<15 years: 46.0%) were more likely to live in poverty than U.S.-born citizens (31.7%). On average, foreign-born adults with shorter residency were younger in age, between ages 21 and 44 years (naturalized citizens: 67.5%, noncitizens: 77.6% vs U.S.-born citizens: 43.8%), and naturalized citizens with longer residency were older (23.1%) than U.S.-born citizens (19.2%). Noncitizens were more likely to have lower education (<15 years: 33.4%, ≥15 years: 51.0%) than U.S.-born citizens (10.9%) and naturalized citizens (<15 years: 13.3%, ≥15 years: 19.8%). In addition, noncitizens were more likely to be uninsured (<15 years: 46.2%, ≥15 years: 47.6%) than U.S.-born citizens (13.4%) and naturalized citizens (<15 years: 18.0%, ≥15 years: 12.8%). Foreign-born adults were more likely to be married, never had smoked, and were less physically active than U.S.-born citizens. Among females, mean allostatic load scores (results not shown) were the highest among U.S.-born citizens (mean=2.40, SE=0.03), followed by naturalized citizens (mean=2.31, SE=0.05) and noncitizens (mean=2.11, SE=0.06). Similarly, male U.S.-born citizens had the highest levels of allostatic load (mean=2.38, SE=0.03) followed by naturalized citizens (mean=2.29, SE=0.06) and noncitizens (mean=2.26, SE=0.05).

Table 2.

Weighted Sample Characteristics by Citizenship Status and Length of U.S. Residency, NHANES 2009–2018

		Citizenship status
Characteristics	Total	U.S. born	Naturalized citizens, shorter residency (<15 years)	Noncitizens, shorter residency (<15 years)	Naturalized citizens, longer residency (≥15 years)	Noncitizens, longer residency (≥15 years)	p-value
Total	21,190
Citizenship status and U.S. residency							—
U.S.-born citizen	15,199 (83.4)	—	—	—	—	—
Naturalized citizens, shorter residency (<15 years)	470 (1.4)	—	—	—	—	—
Noncitizens, shorter residency (<15 years)	1,714 (5.3)	—	—	—	—	—
Naturalized citizens, longer residency (≥15 years)	2,608 (6.7)	—	—	—	—	—
Noncitizens, longer residency (≥15 years)	1,199 (3.2)	—	—	—	—	—
Sex
Female	10,860 (51.4)	7,777 (51.5)	265 (56.5)	865 (47.8)	1,350 (52.9)	603 (48.0)
Male	10,330 (48.6)	7,422 (48.5)	205 (43.5)	849 (52.2)	1,258 (47.1)	596 (52.0)
Race/ethnicity
Non-Hispanic White	8,875 (67.8)	8,485 (77.9)	33 (15.6)	84 (13.2)	224 (22.0)	49 (13.1)	<0.001
Latinx	5,030 (13.7)	1,960 (7.0)	159 (32.3)	849 (50.5)	1,144 (36.8)	918 (71.0)
Black	4,368 (10.4)	3,874 (11.1)	64 (11.9)	129 (6.6)	235 (7.8)	66 (4.4)
Other	2,917 (8.2)	880 (4.0)	214 (40.2)	652 (29.7)	1,005 (33.4)	166 (11.5)
Poverty level							<0.001
No	10,959 (65.3)	8,226 (68.3)	237 (54.0)	649 (40.6)	1,490 (64.1)	357 (35.8)
Yes	10,231 (34.7)	6,973 (31.7)	233 (46.0)	1,065 (59.4)	1,118 (35.9)	842 (64.2)
Age, years							<0.001
21–44	9,031 (45.3)	6,482 (43.8)	282 (67.5)	1,202 (77.6)	624 (32.9)	441 (46.4)
45–64	7,348 (36.5)	5,007 (36.9)	131 (25.7)	381 (18.2)	1,240 (44.0)	589 (44.7)
≥65	4,811 (18.2)	3,710 (19.2)	57 (6.8)	131 (4.2)	744 (23.1)	169 (8.9)
Marital status							<0.001
Married/ lives with partner	12,685 (63.8)	8,480 (62.5)	343 (72.0)	1,209 (70.8)	1,805 (68.8)	848 (72.6)
Widowed/ divorced/ separated	4,601 (18.2)	3,569 (18.8)	51 (8.7)	179 (8.7)	572 (20.7)	230 (15.5)
Never married	3,904 (18.0)	3,150 (18.7)	76 (19.2)	326 (20.5)	231 (10.5)	121 (11.9)
Education							<0.001
Less than high school	4,588 (14.0)	2,474 (10.9)	78 (13.3)	632 (33.4)	685 (19.8)	719 (51.0)
High-school graduate/GED	4,768 (22.4)	3,784 (23.3)	91 (23.5)	277 (17.6)	413 (15.5)	203 (21.0)
Some college	6,606 (32.3)	5,351 (34.4)	139 (28.2)	291 (18.0)	665 (26.9)	160 (15.7)
Bachelor’s degree or more	5,228 (31.3)	3,590 (31.4)	162 (34.9)	514 (31.0)	845 (37.8)	117 (12.3)
Smoking status							<0.001
Never smoked	11,893 (56.1)	7,746 (53.6)	354 (74.0)	1,256 (71.2)	1,763 (67.4)	774 (64.0)
Former smoker	5,098 (25.0)	3,892 (26.0)	78 (15.1)	259 (16.6)	605 (22.8)	264 (22.6)
Current smoker	4,199 (18.9)	3,561 (20.4)	38 (10.8)	199 (12.2)	240 (9.7)	161 (13.4)
Physical activity levels							<0.001
None	6,734 (26.1)	4,509 (24.7)	156 (31.6)	620 (32.3)	964 (33.2)	485 (34.8)
Some activity	8,598 (41.5)	6,199 (41.7)	222 (47.8)	672 (39.2)	1,066 (41.3)	439 (37.6)
Active	5,858 (32.4)	4,491 (33.6)	92 (20.5)	422 (28.5)	578 (25.5)	275 (27.6)
Health insurance							<0.001
No	4,251 (16.3)	2,484 (13.4)	84 (18.0)	793 (46.2)	332 (12.8)	558 (47.6)
Yes	16,939 (83.7)	12,715 (86.6)	386 (82.0)	921 (53.8)	2,276 (87.2)	641 (52.4)

Note:

Means and SEs for population estimates are presented for continuous variables, and n and weighted percentages are presented for categorical variables.

NHANES, National Health and Nutrition Examination Survey.

For analysis with female U.S.-born citizens as the comparison category, Table 3 presents multivariable Poisson regression results. Crude female models demonstrated that naturalized citizens (b= −0.25, 95% CI= −0.36, −0.14) and noncitizens (b= −0.25, 95% CI= −0.31, −0.19) with shorter U.S. residency had significantly lower allostatic load than U.S.-born citizens. After critical covariable adjustment (Model 1), these relationships remained significant for noncitizens with shorter residency (b= −0.19, 95% CI= −0.25, −0.12) and naturalized citizens with shorter residency (b= −0.19, 95% CI= −0.30, −0.08) than U.S.-born citizens.

Table 3.

Poisson Regression Modeling of Allostatic Load Scores, NHANES 2009–2018

	Female n=14,322		Male n=13,383
	Crude	Model 1	Crude	Model 1
Variables	β (95% CI)	β (95% CI)	β (95% CI)	β (95% CI)
Citizenship status and duration
U.S.-born citizen	ref	ref	ref	ref
Naturalized citizen, shorter residency (<15 years)	−0.25*** (−0.36, −0.14)	−0.19** (−0.30, −0.08)	−0.15 (−0.32, 0.03)	−0.12 (−0.28, 0.04)
Noncitizen, shorter residency (<15 years)	−0.25*** (−0.31, −0.19)	−0.19*** (−0.25, −0.12)	−0.16*** (−0.22, −0.11)	−0.12** (−0.19, −0.06)
Naturalized citizen, longer residency (≥15 years)	0.02 (−0.03, 0.07)	−0.04 (−0.09, 0.02)	−0.01 (−0.07, 0.05)	−0.07 (−0.13, 0.00)
Noncitizen, longer residency (≥15 years)	0.07 (0.00, 0.14)	0.01 (−0.06, 0.09)	0.08* (0.01, 0.14)	0.02 (−0.06, 0.10)
Race/ethnicity
Non-Hispanic White		ref		ref
Latinx		0.05* (0.00, 0.09)		0.04 (−0.01, 0.08)
Black		0.09*** (0.05, 0.12)		−0.01 (−0.05, 0.03)
Other		0.07* (0.02, 0.13)		0.04 (−0.01, 0.10)
Poverty
No		ref		ref
Yes		0.03 (−0.01, 0.07)		0.02 (−0.01, 0.06)
Age, years
20–44		ref		ref
45–64		0.29*** (0.25, 0.33)		0.21*** (0.17, 0.24)
≥65		0.29*** (0.24, 0.33)		0.14*** (0.09, 0.18)
Marital status
Married/lives with partner		ref		ref
Widowed/divorced/separated		0.02 (−0.02, 0.05)		0.04 (−0.01, 0.08)
Never married		−0.07* (−0.12, −0.02)		−0.15*** (−0.19, −0.10)
Education
Less than high school		ref		ref
High-school graduate/GED		−0.01 (−0.06, 0.03)		0.03 (−0.02, 0.08)
Some college		−0.02 (−0.07, 0.02)		0.03 (−0.01, 0.07)
College graduate or more		−0.14*** (−0.20, −0.08)		−0.08** (−0.13, −0.04)
Smoking status
Never smoked		ref		ref
Former smoker		0.06** (0.02, 0.10)		0.03 (−0.01, 0.07)
Current smoker		0.11*** (0.07, 0.16)		0.06** (0.02, 0.10)
Physical activity levels
None		ref		ref
Some		−0.06** (−0.09, −0.02)		−0.07* (−0.11, −0.03)
Active		−0.11*** (−0.16, −0.05)		−0.14*** (−0.18, −0.09)
Health insurance
No		ref		ref
Yes		0.00 (−0.04, 0.04)		0.01 (−0.04, 0.06)
Constant	0.88*** (0.86, 0.89)	0.74*** (0.68, 0.80)	.87*** (.85, .89)	0.84*** (0.77, 0.91)
Citizenship status comparisons
	Female		Male
	Crude	Model 1	Crude	Model 1
	β (95% CI)	β (95% CI)	β (95% CI)	β (95% CI)
Compared with naturalized citizen, shorter residency (<15 years)
Noncitizen, shorter residency (<15 years)	0.00 (−0.12, 0.12)	0.01 (−0.11, 0.13)	−0.02 (−0.20, 0.17)	−0.00 (−0.17, −0.17)
Naturalized citizen, longer residency (≥15 years)	0.27*** (0.16, 0.38)	0.16** (0.04, 0.27)	0.14 (−0.04, 0.31)	0.06 (−0.11, 0.22)
Noncitizen, longer residency (≥15 years)	0.32*** (0.19, 0.45)	0.21** (0.08, 0.33)	0.22* (0.04, 0.40)	0.14 (−0.03, 0.31)
Compared with noncitizen, shorter residency (<15 years)
Naturalized citizen, longer residency (≥15 years)	0.27*** (0.20, 0.34)	0.15*** (0.08, 0.22)	0.15*** (0.08, 0.23)	0.06 (−0.01, 0.13)
Noncitizen, longer residency (≥15 years)	0.32*** (0.23, 0.41)	0.20*** (0.11, 0.28)	0.24*** (0.16, 0.32)	0.14** (0.05, 0.23)
Compared with naturalized citizen, longer residency (≥15 years)
Noncitizen, longer residency (≥15 years)	0.05 (−0.03, 0.13)	0.05 (−0.04, 0.13)	0.09* (0.01, 0.17)	0.08 (−0.01, 0.17)

Note: Boldfaces indicate statistical significance (*p<0.05, **p<0.01, and ***p<0.001).

Poisson regression betas and CIs are presented. Model 1 adjusted for race/ethnicity, poverty, age, marital status, education, smoking status, activity levels, and health insurance. Poisson regression models were imputed using MICE algorithms.

MICE, multiple imputation with chained equations; NHANES, National Health and Nutrition Examination Survey.

The authors also compared across female naturalized citizens and noncitizens. For crude female models, naturalized citizens (b=0.27, 95% CI=0.16, 0.38 and b=0.27, 95% CI=0.20, 0.34) and non-citizens (b=0.32, 95% CI=0.19, 0.45 and b=0.32, 95% CI=0.23, 0.41) with longer U.S. residency had significantly higher allostatic load than both naturalized citizens and noncitizens with shorter U.S. residency, respectively. After multivariable adjustment, these relationships held for both female naturalized citizens (b=0.16, 95% CI=0.04, 0.27 and b=0.15, 95% CI=0.08, 0.22) and noncitizens (b=0.21, 95% CI=0.08, 0.33 and b=0.20, 95% CI=0.11, 0.28) with longer U.S. residency compared with naturalized citizens and noncitizens with shorter U.S. residency, respectively.

With male U.S.-born citizens as the comparison category, for crude male models, noncitizens with shorter U.S. residency (b= −0.16, 95% CI= −0.22, −0.11) had lower allostatic load than U.S.-born citizens. After covariable adjustment (Model 1), this relationship remained significant (b= −0.12, 95% CI= −0.19, −0.06). In initial models, noncitizens with longer U.S. residency had higher allostatic load than U.S.-born citizens (b=0.08, 95% CI=0.01, 0.14); however, this became attenuated after multivariable adjustment.

The authors also compared across male naturalized citizens and noncitizens. For crude male models, naturalized citizens with longer U.S. residency (b=0.15, 95% CI=0.08, 0.23) had greater allostatic load than noncitizens with shorter U.S. residency. Noncitizens with longer U.S. residency had greater allostatic load across all categories, including naturalized citizens with longer (b=0.09, 95% CI=0.01, 0.17) and shorter (b=0.22, 95% CI=0.04, 0.40) residency and noncitizens with shorter U.S. residency (b=0.24, 95% CI=0.16, 0.32). After multivariable adjustment, these relationships became attenuated except for non-citizens with longer U.S. residency compared with noncitizens with shorter U.S. residency (b=0.14, 95% CI=0.05, 0.23).

Qualitatively similar results were observed among Latinx adults, compared with the main results (Appendix Table 1, available online). Among Latinx females, the results demonstrated that both naturalized citizens (b= −0.20, 95% CI= −0.38, −0.03) and non-citizens (b= −0.15, 95% CI= −0.23, −0.07) with shorter U.S. residency had lower allostatic load than U.S.-born citizens in the final models. Naturalized citizens (b=0.18, 95% CI=0.00, 0.36 and b=0.13, 95% CI=0.04, 0.22) and noncitizens (b=0.23, 95% CI=0.04, 0.42 and b=0.18, 95% CI=0.08, 0.28) with longer U.S. residency had significantly greater allostatic load than both naturalized citizens and noncitizens with shorter U.S. residency, respectively.

Among Latinx males, noncitizens with shorter U.S. residency (b= −0.15, 95% CI= −0.24, −0.06) had lower allostatic load than U.S.-born citizens, after adjusting for critical covariables. Noncitizens with longer U.S. residency (b=0.13, 95% CI=0.03, 0.23) had higher allostatic load than noncitizens with shorter U.S. residency.

The current study also examined cut points of less than 10 years or greater than 10 years of U.S. residency and citizenship status for allostatic load (results not shown). Female noncitizens with shorter residency (b= −0.19, 95% CI= −0.27, −0.11) had significantly lower allostatic load than U.S.-born citizens after multivariable adjustment. Similarly, male noncitizens with shorter residency (b= −0.10, 95% CI= −0.18, −0.03) had lower allostatic load than U.S.-born citizens.

DISCUSSION

The current study examined different pairings of citizenship status and length of U.S. residency, including U.S.-born citizens, because these analyses provide greater understanding of how citizenship status and length of U.S. residency together have varying implications for chronic stress. This study utilized clinical biomarkers among a representative sample of U.S. adults. The findings demonstrate nuanced relationships between citizenship status and length of U.S. residency on allostatic load, contributing to an underdeveloped area of research.

This study supported the hypotheses in that noncitizens and naturalized citizens (females only) with shorter U.S. residency have lower allostatic load than U.S.-born citizens. In considering length of U.S. residency, a previous empirical study found that documented immigrants with shorter U.S. residency have lower blood pressure than documented immigrants with longer U.S. residency and U.S.-born citizens.³⁷ The current study findings align with these results regarding chronic stress. A well-established area of research has demonstrated that immigrants tend to be healthier than U.S.-born counterparts,^1,2 which may potentially explain some of these findings.

The study also supported the hypothesis in that noncitizens with longer residency have significantly greater chronic stress than noncitizens with shorter residency and naturalized citizens with shorter residency (females only). However, a significant difference in allostatic load was not observed between noncitizens and naturalized citizens with longer residency and U.S.-born citizens. In other words, noncitizens and naturalized citizens have levels of allostatic load or chronic stress similar to those of U.S.-born citizens after residing in the U.S. for more than 15 years and greater levels of chronic stress than their noncitizen and naturalized citizen counterparts who have resided in the U.S. for less than 15 years. A recent study found that noncitizens and undocumented adults had greater allostatic load than U.S.-born adults³⁰; however, this study did not examine differences by length of U.S. residency or by sex. The combination of one’s citizenship status and length of U.S. residency together may indirectly impact allostatic load through sociodemographic factors (e.g., income, education). In bivariate models, noncitizens with longer U.S. residency were more likely to live in poverty and were less likely to either graduate from high school or have health insurance than U.S.-born citizens. Previous research has found that noncitizens are less likely to have health insurance^16,17 and are also less likely to obtain diagnosis and treatment for hypertension and diabetes than U.S.-born citizens.^20,21 Noncitizens living in the U.S. often navigate having a precarious immigration status for many years while managing multiple other stressors (being uninsured, stressful job conditions). A recent study observed high levels of stress among noncitizen temporary visa migrants due to navigating the uncertainties of their limited term visa contracts and/or waiting in limbo to obtain legal residency, in tandem with the backlog of visa processing times (spanning well over a decade).^31,32 For naturalized citizens, despite obtaining citizenship, the stressors and exposures of living in the U.S. may offset the benefits of obtaining citizenship after 15 years. The combination of these factors may contribute to the accumulation of stressors over time and, in turn, allostatic load. The authors invite future empirical studies to further explore these relationships utilizing longitudinal data.

In addition, female naturalized citizens with longer U.S. residency had significantly higher allostatic load than their naturalized citizen counterparts with shorter residency. A recent study among Latinx and Asian immigrants found that one’s legal status is inextricably linked to access to healthcare and employment opportunities.⁵⁷ Respondents often prioritized obtaining more secure legal statuses (including citizenship) and made difficult decisions to forgo advancement in their careers, including maintaining short-term employment and not having access to health care. In other words, many viewed obtaining a more secure legal status as a longer-term solution above and beyond having access to health care or stable employment, leaving them in a state of limbo for a number of years and, inadvertently, contributing to worse health over time. Previous research has found that older migrants experience a rapid and significant deterioration in health later in life owing to working in physically demanding jobs throughout adulthood.^19,58,59

Some notable differences were observed by sex, with more significant relationships noted among females than among males. Female noncitizens and naturalized citizens with longer residency had greater allostatic load than their counterparts with shorter residency. The impact of immigration-related stressors over time may get under the skin over time for females compared with males. Emerging research has found that obtaining a more secure legal status after having a precarious status for a number of years may not always guarantee improved employment and health conditions, including access to health insurance.⁵⁷ These differences could also be due to diverse coping mechanisms employed by sex and/or gender. Females are more likely to maintain larger social networks,⁶⁰ are less likely to be socially isolated,⁶¹ and are more likely to utilize their social networks when experiencing stress than males.62, 63, 64, 65 Together, these factors may have an initial protective effect for immigration-related stressors for females.⁶⁶ Future investigations to explore sex and gender differences for citizenship status and chronic stress are welcomed.

Limitations

There are several noteworthy caveats to this study. NHANES is a cross-sectional study; hence, the findings should not be interpreted as causal relationships. C-reactive protein, a robust measure of inflammation, was not collected in 2011–2014 in the NHANES. Instead, the authors used white blood cell count and asthma diagnosis for inflammation as previously done in prior studies.^41,42 There may be a potential underreporting of asthma among racial/ethnic groups and noncitizens owing to lack of health insurance and/or limited access to health care; thus constraining opportunities for diagnoses for these groups.

Although the NHANES study provides population-level estimates, they do not collect data by country of origin, which are critical for understanding immigrant health disparities and heterogeneity within groups⁶⁷ as previously detailed in the Methods section in the manuscript. The data also did not include detailed noncitizen categories (e.g., visa status, undocumented status), as well as for when respondents obtained their citizenship status or date of arrival to the U.S. Thus, the authors could not parse out estimations by noncitizen categories and changes to their legal status (e.g., when they obtained citizenship) for allostatic load or conduct nuanced investigations of length of U.S. residency (with predetermined categories by years and decades). The reported effect estimates are more than likely to be underestimations of the magnitude of the relationship for noncitizens and allostatic load and, specifically, for those with undocumented status; however, a recent study found that noncitizen adults had higher mean allostatic load than undocumented adults.³⁰ After the implementation of the Affordable Care Act, which contributed to increases in health insurance rates among U.S. adults, changes in insurance status were lower for noncitizens compared with their citizen counterparts.⁶⁸ The authors invite future research to examine age-cohort patterns interlinked with citizenship and length of U.S. residency and more nuanced mechanisms (e.g., mediation, moderation) for how citizenship contributes to health inequities through other social determinant factors (e.g., poverty, education, sex). In other words, citizenship and length of U.S. residency may impact allostatic load through multiple pathways both indirectly and directly, warranting more empirical research in unpacking these mechanisms. The authors also recommend examining these relationships among a robust foreign-born sample with diverse populations (e.g., Asian, Black immigrants) to further disentangle the citizenship status, length of U.S. residency, and allostatic load. Despite some of these caveats, the unique strengths of the NHANES data in utilizing biomarkers rather than self-reported diagnosed chronic conditions offset these limitations by contributing to the understanding of the relationship between citizenship status and chronic stress among a nationally representative sample of U.S. adults.

CONCLUSIONS

This study assessed biosocial pathways of stress or how citizenship gets under the skin using a nationally representative sample of U.S., with the findings providing a foundation for future exploration in unpacking stress–health mechanisms by citizenship status and length of U.S. residency for allostatic load. The authors invite future research to investigate how different legal status categories interlinked with length of U.S. residency contribute to health inequities over time.