Care Needs and Arrangements of Aging Immigrants in the United States

Abstract

Objectives

To examine the need for and arrangements pertaining to personal care assistance among individuals 65 and older, and how life stage at migration impacts nativity differences in aging-related care.

Methods

Using data from the Survey of Income and Program Participation (2001, 2004, and 2008), I examine the odds of needing care assistance, who provides care assistance, and the duration of time care assistance is needed, comparing U.S.-born individuals to migrants who arrived before age 50 (“earlier-life migrants”) and those who arrived after age 50 (“later-life migrants”).

Results

While earlier-life migrants showed similar patterns to U.S.-born, later-life migrants showed higher care needs, were more likely to receive care from an adult child, and were particularly likely to need care for longer durations compared to U.S.-born.

Discussion

Aging later-life migrants have strikingly distinct care needs and arrangements, with implications for individual and family well-being, especially considering their barriers to public support.

Keywords

migration aging caregiving families lifecourse socioeconomic status

Introduction

The rise and diversification of immigration to the United States following the Immigration and Nationality Act of 1965 sparked an interest in research and policy centered around the health and well-being of immigrants (Hagan, 2004; Hagan et al., 2003; Hirschman & Wong, 1981; Ihara, 2011; Yeo, 2017). The 1965 Immigration and Nationality Act granted priority to immediate relatives of U.S. citizens seeking to migrate to the United States, which includes aging parents of U.S. citizens. Family reunification policies and the shifting demographics of recent immigrants have important and relatively understudied links with another rapidly changing demographic process—aging. As more aging parents of U.S. citizens began to migrate through family reunification, the age structure of immigrants began to shift upwards, calling attention to the need for research on the health and aging of immigrants, and distinguishing between migrants who arrived at younger ages and aged in the United States (“earlier-life migrants”) and those who migrated at older ages (“later-life migrants”) (Carr & Tienda, 2013; Batalova, 2012; Leach, 2009; Mizoguchi et al., 2019; Pew Research Center, 2015). While research has demonstrated that age at migration has important implications for well-being broadly (Angel & Angel, 1992; Blakemore, 1999; Garcia & Reyes, 2018; Gubernskaya & Tang, 2017; Maleku et al., 2021; Wakabayashi, 2010), the personal care needs and care arrangements of aging immigrants who migrate at different life stages are relatively unknown. This paper will contribute to the literature on migration and aging by examining the need for care assistance with activities of daily living and instrumental activities of daily living among a sample of non-Hispanic (NH) White, Hispanic, and NH Asian individuals aged 65 and older, comparing U.S.-born individuals to both earlier-life and later-life migrants. In addition, among those who need care assistance, I examine who provides such care (respondent’s adult child, spouse, or some other arrangement), and for how long care is needed, paying particular attention to differences by life stage at migration.

Background

Life Stage at Migration and Aging-Related Health

Life course theory suggests that events and conditions experienced by an individual may have different consequences depending on the individual’s age or life stage (Elder & Johnson, 2003). Under life course theory, migration shapes opportunities and (dis)advantages differently depending on at what age migration occurred. Life stage at migration has important implications for the accumulation of economic resources (Angel et al., 1999), for social connectedness and acculturation (Angel & Angel, 1992), and the ability to navigate U.S. systems and institutions (Alston & Aguirre, 1987), all of which impacts aging-related health and access to long-term care (Angel & Angel, 2006; Findley, 1988; Guo et al., 2019; House et al., 1989; Wakabayashi, 2010; Walsh & Walsh, 1987). For example, earlier-life migrants with longer durations in the United States may benefit from better English language ability, better integration into the U.S. labor market, and more experience with and understanding of U.S. health care systems compared to later-life migrants. However, earlier-life migrants have more exposure to discrimination and other forms of systemic racism and nativism in the United States, which cumulative disadvantage theory suggests may lessen health advantages over time (Angel et al., 2001). While later-life migrants on the other hand have fewer years of exposure to discriminatory conditions in the United States, this comes with less familiarity and access to U.S. labor markets, health care institutions, and social networks, creating disadvantages for well-being in old age. Furthermore, early-life adversities experienced in sending countries, such as war, famine, or exposure to violence, may impact the aging-related health of later-life migrants (Crocker, 2021; Elo & Preston, 1992; Gubernskaya & Kim, 2018; Kovnick et al., 2021). Considering the rising proportion of later-life migrants among both immigrant and aging populations (Carr & Tienda, 2013; Mizoguchi et al., 2019; Pew Research Center, 2015), it is increasingly important to understand how life stage at migration influences the personal care needs and care arrangements of aging immigrants.

Health Status and Selectivity Differences by Life Stage at Migration

There is a large body of literature on the “healthy migrant effect,” or the observation that migrants tend to have better mortality-related health outcomes than non-migrants, despite socioeconomic disadvantage among many immigrant groups (Abraido-Lanza et al., 1999; Liao et al., 1998; Markides & Coreil, 1986; Ruiz et al., 2012; Sorlie et al., 1993). Findings surrounding the healthy migrant effect are often attributed to selectivity, as healthier individuals tend to be the ones to migrate, particularly if migrating for employment reasons (Garcia et al., 2017; Markides & Rote, 2019; Palloni & Arias, 2004). However, the mortality advantage among immigrants has not consistently translated to increased health across the life course. Researchers have found heightened prevalence of functional limitations among immigrant-origin groups, including difficulties with activities of daily living (ADLs, such as eating, bathing, dressing, and getting around inside the home) and instrumental activities of daily living (IADLs, such as using a telephone, driving a car, shopping, preparing meals, doing housework, taking medicine, and handling finances), when compared to their U.S.-born counterparts (Boen & Hummer, 2019; Garcia & Reyes, 2018; Hayward et al., 2014; Melvin et al., 2014).

The relatively high levels of morbidity among older immigrants is typically attributed to longevity, negative acculturation, and cumulative disadvantage (Angel et al., 2001; Boen & Hummer, 2019; Choi, 2012; Markides et al., 2009) but may also be driven by the increase in later-life migrants, who are more likely to migrate for family reunification than for employment, and thus may be less positively selected on health than earlier-life migrants (Choi, 2012; Gubernskaya, 2014; Hill et al., 2012; Markides et al., 2009; Walters, 2002). In fact, individuals who migrate later in life are particularly likely to cite declining health as their reason for migrating (Markides & Rote, 2019; Tienda, 2017; Walters, 2002) and report worse self-rated health and functional limitations in old age compared to their counterparts who migrated earlier in life (Gubernskaya, 2014; Wakabayashi, 2010). Considering these distinctions in health selectivity and health status, examining the aging-related care needs and arrangements of later-life migrants separately from their peers who migrated earlier in life is an important step in addressing the long-term care needs of this growing and heterogeneous population.

Health Status and Selectivity by Region of Origin

Most research on immigrant health selectivity and the healthy migrant effect focuses on Hispanic populations, particularly Mexican-origin individuals (Markides & Rote, 2019), as Mexico represents the top country of origin for immigrants in the United States (Budiman, 2020). However, Asian Americans are now the fastest growing racial/ethnic group in the United States, largely due to immigration (Budiman & Ruiz, 2021), thus their inclusion is critical for understanding the aging-related care needs of the growing and diverse foreign-born population. While research on health and aging among Asian immigrants is sparse (Kim et al., 2010), a few studies have shown mortality advantages among Asian Americans (Lauderdale & Kestenbaum, 2002; Markides & Rote, 2019), but higher rates of disability, particularly among the foreign-born (Markides et al., 2009; Mutchler et al., 2007).

Despite similar mortality advantages, Hispanic and Asian immigrants experience unique migration conditions, opportunities for incorporation, and experiences of racial exclusion upon arrival in the United States, all of which may differently impact aging-related care needs and arrangements. For example, while Hispanic immigrants tend to have lower education levels than U.S.-born individuals, many groups of Asian immigrants are on average more educated than both U.S.-born individuals and non-migrants in sending countries (Tran et al., 2018; Zhou & Bankston, 2020). Thus, U.S. labor markets for immigrant-origin individuals tend to be relatively bifurcated, with Hispanic individuals disproportionately likely to work in low-paid positions and Asian individuals in higher-paid positions (Zhou & Bankston, 2020). For reasons further discussed below, the disproportionate representation of Hispanic individuals in low-paid and low-benefit jobs suggests that Hispanic individuals may have more care needs and fewer options for care arrangements in old age than both Asian and non-Hispanic White individuals, having worked more labor-intensive jobs with less access to health insurance and retirement benefits (Aguila & Zissimopoulos, 2013; Carr & Tienda, 2013; Taubman & Sickles, 1983; Herd et al., 2008).

Compositional Differences Between Earlier-Life and Later-Life Migrants

Scholars have noted important compositional differences between earlier-life and later-life migrants, with implications for aging-related care needs and care arrangements. For example, immigrants who migrate later in life are more likely to experience work disruptions and have labor histories split between two countries, which impacts their ability to receive retirement income, Social Security, and other benefits demonstrated to improve the health and disability status of older adults (Aguila & Zissimopoulos, 2013; Carr & Tienda, 2013; Taubman & Sickles, 1983; Herd et al., 2008). This relative lack of economic security in old age suggests that later-life migrants may be more likely to rely on family members for prolonged care and support (Angel & Angel, 1992; Angel et al., 1999; Gubernskaya, 2014). Finally, older individuals who migrate later in life are less likely to be currently married than those who migrate earlier in life and native-born individuals, as death of a spouse is a common driver of later-life migration (Chevan, 1995; Choi, 1996; Leach, 2009; Markides & Rote, 2019; Tienda, 2017; Walters, 2002). Thus, although spouses are among the more common means of care assistance in old age (Ornstein et al., 2017), later-life migrants are less likely to have spousal support and may be more likely to seek care from their adult children.

In addition, scholars note important nativity differentials in household composition, which are further patterned by life stage at migration. For example, Gubernskaya & Tang (2017) show that co-residency with extended family members is considerably higher for older immigrants who migrated after age 50 than for both older immigrants who migrated at younger ages and older adults in sending countries, a pattern likely driven by U.S. family reunification policies and immigrant selectivity. These distinctions in residential patterns suggest that aging-related care arrangements may be similarly dependent on life stage at migration, as co-residency and proximity to kin are important determinants of family caregiving (Litwak & Kulis, 1987).

Study Purpose and Hypotheses

Given the rise in the proportion of older, foreign-born individuals living in the United States, and prior research suggesting important distinctions in health and aging by life stage at migration, this analysis examines differences in care needs and care arrangements between aging NH White, Hispanic, and NH Asian U.S.-born and foreign-born individuals, separating foreign-born into those who migrated earlier in life (before age 50) and later in life (age 50 or later). I test the following three hypotheses:

○ H1: I hypothesize that while earlier-life migrants will show similar care needs to U.S.-born individuals due to positive health selectivity, later-life migrants will experience greater care needs than their U.S.-born aging peers, as later-life migrants are on average less positively selected on health and have fewer opportunities for acculturation and socioeconomic integration in the United States.

○ H2: I expect that later-life migrants will be especially likely to receive care from their adult children, not only because they are less likely to have a spouse present, but because migration and welfare policies often exclude later-life migrants from public supports that may assist with other forms of long-term care (Fix & Passel, 2002; Gubernskaya & Dreby, 2017; Treas & Batalova, 2009).

○ H3: I expect that later-life migrants will on average need more prolonged care assistance, as declining health is a common reason for migrating in older ages (Markides & Rote, 2019; Tienda, 2017; Walters, 2002), and thus later-life migrants may be particularly likely to migrate with the primary purpose of seeking care assistance from a family member living in the United States.

Methods

Data and Sample

I pooled data from 2001, 2004, and 2008 panels of the Survey of Income and Program Participation (SIPP). The SIPP is a nationally representative survey designed as a continuous series of panels. Respondents include all household members aged 15 and older, including non-English speakers (the data includes an indicator if the interview was conducted in a language other than English). The SIPP collects monthly-level data on topics such as family and household structure, employment and income histories, and insurance coverage and health care utilization, and includes cross-sectional topical modules on several social indicators, including information on migration histories and personal care needs and care arrangements. The SIPP questionnaire has gone through many re-designs since its inception in 1984, and many relevant questions used for this analysis on adult care needs and arrangements were discontinued in the most recent panels (2014 and 2018). Whereas many nationally representative surveys group Asian American respondents into an “other race” category due to sample size constraints, an advantage of the SIPP is the inclusion of Asian American and Pacific Islanders as a stand-alone racial/ethnic group. This inclusion allows for estimates of care needs and arrangements among the two largest racialized immigrant-origin groups in the United States (Hispanic and Asian immigrants), expanding upon prior studies which tend to focus exclusively on Hispanic immigrants (Markides & Rote, 2019).

As the purpose of this study is to understand how life stage at migration impacts the care needs and care arrangements of older immigrants, I used cross-sectional data from the wave of each panel where topical modules on adult care needs and arrangements were asked. Specifically, I used core and topical module data from wave 5 of panel 2001, wave 5 of panel 2004, and wave 6 of panel 2008. In addition, I applied the retrospective question on year migrated from the migration topical module in wave 2 of each panel to distinguish between earlier-life and later-life migrants. I also used retrospective data from the topical module on fertility in wave 2 of each panel to look descriptively at whether an individual has children. Finally, as each SIPP wave covers a four-month period but is designed to track changes at the monthly-level, the survey retrospectively asks respondents to report certain measures such as health insurance coverage and use of public benefits for each month over the four-month reference period. Thus, each individual has four observations per-wave (one for each month of the 4-month reference period). I included only the first observation per person, corresponding to the first month of the four-month reference period. I used the monthly cross-sectional weights at the individual-level provided by the SIPP, so the analysis is representative at the individual-level, rather than the household or family-level.

I created two samples from the data for this analysis—the “full sample” of all adults aged 65 or older, excluding those noted below, and a “sub-sample” of only individuals in need of care assistance (the sub-sample is derived from the full-sample). To create the full sample, I restricted to only respondents aged 65 or older, and thus focused on care needs and arrangements from the perspective of older adults themselves. Furthermore, as the purpose of this analysis is to examine the influence of migration on care needs and arrangements, I only include individuals who self-identify as Hispanic, non-Hispanic (NH) Asian or Pacific Islander, and NH White. NH Black individuals were excluded from this analysis as the sample of foreign-born Black individuals was quite small, and thus Black individuals would be included in the U.S.-born reference group, but not in the immigrant comparison groups. Furthermore, Black/White differences in care needs and arrangements likely reflect deep-rooted histories of racial stratification in the United States rather than mechanisms related to migration, which is outside the scope of this analysis. NH “other race” individuals were also excluded, as the heterogeneity and ambiguity of this group make conclusions difficult to draw. As such, it is important to note that results from this analysis are specific to NH White, Hispanic, and NH Asian individuals only. The full sample consists of 28,332 older adults, including 25,913 U.S.-born NH White, Hispanic, and NH Asian individuals, 1594 NH White, Hispanic, and NH Asian earlier-life migrants, and 825 NH White, Hispanic, and NH Asian later-life migrants. The sub-sample of individuals who need care assistance (n = 4354), includes 3912 U.S.-born NH White, Hispanic, and NH Asian individuals, 245 NH White, Hispanic, and NH Asian earlier-life migrants, and 197 NH White, Hispanic, and NH Asian later-life migrants. The criteria used to identify the sub-sample of individuals who need care assistance is described in the following section.

Outcome Variables

I examine five outcome variables, the first three measuring care needs and the last two measuring care arrangements. The first three outcome variables measuring care needs were analyzed among the full sample, while the last two outcome variables measuring care arrangements were analyzed for the sub-sample of individuals in need of care assistance only. Each outcome variable is described below:

○ Any care assistance: This outcome is a dichotomous variable (yes/no) measuring whether the respondent needs any care assistance, with “no” as the reference group. This measure is derived from a series of questions in the SIPP asking if the respondent “needs the help of another person” for each of the following tasks: getting around inside the house, going outside the house, getting in and out of bed, taking a bath or shower, getting dressed, walking, eating, using or getting to the toilet, keeping track of money or bills, preparing meals, doing light housework, and/or taking medication. For this measure, “yes” indicates the respondent needs assistance with at least one of the listed tasks, and “no” indicates the respondent does not need assistance with any of the tasks. This outcome variable was examined for the full sample only. The “sub-sample” of only individuals who need care assistance used in later analyses includes only those who are coded as “yes” for this measure.

○ Assistance with ADL: This outcome is a dichotomous variable (yes/no) measuring whether the respondent needs care assistance with an ADL-specific task with “no” as the reference group. This measure is derived from the same series of questions used to create the “any care assistance” outcome variable, taking the value “yes” for individuals who need the “help of another person” with an ADL-specific task (getting around inside the house, going outside the house, getting in and out of bed, taking a bath or shower, getting dressed, walking, eating, and/or using or getting to the toilet). This outcome variable was examined for the full sample only.

○ Assistance with IADL: This outcome is a dichotomous variable (yes/no) measuring whether the respondent needs care assistance with an IADL-specific task with “no” as the reference group. This measure is derived from the same series of questions used to create the “any care assistance” outcome variable, taking the value “yes” for individuals who need the “help of another person” with an IADL-specific task (keeping track of money or bills, preparing meals, doing light housework, and/or taking medication). This outcome variable was examined for the full sample only.

○ Who provides care: This outcome is a categorical measure capturing who provides care assistance to the respondent. This measure is drawn from a survey question asking the respondent to identify the person who “generally helps with these activities.” Possible response options in the SIPP include the respondent’s son, daughter, spouse, parent, other relative, friend or neighbor, paid help, other nonrelative, or did not receive help. For sample size purposes, I grouped this measure into three categories: the respondent’s adult child (reference group), spouse, or any other arrangement. I excluded a small number of individuals whose care arrangements did not match their marital status. Specifically, I excluded 5 never-married and 37 previously married individuals who reported a spouse as their primary caregiver. This outcome variable was examined for the sub-sample of individuals in need of care assistance only (those who were coded as “yes” for the variable “any care assistance”).

○ Years needing care assistance: This outcome is a categorical variable measuring the amount of time (years) the individual needed care assistance. This measure is drawn from a survey question asking the respondent “for how long have you needed the help of another person [with the listed task(s)]?” The original survey question in the SIPP included five categories: less than 6 months, 6–11 months, 1–2 years, 3–5 years, and more than 5 years. For sample size purposes, I grouped this variable into three categories: less than one year, 1–5 years (reference group), and more than 5 years. Alternative specifications were checked for robustness. This outcome variable was examined for the sub-sample of individuals in need of care assistance only (those who were coded as “yes” for the variable “any care assistance”).

While paid help would ideally be examined as its own category under “who provides care,” a relatively small proportion (12%) of respondents received paid help, which was particularly small among later-life migrants (n = 25), thus the sample is underpowered to examine paid help as a standalone outcome. Similarly, individuals needing care assistance who reported they “did not receive help” represented only 1.5% of respondents with a care need (n = 67) and were grouped into the “any other arrangements” category rather than examined alone. While these sample size constraints may mask important variation in the “any other arrangements” category, research shows that spouses and adult biological children are the most common caregivers (Wolff et al., 2018), and thus the distribution of these two caregiver types across groups was of particular interest in this analysis.

Main Predictor Variable

○ Life stage at migration: This variable is a categorical measure of life stage at migration with three categories: earlier-life migrants, later-life migrants, and U.S.-born individuals (reference group). Earlier-life migrants are foreign-born individuals who migrated before age 50, and later-life migrants are foreign-born individuals who migrated at age 50 or later. Foreign-born and U.S.-born status was measured using a variable asking whether the respondent was born in the U.S. (yes/no). And among the foreign-born, age at migration was calculated by subtracting the individual’s birth year from the year they migrated. The variable for year migrated in the SIPP was often reported in a 1–2-year range, and the mid-point was used as an estimate of exact year migrated (e.g., an individual who migrated in 1982–1984 was coded as migrating in 1983). The use of age 50 to distinguish later-life and earlier-life migrants was adopted from previous studies (Angel et al., 2002; Gubernskaya & Tang, 2017).

Sociodemographic Covariates

Finally, I examine the extent to which the following sociodemographic measures impact differences in care needs and arrangements by life stage at migration:

○ Racial/ethnic stratification: This measure includes three categories: Hispanic, NH Asian, and NH White (reference group).

○ Gender: This variable measures the respondent’s gender, coded dichotomously as woman/man (reference group).

○ Age group: This variable groups a continuous measure of respondent’s age into four categories: 65–70 (reference group); 71–75; 76–80; and 81+.

○ Marital status: This variable groups respondent’s current marital status into three categories: currently married (reference group), formerly married (which included individuals who are divorced, separated, or widowed and not currently married), and never married.

○ College degree: This variable was coded dichotomously (yes/no) and indicates whether the respondent completed a bachelor’s degree, with “no” as the reference group.

○ Private health insurance: This variable was coded dichotomously (yes/no) and indicates whether the respondent is currently covered by a private health insurance plan, with “no” as the reference group.

○ Medicaid: This variable was coded dichotomously (yes/no) and indicates whether the respondent is currently covered by Medicaid, with “no” as the reference group.

○ Medicare: This variable was coded dichotomously (yes/no) and indicates whether the respondent is currently covered by Medicare, with “no” as the reference group.

○ Social Security payments: This variable was coded dichotomously (yes/no) and indicates whether the respondent is currently receiving any Social Security payments, with “no” as the reference group.

Analytic Strategy

I first ran weighted descriptive statistics of the full sample, as well as the sub-sample of individuals who need care assistance, separately for U.S.-born, earlier-life, and later-life migrants, to compare the distribution of care needs, care arrangements, and sociodemographic characteristics across nativity status.

Next, I ran logistic regression models examining the odds of needing any care assistance from another person (yes/no), as well as needing care specifically with an ADL task(s) and an IADL task(s) among the full sample. I first examine differences in care needs by nativity status, comparing U.S.-born (reference) to earlier- and later-life migrants, controlling for sociodemographic and survey differences including racial/ethnic stratification, gender, age, and survey year. I then examine the extent to which compositional differences including marital status and having a college degree influence differences in care needs, and finally, the extent to which health insurance coverage and receipt of Social Security payments influence differences in care needs.

Among the sub-sample of respondents who need care assistance, I ran multinomial logistic regression models to examine who provides care: the respondent’s adult child (reference), their spouse, or some other arrangement. I ran one set of models among all individuals in need of care assistance (married and unmarried individuals) (n = 4354), and another set of models among only married individuals in need of care assistance (n = 1792), as an individual’s marital status highly influences care arrangements (e.g., individuals without a spouse cannot be cared for by their spouse). I first examine differences in care arrangements between U.S.-born (reference), earlier-life, and later-life migrants, controlling for racial/ethnic stratification, gender, age, and survey year. I then control for compositional differences in education, and finally for health insurance coverage and receipt of Social Security payments, to examine how these sociodemographic indicators influence differences in care arrangements by nativity status.

Finally, among the sub-sample of individuals who need care assistance, I ran multinomial logistic regression models examining the number of years the individual has needed care assistance (n = 4,288, as 66 individuals who needed care assistance (1.5%) had missing data on the number of years care assistance was needed). The outcome variable included three categories: less than 1 year, 1–5 years (reference), and more than 5 years. 1–5 years was chosen as the reference category as it was the most commonly reported category and allows for estimates of both the lowest need group (less than 1 year) and the highest need group (more than 5 years). I ran additional models to test alternative specifications of this outcome variable for robustness. These models follow a similar structure as the first two analyses, first controlling for sociodemographic and survey differences (racial/ethnic stratification, gender, age, and survey year), then compositional differences (marital status, education), and finally health insurance coverage and receipt of Social Security payments.

I used the “svy set” commands in Stata for the descriptive statistics and all analytical models to account for the weighting, clustering, and stratification within the complex survey design of SIPP. Person-level weights were used to allow representativeness at the individual-level, rather than at the household or family-level.

Results

Descriptive Results

Table 1 shows weighted descriptive characteristics of individuals aged 65 and older from the pooled 2001, 2004, and 2008 Survey of Income and Program Participation, separately for U.S.-born, earlier-life migrants, and later-life migrants. While U.S.-born and earlier-life migrants showed similar prevalence of needing personal care assistance (14.9% and 15.1%, respectively), 23.3% of later-life migrants reported having a care need. These patterns held for both ADL and IADL specific care needs, with 20.5% of later-life migrants needing assistance with at least one ADL task, compared to 12.5% and 13.1% of U.S.-born and earlier-life migrants, respectively, and 16.5% of later-life migrants needing assistance with at least one IADL task, compared to 10.7% and 11.2% of U.S.-born and earlier-life migrants, respectively. The majority of U.S.-born (95.6%) and earlier-life migrants (53.6%) identified as White, with 31.1% of earlier-life migrants reporting Hispanic ethnicity, and 15.3% of earlier-life migrants identifying as Asian. Later-life migrants comprised of a relatively even share of Hispanic, NH White, and NH Asian individuals (32.0%, 32.5%, and 35.5%, respectively). Slightly over half the sample in each group were women. The mean age ranged from 73.6 (earlier-life migrants) to 75.0 (later-life migrants), with later-life migrants particularly likely to be above age 75 (46.1%). Among immigrants, the mean age at migration was 32.2 for earlier-life migrants and 62.9 for later-life migrants.¹

Table 1.

Descriptive Characteristics of Individuals Age 65 and Older, by Nativity Status. Survey of Income and Program Participation (SIPP) Pooled Panels 2001, 2004, 2008 (n = 28,332).

	U.S.-born (n = 25,913)	Earlier-life migrants (n = 1594)	Later-life migrants (n = 825)
Needs personal care assistance (%)	14.9	15.1	23.3*
Needs assistance with ADL activities	12.5	13.1	20.5*
Needs assistance with IADL activities	10.7	11.2	16.5*
Racial/ethnic stratification (%)
Hispanic	3.7	31.1*	32.0*
NH White	95.6	53.6*	32.5*
NH Asian	0.7	15.3*	35.5*
Gender (%)
Women	56.8	60.1*	56.8
Men	43.2	39.9*	43.3
Age (mean)	74.5	73.6*	75.0*
Age group (%)
65–70	33.9	39.4*	30.7
71–75	23.1	24.0	23.2
76–80	19.5	18.1	20.4
81+	23.5	18.5*	25.7
Survey year
2001	31.0	34.7*	20.7*
2004	33.0	31.2	20.7*
2008	36.0	34.1	58.6*
Age migrated (mean)	—	32.2	62.9*
Marital status (%)
Married	57.3	58.8	54.5
Previously married	39.2	36.5*	42.1
Never married	3.6	4.7*	3.5
Has biological child(ren) (%)	88.7	86.5*	89.1
College degree (%)	20.5	19.4	20.8
Has private health insurance coverage (%)	81.1	59.1*	24.6*
Has medicaid coverage (%)	6.1	17.4*	47.8*
Has medicare coverage (%)	97.9	94.9*	78.4*
Receives Social Security payments (%)	94.7	90.5*	57.6*

Needs personal care assistance includes individuals needing help with the following task(s): getting around inside the house, going outside the house, getting in and out of bed, taking a bath or shower, getting dressed, walking, eating, using or getting to the toilet, keeping track of money or bills, preparing meals, doing light housework, and taking medication.

Needs assistance with ADL activities includes individuals needing help with the following task(s): getting around inside the house, going outside the house, getting in and out of bed, taking a bath or shower, getting dressed, walking, eating, using or getting to the toilet.

Needs assistance with IADL activities includes individuals needing help with the following task(s): keeping track of money or bills, preparing meals, doing light housework, taking medication.

Means and percentages weighted.

*statistically different from U.S.-born (p ≤ .05).

(or from earlier-life migrants in categories where U.S.-born do not have values).

Later-life migrants were the least likely to be currently married (54.5%) and the most likely to be previously married (42.1%), which includes divorced, separated, and widowed individuals. Over 85% of individuals across each group had at least one biological child, ranging from 86.5% of earlier-life migrants to 89.1% of later-life migrants. Receipt of college degree was largely comparable, with about one-fifth of respondents in each group holding a college degree.

Later-life migrants were by and large the least likely to have private health insurance (24.6%), while over 80% of U.S.-born and about 60% of earlier-life migrants had private health insurance. Later-life migrants were the most likely to be covered under Medicaid (47.8%). While most individuals across all groups received Medicare coverage, later-life migrants were the least likely (78.4%). Furthermore, later-life migrants were by far the least likely to receive Social Security payments (57.6%), compared to 90.5% of earlier-life migrants and 94.7% of U.S.-born older adults.

Table 2 shows weighted descriptive characteristics of the sub-sample of individuals who need care assistance, separately for U.S.-born, earlier-life migrants, and later-life migrants. Across all groups, those in need of care assistance are more likely to need assistance with an ADL task(s) than with an IADL task(s). Adult children were the most common form of care assistance across all groups, particularly for later-life migrants, with nearly half of later-life migrants in need of care assistance receiving care from their adult children. Later-life migrants were also the least likely of any group to receive care from a spouse (16.1%). Later-life migrants were the most likely to report needing care for more than 5 years (39.6%) and the least likely to have needed care for less than 1 year (21.6%). Just under half of respondents across each group reported that their caregiver was a household member.

Table 2.

Descriptive Characteristics of Individuals Age 65 and Older, Among Individuals Who Need Care Assistance, by Nativity Status. Survey of Income and Program Participation (SIPP) Pooled Panels 2001, 2004, 2008 (n = 4354).

	U.S.-born (n = 3912)	Earlier-life migrants (n = 245)	Later-life migrants (n = 197)
Needs personal care assistance (%)	100.0	100.0	100.0
Needs assistance with ADL activities	83.4	86.6	87.6
Needs assistance with IADL activities	71.8	74.4	70.7
Person who generally provides care (%)
Adult child	39.3	41.4	47.8*
Spouse	29.8	22.2*	16.1*
Some other arrangement	30.9	36.5	36.1
Years needing care assistance (%)
Less than 1 year	24.1	24.7	21.6
1–5 years	50.9	45.0	38.8*
More than 5 years	25.0	30.3	39.6*
Caregiver is a household member (%)	47.3	47.1	49.6
Racial/ethnic stratification (%)
Hispanic	5.2	30.1*	35.4*
NH White	94.2	54.2*	29.9*
NH Asian	0.6	15.8*	34.8*
Gender (%)
Women	67.0	69.4	61.4
Men	33.1	30.6	38.6
Age (mean)	78.3	77.9	78.3
Age group (%)
65–70	16.1	16.2	14.3
71–75	17.0	22.0*	17.9
76–80	20.3	19.6	23.6
81+	46.5	42.2	44.2
Survey year
2001	31.4	34.5	25.2
2004	32.4	32.5	22.0*
2008	36.2	33.0	52.8*
Age migrated (mean)	—	35.1	64.4*
Marital status (%)
Married	41.4	38.9	37.5
Previously married	54.5	56.8	57.9
Never married	4.0	4.3	4.6
Has biological child(ren) (%)	88.7	89.5	91.2
College degree (%)	12.5	11.8	12.1
Has private health insurance coverage (%)	72.5	43.8*	12.2*
Has medicaid coverage (%)	13.7	38.8*	74.4*
Has medicare coverage (%)	98.3	96.8	76.6*
Receives Social Security payments (%)	95.4	91.2*	48.5*

Needs assistance with IADL activities includes individuals needing help with the following task(s): keeping track of money or bills, preparing meals, doing light housework, taking medication.

Means and percentages weighted.

*statistically different from U.S.-born (p ≤ .05).

(or from earlier-life migrants in categories where U.S.-born do not have values).

Patterns of racial/ethnic stratification across groups in the full sample largely carried over to the sub-sample of individuals needing care assistance. The gender gap among the sub-sample of individuals needing care assistance was larger than in the full sample, with women making up 67.0% of U.S.-born, 69.4% of earlier-life migrants, and 61.4% of later-life migrants in need of care assistance. Individuals in need of care assistance were on average slightly older than the full sample population, and the average age migrated was slightly older among those who need care than in the full sample. Approximately 40% of individuals across all groups in the sub-sample of individuals in need of care assistance were currently married, which was sizably lower than for the full sample. Nearly 90% of individuals in need of care assistance across each group had at least one biological child. Individuals in need of care assistance were nearly 10 percentage points less likely to have a college degree than those in the full sample across each group. Finally, there were large differences in health insurance coverage between the full sample and the sub-sample of individuals who need care assistance. Sizably lower proportions of those needing care assistance were covered under private health insurance, including 12.2% of later-life migrants, 43.8% of earlier-life migrants, and 72.5% of U.S.-born older adults. Sizably higher proportions of those needing care assistance were covered under Medicaid, including 74.4% of later-life migrants, 38.8% of earlier-life migrants, and 13.7% of U.S.-born older adults. Medicare coverage was similar among individuals in need of care assistance as in the full sample, as was receipt of Social Security payments for all groups except later-life migrants, with only 48.5% of later-life migrants in need of care assistance receiving Social Security payments, compared to 57.6% of later-life migrants in the full sample.

Need for Care Assistance

Table 3 shows results from the logistic regression models predicting the need for care assistance (yes/no). Models 1, 2, and 3 report the odds of needing any type of care assistance, models 4, 5, and 6 report the odds of needing care assistance with an ADL task(s), and models 7, 8, and 9 report the odds of needing care assistance with an IADL task(s). Later-life migrants showed significantly higher odds of needing any care assistance, including with both ADL and IADL-specific tasks, compared to U.S.-born older individuals, while earlier-life migrants showed no significant differences in the likelihood of needing care assistance compared to U.S.-born individuals. These patterns held when accounting for differences in racial/ethnic stratification, gender, age, survey year, marital status, and receipt of college degree. However, health insurance coverage, specifically private health insurance and Medicaid coverage, appeared to explain the difference in care needs between later-life migrants and U.S.-born older adults, as the difference became statistically insignificant once health insurance and Social Security payments were included, with private health insurance and Medicaid coverage being statistically significant predictors. These results provide evidence in support of the first hypothesis (H1) that later-life migrants will experience greater care needs, being particularly likely to migrate in response to declining health and thus less positively selected on health and having accrued fewer opportunities for acculturation and socioeconomic integration after migrating.

Table 3.

Logistic Regression Predicting the Odds of Needing Personal Care Assistance. Survey of Income and Program Participation (SIPP) Pooled Panels 2001, 2004, 2008 (n = 28,332).

	Any care assistance				Assistance with ADL			Assistance with IADL
	Model 1	Model 2	Model 3		Model 4	Model 5	Model 6	Model 7	Model 8	Model 9
Life stage at migration (reference = U.S.-born)
Earlier-life migrant	.94 (.08)	.96 (.08)	.86+ (.07)		1.01 (.09)	1.03 (.09)	.93 (.09)	.97 (.09)	.99 (.09)	.88 (.08)
Later-life migrant	1.40** (.15)	1.41** (.15)	.87 (.11)		1.51*** (.17)	1.52*** (.18)	.94 (.12)	1.31* (.16)	1.32* (.17)	.83 (.12)
Racial/ethnic stratification (reference = NH White)
Hispanic	1.68*** (.15)	1.52*** (.14)	1.10 (.09)		1.60*** (.15)	1.46*** (.14)	1.06 (.10)	1.77*** (.21)	1.61*** (.20)	1.17 (.14)
NH Asian	1.21 (.16)	1.25+ (.16)	.98 (.14)		1.12 (.16)	1.15 (.17)	.89 (.14)	1.18 (.18)	1.21 (.18)	.95 (.15)
Woman	1.51*** (.06)	1.28*** (.05)	1.27*** (.05)		1.59*** (.07)	1.36*** (.06)	1.35*** (.06)	1.31*** (.06)	1.12** (.05)	1.11* (.05)
Age group (reference = 65–70)
71–75	1.67*** (.10)	1.61*** (.09)	1.62*** (.10)	1.65*** (.10)		1.59*** (.10)	1.60*** (.10)	1.72*** (.13)	1.67*** (.13)	1.67*** (.13)
76–80	2.48*** (.14)	2.32*** (.13)	2.39*** (.14)	2.56*** (.16)		2.40*** (.16)	2.47*** (.16)	2.37*** (.16)	2.23*** (.16)	2.27*** (.16)
81+	5.57*** (.29)	4.83*** (.26)	5.13*** (.28)	5.54*** (.29)		4.82*** (.26)	5.10*** (.28)	5.68*** (.35)	4.96*** (.31)	5.21*** (.33)
Survey year (reference = 2001)
2004	.94 (.05)	.94 (.05)	.92 (.05)	.96 (.06)		.96 (.06)	.94 (.06)	.92 (.05)	.92 (.05)	.89* (.05)
2008	.96 (.05)	.99 (.05)	1.00 (.05)	.99 (.05)		1.01 (.05)	1.03 (.05)	.90+ (.05)	.92 (.05)	.94 (.05)
Marital status (reference = married)
Previously married	1.46*** (.05)	1.31*** (.05)		1.44*** (.06)		1.29*** (.05)		1.42*** (.06)	1.27*** (.06)
Never married		1.43*** (.14)	1.21+ (.12)			1.35* (.15)	1.13 (.13)		1.51*** (.16)	1.29* (.14)
College degree		.62*** (.03)	.67*** (.04)			.63*** (.04)	.69*** (.04)		.64*** (.04)	.70*** (.04)
Private health insurance		.75*** (.03)				.76*** (.04)			.76*** (.03)
Medicaid			2.70*** (.18)				2.64*** (.20)			2.66*** (.20)
Medicare			1.06 (.17)				1.03 (.18)			1.24 (.22)
Social Security payments		.93 (.09)				.91 (.09)			.92 (.10)
Constant	.06*** (.00)	.06*** (.00)	.08*** (.01)	.04*** (.00)		.05*** (.00)	.06*** (.01)	.04*** (.00)	.05*** (.00)	.05*** (.01)

Notes: +p≤.1; *p≤.05; **p≤.01; ***p≤.001.

Odds ratios (standard errors) reported.

In addition, table 3 shows notable racial/ethnic and sociodemographic differences in care needs. For example, while Hispanic individuals were more likely to report a care need than NH White individuals, Asian individuals showed no differences in care needs compared to NH White individuals. However, the difference in care needs between Hispanic and NH White older adults became statistically insignificant with the inclusion of health insurance coverage in the model. Women were consistently more likely than men to report a care need across all models, which may reflect increased longevity among women compared to men. Older age groups were associated with higher odds of needing care assistance, as was being formerly or never married compared to being currently married. Individuals with a college degree were significantly less likely to need care assistance, including ADL and IADL specific care, compared to individuals without a college degree, reflecting well-documented educational gradients in health (Conti et al., 2010; Elo, 2009; Kimbro et al., 2008). Private health insurance coverage was associated with a significantly lower likelihood of having a care need, while Medicaid coverage was associated with a significantly higher likelihood of having a care need.

Care Arrangements Among Individuals Who Need Care Assistance

Table 4 presents results from the multinomial logistic regression models examining care arrangements among the sub-sample of individuals who need care assistance, first among all individuals needing care assistance (models 1–3) and then among only married individuals in need of care assistance (models 4–6). Later-life migrants were significantly less likely to receive care from a spouse than from an adult child as compared to U.S.-born individuals, including among married individuals alone. Health insurance coverage, specifically private health insurance and Medicaid coverage, explained much of the variation between later-life and U.S.-born individuals, as this difference became insignificant in models 3 and 6, with Medicaid and private insurance showing significant associations with care arrangements. Earlier-life migrants showed no significant differences in receiving care from a spouse or an adult child compared to U.S.-born individuals. There were no significant nativity differences in receipt of care from some other arrangement. These findings provide evidence in support of the second hypothesis (H2) that later-life migrants in need of care assistance will be especially likely to receive care from their adult children, as later-life migrants are likely to migrate in response to the death of a spouse, and more likely to be excluded from public supports for long-term care following migration, resulting in fewer options for care assistance outside of their adult children.

Table 4.

Multinomial Logistic Regression Predicting Who Provides Care, Among Individuals Who Need Care Assistance. Survey of Income and Program Participation (SIPP) Pooled Panels 2001, 2004, 2008

	All individuals in need of care assistance (n = 4354)			Married individuals in need of care assistance only (n = 1792)
	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
Adult child (reference)
Spouse
Life stage at migration (reference = U.S.-born)
Earlier-life migrant	.86 (.20)	.84 (.19)	1.00 (.23)	.76 (.25)	.75 (.25)	.93 (.29)
Later-life migrant	.48* (.14)	.48* (.14)	1.00 (.34)	.48* (.17)	.48* (.17)	.90 (.35)
Racial/ethnic stratification (reference = NH White)
Hispanic	.45*** (.10)	.48*** (.10)	.74 (.15)	.44* (.15)	.46* (.16)	.58 (.21)
NH Asian	.60+ (.18)	.59+ (.18)	.83 (.28)	.35** (.11)	.35** (.11)	.42* (.15)
Woman	.16*** (.02)	.17*** (.02)	.17*** (.02)	.37*** (.05)	.38*** (.05)	.37*** (.05)
Age group (reference = 65–70)
71–75	.81 (.12)	.80 (.12)	.76+ (.12)	.95 (.21)	.93 (.21)	.92 (.21)
76–80	.52*** (.08)	.51*** (.07)	.44*** (.06)	.69+ (.14)	.68* (.13)	.63* (.12)
81+	.19*** (.02)	.19*** (.02)	.15*** (.02)	.37*** (.08)	.37*** (.08)	.33*** (.07)
Survey year (reference = 2001)
2004	1.06 (.13)	1.05 (.13)	1.02 (.13)	1.56+ (.37)	1.55+ (.37)	1.53+ (.35)
2008	1.02 (.13)	.97 (.13)	.95 (.13)	1.20 (.22)	1.17 (.21)	1.16 (.21)
College degree		1.90*** (.28)	1.70*** (.25)		1.48+ (.32)	1.38 (.30)
Private health insurance			1.99*** (.23)			1.58* (.33)
Medicaid			.51*** (.07)			.57** (.12)
Medicare			1.57 (.65)			.76 (.42)
Social Security payments			1.16 (.30)			1.44 (.48)
Constant	5.71*** (.74)	5.26*** (.68)	2.09* (.72)	9.58*** (2.09)	9.11*** (2.00)	6.62*** (3.24)
Other arrangement
Nativity status (reference = U.S.-born)
Earlier-life migrant	1.26 (.25)	1.25 (.25)	1.21 (.25)	1.20 (.49)	1.18 (.48)	1.16 (.46)
Later-life migrant	1.21 (.27)	1.21 (.26)	1.08 (.26)	.78 (.39)	.77 (.39)	.85 (.52)
Race (reference = NH White)
Hispanic	.79 (.11)	.84 (.12)	.78 (.12)	.84 (.30)	.89 (.32)	.87 (.34)
NH Asian	.56* (.14)	.56* (.15)	.52* (.14)	.73 (.37)	.74 (.37)	.77 (.39)
Woman	.75*** (.07)	.78** (.07)	.78** (.07)	.72* (.12)	.75+ (.13)	.75+ (.13)
Age group (reference = 65–70)
71–75	.83 (.12)	.83 (.12)	.84 (.12)	1.07 (.40)	1.05 (.39)	1.06 (.39)
76–80	.52*** (.07)	.51*** (.07)	.52*** (.07)	1.02 (.31)	1.00 (.30)	.99 (.31)
81+	.63*** (.07)	.63*** (.07)	.64*** (.07)	.95 (.31)	.94 (.30)	.94 (.31)
Survey year (reference = 2001)
2004	1.06 (.11)	1.05 (.11)	1.03 (.11)	.87 (.27)	.86 (.27)	.84 (.27)
2008	.96 (.10)	.93 (.09)	.92 (.10)	.84 (.18)	.81 (.17)	.79 (.17)
College degree		1.82*** (.27)	1.86*** (.29)		1.57 (.50)	1.60 (.50)
Private health insurance			.97 (.10)			.96 (.21)
Medicaid			1.19 (.14)			1.07 (.40)
Medicare			1.30 (.38)			1.81 (1.49)
Social Security payments			.83 (.14)			1.06 (.45)
Constant	1.48** (.19)	1.37* (.18)	1.25 (.33)	.94 (.30)	.88 (.29)	.48 (.42)

Notes: +p≤.1; *p ≤ .05; **p ≤ .01; ***p ≤ .001.

Relative risk ratios (standard errors) reported.

Additionally, table 4 shows that Hispanic individuals were less likely to receive care from a spouse than from an adult child compared to NH White individuals, although this difference became insignificant with the inclusion of health insurance variables. Asian individuals in need of care assistance were marginally less likely (p < .1) to receive care from a spouse than from an adult child compared to NH White individuals in models 1 and 2. Among married individuals in need of care assistance, Asian older adults were significantly less likely to receive care from a spouse than from an adult child across all three models. Furthermore, Asian older adults in need of care assistance were less likely to receive care from some other arrangement than from their adult child. There were no significant differences in care from some other arrangement between married Asian and married NH White older individuals, which may reflect sample size constraints. Hispanic older adults in need of care assistance showed no significant differences in receiving care from some other arrangement than NH White older adults.

Older age was associated with a significantly lower likelihood of receiving care from a spouse. Individuals with a college degree were more likely to receive care from either a spouse or from some other arrangement than from adult children, but not among married individuals alone, which may reflect sample size constraints. Individuals with private health insurance coverage were more likely to receive care from a spouse than from an adult child, while individuals with Medicaid coverage were less likely.

Years Needing Care Among Individuals Who Need Care Assistance

Table 5 presents results from multinomial logistic regression models predicting years needing care assistance among individuals in need of care assistance. While earlier-life migrants showed no significant differences in length of care needs compared to U.S.-born individuals, later-life migrants stand out as being significantly more likely to have long-term care needs, specifically more likely to need care for more than 5 years compared to their U.S.-born peers (reference = 1–5 years). This difference was largely explained by group differences in health insurance coverage and receipt of Social Security payments, as the distinction in long-term care needs between later-life migrants and U.S.-born older adults became statistically insignificant when these variables were controlled for. These findings provide evidence in support of the third hypothesis (H3) that later-life migrants will need more prolonged care assistance, being disproportionately likely to migrate for reasons related to family reunification and declining health.

Table 5.

Multinomial Logistic Regression Predicting Years Needing Care Assistance, Among Individuals Who Need Care Assistance. Survey of Income and Program Participation (SIPP) Pooled Panels 2001, 2004, 2008 (n = 4288).

	Model 1	Model 2	Model 3	Model 1	Model 2	Model 3
	1–5 years (reference)
	Less than 1 year			More than 5 years
Life stage at migration (reference = U.S.-born)
Earlier-life migrant	1.21 (.25)	1.21 (.25)	1.30 (.26)	1.24 (.23)	1.24 (.23)	1.19 (.23)
Later-life migrant	1.21 (.30)	1.19 (.29)	1.34 (.35)	1.82** (.38)	1.77** (.38)	1.35 (.31)
Racial/ethnic stratification (reference = NH White)
Hispanic	.75 (.14)	.74 (.14)	.86 (.17)	1.26 (.18)	1.25 (.18)	1.10 (.17)
NH Asian	1.08 (.34)	1.09 (.34)	1.19 (.38)	1.23 (.34)	1.29 (.35)	1.11 (.30)
Woman	.89 (.08)	.85+ (.08)	.85 (.08)	1.03 (.08)	.91 (.08)	.90 (.07)
Age group (reference = 65–70)
71–75	1.19 (.16)	1.19 (.16)	1.19 (.16)	.86 (.11)	.86 (.11)	.87 (.11)
76–80	1.08 (.14)	1.08 (.14)	1.04 (.14)	.66*** (.08)	.65*** (.08)	.68** (.09)
81+	.81 (.11)	.78+ (.10)	.75* (.10)	.68*** (.07)	.63*** (.06)	.66*** (.07)
Survey year (reference = 2001)
2004	.84 (.10)	.83 (.09)	.82+ (.09)	.95 (.10)	.93 (.09)	.91 (.09)
2008	.97 (.11)	.96 (.10)	.95 (.10)	1.00 (.11)	.97 (.11)	.98 (.11)
Marital status (reference=married)
Previously married		1.12 (.12)	1.18 (.12)		1.42*** (.12)	1.37*** (.12)
Never married		1.51* (.32)	1.63* (.35)		2.31*** (.46)	2.08*** (.41)
College degree		1.06 (.15)	1.00 (.14)		1.14 (.13)	1.15 (.13)
Private health insurance			1.28* (.13)			1.00 (.11)
Medicaid			.74* (.10)			1.28* (.15)
Medicare			1.39 (.44)			1.34 (.38)
Social Security payments			.61* (.14)			.56** (.10)
Constant	.58*** (.09)	.56*** (.08)	.56+ (.19)	.64*** (.08)	.58*** (.08)	.73 (.17)

Notes: +p ≤ .1; *p ≤ .05; **p ≤ .01; ***p ≤ .001.

Relative risk ratios (standard errors) reported.

n = 66 individuals in need of care assistance have missing data on length of care arrangements, and are therefore not included in this analysis.

Those with Medicaid coverage were significantly more likely to need care for more than 5 years and significantly less likely to need care for under a year, compared to needing care for 1–5 years, while individuals who receive Social Security payments were less likely to need care for more than 5 years and less likely to need care for under a year. Older age groups were negatively associated with long-term care needs, likely due to increased mortality, while being previously or never married were positively associated with long-term care needs, consistent with prior research on marital status and health, including at older ages (Goldman et al., 1995; Liu & Umberson, 2008).

The finding that later-life migrants were significantly more likely to need care assistance for over 5 years held across various specifications of the categorical variable measuring time needing care assistance (data not shown). Alternative specifications included the original measure in the SIPP (with categories: less than 6 months, 6–11 months, 1–2 years (reference), 3–5 years, and more than 5 years), and other groupings such as less than 6 months, 6 months-2 years (reference), 3–5 years, more than 5 years, and the grouping less than 1 year, 1–2 years (reference), 3–5 years, more than 5 years. Across all specifications, later-life migrants were significantly more likely to need care for over 5 years than U.S.-born older adults, a distinction that was largely explained by the inclusion of health insurance coverage, particularly Medicaid, or receipt of Social Security payments, as the significant difference between later-life migrants and U.S.-born individuals disappeared with the inclusion of these measures. Furthermore, in all specifications that categorized less than 6 months as the lowest need group, earlier-life migrants were significantly more likely to need care for less than 6 months than for a middle range reference (6 months–2 years, or 1–2 years), a pattern which held with the inclusion of all sociodemographic variables, including health insurance and receipt of Social Security payments. However, when 1 year was used as the cut-off for the lowest need group, earlier-life migrants showed no significant differences in length of care needs compared to U.S.-born older adults.

Discussion

As foreign-born individuals, including those who migrate at older ages, make up an increasing proportion of the aging population in the United States, understanding the care needs and arrangements of older immigrants is an important and timely research and policy goal to ensure the health and well-being of aging immigrants and their families. This paper contributes to literature on migration and aging by examining the need for and arrangements pertaining to personal care assistance among NH White, Hispanic, and NH Asian individuals aged 65 and older, noting differences between U.S.-born older adults, those who migrated under age 50 (“earlier-life migrants”), and those who migrated at age 50 or older (“later-life migrants”). Results show that later-life migrants were particularly likely to need care assistance, to receive care assistance from their adult children, and to need care assistance for longer durations compared to U.S.-born individuals, while earlier-life migrants were not. These findings underscore the importance of life course theory in understanding the health of immigrants in later-life, as aging-related well-being was strikingly distinct between more recent migrants who arrived at older ages and those who migrated younger and aged in the United States.

That health insurance coverage, particularly Medicaid and private health insurance, explains much of the difference in care needs and arrangements between later-life migrants and U.S.-born older adults may be in part due to endogeneity, as health insurance variables may be capturing differences across groups in health status and disability, and thus absorbing variation in the outcome variable. However, private health coverage may reflect access to health care, including long-term care, while Medicaid coverage may serve as an indicator of relative poverty, which is likely to impact the ability to receive long-term care. As later-life immigrants are significantly less likely to have private health insurance and significantly more likely to have Medicaid coverage than U.S.-born older adults, it is reasonable to conclude that differences in care needs and arrangements may be largely due to differences in health-related resources.

In addition, while Hispanic individuals were more likely to report a care need than NH White individuals, Asian individuals were not, which may reflect differences in socioeconomic selectivity and labor market experiences between Hispanic and Asian individuals, factors likely to influence aging-related health (Aguila & Zissimopoulos, 2013; Carr & Tienda, 2013; Taubman & Sickles, 1983; Herd et al., 2008). Furthermore, while Hispanic and Asian older adults in need of care assistance were more likely to receive care from an adult child than from a spouse compared to NH White older adults, Asian older adults were also more likely to receive care from their adult children than from some other arrangement, which may reflect cultural practices and expectations that make older Asian individuals particularly likely to receive care from their adult children (Miyawaki, 2015), or perhaps the large representation of Asian later-life migrants (43% of Asian individuals in this sample were later-life migrants), who are disproportionately likely to migrate for family reunification and thus more likely to have adult children available and willing to provide care support.

These findings have implications for both migration and aging-related policies and programs. Although U.S. migration policy purports to hold family reunification as its cornerstone (Treas & Gubernskaya, 2015; Wolgin, 2018), welfare and other anti-immigrant policies systematically exclude and deny immigrants from support for health and well-being, including restrictions to public health coverage (Fix & Passel, 2002). Legislation has been introduced in Congress to expand Medicaid eligibility for recent immigrants (National Immigration Law Center, 2020), and such policies are crucial to ensure access to needed health care, including long-term care. Related, as private health insurance, Social Security, and other retirement benefits are often hinged on U.S. work experience, efforts to decouple these benefits from the labor market would greatly expand long-term care options for both earlier-life migrants who work in precarious or informal sectors, and later-life migrants with little to no U.S. work experience. Family reunification remains an essential policy goal, as family migration has positive socioeconomic outcomes for both families and society (Gubernskaya & Dreby, 2017), and future research should continue to push to inform policy change that bolster family reunification and support the well-being of aging immigrants and their families.

These findings have implications for family caregivers as well. As this and other studies show, adult children are an important means of care support for immigrants (Angel et al., 2014), particularly for later-life immigrants, and literature on care work shows that individuals who provide unpaid care often face penalties in the labor market, detriments to their own health and well-being, and additional obstacles to caring for both aging parents and young children, largely due to a distressing lack of support for family caregivers (Bauer & Sousa-Poza, 2015; Moussa, 2019; O’Sullivan, 2014; Wakabayashi & Donato, 2005). Caregivers from immigrant families may be particularly impacted, being disproportionately likely to work in lower income or less stable jobs and, depending on their own migration background, less likely to have social supports and resources to help care for aging parents (Flippen, 2014; Gans, 1992; León-Pérez et al., 2021).

There are limitations to this analysis. First, the sample size of foreign-born respondents, particularly those who migrated after age 50, was limited, and thus disaggregation of trends by racial/ethnic stratification, region of origin, or additional groupings of age at migration was not possible. Sample size constraints also limited the ability to examine intersections between racial/ethnic stratification and nativity status. Furthermore, given the extremely small number of later- and earlier-life migrants who identified as Black, Black individuals were excluded from this analysis, as patterns of care needs and arrangements among Black individuals would be reflected in the U.S.-born reference group, but not in the immigrant groups. Given recent increases in Black immigrants living in the United States (Hamilton, 2020), future studies using more recent data should incorporate Black immigrants to better understand both racial and nativity disparities in health and aging. Limited sample size also prevented separate analyses by gender, including gender of caregiver, which are important avenues for future research considering well-documented gender divisions in both care needs and caregiving (Finley, 1989; Glauber, 2017; Lee et al., 1993). Furthermore, while the analysis on care arrangements focuses on identifying the individual who “generally” provides care, research shows that individuals often have more than one caregiver (Andersson & Monin, 2018; Reyes et al., 2021; Xu et al., 2021), and future studies should look at caregiving more holistically, as immigrants may have distinct networks of care providers due to varying availability of family members living in the United States. Finally, future studies should examine how factors such as social supports, health status, and living arrangements impact the care needs and care arrangements of older foreign-born adults, which literature shows to be important mechanisms for aging-related well-being (Angel & Angel, 2006; Hays, 2002; House et al., 1989; Walsh & Walsh, 1987), and that vary greatly depending on life stage at migration (Alston & Aguirre, 1987; Angel & Angel, 1992; Gubernskaya & Tang, 2017; Van Hook & Glick, 2007).

Nevertheless, this research adds to the body of literature on immigrant health and aging by examining specific measures of care needs and care arrangements, with an emphasis on how life course theory and access to health-related resources explain care in older ages. Notably, the SIPP allows for the inclusion of Asian Americans in the analysis, who, despite being the fastest growing immigrant and racial/ethnic group in the United States (Budiman, 2020; Budiman & Ruiz, 2021), are often left out of research or grouped into an “other race” category. Future research should continue to explore how migration status intersects with racial/ethnic stratification to shape aging-related care needs and arrangements, and the impact of social policies on health and well-being for both aging individuals and their families, which, as this and other research demonstrates, is particularly important for groups who are systematically excluded from public supports.

Footnotes

Authors’ Note

An earlier draft of this article was presented at the 2022 Population Association of America Meeting.

Acknowledgments

I am grateful to Chenoa Flippen for guidance and mentorship with this study, as well as Pilar Gonalons-Pons, Emilio Parrado, Courtney Boen, and Mikko Myrskylä for their feedback to strengthen the article. I also thank the editor and three anonymous reviewers for constructive and insightful comments and suggestions.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work received support from the Population Research Training Grant (NIH T32 HD007242) awarded to the Population Studies Center at the University of Pennsylvania by the National Institutes of Health’s (NIH)’s Eunice Kennedy Shriver National Institute of Child Health and Human Development.

Note

ORCID iD

Zohra Ansari-Thomas

References

Abraído-Lanza

A. F.

Dohrenwend

B. P.

Ng-Mak

D. S.

Turner

J. B.

(1999). The latino mortality paradox: A test of the “salmon bias” and healthy migrant hypothesis. American Journal of Public Health, 89(1), 1543–1548. https://doi.org/10.2105/ajph.89.10.1543

Aguila

Zissimopoulos

(2013). Retirement and health benefits for Mexican migrant workers returning from the United States. International Social Security Review, 66(2), 101–125. https://doi.org/10.1111/issr.12014

Alston

L. T.

Aguirre

(1987). Elderly Mexican Americans: Nativity and health access. International Migration Review, 21(3), 626–642. https://doi.org/10.1177/019791838702100310

Andersson

M. A.

Monin

J. K.

(2018). Informal care networks in the context of multimorbidity: Size, composition, and associations with recipient psychological well-being. Journal of Aging and Health, 30(4), 641–664. https://doi.org/10.1177/0898264316687623

Angel

J. L.

Angel

R. J.

(1992). Age at migration, social connections, and well-being among elderly Hispanics. Journal of Aging and Health, 4(4), 480–499. https://doi.org/10.1177/089826439200400402

Angel

J. L.

Angel

R. J.

(2006). Minority group status and healthful aging: Social structure still matters. American Journal of Public Health, 96(7), 1152–1159. https://doi.org/10.2105/AJPH.2006.085530

Angel

J. L..

Buckley

C. J.

Sakamoto

(2001). Duration or disadvantage? Exploring nativity, ethnicity, and health in midlife. Journal of Gerontology: Social Sciences, 56B(5), S275–S284. https://doi.org/10.1093/geronb/56.5.s275

Angel

J. L.

Rote

S. M.

Brown

D. C.

Angel

R. J.

Markides

K. S.

(2014). Nativity status and sources of care assistance among elderly Mexican-origin adults. Journal of Cross-Cultural Gerontology, 29(3), 243–258. https://doi.org/10.1007/s10823-014-9234-9

Angel

R. J.

Angel

J. L.

Lee

G. Y.

Markides

K. S.

(1999). Age at migration and family dependency among older Mexican immigrants: Recent evidence from the Mexican American EPESE. The Gerontologist, 39(1), 59–65. https://doi.org/10.1093/geront/39.1.59

10.

Angel

R. J.

Angel

J. L.

Markides

K. S.

(2002). Stability and change in health insurance among older Mexican Americans: Longitudinal evidence from the hispanic established populations for epidemiologic study of the elderly. American Journal of Public Health, 92(8), 1264–1271. https://doi.org/10.2105/ajph.92.8.1264

11.

Batalova

(2012). Senior immigrants in the United States in 2010. Migration Policy Institute. https://www.migrationpolicy.org/article/senior-immigrants-united-states-2010

12.

Bauer

J. M.

Sousa-Poza

(2015). Impacts of informal caregiving on caregiver employment, health, and family. Population Ageing, 8(3), 113–145. https://doi.org/10.1007/s12062-015-9116-0

13.

Blakemore

(1999). International migration in later life: Social care and policy implications. Ageing and Society, 19(6), 761–774. https://doi.org/10.1017/S0144686X99007503

14.

Boen

C. E.

Hummer

R. A.

(2019). Longer—but harder—lives? The hispanic health paradox and the social determinants of racial, ethnic, and immigrant-native health disparities from midlife through late life. Journal of Health and Social Behavior, 60(4), 434–452. https://doi.org/10.1177/0022146519884538

15.

Budiman

(2020). Key findings about U.S. Immigrants. Pew Research Center. https://www.pewresearch.org/fact-tank/2020/08/20/key-findings-about-u-s-immigrants/

16.

Budiman

Ruiz

N. G.

(2021). Asian Americans are the fastest-growing Racial or ethnic Group in the U.S. Pew Research Center. https://www.pewresearch.org/fact-tank/2021/04/09/asian-americans-are-the-fastest-growing-racial-or-ethnic-group-in-the-u-s/

17.

Carr

Tienda

(2013). Family sponsorship and late-age immigration in aging America: Revised and expanded estimates of chained migration. Population Research and Policy Review, 32(6), 825–849. https://doi.org/10.1007/s11113-013-9300-y

18.

Chevan

(1995). Holding on and letting go: Residential mobility during widowhood. Research on Aging, 17(3), 278–302. https://doi.org/10.1177/0164027595173003

19.

Choi

N. G.

(1996). Older persons who move: Reasons and health consequences. The Journal of Applied Gerontology, 15(3), 325–344. https://doi.org/10.1177/073346489601500304

20.

Choi

S. H.

(2012). Testing health immigrant effects among late life immigrants in the United States: Using multiple indicators. Journal of Aging and Health, 24(3), 475–506. https://doi.org/10.1177/0898264311425596

21.

Conti

Heckman

Urzua

(2010). The education-health gradient. American Economic Review: Papers & Proceedings, 100(2), 234–238. https://doi.org/10.1257/aer.100.2.234

22.

Crocker

R. M.

(2021). The impact of binational barriers to medical care on the care-seeking practices of Mexican immigrants. Qualitative Health Research, 31(6), 1043–1055. https://doi.org/10.1177/1049732321992041

23.

Elder

G. H.

Johnson

M. K.

(2003). The life course and aging: Challenges, lessons, and new directions. In Settersten

R. A.

(Ed.), Inviation to the life course: Toward new understandings of later life (pp. 49–81). Baywood.

24.

Elo

I. T.

(2009). Social class differentials in health and mortality: Patterns and explanations in comparative perspective. Annual Review of Sociology, 35(1), 553–572. https://doi.org/10.1146/annurev-soc-070308-115929

25.

Elo

I. T.

Preston

S. H.

(1992). Effects of early-life conditions on adult mortality: A review. Population Index, 58(2), 186–212. https://doi.org/10.2307/3644718

26.

Findley

S. E.

(1988). The directionality and age selectivity of the health-migration relation: Evidence from sequences of disability and mobility in the United States. International Migration Review, 22(3), 4–29. https://doi.org/10.2307/2546583

27.

Finley

N. J.

(1989). Theories of family labor as applied to gender differences in caregiving for elderly parents. Journal of Marriage and Family, 51(1), 79–86. https://doi.org/10.2307/352370

28.

Fix

Passel

(2002). The scope and impact of welfare Reform’s immigrant provisions. The Urban Institute. https://www.urban.org/sites/default/files/publication/60346/410412-Scope-and-Impact-of-Welfare-Reform-s-Immigrant-Provisions-The.PDF

29.

Flippen

C. A.

(2014). Intersectionality at work: Determinants of labor supply among immigrant Latinas. Gender & Society, 28(3), 404–434. https://doi.org/10.1177/0891243213504032

30.

Gans

H. J.

(1992). Second-generation decline: Scenarios for the economic and ethnic futures of the post-1965 American immigrants. Ethnic and Racial Studies, 15(2), 173–192. https://doi.org/10.1080/01419870.1992.9993740

31.

Garcia

M. A.

Downer

Crowe

Markides

K. S.

(2017). Aging and disability among hispanics in the United States: Current knowledge and future directions. Innovations in Aging, 1(2), 1–11. https://doi.org/10.1093/geroni/igx020

32.

Garcia

M. A.

Reyes

A. M.

(2018). Prevalence and trends in morbidity and disability among older Mexican Americans in the Southwestern United States, 1993-2013. Research on Aging, 40(4), 311–339. https://doi.org/10.1177/0164027517697800

33.

Glauber

(2017). Gender differences in spousal care across the later life course. Research on Aging, 39(8), 934–959. https://doi.org/10.1177/0164027516644503

34.

Goldman

Korenman

Weinstein

(1995). Marital status and health among the elderly. Social Science and Medicine, 40(12), 1717–1730. https://doi.org/10.1016/0277-9536(94)00281-w

35.

Gubernskaya

(2014). Age at migration and self-rated health trajectories after age 50: Understanding the older immigrant health paradox. Journals of Gerontology, Series B: Psychological Science and Social Sciences, 70(2), 279–290. https://doi.org/10.1093/geronb/gbu049

36.

Gubernskaya

Dreby

(2017). U.S. immigration policy and the case for family unity. Journal on Migration and Human Security, 5(2), 417–430. https://doi.org/10.1177/233150241700500210

37.

Gubernskaya

Kim

(2018). Childhood disadvantage and physical health of older immigrants in the U.S. Innovation in Aging, 2(1), 465. https://doi.org/10.1093/geroni/igy023.1739

38.

Gubernskaya

Tang

(2017). Just like in their home country? A multinational perspective on living arrangements of older immigrants in the United States. Demography, 54(5), 1973–1998. https://doi.org/10.1007/s13524-017-0604-0

39.

Guo

Stensland

Dong

Tiwari

(2019). Is migration at older age associated with poorer psychological well-being? Evidence from Chinese older immigrants in the United States. The Gerontologist, 59(5), 865–876. https://doi.org/10.1093/geront/gny066

40.

Hagan

Rodriguez

Capps

Kabiri

(2003). The effects of recent welfare and immigration reforms on immigrants’ access to health care. International Migration Review, 37(2), 444–463. https://doi.org/10.1111/j.1747-7379.2003.tb00144.x

41.

Hagan

J. M.

(2004). Contextualizing immigrant labor market incorporation: Legal, demographic, and economic dimensions. Work and Occupations, 31(4), 407–423. https://doi.org/10.1177/0730888404268921

42.

Hamilton

T. G.

(2020). Black immigrants and the changing portrait of Black America. Annual Review of Sociology, 46, 295–313. https://doi.org/10.1146/annurev-soc-121919-054728

43.

Hays

J. C.

(2002). Living arrangements and health status in later life: A review of recent literature. Public Health Nursing, 19(2), 136–151. https://doi.org/10.1046/j.1525-1446.2002.00209.x

44.

Hayward

M. D.

Hummer

R. A.

Chiu

C. T.

González-González

Wong

(2014). Does the Hispanic paradox in U.S. adult mortality extend to disability? Population Research and Policy Review, 33(1), 81–96. https://doi.org/10.1007/s11113-013-9312-7

45.

Herd

Schoeni

R. F.

House

J. S.

(2008). Upstream solutions: Does the supplemental security income program reduce disability in the elderly? Milbank Quarterly, 86(1), 5–45. https://doi.org/10.1111/j.1468-0009.2007.00512.x

46.

Hill

T. D.

Angel

J. L.

Balistreri

K. S.

Herrera

A. P.

(2012). Immigrant status and cognitive functioning in late-life: An examination of gender variations in the healthy immigrant effect. Social Science and Medicine, 75(12), 2076–2084. https://doi.org/10.1016/j.socscimed.2012.04.005

47.

Hirschman

Wong

M. G.

(1981). Trends in socioeconomic achievement among immigrant and native-born asian Americans, 1960-1976. The Sociological Quarterly, 22(4), 495–514. https://doi.org/10.1111/j.1533-8525.1981.tb00677.x

48.

House

J. S.

Landis

K. R.

Umberson

(1989). Social relationships and health. Science, 241(4865), 540–545. https://doi.org/10.1126/science.3399889

49.

Ihara

E. S.

(2011). Exploring the socioeconomic and immigrant health contexts of Asian Americans. Journal of Human Behavior in the Social Environment, 21(5), 521–539. https://doi.org/10.1080/10911359.2011.580239

50.

Kim

Chiriboga

D. A.

Jang

Lee

Huang

Parmelee

(2010). Health status of older asian Americans in California. Journal of the American Geriatrics Society, 58(10), 2003–2008. https://doi.org/10.1111/j.1532-5415.2010.03034.x

51.

Kimbro

R. T.

Bzostek

Goldman

Rodríguez

(2008). Race, ethnicity, and the education gradient in health. Health Affairs, 27(2), 361–372. https://doi.org/10.1377/hlthaff.27.2.361

52.

Kovnick

M. O.

Young

Tran

Teerawichitchainan

Tran

T. K.

Korinek

(2021). The impact of early life war exposure on mental health among older adults in northern and central Vietnam. Journal of Health and Social Behavior, 62(4), 526–544. https://doi.org/10.1177/00221465211039239

53.

Lauderdale

D. S.

Kestenbaum

(2002). Mortality rates of elderly asian American populations based on medicare and social security data. Demography, 39(3), 529–540. https://doi.org/10.1353/dem.2002.0028

54.

Leach

M. A.

(2009). America’s older immigrants: A profile. Generations, 32(4), 34–39.

55.

Lee

G. R.

Dwyer

J. W.

Coward

R. T.

(1993). Gender differences in parent care: Demographic factors and same-gender preference. Journal of Gerontology: Social Sciences, 48(1), S9–S16. https://doi.org/10.1093/geronj/48.1.s9

56.

León-Pérez

Richards

Non

A. L.

(2021). Precarious work and parenting stress among Mexican immigrant women in the United States. Journal of Marriage and Family, 83(1), 881–897. https://doi.org/10.1111/jomf.12761

57.

Liao

Cooper

R. S.

Cao

Durazo-Arvizu

Kaufman

J. S.

Luke

McGee

D. L.

(1998). Mortality patterns among adult hispanics: Findings from the NHIS 1986 to 1990. American Journal of Public Health, 88(2), 227–232. https://doi.org/10.2105/ajph.88.2.227

58.

Litwak

Kulis

(1987). Technology, proximity, and measures of kin support. Journal of Marriage and Family, 49(3), 649–661. https://doi.org/10.2307/352210

59.

Liu

Umberson

D. J.

(2008). The times they are a changin’: Marital status and health differentials from 1972 to 2003. Journal of Health and Social Behavior, 49(3), 239–253. https://doi.org/10.1177/002214650804900301

60.

Maleku

España

Jarrott

Karandikar

Parekh

(2021). We are aging too! Exploring the social impacts of late-life migration among older immigrants in the United States. Journal of Immigrant and Refugee Studies, 20(3), 1–18. https://doi.org/10.1080/15562948.2021.1929643

61.

Markides

K. S.

Coreil

(1986). The health of hispanics in the southwestern United States: An epidemiologic paradox. Public Health Reports, 101(3), 253–265. https://www-jstor-org-443.web.bisu.edu.cn/stable/4627869

62.

Markides

K. S.

Rote

(2019). The healthy immigrant effect and aging in the United States and other western countries. The Gerontologist, 59(2), 205–214. https://doi.org/10.1093/geront/gny136

63.

Markides

K. S.

Salinas

Sheffield

(2009). The health of older immigrants. Generations: Journal of the American Society on Aging, 32(4), 46–52. https://www-jstor-org-443.web.bisu.edu.cn/stable/26555616

64.

Melvin

Hummer

Elo

Mehta

(2014). Age patterns of racial/ethnic/nativity differences in disability and physical functioning in the United States. Demographic Research, 31(17), 497–510. https://doi.org/10.4054/DemRes.2014.31.17

65.

Miyawaki

C. E.

(2015). A review of ethnicity, culture, and acculturation among Asian caregivers of older adults (2000-2012). Sage Open, 5(1), 1–18. https://doi.org/10.1177/2158244014566365

66.

Mizoguchi

Walker

Trevelyan

Ahmed

(2019). The older foreign-born population in the United States: 2012-2016. U.S. Census Bureau. American Community Survey Reports, ACS-42 https://www.census.gov/content/dam/Census/library/publications/2019/acs/acs-42.pdf

67.

Moussa

M. M.

(2019). The relationship between elder care-giving and labour force participation in the context of policies addressing population ageing: A review of empirical studies published between 2006 and 2016. Ageing & Society, 39(6), 1281–1310. https://doi.org/10.1017/S0144686X18000053

68.

Mutchler

J. E.

Prakash

Burr

J. A.

(2007). The demographic of disability and the effects of immigrant history: Older asians in the United States. Demography, 44(2), 251–263. https://doi.org/10.1353/dem.2007.0015

69.

National Immigration Law Center . (2020). The HEAL Act expands access to health care Regardless of immigration status. https://www.nilc.org/2020/05/20/heal-act-expands-access-to-health-care-regardless-of-immigration-status/

70.

Ornstein

K. A.

Kelley

A. S.

Bollens-Lund

Wolff

J. L.

(2017). A national profile of end-of-life caregiving in the United States. Health Affairs, 36(7), 1184–1192. https://doi.org/10.1377/hlthaff.2017.0134

71.

O’Sullivan

(2014). Pulled from all sides: The sandwich generation at work. Work, 50(3), 491–494. https://doi.org/10.3233/WOR-141959

72.

Palloni

Arias

(2004). Paradox lost: Explaining the Hispanic adult mortality advantage. Demography, 41(3), 385–415. https://doi.org/10.1353/dem.2004.0024

73.

Pew Research Center . (2015). Modern immigration wave brings 59 million to U.S., driving population growth and change through 2065. https://www.pewresearch.org/hispanic/2015/09/28/modern-immigration-wave-brings-59-million-to-u-s-driving-population-growth-and-change-through-2065/

74.

Reyes

A. M.

Schoeni

R. F.

Freedman

V. A.

(2021). National estimates of kinship size and composition among adults with activity limitations in the United States. Demographic Research, 45(36), 1097–1114. https://doi.org/10.2105/AJPH.2012.301103

75.

Ruiz

J. M.

Steffen

Smith

T. B.

(2012). Hispanic mortality paradox: A systematic review and meta-analysis of the longitudinal literature. American Journal of Public Health, 103(3), e52–e60. https://doi.org/10.2105/AJPH.2012.301103

76.

Sorlie

P. D.

Backlund

Johnson

N. J.

Rogot

(1993). Mortality by hispanic status in the United States. Journal of the American Medical Association, 270(20), 2464–2468. https://doi.org/10.1001/jama.1993.03510200070034

77.

Taubman

P. J.

Sickles

R. C.

(1983). Supplemental social insurance and the health of the poor. NBER, p. 1062.

78.

Tienda

(2017). Multiplying diversity: Family unification and the regional origins of late-age U.S. immigrants. International Migration Review, 51(3), 727–756. https://doi.org/10.1111/imre.12241

79.

Tran

V. C.

Lee

Khachikian

Lee

(2018). Hyper-selectivity, racial mobility, and the remaking of race. RSF: The Russell Sage Foundation Journal of the Social Sciences, 4(5), 188–209. https://doi.org/10.7758/rsf.2018.4.5.09

80.

Treas

Batalova

(2009). Immigrants and aging. In Uhlenberg

(Ed.), International handbook of population aging (pp. 365–394). Springer Netherlands.

81.

Treas

Gubernskaya

(2015). Policy contradictions and immigrant families. Public Policy & Aging Report, 25(3), 107–112. https://doi.org/10.1093/ppar/prv010

82.

Van Hook

Glick

J. E.

(2007). Immigration and living arrangements: Moving beyond economic need versus acculturation. Demography, 44(2), 225–249. https://doi.org/10.1353/dem.2007.0019

83.

Wakabayashi

(2010). Effects of immigration and age on health of older people in the United States. Journal of Applied Gerontology, 29(6), 697–719. https://doi.org/10.1177/0733464809353602

84.

Wakabayashi

Donato

K. M.

(2005). The consequences of caregiving effects on women’s employment and earnings. Population Research and Policy Review, 24(5), 467–488. https://doi.org/10.1007/s11113-005-3805-y

85.

Walsh

P. A.

(1987). Social support, assimilation, and biological effective blood pressure levels. International Migration Review, 21(3), 577–591. https://doi.org/10.1177/019791838702100307

86.

Walters

W. H.

(2002). Later-life migration in the United States: A review of recent research. Journal of Planning Literature, 17(1), 37–66. https://doi.org/10.1177/088541220201700103

87.

Wolff

J. L.

Mulcahy

Huang

Roth

D. L.

Covinsky

Kasper

J. D.

(2018). Family caregivers of older adults, 1999-2015: Trends in characteristics, circumstances, and role-related appraisal. The Gerontologist, 58(6), 1021–1032. https://doi.org/10.1093/geront/gnx093

88.

Wolgin

P. E.

(2018). Family reunification is the bedrock of U.S. Immigration policy. Center for American Progress. https://www.americanprogress.org/article/family-reunification-bedrock-u-s-immigration-policy/

89.

Liu

P. J.

Beach

(2021). Multiple caregivers, many minds: Family discord and caregiver outcomes. The Gerontologist, 61(5), 661–669. https://doi.org/10.1093/geront/gnaa067

90.

Yeo

(2017). Healthcare inequality issues among immigrant elders after neoliberal welfare reform: Empirical findings from the United States. European Journal of Health Economics, 18(5), 547–565. https://doi.org/10.1007/s10198-016-0809-y

91.

Zhou

Bankston

C. L.

(2020). The model minority stereotype and the national identity question: The challenges facing Asian immigrants and their children. Ethnic and Racial Studies, 43(1), 233–253. https://doi.org/10.1080/01419870.2019.1667511