Family Relations,Social Connections,and Mental Health Among Latino and Asian Older Adults

The Mental Health of Latino and Asian Older Adults in the United States

Latino and Asian older adults have different mental health needs than other older Americans. Many older Latinos have limited English language skills, which is a major barrier to their education, employment, and access to health and social services (Administration on Aging [AOA], 2010b). Latinos in general and older Latinos in particular continue to have difficult and stressful acculturation experiences, sometimes accompanied by feelings of being second class (Gonzalez, 2007). All of these factors have a pervasive influence on their daily lives, contributing to mental distress. National and regional studies have consistently reported higher past-week (13.2%–30%) and 12-month (8.6%–10.8%) prevalence rates of depression among Latinos than non-Latino Whites (past week: 9%–16%; 12 month: 5.3%–7.8%; Black, Markides, & Miller, 1998; Dunlop, Song, Lyons, Manheim, & Chang, 2003; González, Haan, & Hinton, 2001; Jang et al., 2008; Moscicki, Locke, Rae, & Boyd, 1989; Vernon & Roberts, 1982). However, the lifetime prevalence of depression seems to be comparable for Latinos (15.2%–17%) and Whites (16.8%–16.9%; Jimenez, Alegría, Chen, Chan, & Laderman, 2010; Woodward et al., 2012). Studies on depression in older Latinos have identified some risk factors, including low income, a low level of acculturation, caregiving burden, poor support networks, and exposure to trauma such as high crime rates and political violence (L. Hinton & Areán, 2008).

Similar to their Latino counterparts, older Asians often experience a great deal of stress associated with immigration, acculturation, discrimination, and prejudice (Kuo et al., 2008). Early Asian immigrants, such as the Japanese, the Chinese, and Filipinos, shared the experience of racism, isolation, and laws against Asians that characterized the early 20th century (AOA, 2010a). More recent refugees from Southeast Asia are at greater risk for PTSD as a result of political torture and loss of significant others before and after immigrating to the United States (Moon & Cho, 2012). These demeaning experiences may be internalized, having negative consequences on mental health in later life. A review of 24 studies of depression in older Asian immigrants (OAIs) reported higher prevalence rates of depression among OAIs (ranging from 4.2% to 44.8%) than community-dwelling general older adults (18%–20%; Kuo et al., 2008). The common correlates of depression in OAIs include female gender, poor physical health, economic strain, social isolation, recent immigration, and a lack of English proficiency (Kuo et al., 2008).

Research on anxiety in both Latino and Asian older adults is sparse. To date, only two recent studies have examined the issue using data from national epidemiologic surveys (Jimenez et al., 2010; Woodward et al., 2012). Both studies reported a higher lifetime prevalence of any anxiety disorders, GAD, and social phobia in Whites and Latinos than in Asians and African Americans. The racial/ethnic differences were less consistent for other anxiety disorders, such as panic disorder, PTSD, and agoraphobia. Overall, Latinos tended to have a higher prevalence of agoraphobia, and Asians had the lowest rates of all anxiety disorders among the four racial/ethnic groups. Although the two studies shed important light on the mental health of older ethnic minorities, both focused on younger people (aged 50+ in Jimenez et al., 2010; aged 55+ in Woodward et al., 2012), instead of using more commonly used age cutoffs for older populations (e.g., 60+ or 65+).

It is noteworthy that the social categories of Latino and Asian both represent enormous internal heterogeneity based on demographic and social characteristics such as ethnic identity, nativity, socioeconomic status (SES), immigration history, acculturation, religion, language, and life experience (Choi, 2000; L. Hinton & Areán, 2008; Woodward et al., 2012). Such internal heterogeneity, however, does not contradict the fact that both Latinos and Asians share common values and norms, such as strong extended family networks, collectivism, respect for elders, and greater interdependence among family members (Kim & McKenry, 1998). In one study, when SES factors were controlled, there were more similarities than differences in family support and social network across ethnic minority groups (Kim & McKenry, 1998). We now discuss the strong family and social ties shared by older Latinos and Asians, which can be important predictors of mental health in later life.

Family Relations and Social Connections Among Latino and Asian Older Adults

Ecological system theory depicts the mutual accommodation between individuals and the changing environment, which is composed of nested structures, including micro-, meso-, exo-, macro-, and chronosystems (Bronfenbrenner, 1979). Family and neighborhood represent the microsystem that constitutes the immediate environment in which individuals engage in face-to-face interactions. They powerfully shape one’s psychological well-being through intimate interpersonal relationships (Bronfenbrenner, 1979, 1986). As older adults retire from work, family and neighborhood become the most important microsystem settings in which they participate in daily activities, play different roles, and maintain the most immediate relationships. Perceptions of these relationships, either positive or negative, substantially affect older adults’ adaptation to the environment and consequently shape their mental health outcomes. Because of cultural preferences, economic constraints, and linguistic isolation, families, friends, and ethnic communities are inextricably linked with the well-being of older immigrants/ethnic minorities (Treas, 2008). Close and supportive family relations and social connections help protect the mental health of older immigrants/ethnic minorities by mitigating stress, reducing social isolation, and enhancing access to social and health services (Krause, 2001).

Both Latino and Asian older adults tend to have larger family networks than Whites and African Americans, and they are more likely to seek assistance from their children in an emergency (de Leon & Glass, 2004; Kim & McKenry, 1998). Latino families are often portrayed as cohesive, with a deep sense of reciprocal family obligations in multigenerational households (Bermúdez et al., 2010). Asian families share these interdependent family norms with their Latino counterparts. For instance, compared to other ethnic groups, Chinese and Korean elders have fewer sources of support than their adult children and tend to turn first to their children for support (Wong, Yoo, & Stewart, 2007).

Maintaining close and supportive family relations is associated with lower risks of psychiatric disorders in both Latino and Asian older adults (Lopez et al., 2004; Mulvaney-Day, Alegria, & Sribney, 2007). However, little is known about whether the greater interdependence of family members in the two groups can be a source of tension, conflict, or disappointment or whether it can compromise the mental health of older adults. Among older immigrants in particular, family conflicts often occur as a result of cultural differences in values and lifestyles across the generations (Glick, 2010). Not exposed to the acculturating experience of work and school as their offspring are, older adults face more difficulties learning a new language and adapting to new cultural norms. Family cultural conflicts caused by different acculturation levels across generations have a detrimental influence on the mental health of both Latino and Asian older adults (Lee, Choe, Kim, & Ngo, 2000; Mulvaney-Day et al., 2007; Rivera et al., 2008). Power and authority dynamics also shift in immigrant/ethnic minority families as a result of the acculturation process. Older parents often lose authority that they used to have and tend to become peripheral to the nuclear family (Wong, Yoo, & Stewart, 2006). For instance, discord in values between older parents and their children lead older Asians to feel stripped of their role as “wisdom givers” and transmitters of cultural heritage (Guglani, Coleman, & Sonuga-Barke, 2000).

Friends compose another important dimension of social support for older adults. Support from friends is crucial to older adults who are experiencing challenges and difficulties related to immigration (Wong et al., 2006). Having same-age friends helped Chinese older immigrants to establish support networks for mutual help and information sharing and to maintain customs and values in the new country (Tsai & Lopez, 1998). Research similarly showed that a lack of friends or support from friends was directly related to depression in Latino immigrants (Almeida, Subramanian, Kawachi, & Molnar, 2011; Vega, Kolody, & Valle, 1987). Findings were inconsistent regarding whether friend or family support was more powerful in predicting depression in Latino immigrants (Almeida et al., 2011; Vega et al., 1987).

Immediate neighborhoods are important for older adults because of their decreased physical and cognitive functioning, increased discomfort with driving, and decreased contact with social network members (Yen, Michael, & Perdue, 2009). A cohesive and supportive neighborhood is particularly important for older immigrants who spend most of their time in the ethnic community. Latino enclave communities offer social resources to their members and have a salutary impact on health (Eschbach, Ostir, Patel, Markides, & Goodwin, 2004). Similarly, ethnic communities facilitate the adjustment of older Asian American immigrants to the new environment and increase their sense of well-being by providing religious and health care services, outlets for socialization and recreation with peers, and informational and instrumental support (Choi, 2000).

Data

This study used data from a nationally representative survey, the National Latino and Asian American Study (NLAAS, 2002–2003), which aimed to understand the prevalence of psychiatric disorders, family and social factors associated with these disorders, and mental health service utilization among adult Latinos and Asian Americans. The sampling design of NLAAS included three components: a nationally representative sample of primary and secondary sampling units, a high-density supplemental sampling of geographic areas where Asian or Latino ethnic groups made up more than 5% of the population, and a second respondent sampling from households in which a primary respondent had been interviewed (Alegría et al., 2004; Heeringa et al., 2004).

To qualify for the study, individuals had to have been 18 years of age or older, to self-identify as Latino or Asian American, and to reside in households in the 50 states or Washington, DC. Race and ethnicity were assessed by self-report responses to categories identical to those used in the 2000 Census. Trained interviewers with linguistic and cultural backgrounds similar to those of the respondents administered the survey face-to-face in the respondents’ preferred language. On average, the survey took 2.5 hr to complete. The final NLAAS sample consisted of 4,649 adults, of whom 2,095 were Asian American. The weighted response rates were 73.2% for the total sample, 75.5% for the Latino sample, and 65.6% for the Asian sample (Heeringa et al., 2004). To ensure that the sample was nationally representative, the NLAAS survey team developed sampling weights based on factors such as unequal probabilities of selection, characteristics of nonresponse, and poststratification. For the purpose of this study, we restricted our analytical sample to individuals aged 60 and older (N = 616), which included 360 Latinos and 256 Asians.

Measures

Analysis

We first estimated the weighted sample characteristics of Latino and Asian older adults followed by examinations of the lifetime and 12-month prevalence of the six anxiety and the two mood disorders. Given that family relations and social connections change over the lifetime, we conducted multistage logistic regressions to examine the associations of 12-month prevalence of anxiety and mood disorders, respectively, using the combined sample of Asians and Latinos. The three sets of variables mentioned previously were entered sequentially in regressions. This sequence enabled us to evaluate the unique contribution of family relations and social connections on individual well-being while segregating the possible effects triggered by sociodemographic and immigration-related variables. We normalized the indicators of family relations and social connections to have a mean of 0 and a standard deviation of 1 in the sample. All analyses took into account design effects (e.g., stratification and clustering) by using the svy family commands of Stata 11.0 (StataCorp, 2011).

The Prevalence of Anxiety/Mood Disorders Among Older Latinos and Asians

The prevalence of mental disorders differed significantly between older Latinos and Asians. Approximately one in seven (14.3%) older Latinos had experienced at least one anxiety or mood disorder in the past year, and more than one third (34.5%) had had at least one of these disorders in their lifetime. These figures were almost double those among older Asians (7.4% and 17.7%, respectively). Caution should be taken when comparing the findings of this study to other national statistics, given the differences in sampling (regional or national), age criteria (50+, 55+, 60+, or 65+), and assessment tools used (different versions of the DSM). To the best of our knowledge, only one national study has used identical age criteria (60+) and assessment tools (DSM–IV; Kessler et al., 2005). Kessler and colleagues’ (2005) study reported a lower lifetime prevalence of GAD (3.6%), social phobia (6.6%), panic disorder (2.0%), agoraphobia (1.0%), major depressive episode (10.6%), and dysthymia (1.3%) in the general older population than the older Latinos in our sample. In contrast, older Asians had a lower lifetime prevalence of most conditions (i.e., panic disorder, social phobia, agoraphobia, and major depressive episode) than the general older population. Older Latinos (2.4%) and Asians (2.3%) in our study seemed to have comparable levels of PTSD compared to the general older population (2.5%).

The results of regression analyses revealed that the favorable mental health status of Asians could be partially attributed to their higher SES and better physical health status. When these variables were controlled for, older Latinos were as likely as Asians to have anxiety or mood disorders. This finding is similar to that of Dunlop, Song, Lyons, Manheim, and Chang (2003), who found that Latinos and Whites exhibited similar rates of depression after confounders were controlled for.

It is noteworthy that although depression is the most studied psychiatric disorder in older adults, the most common mood/anxiety symptom experienced by both Latino and Asian elders was panic attack, which was assessed in this study mainly by somatic symptoms such as sudden feelings of being frightened, anxious, or uneasy; problems of shortness of breath, pounding heart, or feeling dizzy; and being afraid of dying, going crazy, or losing one’s mind. It is likely that the greater social acceptability of somatic expressions of mental disorders (in contrast to affective expressions) accounted for the high levels of panic attack diagnosis in the two groups. Previous studies have shown that culture influences socially acceptable patterns of emotion expression and that many minority cultures tend to deny or minimize the affective components of distress in favor of more socially acceptable somatic symptoms (D. Hinton & Good, 2009; Kirmayer, 2001). Culture-specific symptoms may also have contributed to the elevated number of panic attacks among the two groups. For instance, Latinos are more likely to experience cultural symptoms of ataque de nervios (attack of nerves), a greatly feared syndrome that may cause loss of control and death (D. Hinton & Good, 2009). In Khmer and Cambodian refugees, kyol goeu (wind overload) is a commonly experienced panicked state associated with dizziness and nausea (D. Hinton, Um, & Ba, 2001). Our findings suggest that focusing on depression alone does not provide an adequate assessment of the mental health risks of older Latinos and Asians. More studies are needed to understand the cross-cultural presentation of panic attacks among the two groups.

Family Relations and Anxiety/Mood Disorders Among Older Latinos and Asians

Among the various family relation and social connection variables investigated, only family cultural conflict and family cohesion were found to be associated with the 12-month prevalence of anxiety and mood disorders, respectively. This finding supports the assumption of the individual-level healthful effect; that is, the benefit of social relations on individuals is primarily experienced through individual- rather than community-level social connections (Mulvaney-Day et al., 2007). The finding also reflects the predominant importance of families in meeting older minorities’ social support needs. Whereas the family lives of other older Americans may involve a delicate balance of intimacy and independence, the families of older ethnic minorities are often characterized by closer geographical proximity and greater interdependence (Kritz, Gurak, & Chen, 2000). Older minorities contribute in various ways to their families through helping with child care and household chores. One study found that family demands could be so consuming for some older ethnic minorities that they had little time or energy for friends and social ties outside their families (Treas & Mazumdar, 2002). This may in part explain the greater influence of family relations than nonkin social connections on the older adults in this study.

In addition, a differential effect of family relations on anxiety and mood disorders was found: Family cohesion was a risk factor for depression only, and family cultural conflict was associated with elevated prevalence of anxiety but not depression. These findings are in line with previous research that has shown that smaller social networks and a lack of supportive relationships are risk factors associated with the prevalence of depression in older adults, whereas anxiety may be more strongly related to negative life events that present actual and more imminent threats (Vink, Aartsen, & Schoevers, 2008). Our finding underlines the importance of studying multidimensional family dynamics, as they are related to various mental health outcomes. This finding can also help clinicians differentiate between older adults at risk for anxiety or depression.

It is important to note that the common notions of strong family ties among Latinos and Asians may obscure the fact that family cultural conflict in the two groups can be an important source of distress and anxiety for older people. Similar findings have been reported in general populations—although positive relations tend to have moderate benefits on individual well-being, negative interactions appear to be more detrimental to mental health (see Finch, Okun, Pool, & Ruehlman, 1999, for a review). In other words, the absence of negative family interactions may be more important to an individual’s mental health than the presence of supportive interactions.

Several possible mechanisms may underlie this finding. First, as family ties cannot be terminated easily as other social ties, negative interactions in family relationships may be a relatively persistent and ongoing problem (Krause & Rook, 2003). In other words, family conflict may be fairly stable over long periods of time, being particularly detrimental to the mental health of older adults. Second, the negative feelings in a family may detract from the positive effects of close and supportive family relations (Fingerman, Pitzer, Lefkowitz, Birditt, & Mroczek, 2008). Although the majority of our sample either was U.S.-born or had lived in the United States for more than 20 years, 66.9% of the Latinos and 50% of the Asians had poor or fair English proficiency. It is likely that many older adults in the study relied heavily on their family members for instrumental or even monetary support. Having conflicts with family members would increase levels of stress and anxiety and trigger feelings of dependence and indebtedness upon the receipt of support from family members. Family conflicts may also inhibit family members from providing practical support to older adults, such as monitoring medication adherence, providing transportation to the clinic, and helping with translation, which in turn may compromise older persons’ mental health.

The negative consequences of having family conflicts may be particularly strong for Asians and Latinos, who both strongly endorse traditional family norms and collectivist ideologies. In more individualist cultures, relationships often take an independent form and come with relatively few obligations (Adams & Plaut, 2003). By contrast, in more collectivist cultures, such as Latino and Asian cultures, people emphasize the values of family obligation and subordination of individual interest to the group. In these cultures, individuals may be more cautious about potential negative family relations that disrupt family harmony (Goebert, 2009). Having family cultural conflicts is likely to result in disappointment and unhappiness among older Latinos and Asians due to unrealized aspirations for traditional values of family closeness, companionship, and interdependence (Treas & Mazumdar, 2002).

This study has several limitations. Because of insufficient representation of subethnic groups in the data set, intragroup differences among Latinos and Asians were not explored. Larger samples are necessary to differentiate the psychiatric vulnerabilities in subethnicities of the two groups. In addition, the cross-sectional design of this study precluded the identification of causal relationships. For instance, the associations between family conflict and the higher prevalence of anxiety disorders may also be explained by the fact that older adults who had poor mental health were more likely to have conflicts with family members. Longitudinal studies would provide a superior conceptual stand in testing causal relationships. Finally, social stigmas associated with mental illness in both groups (AOA, 2010a, 2010b) and the desire for social approval might have inhibited older Latino and Asian adults from overtly acknowledging mental disorders and negative family interactions. Hence, the findings on prevalence rates and family cultural conflicts should be interpreted with caution.