Physical and Mental Health Consequences of Aging in Place and Aging Out of Place Among Black Caribbean Immigrants

RESEARCH IN HUMAN DEVELOPMENT, 2(4), 229 244 Copyright 2005, Lawrence Erlbaum Associates, Inc. Physical and Mental Health Consequences of Aging in Place and Aging Out of Place Among Black Caribbean Immigrants James S. Jackson and Toni C. Antonucci Institute for Social Research University of Michigan Our purpose in this article was to examine the influences of age group, timing of immigration, and race/ethnicity on self-reported health, self-esteem, and serious mental disorders among Black Caribbean immigrants to the United States. Analyses of the recently collected National Survey of American Life (Jackson et al., 2004) revealed that immigrants have better self-reported physical health and mental health than nonimmigrant ancestry group members. Age group, however, moderates and complicates this relationship. Black Caribbeans who immigrated more recently tend to have better self-reported health; older, more recent immigrants have lower selfreported health than those who are older and have lived longer in the United States. Similar results were found for the prevalence rates of reported any lifetime or 12 month Diagnostic and Statistical Manual of Mental Disorders (4th ed.; American Psychiatric Association, 1994) mental disorders. Individuals who immigrate late in life, perhaps to join their children, or who have themselves immigrated years earlier with their children, will age as immigrants (Jackson, 2003; Ruggles, 1994). Increased life expectancy and reduced numbers of children among native members of advanced industrial countries may result in both more immigrating to age as well as more aging in place as immigrants among diverse groups (Jackson, 2003). Additional research on the intersection of immigration, birth cohort, and aging-related processes is necessary to develop programs that may maximize the positive, and minimize the negative, life-course experiences of both immigrants and native members of their host countries Correspondence should be sent to James S. Jackson, 5010 Institute for Social Research, University of Michigan, 426 Thompson Street, Ann Arbor, MI 48106 1248. E-mail: jamessj@umich.edu

230 JACKSON AND ANTONUCCI (Andersson & Oberg, 2004; Torres, 2004; Waters & Jimenez, 2005). In this article, we utilize data from the National Survey of American Life (NSAL; Jackson et al., 2004) to examine the self-reported health, self-esteem levels, and the prevalence of having any serious lifetime and 12-month mental disorders among different aged, United States born, and immigrant Black Caribbeans who have spent varying numbers of years in the United States. Native-born African Americans provide a useful contrast group. People in most advanced industrialized countries are living longer and life expectancy is increasing (United Nations, 2002). At the same time because of lowered native-born birth rates, the demographic distribution of the population in many nations has changed from a pyramid shaped distribution with greater numbers of younger than older people to a beanpole shaped population distribution with relatively equal numbers of people in each generation (United Nations, 2002). One by-product of this new beanpole shaped population is that older people have fewer children to whom they can turn when they reach old age or when they are facing the chronic or acute challenges associated with aging (Andersson & Oberg, 2004). It has traditionally been the case that older people turn to their children for help when they face the difficulties that often accompany aging. In the past, when people had many children, it was very likely that several of these children lived nearby, were available to help, and could share the provision of care among a number of siblings and other family members. Now, with fewer children available, it is more likely that parents seeking to be close to their children will have fewer to choose from and will either have to immigrate to be with their children who have immigrated previously or remain with their children in countries to which they had immigrated years earlier. The adaptation and expectations of these immigrants, both older individuals and their families, represent important new challenges facing the field of gerontology specifically and society more generally (Castles & Miller, 1998; Palloni & Ewbank, 2004; Torres, 2004; Waters & Jimenez, 2005). In this article, we primarily focus on Black Caribbeans, recognizing that the reception these immigrants receive will significantly influence both how well they adapt to their new country and how successfully they age (Model, 1995; Waters, 1999). We capitalize on unique data that provide the first national estimates on the Black Caribbean population of both immigrants and those of third-generation (and more) and second-generation ancestry (Jackson et al., 2004). Thus, we are able to compare within ethnic background individuals who are first-generation immigrants with those born in the United States and whose parents were born outside the United States (second generation) with those who were born in the United States and whose parents and grandparents were also born in the United States (third generation and beyond). We believe that this is a much more appropriate comparison than to the general population of native-born U.S. citizens (Jasso,

AGING AND BLACK IMMIGRANT HEALTH 231 Massey, Rozsenzweig, & Smith, 2004). At some points in the analyses, we compare native and immigrant Black Caribbeans with African Americans, a fully native-born population group that shares racial but not necessarily cultural characteristics with Caribbean Blacks (Model, 1995). Our purpose in this article is to examine for one particular group, Caribbean Blacks, the nature and distribution of physical and mental health among both firstgeneration immigrants and U.S. natives with ancestry from the Caribbean region. Our focus is on the physical and mental health outcomes and the ways in which the macro social context (age, age of entry, generation position, and years in the country), controlling for age, family income, education, and gender, may influence physical and mental health outcomes for Black Caribbean second- and thirdgeneration U.S. natives and first-generation immigrants (Vega & Rumbaut, 1991; Waters, 1999). Overall, we predicted that first-generation immigrants would show better physical and mental health outcomes than Caribbean U.S. natives (Vega & Rumbaut, 1991). Because race is a powerful stratification factor in U.S. society, we hypothesized that second- and third-generation Blacks would have experienced similar life circumstances and exposures to race-related treatment as African Americans (Model, 1991, 1995; Small, 1994). Simultaneously, we also predicted that those who were older when they immigrated to the United States would demonstrate poorer adjustment reflected in poorer physical and mental health than similarly aged immigrants who had migrated earlier in their lives, even accounting for their longer exposure to racialized treatment (Nazroo, 2004; Ruggles, 1994; Small, 1994). METHODS Sample The NSAL (Jackson et al., 2004) is the most comprehensive and detailed study of mental disorders and the mental health of Americans of African descent ever completed. The study was part of the National Institute of Mental Health (NIMH) Collaborative Epidemiology Survey initiative that included three national representative surveys the NSAL, the National Comorbidity Survey Replication, and the National Latino and Asian American Study (NIMH, 2004). The African American sample, the core sample of the NSAL, is a national representative multistage probability sample of households located in the 48 coterminous states with at least one Black adult 18 years or over who did not identify ancestral ties in the Caribbean. Of the adults in the African American sample, 50 were born in Africa. The Caribbean Black sample was selected from two area probability sample frames: A total of 265 came from the households in the core sample, whereas

232 JACKSON AND ANTONUCCI 1,356 came from an area probability sample of housing units from geographic areas with a relatively high density of persons of Caribbean descent. We designed the NSAL analysis weights for the African American and Caribbean Black samples to provide population representation for these populations in the 48 coterminous states. Although not used in this article, the White sample was a stratified, disproportionate sample of non-hispanic White adults residing in households located in census tracts and blocks that had 10% or greater African American population (Heeringa, Wagner, Torres, Duan, Adams, & Berglund, 2004). The sample design and analysis weights for this sample were designed to be optimal for comparative analyses in which residential, environmental, and socioeconomic characteristics are controlled in the Black White statistical contrasts. Most of the Caribbean sample (70.1%) had roots in the English-speaking areas of the Caribbean, 18.6% from Haiti, and 11.3% from the Spanish-speaking Caribbean region. The final NSAL adult sample was an integrated, national household, probability sample of 3,570 African Americans, 891 non-hispanic Whites, and 1,621 Blacks of Caribbean descent for a total of 6,082 individuals aged 18 and over. As noted previously, we selected the African American and White samples exclusively from geographic segments in proportion to the African American population; we selected the Caribbean Black sample primarily from the African American segments and metropolitan segments (500) with more than 10% Blacks of Caribbean descent, making this the first national sample of people of different race and ethnic groups who live in the same contexts and geographical areas as Blacks are distributed. Most (95%) of the interviews were conducted face-to-face using a computerassisted instrument and averaged 2 hr and 20 min; a small subset (5%) was interviewed by telephone; no mode of administration differences have been found. Data collection was completed between February 2001 and June 2003. The final overall response rate was 72%. The rates varied by race and ethnicity: a total of 71% for the African American sample, 78% for Caribbean Blacks, and 70% for non-hispanic Whites. The demographic data in Table 1 indicate that compared to the African American population, Black Caribbeans had higher levels of education and income and were more likely to be married, employed, and reside in the Northeast and in major metropolitan areas. As shown in Table 2, the average age of third-generation respondents (41.31 years) was 10 years older than the second (30.74). Among immigrants, those who immigrated more than 20 years ago averaged 52.04 years in comparison to those who immigrated between 11 and 20 years (40.10) or 0 to 10 years ago (33.00). In general, immigrants were of higher average age (43.50) than Caribbean Blacks born in the United States (33.71) and of African Americans, reflecting both U.S. immigration policies changes and family reunification processes (Castles & Miller, 1998; Ruggles, 1994).

AGING AND BLACK IMMIGRANT HEALTH 233 TABLE 1 Demographic Characteristics of Black Caribbean and African American National Survey of American Life Samples Overall Demographic Characteristics African American Black Caribbean Age 42.3 years 40.2 years Education 12.4 years 12.9 years Family income $37,597 $47,099 Female 55.9% 49.0% Employed 66.8% 74.7% Home owner 49.8% 45.1% Married 32.9% 37.3% U.S. born 97.7% 35.1% TABLE 2 Average Age of Black Caribbean Sample by Ancestry and Immigration Status Ancestry and Immigration Status N M Age at Time of Interview (Years) Born in United States Third generation 125 41.31 Second generation 320 30.74 Total born in United States 445 33.71 Immigration experience Immigrated more than 20 years ago 504 52.04 Immigrated between 11 to 20 years ago 370 40.10 Immigrated 0 to 10 years ago 290 33.00 Total immigrants 1,164 43.50 MEASURES Self-Reported Health We employed the measure of self-reported health used in most health surveys (Jasso et al., 2004) assessed on a 4-point scale ranging from 1 (very poor) to4 (very good). Self-Esteem The self-esteem measure is a modified Rosenberg (1989) 10-item, 4-point response scale used to assess the self-acceptance dimension of self-evaluations. Items were presented in both negative and positive formats. All responses to the negative items were reversed such that 4 (strongly agree) represented positive self-esteem and 1 (strongly disagree) represented low self-esteem. The 10 items

234 JACKSON AND ANTONUCCI were summed; overall, the scale had very acceptable internal consistency (α =.78) across the total sample. Examining the internal consistency by age (three age groups) and race/ethnicity (three groups) indicated a range of a low of.74 for African Americans 55 years of age and older and a high of.83 among Black Caribbeans 18 to 34 years of age. Thus, for all age by ethnicity groups, there was a remarkable similarity in the internal consistency assessment of reliability of the self-esteem measure. Diagnostic Assessment We used the Diagnostic and Statistical Manual of Mental Disorders (4th ed. [DSM IV]; American Psychiatric Association, 1994) World Mental Health Composite Interview (World Health Organization [WHO] Composite International Diagnostic Interview), a fully structured diagnostic interview, to assess a wide range of serious mental disorders. The mental disorders sections used for the NSAL were slightly modified versions of those developed for the World Mental Health project initiated in 2000 (Kessler & Ustum, 2004; Pennell et al., 2004). The mental disorders we assess and present in this article among African Americans and Caribbean Blacks include anxiety disorders (panic disorder, agoraphobia, social phobia, generalized anxiety disorder, obsessive compulsive disorder, posttraumatic stress disorder), mood disorders (major depressive disorder, dysthymia, Bipolar I and II disorders), substance disorders (alcohol abuse, alcohol dependence, drug abuse, drug dependence), childhood disorders (separation anxiety disorder, oppositional defiant disorder, conduct disorder, attention deficit hyperactivity disorder, with the last three asked only of respondents in the 18 44 year age range), and eating disorders (anorexia, bulimia). We assessed the lifetime disorders measure as one or more disorders out of the 19 total assessed for African Americans and Caribbean Blacks; the 12-month measure was having any disorder in the past 12 months out of the 18 total assessed in the survey. Social Demographics Table 2 shows the Ns and mean age at time of interview for the Black Caribbean and African American population samples. As can be seen from this table, overall, Caribbean Blacks had lower average age, greater income, higher average education, better sex distribution, better employment characteristics, slightly lower home ownership rates, and higher marriage rates than did African Americans (Model, 1991, 1995). A number of these demographic and economic variables served as controls in the analyses. We assessed education in years. We assessed family income directly, and we imputed it when necessary (less than 12.7% of the total sample was imputed). We assessed the sex of respondents through direct observation or in the small proportion of telephone interviews, through a direct

AGING AND BLACK IMMIGRANT HEALTH 235 question. We calculated age based on year and date of birth reported by each respondent. For purpose of assessing differences in large age cohorts roughly representing young, middle-age, and older groups, we categorized the sample by 18 to 34, 35 to 54, and 55 and older. Because these are large categories representing large numbers of yearly birth cohorts, we also controlled individual age of respondents within age categories in analyses on self-esteem and self-rated health. Analysis Strategy Cross-tabulations (see Table 3) represent differences on ancestry/timing of immigration differences in self-reported health, self-esteem, and percentage having at least one lifetime disorder out of the 19 that we assessed. The means and percentages represent weighted proportions based on the sample s race-adjusted weight measure; the standard errors reflect the recalculation of variance using the study s complex design; and in the case of the percent mental disorder variable, the Rao Scott chi-square represents a complex design-corrected measure of association (SAS Institute, Inc., 2005). All the means shown have been adjusted for individual age, gender, education, and family income. We used least squares regressions to estimate the means and significance of the ancestry/immigration and age category differences in selfreported health and self-esteem, and we used logistic regression analysis to examine these same differences in prevalence rates of at least one disorder among the different ancestry/immigration and age category groups. Throughout the analyses, the.05 level of a two-tailed test of significance represented the cutoff for assessing statistical significance. We conducted all analyses using SAS Version 9.0, which uses the Taylor expansion approximation technique for calculating the complex-design based estimates of variance (SAS Institute, Inc., 2005). RESULTS In this section, we summarize the differences in ancestry/years of immigration factors and the ancestry/years of immigration factor by age category differences among Caribbean Blacks. Table 3 presents the means and percentages of the set of criterion variables we used in these analyses. Self-Reported Health As shown in Table 3, for the total immigrant and the total nonimmigrant Caribbean samples, immigrants (2.86) overall had better average self-reported health than those Caribbean Blacks of second- or third-generation status (2.33), t(1, 1166) = 14.54, p <.001. African Americans (see Table 4) a native-born, non-

TABLE 3 Self-Reported Health, Self-Esteem, and Percent Having Any DSM IV Lifetime and 12-Month Mental Disorder Among Black Caribbean Sample by Ancestry, Immigration Status, and Age Group Ancestry and Immigration Status of Black Caribbeans Residing in the United States Self- Reported Health Self-Esteem DSM IV Lifetime Disorder DSM IV 12-Month Disorder N M SE N M SE N % SE N % SE Born in United States Third generation (all ages) 124 2.27.05 124 3.52.12 122 60 6.53 122 36 6.73 18 34 years of age 55 2.38.11 55 3.52.14 53 66 4.92 53 31 18.10 35 54 years of age 54 2.49.14 54 3.62.09 54 61 19.78 54 38 20.47 55 years of age and older 15 1.89.27 15 3.44.19 15 51 17.30 15 39 24.78 Second generation (all ages) 316 2.44.23 316 3.66.03 310 48 6.92 310 28 5.57 18 34 years of age 210 2.53.23 210 3.65.02 204 53 7.42 204 34 6.87 35 54 years of age 76 2.37.18 76 3.68.04 76 43 12.48 76 17 6.59 55 years of age and older 30 2.35.40 30 3.60.12 30 8 5.37 30 4 2.89 Total born in United States (all ages) 440 2.39.18 440 3.62.06 432 51 5.65 432 31 5.90 Timing of immigration experience Immigrated more than 20 years ago (all ages) 505 2.71.14 505 3.69.03 494 23 2.86 494 13 1.30 18 34 years of age 47 2.72.24 47 3.75.08 46 27 12.01 46 20 10.58 35 54 years of age 258 2.73.14 258 3.65.03 256 31 2.84 256 19 1.66 55 years of age and older 200 2.50.07 200 3.74.04 192 11 3.18 192 3 1.47 Immigrated 11 to 20 years ago (all ages) 372 3.02.12 372 3.62.02 362 18 1.86 362 11 1.10 18 34 years of age 139 2.92.08 139 3.56.09 135 36 3.41 135 25 4.04 35 54 years of age 197 3.09.21 197 3.66.06 191 10 3.28 191 5 1.54 55 years of age and older 36 2.77.35 36 3.64.14 36 9 4.92 36 2 0.98 Immigrated 0 to 10 years ago (all ages) 290 2.98.17 290 3.61.05 286 33 4.97 286 17 2.99 18 34 years of age 171 3.10.21 171 3.62.07 171 39 11.31 171 21 5.40 35 54 years of age 98 2.98.18 98 3.66.08 94 18 7.48 94 9 4.70 55 years of age and older 21 2.27.29 21 3.13.32 21 43 15.14 21 6 5.45 Total immigrants 1,167 2.88.14 1,167 3.65.03 1,142 23 3.07 1,142 13 1.66 Note. DSM IV = Diagnostic and Statistical Manual of Mental Disorders (4th ed.); NSAL = National Survey of American Life; SE = Standard error of estimate. A total of 20 DSM IV lifetime and 18 of the 12-month disorders were assessed in the NSAL survey; all means adjusted for age, education, family income, and gender. 236

AGING AND BLACK IMMIGRANT HEALTH 237 Caribbean, Black sample had overall average self-reported health that was intermediate (2.51) to that of native-born, Black Caribbeans (2.39) and Caribbean Blacks immigrants (2.88). These overall trends, however, hide a great deal of variation among different age groupings. For example, among Caribbean Blacks born in the United States, those 55 years of age and older, both in the third and second generations, tended to have worse average self-reported health than those who were in the other two age groups. This is most noticeable among the second generation of native-born Caribbean Blacks. As also shown in Table 3, among immigrants, there was very little average differences among the three age groups for those who immigrated 11 to 20 and more than 20 years ago. Among those who immigrated less than 10 years ago, however, those who were over 55 years of age at the time of the interview reported poorer average health than those who were 18 to 34 and 35 to 54 years of age. The worse average self-reported health in the table was reported by those third-generation Caribbean Blacks over 55 years of age (1.89). Self-Esteem Examining mean self-esteem levels (Table 3) reveals a remarkable lack of difference by age group, ancestry, or time of immigration. There were no nativity differences and no differences across age groups by time of immigration. As shown in Table 4, African Americans also differed little from Caribbean Blacks among any of the age by time of immigration by nativity cells in the tables. The consistency of these means are noteworthy, especially in light of the individual differences in immigration as well as life experiences of individuals in each of these groups and as well as their differences in all other categories of interest. Mental Disorders We observed the largest health disparities in this analysis for differences in mental disorders. As shown in column 9 of Table 3, we found large differences in the average reports of any lifetime disorders between native-born (51%) and immigrant (23%) Black Caribbean groups, collapsing across age groups, t(1) = 37.48, p <.001. This overall difference, however, masks important differences by nativity and age group. For example, third-generation Caribbean Blacks had larger average percentage reports of lifetime mental disorders (60%) than did those of the second generation (48%). Although the tendency is for these generation differences to replicate across the three large age groups, the second generation 55 years of age and older were 7.5 times less likely (8%) than the 55 years of age or older third generation (50%) to report a lifetime disorder. As shown in Table 3, among immigrants, those who came most recently, regardless of age, tended to have higher prevalence rates of any disorder (33%) than those who came either 11

TABLE 4 Self-Reported Health, Self-Esteem, and Percent Having Any DSM IV Lifetime or 12-Month Mental Disorder Among African American Sample by Age Group Age Group M Self-Reported Health M Self-Esteem DSM IV Lifetime Disorder DSM IV 12-Month Disorder N N N % SE N % SE 18 34 years of age 1,218 2.73 1,233 3.64 1,233 43 2.20 1,199 19 1.54 35 54 years of age 1,378 2.47 1,448 3.60 1,448 39 1.61 1,446 18 1.28 55 years of age and older 666 2.22 709 3.60 710 23 1.87 785 9 1.11 African Americans of all ages 3,263 2.51 3,390 3.62 3,392 38 1.21 3,430 16 0.76 Note. DSM IV = Diagnostic and Statistical Manual of Mental Disorders (4th ed.). 238

AGING AND BLACK IMMIGRANT HEALTH 239 to 20 years ago (18%) or more than 20 years ago (23%). This is in stark contrast to self-reported health in which there are small differences among the three immigrant categories (2.98, 3.02, and 2.71, respectively). Again among immigrants, those who came the most recently and were 55 years of age or older at the time of the interview (who were over 45 years of age when they came to the United States) had greater prevalence rates of any mental disorder (43%) than those who are 55 years of age and older and came 11 to 20 years ago (9%) or more than 20 years ago (11%). This effect of age group within immigrant status is similar to that found for self-reported health in which 55 years of age and older most recent immigrants had lower self-reported health than those either 11 to 20 years in the United States or more than 20 years. All of these differences among Caribbean Blacks were higher than the prevalence rates for African Americans who had lower rates overall (38%) than the two U.S. born Caribbean groups (51%) but higher than the rate overall for Caribbean immigrants (23%). It is possible that the lifetime disorders being reported had their onset in the respective countries of origin. We were interested, however, in the impact of timing of immigration on the onset and more recent experiences of symptoms meeting criteria for DSM IV categorization of a disorder. Thus, we examine in column 12 (Table 3) the effects of age group and nativity on reports of any 12-month disorders. For the most part, the effects reported for 12-month disorders replicated those found for lifetime disorders. Overall, among all immigrants, the rates of disorders in the last 20 months (13%) was lower than the average of the two nativeborn groups (31%) but only slightly lower than the rates for African Americans (16%). As found for lifetime estimates, second-generation Caribbeans had lower rates (29%) than the third generation. Similarly, the second generation showed a declining pattern of 12-month rates with increasing older age groupings (34%, 17%, and 4%, respectively). The third generation actually showed slightly increasing (no significant differences) rates with age group (31%, 38%, and 39%, respectively). In comparison, in Table 4, African Americans showed declining prevalence rates with increasingly older age groupings for both lifetime (43%, 39%, and 23%, respectively) and 12-month disorders (19%, 18%, and 9%, respectively). Among immigrant groups, each age group revealed declining rates with older age. Among those who were 55 years of age, however, those who immigrated most recently (10 or less years) had higher rates (6%) than those who immigrated 11 to 20 years ago (2%) or more than 20 years ago (3%). Table 5 presents a summary of the analyses for self-reported health, selfesteem, and lifetime and 12-month mental disorder rates. The entries in the cells represent the rank orders for each outcome in which 1 represents the best and 5 indicates the worse rank order of the size of the effects. The ranks in parentheses represent differences among immigrant age groups only. For mean self-reported health, it is clear that the two ancestry groups fared the worst. This is generally true for all the outcomes in Table 5, especially for the third-generation group.

240 JACKSON AND ANTONUCCI TABLE 5 Summary of Rankings of Criterion Variables by Ancestry/Immigration Status and Age Group Ancestry and Immigration Status M Self-Reported Health M Self-Esteem Percent DSM IV Lifetime Disorder Percent DSM IV 12-Month Disorder Born in United States Third generation (all ages) 5 5 5 5 18 34 years of age 5 5 5 4 35 54 years of age 5 5 5 5 55 years of age and older 5 4 5 5 Second generation (all ages) 4 2 4 4 18 34 years of age 4 2 4 5 35 54 years of age 4 1 4 3 55 years of age and older 3 3 4 3 Timing of immigration experience Immigrated more than 20 years ago (all ages) 3 (3) 1 (1) 2 (2) 2 (2) 18 34 years of age 3 (3) 1 (1) 1 (1) 1 (1) 35 54 years of age 3 (3) 3.5 (2.5) 3 (3) 4 (3) 55 years of age and older 3 (2) 1 (1) 3 (2) 2 (2) Immigrated 11 to 20 years ago (all ages) 1 (1) 3 (2) 1 (1) 1 (1) 18 34 years of age 2 (2) 4 (3) 2 (2) 3 (3) 35 54 years of age 1 (1) 3.5 (2.5) 1 (1) 1 (1) 55 years of age and older 1 (1) 2 (2) 2 (1) 1 (1) Immigrated 0 to 10 years ago (all ages) 2 (2) 4 (3) 3 (3) 3 (1) 18 34 years of age 1 (1) 3 (2) 3 (3) 2 (2) 35 54 years of age 2 (2) 2 (1) 2 (2) 2 (2) 55 years of age and older 4 (3) 5 (3) 4 (3) 4 (3) Note. 1 = best rank; 5 = worst rank; value in parentheses is the ranking among immigrants. When it was not true, for the most part, it was the older age group (55 years or older) who immigrated most recently (0 10 years ago) who showed the worst outcomes: self-reported health (rank 4 overall and rank 3 among all immigrants over 55), mean self-esteem (rank 5 overall and rank 3 among all immigrants 55 years of age and older), lifetime disorders (rank 4 overall and rank 3 among all immigrants over 55 years of age), and 12-month disorders (rank 4 and rank 3 among all immigrants over 55 years of age and older). DISCUSSION These analyses reveal that there was considerable variation among Black Caribbean ancestry and immigrant groups in physical and mental health outcomes. Perhaps due to longer exposure as a racialized group (Small, 1994) in this country as

AGING AND BLACK IMMIGRANT HEALTH 241 well as the timing of the immigration of their grandparents, third-generation Caribbean Blacks revealed worse outcomes than second-generation or immigrant Caribbean Blacks on all the health outcomes we examined. In fact, thirdgeneration Caribbean Blacks uniformly displayed worse health outcomes than native African Americans, a group that they should resemble given their families time in the country. Thus, these results suggest collapsing across all ancestry groups in comparisons with immigrants from the same ethnic/racial groups may obscure a great deal of variation (Waters, 1999; Waters & Jimenez, 2005). As noted, large differences were found in health and mental health outcomes among second- and third-generation Black Caribbean groups. It is our belief that it is these comparisons within ethnic/racial groups (Butcher, 1994) rather than comparisons between immigrant and all native groups in a society (e.g., Jasso et al., 2004) that are most relevant for uncovering possible immigration influences. As these data clearly illustrate, those Caribbean Blacks who were of third and higher generation backgrounds were much worse off than second-generation Caribbean Blacks in all the outcomes that we examined. In a recent chapter, Jasso et al. (2004) concluded just the opposite when comparing ethnic and racial immigrants to the general native U.S. population, whereas others (e.g., McKay, Macintyre, & Ellaway, 2003) have reported similar results to those found in this article (e.g., Vega & Raumbaut, 1991). We believe greater specificity concerning both immigrant and ancestry status provides a more accurate assessment of the lived experiences and health and mental health statuses of these individuals. The only exception to the general overall results of Black Caribbean immigrants having more positive physical and health statuses than ancestral natives in the United States was for self-esteem in which no differences were found. This unique finding suggests that one s self-evaluation may not be affected in the same way as health and mental health statuses. Self-esteem is undoubtedly predicated on the bases or contingencies used to form one s self-esteem (Antonucci & Jackson, 1983; Crocker & Wolfe, 2001). There is no reason why immigrant or ancestry group members lack the capability to uncover and use information in the environment to support high, positive self-evaluations (Crocker & Wolfe, 2001). These findings are similar to what has been found consistently in comparisons among African American and non-hispanic White populations groups (e.g., Antonucci & Jackson, 1983). The results regarding the effects of time of immigration were mixed, although it is clear that for every outcome, being older at the time of immigration to the United States was associated with more negative outcomes, a function we believe of longer times for attachment to sending countries and thus greater difficulties in adjusting to the demands and rigors of immigrating to the United States (Ruggles, 1994). The growth of aging populations worldwide is creating a demand for immigration among people of all ages but increasingly so among the elderly as they seek to obtain or maintain proximity to their ever decreasing numbers of children

242 JACKSON AND ANTONUCCI (Ruggles, 1994; United Nations, 2002). Unfortunately, it appears that immigrating at older ages may produce risks for negative physical and mental health outcomes, perhaps because of inadequate time to adjust to a new country prior to new demands of aging-related changes (Cohen, Berment, & Magai, 1999; Ruggles, 1994). It is possible that members of advanced industrialized nations, such as those in the United States, may become more intolerant of increased immigration, especially as population aging takes place and costs for providing health and social services continue to rise (Noiriel, 1994). Additional research is needed to examine the direct linkages between aging populations and immigrant attitudes and immigrant physical and mental health. Similarly, additional attention is needed from family, social gerontology, and human development researchers in examining the increased pressure for family reunification. Such pressure may be leading older people to immigrate to reunite with their children. Because this is most likely to be from less to more economically and socially resourceful countries, the potential for negative attitudes and receptivity of these host countries may have enormous detrimental influences on the broader societal context and the adjustment and subsequent health profiles for both immigrant individuals and their families (Hammer, Benjamins, & Rogers, 2004; Ruggles, 1994). ACKNOWLEDGMENTS Portions of this article were presented at the meeting of the Society for the Study of Human Development, Boston, Massachusetts, November 2003. The research on which this article is based was funded by grants from the National Institute of Mental Health (5 U01 MH057716), the Office of Behavioral and Social Science Research, and the National Institute on Aging (3 R01 AG 020282). The data presented and views expressed are solely the responsibility of the authors. We thank the many interviewers, supervisory field staff, and participants in this national study and research assistants Julie Sweetman and Ray Baser for their contributions to data analyses. REFERENCES American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author. Andersson, L., & Oberg, P. (2004). Diversity, health and ageing. In S. O. Daatland & S. Biggs (Eds.), Ageing and diversity (pp. 45 60). Bristol, England: The Policy Press. Antonucci, T. C., & Jackson, J. S. (1983). Physical health and self-esteem. Family and Community Health, 6(2), 1 9.

AGING AND BLACK IMMIGRANT HEALTH 243 Butcher, K. F. (1994). Black immigrants in the United States: A comparison with native blacks and other immigrants. Industrial and Labor Relations Review, 47, 265 284. Castles, S., & Miller, M. J. (1998). The age of migration: International population movements in the modern world. London: Macmillan. Cohen, C. I., Berment, F., & Magai, C. (1999). A comparison of U.S.-born African American and African-American Caribbean psychiatric outpatients. Journal of the National Medical Association, 89, 117 123. Crocker, J., & Wolfe, C. T. (2001). Contingencies of self-worth. Psychological Review, 108, 593 623. Hammer, R. A., Benjamins, M. A., & Rogers, R. C. (2004). Racial and ethnic disparities in health and mortality among the U.S. elderly population. In N. B. Anderson, R. A. Bulatao, & B. Cohen (Eds.), Critical perspectives on racial and ethnic disparities in health in later life (pp. 53 94). Washington, DC: National Research Council. Heeringa, S. G., Wagner, J., Torres, M., Duan, N., Adams, T., & Berglund, P. (2004). Sample designs and sampling methods for the Collaborative Psychiatric Epidemiology Studies. International Journal of Methods in Psychiatric Research, 13(4), 221 240. Jackson, J. S. (2003). Conceptual and methodological linkages in cross-cultural groups and crossnational aging research. Journal of Social Issues, 58(4), 825 835. Jackson, J. S., Torres, M., Caldwell, C. H., Neighbors, H. W., Nesse, R. M., Taylor, R. J., et al. (2004). The National Survey of American Life: A study of racial, ethnic and cultural influences on mental disorders and mental health. International Journal of Methods in Psychiatric Research, 13(4), 196 207. Jasso, G., Massey, D. S., Rozsenzweig, M. R., & Smith, J. P. (2004). Immigrant health: Selectivity and acculturation. In N. B. Anderson, R. A. Bulatao, & B. Cohen (Eds.), Critical perspectives on racial and ethnic disparities in health in later life (pp. 227 268). Washington, DC: National Research Council. Kessler, R. C., & Ustum, T. B. (2004). The World Mental Health (WMH) Survey initiative version of the World Health Organization (WHO) Composite International Diagnostic Interview (CIDI). International Journal of Methods in Psychiatric Research, 13, 93 121. McKay, L., Macintyre, S., & Ellaway, A. (2003). Migration and health: A review of the international literature (MRC Social & Public Health Sciences Unit, Occasional Paper No. 12). Glasgow, Scotland: MRC Social & Public Health Sciences Unit. Retrieved July 24, 2005, from www.msocmrc.gla.ac.uk Model, S. (1991). Caribbean immigrants: A black success story? International Migration Review, 25, 248 276. Model, S. (1995). West Indian prosperity: Fact or fiction? Social Problems, 42, 535 553. Nazroo, J. Y. (2004). Ethnic disparities in aging health: What can we learn from the United Kingdom. In N. B. Anderson, R. A. Bulatao, & B. Cohen (Eds.), Critical perspectives on racial and ethnic disparities in health in later life (pp. 677 702). Washington, DC: National Research Council. NIMH. (2004). Guest editorial. International Journal of Methods in Psychiatric Research, 13(4), 193 195. Noiriel, G. (1994). Civil rights policy in the United States and the policy of integration in Europe: Divergent approaches to a similar issue. Journal of Policy History, 6, 120 139. Palloni, A., & Ewbank, D. C. (2004). Selection processes in the study of racial and ethnic differentials in adult health and mortality. In N. B. Anderson, R. A. Bulatao, & B. Cohen (Eds.), Critical perspectives on racial and ethnic disparities in health in later life (pp. 171 226). Washington, DC: National Research Council. Pennel, B., Bowers, A., Carr, D., Crardoul, S., Cheung, G., Dinkelman, K., et al. (2004). The development and implementation of the National Comorbidity Survey Replication, the National Survey of American Life, and the National Latino and Asian American Survey. International Journal of Methods in Psychiatric Research, 13, 241 269.

244 JACKSON AND ANTONUCCI Rosenberg, M. (1989). Society and the adolescent self-image (Rev. ed.). Middletown, CT: Wesleyan University Press. Ruggles, S. (1994). The origins of African-American family structure. American Sociological Review, 59, 163 151. SAS Institute, Inc. (2005). SAS/STAT user s guide. Version 9.1. Cary, NC: Author. Small, S. (1994). Racialized barriers: The black experience in the United States and England in the 1980s. London: Routledge. Torres, S. (2004). Making sense of the construct of successful ageing: The migrant experience. In S. O. Daatland & S. Biggs (Eds.), Ageing and diversity (pp. 125 140). Bristol, England: The Policy Press. United Nations. (2002). World population ageing: 1950 2050. New York: Author. Vega, W. A., & Rumbaut, R. G. (1991). Ethnic minorities and mental health. Annual Review of Sociology, 17, 351 358. Waters, M. C. (1999). Black identities: West Indian immigrant dreams and American realities. New York: Russell Sage Foundation. Waters, M. C., & Jimenez, T. R. (2005). Assessing immigrant assimilation: New empirical and theoretical challenges. American Review of Sociology, 31, 105 125.