Falling Through the Cracks: Lack of Health Insurance Among Elderly Foreign- and Native-Born Blacks

Transcription

1 DOI /s ORIGINAL PAPER Falling Through the Cracks: Lack of Health Insurance Among Elderly Foreign- and Native-Born Blacks Karyn A. Stewart Andrew S. London Ó Springer Science+Business Media New York 2014 Abstract Little research examines lack of health insurance among elderly Black immigrants in the US. We use data from the 2008 American Community Survey to describe variation in insurance coverage and conduct multivariate logistic regression analyses of uninsurance. Among elderly Blacks, 1.7 % of the US-born were uninsured, compared to 8.4 % of the Latin American and Caribbean-born, 23.2 % of the African-born, and 9.3 % of those born in other regions. In multivariate models, relative to the US-born, the odds of being uninsured were significantly higher among each immigrant group. Among immigrants, the odds of being uninsured were 3.80 times higher among African-born than Latin American and Caribbean-born immigrants net of demographic and socioeconomic controls. This difference was explained by the inclusion of either year of immigration or length of residence. Relative to Latin America and Caribbean-born immigrants, the odds of being uninsured were significantly higher among immigrants from other regions only in the model that included the immigration-related variables. This suppression effect was evident when either length of residence or citizenship was controlled. Recently-arrived, elderly Black immigrants fall through the cracks of insurance coverage. Results are discussed in relation to public and private safety net options. K. A. Stewart (&) African and African American Studies Program, Washington University in St. Louis, One Brookings Drive, Campus Box 1109, St. Louis, MO , USA kstewart@artsci.wustl.edu A. S. London Department of Sociology, Center for Policy Research, Aging Studies Institute, Syracuse University, Syracuse, NY, USA Keywords Health insurance Elderly Black immigrants African Americans Introduction Currently, we know relatively little about health insurance coverage among elderly Black immigrants in the US and how it varies. Even though sponsors pledge to meet the financial needs of parents and other relatives who enter under family reunification provisions, private insurance is an expensive alternative that many families ultimately cannot afford and access to publicly-funded programs is restricted. Thus, some elderly Black immigrants may be falling through the cracks of health insurance coverage as a result of not meeting Medicare qualifications, lack of information regarding Medicare or Medicaid eligibility, and inadequate personal and familial resources to purchase insurance. Given the increased immigration of Blacks from Latin America, the Caribbean, and Africa in recent decades [1, 2] and the potential future increase as a result of family reunification, it is important to consider the scope of uninsurance among elderly Black immigrants and the factors that are associated with it. The number of elderly immigrants living in the US increased from 2.7 million in 1990 to 4.3 million in 2006, and is projected to quadruple by 2050 [3]. While many currently elderly immigrants entered young, aged in the US, worked, became citizens, and ultimately qualified for Medicare, the invited elderly are often later-life migrants who move to be closer to their adult children and grandchildren [4]. The invited elderly are more likely than other migrants to be in poor psychological and physical health, to be financially dependent, to experience linguistic and cultural barriers to social integration, and to need care and

2 support, even if they can make contributions to the family and household in terms of child care or domestic labor [4 6]. The invited elderly are particularly vulnerable to being uninsured because of Medicare eligibility requirements that disqualify non-citizens and those with less than 40 quarters of work experience with a Medicare-taxpaying American employer [4, 7 9]. This places the invited elderly at risk for having unmet medical needs, may increase emergency room utilization, and places financial burden on some immigrant families that pay outof-pocket for elderly uninsured parents needed medical care [4, 7, 8, 10]. The Kaiser Commission on Medicaid and the Uninsured/Urban Institute estimates that 646,000 adults aged 65 and over were uninsured in 2008 [11]. This represents a twofold increase from the estimated 252, ,000 uninsured elderly in 2000 [12, 13]. Among the elderly, higher rates of being uninsured have been documented among: the relatively young (65 74 years old); non-citizens; Blacks and Hispanics; those with less than a high school diploma; the poor; the widowed, formerly-, and never-married; those living in the South; and those reporting fair or poor health [12, 13]. In 2000, 56 % of the uninsured elderly were born outside of the US [12]. Among the foreign-born elderly uninsured, approximately 6 and 18 % had arrived within the past 1 and 5 years, respectively [12]. These results highlight the potential importance of recent immigration for lack of insurance among the elderly, and suggest that uninsurance may increase in the first few years of residence in the US. The provisions of the 1965 Hart Cellar Act contributed to an increased flow of immigrants from Latin America, the Caribbean, and Asia, as well as an increase in the number of invited elderly immigrants [2, 14, 15]. From 1960 to 2005, the Black immigrant population including immigrants from Sub-Saharan Africa increased from 125,000 to 2,815,000 primarily because of family reunification, diversity visas, and individuals seeking refugee status [2]. Despite this increase in the size and diversification of the Black immigrant population, little research focuses specifically on the insurance status of elderly Black immigrants. Using data from 2000, one study documents that approximately 10 % of foreign-born elderly Blacks and 35 % of elderly Black non-citizens were uninsured compared to about 2 % of native-born elderly Black citizens [12]. Although Black immigrants generally tend to be positively selected on health and health-related factors [16 20] and demonstrate better socioeconomic profiles than US-born Blacks [12, 21, 22], health conditions are emergent as people age, family resources and needs change, and private insurance costs are high and increasing. Based on existing research, it is unclear to what extent older Black immigrants are able to access health insurance in the US. Moreover, it is likely that use of public and private health insurance varies among subgroups of the older Black immigrant population in the short-term and over time in ways that have not been systematically documented. In this paper, we use data from the 2008 American Community Survey to address three research questions: (1) What is the prevalence of uninsurance among native- and foreign-born elderly Blacks in the US? (2) Do demographic and socioeconomic variables explain observed nativity differences in uninsurance? and (3) Among immigrants, do demographic, socioeconomic, and immigration-related variables (i.e., year of immigration, duration of residence, citizenship status, and English language proficiency) explain region-of-origin differences in uninsurance? Prior research demonstrates a significant association between place of birth and health insurance status among Blacks [23]. Contemporary research also suggests that citizenship status and timing of migration are both important to the likelihood of being uninsured [24]. The ability to speak English is related to the capacity to work, seek citizenship, and understand eligibility for public insurance, which may be related to insurance coverage. Based on the available evidence, we expect that foreign-born elderly Blacks will be significantly more likely than the US-born elderly Blacks to be uninsured, and that observed differences will to some extent be explained by demographic and socioeconomic characteristics that are associated with nativity status and insurance coverage. Furthermore, among the foreign-born, we expect that African-born immigrants will be significantly more likely than immigrants from Latin America and the Caribbean or other regions to be uninsured, observed region-of-origin differences in uninsurance will be explained to some extent by between-group differences in demographic and socioeconomic characteristics, and immigration-related variables will be additionally important for understanding lack of insurance among elderly Black immigrants. Data and Methods Data This study uses publicly-available data from the 2008 American Community Survey (ACS) [25], which is collected annually by the US Census Bureau. We focus on the Black population aged 65 and over and include all persons who identified themselves as Black, including persons who identified themselves as Black and another race. The analytic sample includes 35,635 elderly Blacks, of whom 32,949 are US-born and 2,686 are immigrants.

3 Dependent Variable The primary dichotomous dependent variable measures insurance status, with no insurance = 1. For descriptive purposes, we define two other variables that measure insurance coverage. One identifies those with private insurance only, public insurance only, and both public and private insurance. A second identifies specific types of private and public insurance: employer-based; purchased directly; TRICARE; Indian Health Service; Medicaid; Medicare; and Veterans Affairs. Independent Variable Our primary independent variable measures nativity status. Respondents reported their nativity status and, if they were foreign-born, their country of birth. We constructed a fourcategory region of origin variable: US, Latin America and the Caribbean, Africa, and other. Demographic, Socioeconomic, and Immigration- Related Controls Variables We include in our analyses a set of demographic and socioeconomic control variables that are likely to be associated with nativity, region of origin, and insurance coverage. We code age categorically: 65 74, 75 84, and 85 or more years. Sex is dichotomous (male = 1). We code marital status as never married, married, separated, divorced, and widowed, and education as no school, less than high school, high school diploma, some college, and college degree or more. Employment status differentiates those who are employed, not in the labor force, and unemployed. The ACS provides a measure of income-toneeds, which we recode to reference below poverty, % of poverty, % of poverty, and 300 % or more of poverty. We code region of current US residence as Northeast, Midwest, South, and West. In analyses focused on immigrants, we also include immigration-related variables. We code citizenship status as naturalized citizen and non-citizen; persons born abroad to American parents (N = 28) are coded as US-born. We code English language proficiency as does not speak English, speaks English well or very well, and speaks only English. Length of residence in the US is coded as 0 5 years, 6 10 years, years, and 16 years or more. Year of immigration is coded as before 1980, , and These categories demarcate periods during which immigration policies were passed that increased and changed migration flows from Africa and the Caribbean [2]. Specifically, the Refugee Act of March 17, 1980 led to increased immigration from the horn of Africa, the 1986 Immigration Reform and Control Act promoted immigration from both the Caribbean and Africa, and the Immigration Act of 1990 emphasized diversity visas and promoted employment-related migration of highly-skilled Africans. Analysis We begin by describing the sample overall, and demographic and socioeconomic differences between nativeand foreign-born Blacks. Next, we estimate the level of uninsurance and the types of insurance used by those with insurance, overall and by region of origin. Design-based F tests are used to evaluate the significance of between-group differences. For the total sample, we then estimate two hierarchical multivariate logistic regression models predicting lack of insurance. The first model includes nativity status only; the second adds demographic and socioeconomic variables that may explain nativity status differences in uninsurance. Finally, for the immigrant subsample, we estimate three hierarchical multivariate logistic regression models of lack of insurance. The first model includes the region of origin variable only, the second adds controls for demographic and socioeconomic characteristics, and the third adds the set of immigration-related control variables. Supplemental analyses were also conducted and are reported in the text. We conducted all analyses using the SVY commands in Stata. In the multivariate analyses, the reference category for each control variable is the modal category based on the full sample or, for the immigration-related variables, the immigrant subsample. We weight all analyses and adjust standard errors for the complex sampling design. Results Sample Description Table 1 provides a description of the Black elderly population in 2008, overall and by region of origin. Overall, 91.3 % were US-born, 6.4 % were born in Latin America and the Caribbean, and roughly 1 2 %, respectively, were African-born or born in other regions. More than half were aged 65 74, one-third were 75 84, and about 12 % were 85 or older. Two-thirds were female. Approximately threequarters had a high school education or less and 84.7 % were not in the labor force. One-fourth lived in households below the poverty threshold, and another 26 % lived in households with income-to-needs ratios % of the poverty threshold. Most were either married or widowed (35.0 and 36.5 %, respectively). Over half lived in the South. Most of the foreign-born immigrated before 1980

4 Table 1 Descriptive statistics, elderly Blacks by nativity, 2008 American Community Survey Variable Total sample 35,635 US-born 32,949 Foreign-born 2,686 p level % Number % Number % Number Region of origin United States ,949 Latin America and the Caribbean 6.4 2,014 Africa Other Age years , , ,688 *** years , , ? years , , Sex Female , , ,602 Male , , ,084 Education No school 3.9 1, , *** \High school , , High school graduate , , ,072 Some college , , College graduate or more , , Employment status Employed , , *** Not in labor force , , ,085 Unemployed Income-to-needs ratio Below poverty , , *** % of poverty , , % of poverty , , ? % of poverty , , ,130 Marital status Never married 8.6 2, , *** Married , , ,185 Separated 3.9 1, , Divorced , , Widowed , , Region of US residence Northeast , , ,336 *** Midwest , , South , , ,108 West 9.0 3, , Year of immigration (foreign-born only) Before , Citizenship status (foreign-born only) Naturalized ,005 Non-citizen English language proficiency (foreign-born only) Does not speak English

5 Table 1 continued Variable Total sample 35,635 US-born 32,949 Foreign-born 2,686 p level % Number % Number % Number Speaks English well or very well Speaks only English ,565 Length of residence (foreign-born only) 0 5 years years years ? years ,200 Table 2 Percentage uninsured, insured and type of insurance, ACS 2008 Region of birth Total (%) United States (%) Latin America and The Caribbean (%) Africa Other p level a The total percentages for type of insurance will not sum to 100 because the insurance type categories are not mutually exclusive Panel A Insurance *** No insurance Privately insured only Publicly insured only Both public and private Insurance Panel B Type of insurance among the insured a Private Employer *** Purchased *** directly TRICARE *** Public Indian health service Medicaid Medicare *** Veterans affairs *** (56.2 %), were naturalized citizens (70.2 %), spoke only English (54.6 %), and had been in the US for 16 years or more (78.2 %). As seen in Table 1, the US-born and foreign-born Black elderly differed significantly on each of the demographic and socioeconomic variables except sex. Compared to the US-born, the foreign-born were more likely to be: younger; without education, high school graduates, and with collegeeducation or more; in the labor force; in households with income-to-needs ratios 300? % of poverty; never- and currently married; and living in the Northeast. Insurance Status Table 2 describes the health insurance coverage of the Black elderly in As shown in Panel A, 2.5 % of the Black elderly were uninsured. The majority had public insurance only (47.2 %) or was covered by both public and private insurance (43.7 %); only 6.7 % had private insurance only. Immigrants were significantly more likely than the US-born to be uninsured; uninsurance rates were 1.7 % among the US-born, 8.4 % among immigrants from Latin America and the Caribbean, 23.2 % among immigrants

6 from Africa, and 9.3 % among immigrants from other regions. There was also variation in type of insurance by region of origin. Both immigrants from Latin America and the Caribbean (10.1 %) and African immigrants (12.1 %) had higher rates of private insurance only compared to the US-born (6.4 %); immigrants from Latin America and the Caribbean also had a higher rate of public insurance only (53.6 vs % among the US-born). Compared to the US-born, immigrants from all three regions had lower rates of combined private and public insurance. Table 2, Panel B provides additional detail regarding type of insurance coverage among those with insurance and how that varied by region of origin. Among the insured, the most common form of health insurance was Medicare (86.4 %). Approximately one-third had employer-based insurance, 27.4 % were covered by Medicaid, 23.5 % purchased private insurance directly, 6.5 % had insurance coverage through Veterans Affairs, and smaller percentages had other forms of health insurance. Significant variation by region of origin was evident for all three types of private insurance and for two types of public insurance Medicare and Veterans Affairs. Immigrants from each of the three regions had lower rates of employer-based, directly purchased private insurance, and TRICARE than the US-born. Immigrants from Latin America and the Caribbean and from Africa, respectively, had very low rates of coverage under TRICARE. Medicare coverage ranged from a high of 87.8 % for US-born Blacks to a low of 45.6 % among African-born Blacks. Rates of Medicaid coverage did not vary significantly by region of origin. All three immigrant groups had lower rates of coverage by Veterans Affairs than the US-born, particularly immigrants from Latin America and the Caribbean and from Africa. Although rates of uninsurance are substantially lower among the native-born than each of the foreign-born subgroups, the native-born population is larger. Thus, there is a larger number of uninsured native-born than foreign-born elderly Blacks. In a supplemental analysis (not shown), we estimated the number uninsured, overall and by region of origin. Overall, we estimated that 81,255 elderly Blacks were uninsured in The number of US-born Blacks who were uninsured was 51,934 (63.9 % of total), compared to 17,873 Latin America and Caribbean-born Blacks (22.0 % of total), 7,492 African-born Blacks (9.2 % of total), and 3,956 Black born in other regions (4.9 % of total). Multivariate Analyses Foreign-Born Disadvantage Table 3 presents results from logistic regression analyses modeling the likelihood of being uninsured in the full Table 3 Logistic regression models of non-insurance among nativeborn Blacks and Black immigrants, ACS 2008 Variable (reference category) No insurance Model 1 Model 2 OR 95 % CI OR 95 % CI Region of origin (US) Latin America and the Caribbean 5.26*** *** Africa 17.35*** *** Other 5.89*** *** Sex (female) Male 1.35** Age (65 74 years) years 0.71** ? years Education (high school) No school \High school 1.27* Some college College graduate or more Employment status (not in labor force) Employed Unemployed 2.97*** Income-to-needs (300? % of poverty) Below poverty 1.96*** % of 1.34* poverty % of poverty Marital status (widowed) Married 0.71** Separated Divorced Never married/ single 1.91*** Region of US residence (south) Northeast 0.58*** Midwest West Constant 0.02*** *** Un-weighted N: 35,635 Significance levels: * p \ 0.05; ** p \ 0.01; *** p \ sample; the focus is on estimating the scope of the foreignborn disadvantage relative to the native-born. Model 1 includes the region of origin variable only; Model 2 adds the demographic and socioeconomic controls.

7 Model 1 indicates that the odds of being uninsured were significantly higher for each group of foreign-born Blacks relative to US-born Blacks. Interestingly, observed differences increased once controls were included in the models. Controlling for demographic and socioeconomic variables, compared to the US-born, the odds of being uninsured were 6.65 times higher for the Latin American and Caribbeanborn, times higher among the African-born, and 7.54 times higher among Blacks born in other regions. In addition to differences by region of origin, the odds of being uninsured were significantly higher among: men relative to women; those with less than high school education relative to high school graduates; those who were unemployed relative to those not in the labor force; those with income-to-needs ratios below poverty and % of poverty, respectively, relative to those with income-toneeds ratios 300? % of poverty; and the never married relative to the widowed. The odds of being uninsured were significantly lower among: those aged relative to those aged 65 74; the married relative to the widowed; and those living in the Northeast relative to those living in the South. Uninsurance Among Immigrants Table 4 presents results from logistic regression analyses modeling the likelihood of being uninsured among immigrants; the focus is on estimating the contribution of immigration-related variables to variation in uninsurance. Model 1 includes only the region of birth variable, Model 2 adds the demographic and socioeconomic controls, and Model 3 adds the immigration-related variables. In Model 1, the odds of being uninsured were 3.30 times higher among African-born immigrants than among immigrants born in Latin America and the Caribbean. Adding demographic and socioeconomic variables (Model 2) increased the difference between these two groups slightly. Immigrants from other regions were not different from immigrants from Latin America and the Caribbean in either Model 1 or Model 2. Model 3 includes the four immigration-related variables. The difference between African-born immigrants and immigrants born in Latin America and the Caribbean is fully explained by these variables, while the odds of being uninsured among immigrants born in regions other than Latin America, the Caribbean, and Africa were 2.16 and now significantly higher than among Latin America and Caribbean-born immigrants. In Model 3, relative to naturalized citizens, the odds of being uninsured were significantly higher among non-citizens. Compared to those who had been in the US. 16 or more years, the odds of being uninsured were significantly higher among those who had been in the US 0 5 years. Examination of Models 2 and 3 indicate that few other variables are associated with uninsurance among elderly Black immigrants. Adding the immigration-related variables explains the difference between those with less than high school education and high school graduates, and those who live in the West relative to those who live in the South, but does not explain the differences between those who are never-married compared to those who are widowed, and those who reside in the Northeast compared to those who reside in the South, respectively. In supplemental analyses (not shown), we added each immigration-related variable individually. Net of other variables, each of the four variables was significantly associated with uninsurance when entered independently. Compared to naturalized citizens, the odds of being uninsured were significantly higher among non-citizens (OR 12.05, p \ 0.001). Compared to those who speak only English, the odds of being uninsured were significantly higher among those who did not speak English (OR 1.59, p \ 0.05). Compared to those with 16 or more years of residence, the odds of being uninsured were significantly higher among those with 0 5 (OR 19.53, p \ 0.001), 6 10 (OR 6.82, p \ 0.001), and (OR 5.39, p \ 0.001) years of residence, respectively. Compared to those who immigrated prior to 1980, the odds of being uninsured were significantly higher among those who immigrated between 1980 and 1989 (OR 2.18, p \ 0.05) and between 1990 and 2008 (OR 9.86, p \ 0.001), respectively. Inclusion of either year of immigration or length of residence reduced the difference between African-born and Latin American and Caribbean-born immigrants to non-significance. The region-of-origin difference was not explained by the inclusion of citizenship status or English language proficiency alone. Inclusion of either citizenship status or length of residence resulted in the emergence of the significant difference between immigrants from other regions and Latin American and Caribbean-born immigrants. Discussion Limited population-based research examines lack of health insurance among elderly immigrants, and no studies of which we are aware specifically focus on the elderly Black population. We use data from the 2008 American Community Survey to describe variation in insurance coverage by region of origin (US, Latin America and the Caribbean, Africa, and other regions) and conduct multivariate logistic regression analyses of uninsurance. We found that 1.7 % of the US-born were uninsured, compared to 8.4 % of the Latin American and Caribbean-born, 23.2 % of the African-born, and 9.3 % of those born in other regions. Interestingly, although rates of uninsurance were significantly

8 Table 4 Logistic regression models of non-insurance among Black immigrants, ACS 2008 Variable (reference category) No insurance Model 1 Model 2 Model 3 OR 95 % CI OR 95 % CI OR 95 % CI Region of origin (Latin America and the Caribbean) Africa 3.30*** *** Other ** Sex (female) Male Age (65 74 years) years ? years Education (high school) No school \High school 2.06*** Some college College graduate or more Employment status (not in the labor force) Employed Unemployed Income-to-needs ratio (300? % of poverty) Below poverty % of poverty % of poverty Marital status (widowed) Married Separated Divorced Never married/single 2.00* * Region of US residence (south) Northeast 0.35*** *** Midwest West 0.38* Citizenship status (naturalized citizen) Non-citizen 6.44*** English language proficiency (speaks only English) Does not speak English Speaks english well or very well Length of residence (16? years) 0 5 years 6.43** years years Year of immigration (before 1980) Constant 0.09*** *** *** Un-weighted N: 2,686 Significance levels: * p \ 0.05; ** p \ 0.01; *** p \ 0.001

9 higher among each immigrant group than among the native-born, the majority (63.9 %) of the estimated 81,255 uninsured elderly Blacks in the US in 2008 was native-born because of the relatively large size of the native-born population. Multivariate models that included demographic and socioeconomic controls indicated that the odds of being uninsured were substantially and significantly higher among each of the immigrant groups relative to the USborn. Among immigrants, the odds of being uninsured were 3.80 times higher among African-born than Latin American and Caribbean-born immigrants net of controls, but this difference was explained by the inclusion of either year of immigration or length of residence. Additionally, once immigration-related variables were taken into account, immigrants from regions other than Latin America, the Caribbean, and Africa were significantly more likely than Latin America and Caribbean-born immigrants to be uninsured. Taken together, these results suggests that some native- and foreign-born elderly Blacks fall through the cracks of health insurance, although the odds of doing so are substantially higher among recently arrived foreignborn non-citizens and immigrants from other regions. Primarily because of the public safety net provided by Medicare and Medicaid, almost all US-born elderly Blacks had health insurance; only 6.4 % relied on private insurance only, while Latin American and Caribbean- and Africanborn immigrants were more-represented in this group. Among both the native- and foreign-born elderly, lack of insurance may reflect lack of knowledge of eligibility or lack of eligibility because of failure to meet minimum public program requirements. Lack of knowledge about eligibility is consequential because it likely affects utilization of health care, unmet need, and the allocation of resources for out-ofpocket payments. It has material consequences in the lives of elderly Blacks, as well as those who lives are linked to them. Research efforts to better understand the factors that contribute to lack of coverage and its consequences among the Black elderly both native- and foreign-born would be valuable, and might inform targeted efforts to promote awareness about eligibility, the enhancement of the public safety net, or the development of alternatives for those who are truly ineligible. It is likely that a substantial portion of the lack of insurance among elderly immigrants that we documented in this study stems from the fact that they are among the invited elderly, who are ineligible for Medicare and other publicly-funded insurance programs for some period of time, if not indefinitely. Although some analysts have argued for Medicaid expansions that would permit the coverage of non-citizens [26], which might alleviate some of the uninsurance problem among elderly Black immigrants, the fact that family reunification provisions require hosts to guarantee that they will take financial responsibility for their family members makes it unlikely that the public safety net will be extended to cover recent immigrants. Moreover, in the context of debates over immigration reform, one analyst recently argued that it would be prudent for the Affordable Care Act (ACA) to require immigrants entering the country under family reunification provisions to purchase health care [27]. The rationale for implementing this requirement is that the inability of families to privately fund health insurance puts undue pressure on already strained publicly-funded health care resources. Essentially, the argument goes, we the tax-paying American public should not be set up to be responsible for funding the care of foreign-born elders who have not, and likely will not, contribute via employment to the economic well-being of the nation. The policy nexus that allows for the possibility of inviting elders to join their families in the US, but privatizes all responsibility for them is complex. Some may hope that the ACA will assist families in meeting the health insurance needs of older adult immigrants they have sponsored. However, even in the context of the ACA, financial responsibility for invited elderly immigrants remains firmly tied to the private resources of the sponsoring family. Access to Medicare and Medicaid for elderly, legal immigrants remains unchanged under the ACA [28]. Thus, while it is possible that families might be able to access health insurance for an invited older immigrant family member through the health care exchanges established by the ACA, the options available through those exchanges are likely to be limited and have very high deductibles. Generally, insurers do not develop comprehensive private plans for the older adult population because they assume that older adults have access to Medicare. Possibly, lower-cost insurance for other family members may free-up resources that can be used to purchase health care for an uninsured elder. Perhaps, the tax penalty provision of the ACA could compel some families to purchase health insurance for an older adult legal immigrant. But, economic constraint already limits the capacity of some families to do that even though they are, under the provisions of sponsorship, responsible to do so. While future research should evaluate the extent to which the passage of the ACA has reduced or slowed the increase in uninsurance among elderly immigrants, we think it is unlikely that it will have much effect in the absence of specific actions to address the high cost of purchasing private health insurance for older adults. Given current fiscal and political realities, it is unlikely that the government will assume responsibility for insuring the invited elderly. But, perhaps, there is another way. In the past, immigrant aid societies and other voluntary organizations were prevalent in the US [29 31]. Such organizations addressed a substantial number of needs, but generally aimed to address needs that individuals could not manage and the government did, could, or would not meet. By doing

10 so, voluntary organizations mitigated some of the strains recently immigrated individuals and their family members faced. With increased awareness of the uninsurance problem among elderly immigrants, which might arise from the publication of research such as this, it is conceivable that Americans immigrants and the native-born might organize voluntarily within churches, communities, and other institutions to collect private resources that could be channeled toward meeting the needs of elderly immigrants as well as native-born elderly who are currently falling through the cracks. References 1. Logan JR. Who are the other African Americans? Contemporary African and Caribbean immigrants in the United States. In: Shaw- Taylor Y, Tuch SA, editors. The other African Americans: contemporary African and Caribbean immigrants in the United States. Lanham: Rowman & Littlefield Publishers; p Kent MM. Immigration and America s Black population. Pop Bull. 2007;62(4): Leach MA. America s older immigrants: a profile. Generations. 2008/2009;32(4): Choi S. Insurance status and health service utilization among newly-arrived older immigrants. J Immigr Minor Health. 2006;8(2): Kim J, Lauderdale DS. The role of community context in immigrant elderly living arrangements. Res Aging. 2002;24(6): Min PG. Changes and conflicts: Korean immigrant families in New York. Boston: Allyn & Bacon; Tan J. Older immigrants in the United States: the new old face of immigration. Bridgewater Rev. 2011;30(2): Smith LS. Health of America s newcomers. J Community Health Nurs. 2001;18(1). 9. Siddarthan K. Health insurance coverage of the immigrant elderly. Inquiry. 1991;28(4): Lee S, Choi S. Disparities in access to health care among noncitizens in the United States. Health Sociol Rev. 2009;18(3): KCMU/Urban Institute analysis of 2009 ASEC Supplement to the CPS. 12. Mold JW, Fryer GE, Thomas CH. Who are the uninsured elderly in the United States? J Am Geriatr Soc. 2004;52: Okoro CA, Young SL, Strine TW, Balluz LS. Uninsured adults aged 65 years and older: is their health at risk? J Health Care Poor Underserved. 2005;16(3): Brown SK, Bachmeier JD, Bean FD. Aging societies and the changing logic of immigration. Generations. 2009;32(4): Portes A, Rumbaut RG. Immigrant America. Berkeley: University of California Press; Hummer RA, Rogers RG, Nam CB, LeClere FB. Race/ethnicity, nativity, and U.S. adult mortality. Soc Sci Q. 1999;80(1): Singh GK, Siahpush M. Ethnic-immigrant differentials in health behaviors, morbidity, and cause-specific mortality in the United States: an analysis of two national data bases. Hum Biol. 2002;74(1): Read JG, Emerson MO, Tarlov A. Implications of Black immigrant health for U.S. racial disparities in health. J Immigr Health. 2005;7(3): Read JG, Emerson MO. Racial context, Black immigration and the US Black/White health disparity. Soc Forces. 2005;84(1): Heron MP, Schoeni RF, Morales L. Mimeo, Florida State University Moore AR, Amey FK. Earnings differentials among male African immigrants in the United States. Equal Oppor Int. 2002;21(8): Bideshi D, Kposowa AJ. African immigrants and capital conversion in the U.S. labor market: comparisons by race and national origin. West J Black Stud. 2012;36(3): Lucas JW, Barr-Anderson DJ, Kington RS. Health status, health insurance, and health care utilization patterns of immigrant Black men. Am J Public Health. 2003;93(10): Gubernskaya Z, Bean FD, Van Hook J. (Un) Healthy immigrant citizens naturalization and activity limitations in older age. J Health Soc Behav. 2013;54(4): Ruggles S, Trent AJ, Katie G, Ronald G, Schroeder MB, Sobek M. Integrated public use microdata series: version 5.0 [machinereadable database]. University of Minnesota, Minneapolis (2010). 26. Nam Y, Kim W. Welfare reform and older immigrant adults medicaid and health insurance coverage: changes caused by chilling effects of welfare reform, protective citizenship, or distinct effects of labor market condition by citizenship? J Aging Health. 2012;24(4): Tienda M: Fix immigration bill to reduce health-care costs New Immigrants Medical Insurance. new-immigrants-medical-insurance-plans/. 29. Gunn SM, Platt PS. Voluntary health agencies: an interpretive study. New York: Ronald Press; Hamlin RH. Voluntary health and welfare agencies in the United States: an exploratory study. Schoolmasters Press (1961). 31. Shilts R. And the band played on: people, politics, and the AIDS epidemic. St. Martins, New York (1987).