Ⅰ. Introduction

In 2011, there were 39.9 million immigrants, which is the highest number recorded in U.S. history (Passel & Cohn, 2012). By 2020, 33% of children will be children of immigrants (Mather, 2009). Children in immigrant families adjust to the host country’s language and culture more quickly than adults who have been fully socialized into their original culture (American Psychological Association, 2012). However, adolescents who experience additional challenges with schools and friends in the new social context could experience more distress than younger children (American Psychological Association, 2012).

More than one in five U.S. adolescents had mental disorders in 2010 (Merikangas et al., 2010), and about half of all adult mental disorders emerge in teenage years (Jones, 2013). Likewise, adolescence is an important transitional period between childhood and adulthood that is associated with a heightened risk of mental health problems for both immigrant and non-immigrant adolescents (Hoare, Skouteris, Fuller-Tyszkiewicz, Millar, & Allender, 2014). Furthermore, immigrant adolescents have to cope with distress produced while they are making efforts to be socialized to the new culture, which is conceptualized as acculturation (Kim & Abreu, 2001). In addition, during this acculturation process, they are often exposed to stressful circumstances that adolescents in the general population do not experience at their ages; for example, many of them take on their parents’ roles and responsibilities in daily living due to the limited English proficiency of their parents (Kao, 1999). Therefore, immigrant adolescents may be even more vulnerable to mental health problems than those in the general population.

In fact, there has been an agreement that acculturating process may cause an increased risk of health behaviors among immigrant adolescents (Unger, Gallaher, Shakib, Ritt-Olson, Palmer, & Johnson, 2002). Studies also have found that acculturative stress was significantly associated with internalizing symptoms such as anxiety, depression, and somatic pains among immigrant adolescents (Alegría, Mulvaney-Day, Torres, Polo, Cao, & Canino, 2007; Katsiaficas, Suárez-Orozco, Sirin, & Gupta, 2013; Kim, Cain, & McCubbin, 2006; Mendoza, Javier, & Burgos, 2007). Furthermore, non U.S.-born immigrant adolescents experience greater acculturative stress and mental health problem symptoms than those born in U.S. (Sirin, Ryce, Gupta, & Rogers-Sirin, 2013) because migration causes significant mental distress for adolescent immigrants (Torres & Wallace, 2013). Non U.S.-born adolescents have to maneuver norms between the origin culture and the host culture, and the bicultural stress may negatively influence mental well-being for immigrant adolescents (Romero, Carvajal, Valle, & Orduña, 2007). To assist immigrant adolescents who are new to the U.S. culture, it is important to investigate the mental health status of non U.S.-born adolescents and to know what contributes to their mental health problems.

Ⅱ. Literature Review

According to Bronfenbrenner’s (1979) social ecological model individuals are influenced by intrapersonal, interpersonal, and community level factors. Immigrant adolescents have to adapt to the new U.S. culture, negotiating risk and protective factors at all of these levels. At the individual level, age may be associated with mental health symptoms; older Asian immigrant adolescents reported higher levels of general mental health symptoms (Yeh, 2003). Different types of stress related to acculturation were also significantly associated with high risk of mental health problems in different studies (Katsiaficas, Suaréz-Orozco, Sirin, & Gupta, 2013; Potochnick & Perreira, 2010; Sirin et al., 2013). Increased acculturative stress predicted higher levels of internalizing behavior problems among immigrant adolescents (Katsiaficas et al., 2013; Sirin et al., 2013). Migration stressors, among Latino immigrant adolescents, were associated with increased risk of depressive symptoms and anxiety (Potochnick & Perreira, 2010). In addition, when immigrant youth became more familiar with U.S. culture, they tend to have fewer mental health problems. For example, greater identification with American culture is associated with fewer symptoms of depression and anxiety (Yeh, 2013) and longer stay in the U.S. is associated with reduced risk for anxiety and depression (Potochnick & Perreira, 2010). However, when gender and age are controlled, stay in the U.S. becomes marginally significant (Potochnick & Perreira, 2010).

Resources at the family level also impact mental health among immigrant adolescents. Support from one's family decreased the risk of depressive symptoms and anxiety (Potochnick & Perreira, 2010; Yeh, 2003). Higher levels of family-based acculturation stress was associated with higher level of depression among Latino immigrant adolescents (Kam & Lazarevic, 2013). Although primary language (English) spoken is considered to be a reliable proxy measure of acculturation for foreign born minorities (Lee, Nguyen, & Tsui, 2011), Yu and colleagues (2003) found that when immigrant adolescents used English as a primary language at home rather than other languages, they demonstrated lower risk for psychosocial development problems such as alienation from classmates and being bullied.

As a school level factor, support from teacher is associated with lower levels of depressive symptoms and anxiety (Potochnick & Perreira, 2010). Additionally, emotional support and academic support from significant others were each related to lower levels of internalizing behavior symptoms (Katsiaficas et al., 2013). Furthermore, although higher levels of acculturative stress were associated with higher levels of internalizing behavior problems, the relationship was moderated by both emotional and academic supports. The finding indicates that social supports may play a critical role during the process of acculturation (Katsiaficas et al., 2013).

In sum, previous studies on immigrant adolescents have found different factors contributing to their mental health problems. However, the studies mostly have focused on the relationships between acculturative factors and mental health problems. Although there were some studies that investigated the relationships between one or more specific social environmental factors and mental health problems (Katsiaficas et al., 2013; Potochnick & Perreira, 2010), there has been no study that examined multiple factors related to mental health problems with ecologically based approach. Acculturation occurs through contacts with various social and environmental sources, for example, parents, peers, formal education, and even exposure to media (Gfroerer & Tan, 2003); therefore, it is important to investigate factors of mental health problems with a social ecological perspective. In addition, few studies have focused on mental problems of non U.S.-born immigrant adolescents despite the fact that acculturation experiences between those who were born in the U.S. and those who were born outside U.S. are significantly different. Therefore, utilizing Bronfenbrenner’s (1979) social ecological model, this study examined individual, social, and environmental factors of mental distress, being defined as a wide range of mental health problems from a disturbing or unpleasant emotional state to chronic and severe mental disorders (www.change.org), focusing on non U.S.-born immigrant adolescents.

Ⅲ. Method

This secondary data analysis used data from the 2011-2012 California Health Interview Adolescent survey (CHIS 2011-2012); 2,799 adolescents aged 12 to 17 were recruited using random-digit dial telephone survey and interviewed in one of five languages: English, Spanish, Chinese (Mandarin and Cantonese dialects), Vietnamese, and Korean. An analysis of the 2000 Census data identified the interview languages that might cover the largest number of residents with limited English speaking skills in the CHIS sample. To encourage participation, an advance letter in five languages explaining the survey process with a $2 bill was mailed to participants. For the Adolescent Survey, the completion rate was reported as 42.5%, and the average time of the interview was 21.5 minutes. Detailed information is available elsewhere (http://healthpolicy.ucla.edu/Pages/home.aspx).

Ⅳ. Results

With regard to language use, 238 (68.4%) of non U.S.-born adolescents were interviewed in English, but only 95 (27.3%) reported that they spoke English at home (See Table 1). Approximately half (48.6%) have been living in the U.S. for more than 10 years. The adolescents reported that they rarely have adult supervision after school hours (M = 1.8, SD = 1.0); but they felt relatively high levels of support or care from teachers or adults in school (M = 19.4, SD = 3.7). Moreover, one in 10 students (n = 34, 9.8%) had experience of peer threatening in school. Adolescents also reported average scores of 8.8 for neighborhood cohesion indicating relatively positive feeling about their neighborhood. Among the non U.S.-born adolescents, the average K6 score was 3.8 (SD = 3.2), and compared to the average mental distress scores (M = 4.0, SD = 3.6) for the U.S.-born adolescents in the 2011-2012 CHIS Adolescent survey, the level of mental distress of non-U.S. born adolescents was not significantly different from that of U.S.-born adolescents (t = .892, p = .372). Also, there was no difference in terms of the proportion of having clinical mental distress, as indicated by scores of 13 or greater on the K6, between the U.S.-born (n = 83, 3.4%) and the non U.S.-born adolescents (n = 7, 2.0%), χ² (1, N = 2799) = 1.899, p = .168.

A hierarchical multiple regression was run to examine predictors in each level related to mental distress among non U.S.-born adolescents (See Table 3) for full details on each regression model). Models 1 and 2, with the individual and family predictors were not significant; however, Model 3 was significant (F = 6.15, p < .001) when the school predictors were entered and Model 4 remained significant when the community predictor was added (F = 5.90, p < .001; see Table 3). The final model accounted for 15% (adjusted R² = .12) of the variance in mental distress. As shown in Table 3, the school level predictors accounted for the largest percentage of variance in mental distress(10.1%).

In the final model (Model 4) including all predictors at individual, family, school, and community levels, interview language, home language, school support, peer threatening, and neighborhood cohesion were significantly related to mental distress among non U.S.-born adolescents. At the individual level, those who completed the interview in a language other than English (β = -.21, p = .001) had lower mental distress scores than those who completed the interview in English. Other individual level predictors were not significantly associated with mental distress. At the family level, using English at home (β = .10, p = .057) was related to lower mental distress scores. At the school level, youth with more school support (β = -.26, p < .001) and those who had not experienced peer threatening (β = .16, p = .003) had lower mental distress scores. Last, at the community level, adolescents who perceived higher levels of neighborhood cohesion (β = -.11, p = .049) were related to lower levels of mental distress.

Ⅴ. Discussion

We identified experiences regarding immigration and mental distress among a sample of non U.S.-born adolescents. Although more than two-thirds of the sample completed the interview in English, only one-quarter indicated they spoke English at home. This gap has been found in previous research and might be due to the fact that non U.S.-born adolescents learn English faster than their parents and may be in transition to using English as their primary language, whereas their parents may need to or prefer to use their first language rather than English at home (Kao, 1999; Mather, 2009). Over 90% of English interviewees in this study have been living in the U.S. for more than 5 years, which is consistent with findings that language acquisition is related to length of exposure to a new culture (Kharkhurin, 2008; Whatley & Batalova, 2013).

In addition, profiles of non U.S.-born adolescents were not different from those of U.S.-born adolescents. These findings are inconsistent with prior studies that found that immigrant adolescents are often disadvantaged (Escobar, Nervi, & Gara, 2000, Gupta, Rogers-Sirin, Okazaki, Ryce, & Sirin, 2014; Stolk, Kaplan, & Szwarc, 2014). However, Feliciano (2001) suggested that non U.S.-born adolescents might have benefits and resources from both the immigrant community and mainstream society in U.S. Thus, these findings provide an explanation that non U.S.-born adolescents might have benefits and resources regarding mental health as well as increased mental health problems due to bicultural stress. It is also possible that the sample in this study had stayed longer and were more acculturated than others in previous studies that found significant difference on mental health status between U.S.-born and non U.S.-born adolescents studies (e.g., Sirin et al., 2013). For further study, length of stay in the U.S., other socioeconomic aspects, and substance abuse need to be considered to examine their effects on mental health of non U.S.-born adolescents.

Five predictors related to increased mental distress emerged; English interview language at the individual level, non-English home language at the family level, lower support and experience being threatened by peers at the school level, and lower neighborhood cohesion at the community level were associated with high levels of metal distress. Interestingly, language use as an important acculturation factor had different influences in mental distress at the individual and family levels. There were possible explanations for this difference. First, non U.S.-born adolescents selecting English as an interview language may have more benefits than those selecting non-English. In this study, interview language was significantly related to years lived in U.S., but home language was not. Over 90% of those selecting English as an interview language had been living in U.S. more than five years. In this period that foreign-born adolescents learn English, they might have more attention and support from family or school and lower levels of expectations for academic success (Yeh, 2003); thus, this can cause a reduced mental distress.

An alternative explanation might be because of bicultural stress. Children need less time to adjust in mainstream society than parents (Lau, McCabe, Yeh, Garland, Wood, & Hough, 2005; Portes & Rumbaut, 2001); this acculturation gap leads to family conflict or poor mental health (Kim & Park, 2011). Acculturation is important at the family level as well as at the individual level because of effects on the relationship between adolescents and parents (Schwartz et al., 2013). Family cultural conflict or intergenerational conflict may trigger or aggravate mental disorders among immigrant adolescents (Cook, Alegria, Lin, Guo, 2009; Wu & Chao, 2011).

Given the study findings, this study suggests that acculturation at the individual and family levels might be important for non U.S.-born adolescents' mental distress. For example, it may be helpful for parents to speak English or mixed languages at home so that they will better understand their children. On the other hand, if parents do not know English, they may be stressed by learning and using English as a primary language and they might want to maintain their children's knowledge of the language of the country of origin. Consequently, parents may need support learning and using English in comfortable environments such as their home. Furthermore, as previous studies examined (Gonzales, Deardorff, Formoso, Barr, & Barrera, 2006), parent-adolescent communication may promote the mental health among non U.S.-born adolescents. Therefore, it is meaningful to examine specific language brokering or bicultural gap in family to mental distress among non U.S.-born adolescents for future study using moderate and mediate analyses.

The findings of this study suggest that factors at the school and community levels are associated with mental distress for non U.S.-born adolescents. First, highlighted factors in the school environment or climate for immigrant adolescents are consistent with previous studies. These factors include perceived discrimination, relationship with or support from teachers, alienation from or being bullied by peers (Walsh, Harel-Fisch, & Fogel-Grinvald, 2010), school safety, and belonging and engagement (McNeely & Falci, 2004). The findings suggest that school level factors might have greater impact on mental health outcomes among non U.S.-born adolescents than any other level factor. Also, this implies that teachers can improve non U.S.-born adolescents’ mental health by paying attention to them. In addition, supporting building positive peer-relationships can reduce mental distress among foreign-born adolescents under less familiar school environments compared to U.S.-born counterparts. For future research, examinations of the association between acculturation and school supports or bullying should be conducted, and intervention research using relevant school services and programs will be meaningful for immigrant adolescents’ mental health.

Second, better perceptions of neighborhood cohesion were associated with higher levels of mental distress for non U.S.-born adolescents in this study. This finding is consistent with previous studies that higher levels of neighborhood cohesion are associated with better mental health outcomes among immigrant adolescents because they might have better networks, supports, and resources in communities to cope with conditions that generate mental health problems (Curtis, Dooley, & Phipps, 2004; Leventhal & Brooks-Gunn, 2000). This study suggests that perceptions about neighborhood cohesion might be important for immigrant adolescents' mental health. Future studies on collaboration to promote mental health of immigrant adolescents among their families, schools, and communities as a part of a complex environmental system should be considered.

In spite of expanded knowledge and explanations about acculturation and environmental factors influencing the mental distress of non U.S.-born adolescents, the current study has several limitations to be acknowledged. First, the study sample is geographically limited to California. Also, combining all different ethnic groups into non U.S.-born adolescents can cause a generalized bias by ignoring differences among minorities. In future studies, the different ethnic groups within non U.S.-born adolescents should be examined. Second, the findings are based on cross-sectional survey data and therefore causal relationships cannot be tested. In addition, the findings of this study may not reflect up-to-date characteristics of immigrant youth due to use of 2011-2012 CHIS data. Thus, longitudinal data or intervention research should be conducted to discover causes of mental disorders among immigrant adolescents. Lastly, because of secondary data analyses, this study has a limited ability to explain acculturation using proxy measures such as interview language, home language, and length of stay in the U.S. Considering the finding that about 50% of the sample stayed in the U.S. for more than 10 years, developmental factors of individuals such as self-esteem, school performance, and issues related to dating may influence mental distress of the youth. However, this study also was not able to include developmental factors due to use of secondary data.

In conclusion, based upon the findings, this study has enriched the existing body of knowledge and theory. Specifically, it explicitly highlighted language is an important acculturation indicator. Acculturation gaps in family may cause communication difficulties among family members and have a negative impact on mental health status of non U.S.-born adolescents. Also, this study provided insight into how support from school and neighborhood are important for non U.S.-born adolescents using the ecological model. Schools and communities may be the channels to help immigrant adolescents’ acculturation process and ultimately to reduce their mental distress. Based on the findings and implications of this study, various service intervention studies should be conducted in the future and treatments to reduce the mental health problems of immigrant youth at the family, school, and community levels should be considered for practice.