Limited English Proficiency and Trajectories of Depressive Symptoms Among Mexican American Older Adults

Limited English Proficiency and Trajectories of Depressive Symptoms Among Mexican American Older... Abstract Background and Objectives This study examined the effect of limited English proficiency (LEP) on trajectories of depressive symptoms among Mexican American older adults in the United States. Research Design and Methods The sample was drawn from Waves 1 to 6 (1993–2007) of the Hispanic Established Population for Epidemiological Studies of the Elderly (H-EPESE). A total of 2,945 Mexican American older adults were included in the analyses. A latent growth curve modeling was conducted. Results After adjusting for covariates, results show that Mexican American older adults with LEP had higher levels of depressive symptoms than those with English proficiency (EP) at baseline as well as over a 14-year period. Differential trajectories were observed between those with LEP and EP over time, indicating that those with LEP had a significantly steeper curve of depressive symptom trajectories over time than those with EP. Discussion and Implications These results suggest that LEP is a risk factor not merely for greater depressive symptomatology at each time point, but for an accelerated trajectory of depressive symptoms over time among Mexican American older adults. Overall, the findings emphasize the need to assist not only the informal support system of Mexican American older adults with LEP, but also the formal system. Depressive symptoms, Depression trajectories, Latent growth curve modeling, Limited English proficiency, Mexican American older adults The Hispanic population—defined as “people of Cuban, Mexican, Puerto Rican, South or Central American or other Spanish culture or origin regardless of race” (Humes, Jones, & Ramirez, 2011)—is not only the largest racial/ethnic minority group in the United States, but the population with the largest proportion of persons with limited English proficiency (LEP) (Zong & Batalova, 2015). According to the U.S. Census Bureau, 25.3 millions of Americans, or 8.6% of the U.S. population, speak English less than “very well” and are classified as having LEP (U.S. Census Bureau, 2005; Pandya, McHugh, & Batalova, 2011). The proportion of persons with LEP is estimated at 8.3% of immigrants aged 60 years or older in the United States (U.S. Census Bureau, 2010). The highest proportions of LEP are found in the Hispanic population (Zong & Batalova, 2015), and it is of particular relevance to this paper that across all racial/ethnic groups, older Hispanics are the group most likely to report LEP (Krogstad & Gonzalez-Barrera, 2015, Krogstad, Stepler, & Lopez, 2016). LEP’s impact is pervasive for immigrants living in the United States and other English-speaking countries. First, LEP has been reported as a risk factor for poor health conditions (Derose, Escarce, & Lurie, 2007; Dubard & Glizlice, 2008; Gee & Ponce, 2010; Jang, Yoon, Park, & Chiriboga, 2016; Kandura, Lauderdale, & Baker, 2007; Kim, Aguado Loi, et al., 2011; Kim, Worley, et al., 2011; Mui, Kang, Kang, & Domanski, 2007; Ponce, Hays, & Cunningham, 2006; Sentell & Braun, 2012). Individuals with LEP tend to have poorer physical health status (e.g., poorer self-rated health, poorer health-related quality of life), as well as poorer mental health status (e.g., greater depressive symptoms, greater diagnosis for major depression and anxiety, and greater psychological distress) compared to those who are proficient in English or native English speakers (Derose et al., 2007; Dubard & Glizlice, 2008; Gee & Ponce, 2010; Kandula et al., 2007; Kim, Worley, et al., 2011; Mui et al., 2007; Ponce et al., 2006; Sentell & Braun, 2012). Second, literature also suggests that LEP acts as a significant barrier to access to and utilization of physical and mental health care (Abe-Kim et al., 2007; Kim, Aguado Loi, et al., 2011; Kim, Worley, et al., 2011; Ponce et al., 2006; Sentell, Shumway, & Snowden, 2007). This is mainly due to health care settings’ large dependence on verbal communications required for diagnosis and treatment of illness (Dubard & Gizlice, 2008; Kim, Worley, et al., 2011; Sentell et al., 2007). Lastly and more related to the present study, LEP is frequently reported to be associated with greater depression as language barriers due to LEP could be a significant source of stress, which can exacerbate the acculturative stress among immigrants (Constantine, Okazaki, & Utsey, 2004; Lin & Yi, 1997; Nwadiora & McAdoo, 1996; Sadule-Rios, 2012). For instance, Krause & Goldenhar (1992) reported that older Mexican Americans with LEP tended to show higher risk for depressive symptoms, as compared to those with greater English proficiency (EP). Researchers suggest that LEP itself could be an important social status variable that can directly influence depressive symptoms (Berry, 1997; Zhang, Hong, Takeuchi, & Mossakowski, 2012). Specifically, the LEP status can create inconveniences and discomforts in daily lives in English-speaking countries, especially when communicating with and developing social networks with English-speaking individuals, as well as accessing useful information, which may lead to increased levels of stress (Zhang et al., 2012). Given LEP’s relation to depressive symptoms, researchers have further questioned LEP’s effects on longitudinal courses of depressive symptoms. For example, Nguyen, Rawana, and Flora (2011) examined the effect of immigration factors (i.e., host language proficiency and years since immigration) on the trajectories of depressive symptoms among immigrant adolescents in Canada. They discovered that proficiency in host language significantly influenced on the quadratic trajectories of depressive symptoms; adolescents who were proficient in host language showed steeper increases in depressive symptoms until they diverge at about mid-adolescence and exhibited higher rates of depressive symptoms in early adulthood, as compared to their counterparts with no proficiency in English. To our knowledge, there is no literature on the effect of EP on depressive symptoms among older adults from a longitudinal perspective; therefore, we will attempt to explore the longitudinal effect of LEP among older adults in the present study. The present study was guided by two theories to conceptualize the potentially cumulative disadvantageous effects of LEP: the cumulative disadvantage theory (Dannefer, 2003) and the double jeopardy theory (Dowd & Bengtson, 1978). The cumulative disadvantage theory describes how early advantage or disadvantage is critical to how cohorts become differentiated over time (Dannefer, 2003). Many health disparities researchers employed the cumulative disadvantage theory to examine longitudinal effects of risk factors on minority older adults’ health (e.g., Carreon & Noymer, 2011). In the case of the present sample, given the generally very low socioeconomic status of LEP Hispanic older adults, the majority of them have faced disadvantages across their live courses. The double jeopardy theory (Dowd & Bengtson, 1978) has also been widely used in minority aging research. According to the double jeopardy theory (Dowd & Bengtson, 1978), racial/ethnic minority older adults may suffer from the double disadvantage to health due to the interactive effects of age and race/ethnicity. In the case of the present sample, Mexican Americans have in general faced lives with greater disadvantages than those faced by non-Hispanic White Americans. As Mexican Americans reach later life and begin to experience many physical and social and behavioral problems associated with the aging process, this is yet another jeopardy, which supports the double jeopardy framework. Taken together the abovementioned theoretical frameworks, we view that racial/ethnic minority older adults with LEP may be in a disadvantaged position and this disadvantaged effect influences cumulatively and longitudinally compared to those who are proficient in English. The purpose of the present study is to assess the effect of LEP on trajectories of depressive symptoms among Mexican American older adults in the United States. Our overarching research questions are whether LEP is harmful for mental health and whether the cumulative disadvantage arising from the lack of language skills influences mental health among minority older adults. Based on prior literature, we hypothesized that (1) Mexican American older adults with LEP would have higher levels of depressive symptoms than those with EP at baseline as well as over time, and (2) Mexican American older adults with LEP would have a steeper curve of depressive symptom trajectories over time compared with those with EP. Methods Sample The sample was drawn from the first six waves of the Hispanic Established Population for Epidemiological Studies of the Elderly (Hispanic EPESE). The Hispanic EPESE is a multistage stratified probability study of community-dwelling Mexican Americans aged 65 years and over who reside in the five southwestern states that contain the majority of Mexican Americans: Texas, California, New Mexico, Colorado, and Arizona. The first wave of interviews (n = 3,050) was conducted between September 1993 and June 1994. Respondents were interviewed at a place of their choosing including their homes by trained interviewees and the interviews were conducted in English or Spanish depending upon respondents’ preference. Follow-up interviews were then conducted approximately every 2 to 3 years, with a supplemental sample of 902 respondents from the same cohorts as the original sample added at Wave 5 (2004–2005). All of the new respondents were aged 75 years or older, were identified using the same multistage sampling procedures and again were interviewed at a place of their choosing. For the present analyses, among a total of 3,952 respondents from Waves 1 to 6 (1993–2007), 2,945 Mexican Americans aged 65 years and older, who have a depression score at least at one time point for the first four waves of data were included in further analyses. As mentioned above, given that there was a large, new group of respondents (n = 902) added at Wave 5, we removed those newly added cases for our analysis but retained the original group of respondents only in the current study. Measures Depressive Symptoms Depressive symptoms were measured with the Center for Epidemiologic Studies Depression scale (CES-D), a 20-item self-administered questionnaire (Radloff, 1977). Respondents were asked to assess the frequency of depressive symptoms experienced during the past week, based on a 4-point scale with categories in the subsequent order: rarely or none of the time (0), some or a little of the time (1), much of the time (2), and most or all of the time (3). The total scores for 20 items potentially ranged from 0 to 60, with higher scores indicating more depressive symptoms. A score of 16 or higher is generally deemed as a sign of probable clinical depression (Zarit & Zarit, 1998). An internal consistency of the CES-D was reported as moderately high (Cronbach’s α = .88) in this Hispanic population. LEP Respondents were asked to respond how well they speak English using the following rating scale: 1 (not at all), 2 (not too well), 3 (pretty well), and 4 (very well). Based on previous studies (Kim, Aguado Loi, et al., 2011; Kim, Worley, et al., 2011; Ponce et al., 2006), LEP was determined if the responses were not at all or not too well and coded as 1 for further analysis (1 for LEP; 0 for non-LEP). Covariates Respondents’ sociodemographic background characteristics (i.e., age, gender [1 for female; 0 for male], marital status [1 for married; 0 for else including separated, divorced, widowed, and never married], educational attainment), nativity (1 for U.S. born; 0 for non-U.S. born), and interview language (1 for Spanish; 0 for English) were used as covariates. These variables were selected as previous studies reported significant correlations with depressive symptoms (Akhtar-Danesh & Landeen, 2007; Angel, Frisco, Angel, & Chiriboga, 2003; Bebbington et al., 1998; Blazer, Burchett, Service, & George, 1991; Cole & Dendukuri, 2003; Jang, Haley, Small, & Mortimer, 2002). Data Analysis The present study used the Full Information Maximum Likelihood estimation method (FIML) to include all available cases rather than deleting those with missing data as FIML estimation has shown superior performance for estimating the parameters compared to other traditional missing data analysis (e.g., listwise deletion, pairwise deletion) (Enders & Bandalos, 2001; Little & Rubin, 2002). To examine the longitudinal trajectories of depressive symptoms, we employed a latent growth curve model (Preacher & Hayes, 2004), which takes into account the between-individual difference as well as within-individual variance. In latent growth models, each growth related term is treated as a latent construct having the observed measures as indicators (see an example in Supplementary Figure S1). Mplus 7.1 (Muthen & Muthén, 2010) with ESTIMATOR=MLR was used to analyze the latent growth curve models by handling the missing data as well as adjusting for the standard errors given the non-normal data. To search for the best fitting growth trajectory to our data, we first analyzed an unconditional growth model, which has no predictors or covariates in the model. To search for the optimal growth trajectory, we then fit a series of four unconditional growth models: including intercept-only, linear growth, quadratic growth, and cubic growth. For the within-subject error variance structure, all possible variance-covariance components were freely estimated with no constraints (i.e., UN structure) given that UN structure is best suited in the search for the optimal growth trajectory (Kim, Kwok, Yoon, Willson, & Lai, 2016). The Likelihood Ratio Test, the Akaike Information Criteria (AIC), and the Bayesian Information Criteria (BIC) were then used to compare the competing models. If there is no significant improvement in LRT, the simpler model is considered to be a better fitting model over the more complex model. Likewise, a smaller AIC and BIC represents a better fitting model over the competing model. Results Background Characteristics of the Sample As summarized in Table 1, age at baseline ranged from 65 to 107 with an average age of 72.83 years (SD = 6.57 years). Over half were female and married. The majority had attained education less than high school (less than 11 years) with an average of 4.85 years of education. Over half (55.9%) reported that they were born in the United States, whereas 78% were interviewed in Spanish (22% in English). Regarding our variable of interests, over half (60.9%) appeared to have LEP by assessing their English-speaking ability as not at all or not too well. All demographic characteristics were significantly different between LEP and EP groups (all ps < .001). The missing data rate for longitudinal depressive symptom variables ranged from 4% for Wave 1 and 72% for Wave 6 (see Table 1). Table 1. Descriptive Statistics for the Demographic Characteristics (n = 2,945) Variable M (SD) or % Total sample LEP group EP group Age (years)** 72.83 (6.57) 73.22 (6.92) 72.28 (5.95) Female** 57.9% 60.3% 53.7% Educational Attainment**  Less than high school (0–11 years) 90.1% 98.3% 77.1%  High school (12 years) 6.8% 1.2% 15.8%  Less than college (13–15 years) 1.6% .1% 3.8%  College (16 years) 1.0% .2% 2.3%  More than college (>17 years) .5% .2% 1.0% Nativity**  U.S. Born 55.9% 41.1% 80.3%  Non-U.S. Born 44.0% 58.9% 19.7% Marital status**  Married 55.8% 52.6% 60.9%  Separated 3.0% 3.6% 1.9%  Divorced 4.7% 3.8% 6.3%  Widowed 30.8% 34.0% 25.9%  Never married 5.6% 6.0% 5.0% Interview language**  English 22.0% 8.2% 43.9%  Spanish 78.0% 91.8% 56.1% Depressive symptoms (CES-D)  Wave 1 (n = 2,823)** 9.94 (9.59) 10.60 (9.73) 8.75 (9.07)  Wave 2 (n = 2,241)* 7.10 (8.41) 7.47 (8.66) 6.58 (7.95)  Wave 3 (n = 1,813)** 8.44 (9.02) 8.93 (9.50) 7.56 (8.07)  Wave 4 (n = 1,587) 7.22 (7.78) 7.33 (7.87) 7.09 (7.72)  Wave 5 (n = 1,060)* 9.58 (8.94) 10.06 (9.45) 8.82 (8.07)  Wave 6 (n = 812) 9.35 (8.94) 9.94 (8.91) 8.30 (8.91) Variable M (SD) or % Total sample LEP group EP group Age (years)** 72.83 (6.57) 73.22 (6.92) 72.28 (5.95) Female** 57.9% 60.3% 53.7% Educational Attainment**  Less than high school (0–11 years) 90.1% 98.3% 77.1%  High school (12 years) 6.8% 1.2% 15.8%  Less than college (13–15 years) 1.6% .1% 3.8%  College (16 years) 1.0% .2% 2.3%  More than college (>17 years) .5% .2% 1.0% Nativity**  U.S. Born 55.9% 41.1% 80.3%  Non-U.S. Born 44.0% 58.9% 19.7% Marital status**  Married 55.8% 52.6% 60.9%  Separated 3.0% 3.6% 1.9%  Divorced 4.7% 3.8% 6.3%  Widowed 30.8% 34.0% 25.9%  Never married 5.6% 6.0% 5.0% Interview language**  English 22.0% 8.2% 43.9%  Spanish 78.0% 91.8% 56.1% Depressive symptoms (CES-D)  Wave 1 (n = 2,823)** 9.94 (9.59) 10.60 (9.73) 8.75 (9.07)  Wave 2 (n = 2,241)* 7.10 (8.41) 7.47 (8.66) 6.58 (7.95)  Wave 3 (n = 1,813)** 8.44 (9.02) 8.93 (9.50) 7.56 (8.07)  Wave 4 (n = 1,587) 7.22 (7.78) 7.33 (7.87) 7.09 (7.72)  Wave 5 (n = 1,060)* 9.58 (8.94) 10.06 (9.45) 8.82 (8.07)  Wave 6 (n = 812) 9.35 (8.94) 9.94 (8.91) 8.30 (8.91) Note: **p < .001, *p < .05 based on results from a chi-square test or t tests comparing between LEP and EP groups. CES-D = Center for Epidemiologic Studies Depression scale; EP = English proficiency; LEP = Limited English proficiency. View Large Table 1. Descriptive Statistics for the Demographic Characteristics (n = 2,945) Variable M (SD) or % Total sample LEP group EP group Age (years)** 72.83 (6.57) 73.22 (6.92) 72.28 (5.95) Female** 57.9% 60.3% 53.7% Educational Attainment**  Less than high school (0–11 years) 90.1% 98.3% 77.1%  High school (12 years) 6.8% 1.2% 15.8%  Less than college (13–15 years) 1.6% .1% 3.8%  College (16 years) 1.0% .2% 2.3%  More than college (>17 years) .5% .2% 1.0% Nativity**  U.S. Born 55.9% 41.1% 80.3%  Non-U.S. Born 44.0% 58.9% 19.7% Marital status**  Married 55.8% 52.6% 60.9%  Separated 3.0% 3.6% 1.9%  Divorced 4.7% 3.8% 6.3%  Widowed 30.8% 34.0% 25.9%  Never married 5.6% 6.0% 5.0% Interview language**  English 22.0% 8.2% 43.9%  Spanish 78.0% 91.8% 56.1% Depressive symptoms (CES-D)  Wave 1 (n = 2,823)** 9.94 (9.59) 10.60 (9.73) 8.75 (9.07)  Wave 2 (n = 2,241)* 7.10 (8.41) 7.47 (8.66) 6.58 (7.95)  Wave 3 (n = 1,813)** 8.44 (9.02) 8.93 (9.50) 7.56 (8.07)  Wave 4 (n = 1,587) 7.22 (7.78) 7.33 (7.87) 7.09 (7.72)  Wave 5 (n = 1,060)* 9.58 (8.94) 10.06 (9.45) 8.82 (8.07)  Wave 6 (n = 812) 9.35 (8.94) 9.94 (8.91) 8.30 (8.91) Variable M (SD) or % Total sample LEP group EP group Age (years)** 72.83 (6.57) 73.22 (6.92) 72.28 (5.95) Female** 57.9% 60.3% 53.7% Educational Attainment**  Less than high school (0–11 years) 90.1% 98.3% 77.1%  High school (12 years) 6.8% 1.2% 15.8%  Less than college (13–15 years) 1.6% .1% 3.8%  College (16 years) 1.0% .2% 2.3%  More than college (>17 years) .5% .2% 1.0% Nativity**  U.S. Born 55.9% 41.1% 80.3%  Non-U.S. Born 44.0% 58.9% 19.7% Marital status**  Married 55.8% 52.6% 60.9%  Separated 3.0% 3.6% 1.9%  Divorced 4.7% 3.8% 6.3%  Widowed 30.8% 34.0% 25.9%  Never married 5.6% 6.0% 5.0% Interview language**  English 22.0% 8.2% 43.9%  Spanish 78.0% 91.8% 56.1% Depressive symptoms (CES-D)  Wave 1 (n = 2,823)** 9.94 (9.59) 10.60 (9.73) 8.75 (9.07)  Wave 2 (n = 2,241)* 7.10 (8.41) 7.47 (8.66) 6.58 (7.95)  Wave 3 (n = 1,813)** 8.44 (9.02) 8.93 (9.50) 7.56 (8.07)  Wave 4 (n = 1,587) 7.22 (7.78) 7.33 (7.87) 7.09 (7.72)  Wave 5 (n = 1,060)* 9.58 (8.94) 10.06 (9.45) 8.82 (8.07)  Wave 6 (n = 812) 9.35 (8.94) 9.94 (8.91) 8.30 (8.91) Note: **p < .001, *p < .05 based on results from a chi-square test or t tests comparing between LEP and EP groups. CES-D = Center for Epidemiologic Studies Depression scale; EP = English proficiency; LEP = Limited English proficiency. View Large Results From Growth Curve Modeling Analysis Depressive symptom scores over six time points are displayed in the bottom rows of Table 1. The mean depressive symptom score at each wave shows that there is some fluctuation on depressive symptom scores over time. These changes in depressive symptoms over time provide justification for searching for the appropriate growth trajectory. As displayed in Table 2, different growth models were compared in order to find the best fitting model. Based on model fit statistics presented in Table 2, both quadratic and cubic growth models fit the best to the data to represent the growth trajectory of depressive symptoms across six waves. We have, therefore, plotted the two competing growth models (i.e., quadratic vs cubic growth model) to compare the overall trajectory and have found that they have almost equivalent growth patterns across times. Based on this equivalence, we selected the more parsimonious quadratic growth trajectory model for further analyses. After comparing the model having the interaction effect between the time and LEP to the model without the interaction, we selected the interaction model as our final model and interpreted the results based on the corresponding model. Table 2. Model Specification Model specification X2 df AIC BIC Intercept-only 278.176 5 73,212 73,344 Linear growth 274.069 4 73,210 73,348 Quadratic growth 139.284 3 73,078 73,221 Cubic growth 115.978 2 73,056 73,206 Quadratic growth with a predictor LEP and covariates 206.739 38 70,991 71,176 Quadratic growth with a moderator LEP and covariates 198.851 36 70,987 71,184 Model specification X2 df AIC BIC Intercept-only 278.176 5 73,212 73,344 Linear growth 274.069 4 73,210 73,348 Quadratic growth 139.284 3 73,078 73,221 Cubic growth 115.978 2 73,056 73,206 Quadratic growth with a predictor LEP and covariates 206.739 38 70,991 71,176 Quadratic growth with a moderator LEP and covariates 198.851 36 70,987 71,184 Note: AIC = Akaike Information Criteria; BIC = Bayesian Information Criteria; LEP = Limited English proficiency. View Large Table 2. Model Specification Model specification X2 df AIC BIC Intercept-only 278.176 5 73,212 73,344 Linear growth 274.069 4 73,210 73,348 Quadratic growth 139.284 3 73,078 73,221 Cubic growth 115.978 2 73,056 73,206 Quadratic growth with a predictor LEP and covariates 206.739 38 70,991 71,176 Quadratic growth with a moderator LEP and covariates 198.851 36 70,987 71,184 Model specification X2 df AIC BIC Intercept-only 278.176 5 73,212 73,344 Linear growth 274.069 4 73,210 73,348 Quadratic growth 139.284 3 73,078 73,221 Cubic growth 115.978 2 73,056 73,206 Quadratic growth with a predictor LEP and covariates 206.739 38 70,991 71,176 Quadratic growth with a moderator LEP and covariates 198.851 36 70,987 71,184 Note: AIC = Akaike Information Criteria; BIC = Bayesian Information Criteria; LEP = Limited English proficiency. View Large Table 3 summarizes the estimated parameters from the quadratic growth model with LEP as a moderator for the growth trajectories while controlling for covariates. The intercept coefficient (β00 = 8.70, p < .01) represents the average level of depression symptoms when all other variables are constrained to be zero. In other words, depressive symptoms at baseline averaged 8.70 for those who had LEP, were average aged, male, have no education, were born outside of the United States, not married, and were interviewed in Spanish. As shown in Table 3, Mexican American older adults with LEP demonstrated greater depressive symptoms than those with EP at baseline when controlling for background characteristics (β01 = 1.10, p = .01). By switching the time coding, we examined the difference on the level of depression symptoms between older adults with LEP and those with EP at each time point. While the significant difference diminished from the second wave to fifth wave, respondents with LEP significantly increased in their depressive affect at the sixth wave than their counterparts. Table 3. Parameter Estimates for Quadratic Growth Model for Depressive Symptoms Parameter Est. SE p-value Intercept β00 8.70 .51 .00 Time β10 −1.07 .25 .00 Time2 β20 .29 .05 .00 LEP β01 1.10 .41 .01 LEP × Time β11 −.88 .32 .01 LEP × Time2 β21 .19 .07 .01 Age (centered) β02 .11 .02 .00 Female β03 2.27 .27 .00 Educational Attainment β04 −.11 .04 .01 Born outside the United States β05 .70 .27 .01 Interviewed in Spanish β06 −1.00 .33 .00 Marital Status β07 −.48 .27 .08 Parameter Est. SE p-value Intercept β00 8.70 .51 .00 Time β10 −1.07 .25 .00 Time2 β20 .29 .05 .00 LEP β01 1.10 .41 .01 LEP × Time β11 −.88 .32 .01 LEP × Time2 β21 .19 .07 .01 Age (centered) β02 .11 .02 .00 Female β03 2.27 .27 .00 Educational Attainment β04 −.11 .04 .01 Born outside the United States β05 .70 .27 .01 Interviewed in Spanish β06 −1.00 .33 .00 Marital Status β07 −.48 .27 .08 Note: LEP = Limited English proficiency. View Large Table 3. Parameter Estimates for Quadratic Growth Model for Depressive Symptoms Parameter Est. SE p-value Intercept β00 8.70 .51 .00 Time β10 −1.07 .25 .00 Time2 β20 .29 .05 .00 LEP β01 1.10 .41 .01 LEP × Time β11 −.88 .32 .01 LEP × Time2 β21 .19 .07 .01 Age (centered) β02 .11 .02 .00 Female β03 2.27 .27 .00 Educational Attainment β04 −.11 .04 .01 Born outside the United States β05 .70 .27 .01 Interviewed in Spanish β06 −1.00 .33 .00 Marital Status β07 −.48 .27 .08 Parameter Est. SE p-value Intercept β00 8.70 .51 .00 Time β10 −1.07 .25 .00 Time2 β20 .29 .05 .00 LEP β01 1.10 .41 .01 LEP × Time β11 −.88 .32 .01 LEP × Time2 β21 .19 .07 .01 Age (centered) β02 .11 .02 .00 Female β03 2.27 .27 .00 Educational Attainment β04 −.11 .04 .01 Born outside the United States β05 .70 .27 .01 Interviewed in Spanish β06 −1.00 .33 .00 Marital Status β07 −.48 .27 .08 Note: LEP = Limited English proficiency. View Large The LEP status associated with both time factors (i.e., β11, β21) was also statistically significant, suggesting that there is a differential effect of LEP on growth trajectories of depressive symptoms. As hypothesized and displayed in Figure 1, those with LEP revealed a steeper curve of depressive symptom trajectory over time than those with EP. Figure 1. View largeDownload slide Quadratic growth trajectories of depressive symptoms over six waves by English proficiency. CES-D = Center for Epidemiologic Studies Depression scale; EP = English proficiency; LEP = Limited English proficiency. Figure 1. View largeDownload slide Quadratic growth trajectories of depressive symptoms over six waves by English proficiency. CES-D = Center for Epidemiologic Studies Depression scale; EP = English proficiency; LEP = Limited English proficiency. Among covariates adjusted, several demonstrated independent contributions. As shown in Table 3, age was significantly related with depressive symptom level (β02 = .11, p < .01), indicating that every year older was associated with a .11 increase in the level of depressive symptoms. Females exhibited higher levels of depressive symptoms than males on average at the baseline (β03 = 2.27, p < .01). Years of education were also significantly related with depressive symptoms (β04 = −.11, p < .01), meaning that every year increase in education was associated with a .11 decrease in the level of depression symptoms. Respondents who were born outside the United States showed significantly more depressive symptoms (β05 = .70, p < .01) while respondents interviewed in Spanish showed less depressive symptoms (β06 = 1.00, p < .01) at the baseline. Although marital status was not statistically significant at p < .05 level, it showed a marginally significant relationship with depressive symptoms (β07 = −.48, p = .08), indicating that married people reported less on their depressive symptoms at the baseline. Discussion Given the reported vulnerability of Hispanic older adults with LEP (Kim, Aguado Loi, et al., 2011; Kim, Worley, et al., 2011; Krause & Goldenhar, 1992) and LEP’s potentially harmful effect on mental health (Mui et al., 2007; Takeuchi et al., 2007), we attempted to explore the effect of LEP on longitudinal trajectories of depressive symptoms among Mexican American older adults over a 14-year period. To our knowledge, this was the first study examining the longitudinal effect of LEP on mental health among older Hispanics in the United States. As hypothesized, LEP was associated not merely with greater depressive symptoms, but with an accelerated trajectory of depressive symptoms among older Mexican Americans over 14 years. This finding also supports what would be expected on the basis of the double jeopardy and cumulative disadvantage theories. Our findings suggest the need for assistance for those with LEP in order to reduce depressive symptoms, which may also help reduce disparities arising from language barriers. The most intriguing finding was the differential growth trajectories of depressive symptoms between with LEP and EP groups over 14 years, especially showing a steeper curve for those with LEP. Despite the nonsignificant differences in depressive symptom levels from 2 to 5 waves, the overall growth trajectory curve of depressive symptoms was significantly steeper for those with LEP compared to those with EP. Given that those with LEP tend to have difficulty navigating the proper health care system and understanding health information and treatment (Flores, 2006; DuBard & Gizlice, 2008; Kim, Aguado Loi, et al., 2011), it seems natural to speculate that poor access to and quality of health care that LEP individuals often experience in health care settings may lead to their poor mental health status both cross-sectionally and longitudinally. Given the availability of variables on health care access and use in the Hispanic EPESE data set, future mediation analyses focusing on health care access and use as potential mediators should be helpful in terms of determining where to target interventions for Hispanic older adults. Given the shortage of bilingual providers and the lack of affordability of mental health care due to the lack of mental health parity in Medicare that had persisted during the study period, LEP older adults are even less likely to access mental health care (Bauer, Chen, & Alegría, 2010; Sentell et al., 2007; Snowden, Masland, & Guerrero, 2007). What we did not know until now was LEP’s long-term impact on poor mental health status through more than ten years of follow-up. Future research should investigate mechanisms deriving long-term vulnerability of LEP older adults. It is worth pointing out opposing growth trajectories of depressive symptoms by host language proficiency found in a previous study (Nguyen et al., 2011). The main difference in the present study and Nguyen and colleagues’ study (2011) was the target population: adults aged 65 years and older versus adolescents aged 12–23 years. Nguyen and colleagues (2011) reported that becoming fluent in the mainstream language (i.e., English or French) was associated with a risk for steeper increases in depressive symptoms among adolescents from immigrant backgrounds in Canada, which was a completely opposite direction observed in the present study. While Nguyen and colleagues (2011) speculated that becoming fluent in the mainstream language often accompanied with a loss of heritage language might affect the parent-child cohesion and closeness negatively, this was conflicting to many other studies including the current study conceptualizing LEP as a marker for less acculturation and therefore as a source of stress. The findings from the current study suggest the need that LEP’s effect should be investigated further across the life span. Findings from this study provide implications for research, clinical practice, and policy pertaining geriatric mental health disparities. Mental health professionals and clinicians working with older Mexican Americans with LEP should be aware of the risk that they tend to have higher levels of depressive symptoms in general, as well as long-term disadvantage to mental health status and high stress relating to communicating with others. It is worth noting regarding clinical relevance that mean depressive symptom scores for both LEP and EP groups were well below the level for probable depression. However, given that previous studies reported standard clinical cut-offs of the CES-D may not be good indicators of depression and significantly higher CES-D scores were found among Hispanic older adults than those of other racial/ethnic elderly groups (e.g., Guarnaccia, Good, & Kleinman, 1990; Kim, Chiriboga, & Jang, 2009), clinicians and researchers should be cautious when interpreting the CES-D scores. In addition, clinicians and researchers should recognize that the CES-D results may need to be interpreted differently cross-culturally. Direct interventions to increase English skills may not be effective for this geriatric minority population. Instead, increasing support from informal and formal network that can build a system to help LEP older adults should be needed. Policy-makers need to focus on educating and training health/mental health care providers in order to increase bilingual workforce, especially in health and mental health professional shortage areas, which may be an effective step for policy intervention. For example, student training fellowships that the Health Resources and Services Administration (HRSA) has awarded fellowships to multiple health professions institutions and students from underrepresented minority groups in order to enhance the workforce diversity of health care professions such as physicians, social workers, and psychologists (Council on Graduate Medical Education, 2016). Policy-makers can also consider developing formal programs to build or invigorate a formal support system for LEP patients. These supports might focus on day care, respite care, meals on wheels and similar home and community-based services, but would require Spanish-language proficiency on the part of the providers. Also, providers can engage someone from informal support system identified by the LEP patient as a translator to help to overcome LEP barriers during treatment. As an extension of treatment, providers can use someone from informal network to help motivate the patient to engage in health promotion, to do homework in between treatment sessions, to take the medications, and to keep the LEP patient coming back to therapy. The present study is not without limitations. First and foremost, given that this study solely focused on Mexican Americans, generalizability is limited. Future research should be extended to other Hispanic subgroups as well as other racial/ethnic minority populations. Second, there may be the heterogeneity issue even within the Mexican American elderly sample. Even if some potentially important covariates were controlled in our analyses, subgroup differences might have existed. Third, like many longitudinal studies, attrition over time may have affected the findings. Fourth, we were not able to control potentially stressful life events such as the death of a spouse that older adults often experience in later life. Lastly, given that measurement equivalence of our outcome measure (i.e., the CES-D) by EP has not yet been established, the CES-D may not have functioned equally across linguistic groups. Given that prior research suggests evidence of measurement bias of a psychological distress measure (K6) when language use is considered (Kim, Decoster, Bryant, & Ford, 2016), further research should test the potential bias of the CES-D by EP. In conclusion, the current longitudinal analysis revealed LEP as a risk factor for an elevated depressive symptom trajectory over a 14-year period in the Hispanic elderly population. Supporting the double jeopardy theory was that those with LEP exhibited greater depression at all waves than did those with EP. Supporting the cumulative disadvantage theory was that those with LEP had a steeper curve of depressive symptom trajectory than those with EP over time. What was particularly noteworthy in the present study was the cumulative disadvantage of LEP to mental health that Mexican American older adults had experienced over 14 years. Overall, the findings emphasize the need to assist the informal support system of Mexican American older adults with LEP, but the formal system may help reduce depressive symptoms arising from language barriers. Focusing on those experiencing language barriers may assist with the goal of improving the mental health status of vulnerable elderly populations. Supplementary Data Supplementary data are available at The Gerontologist online. Funding The Hispanic EPESE was supported by the National Institute on Aging (R01AG10939, PI: Markides, K.). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Conflict of Interest None reported. References Abe-Kim , J. , Takeuchi , D. T. , Hong , S. , Zane , N. , Sue , S. , Spencer , M. S. , & Alegría , M . ( 2007 ). Use of mental health-related services among immigrant and US-born Asian Americans: Results from the National Latino and Asian American Study . American Journal of Public Health , 97 , 91 – 98 . doi: 10.2105/AJPH.2006.098541 Google Scholar CrossRef Search ADS PubMed Akhtar-Danesh , N. , & Landeen , J . ( 2007 ). Relation between depression and sociodemographic factors . International Journal of Mental Health Systems , 1 , 4 . doi: 10.1186/1752-4458-1-4 Google Scholar CrossRef Search ADS PubMed Angel , R. J. , Frisco , M. , Angel , J. L. , & Chiriboga , D. A . ( 2003 ). Financial strain and health among elderly Mexican-origin individuals . Journal of Health and Social Behavior , 44 , 536 – 551 . doi:10.2307/1519798 Google Scholar CrossRef Search ADS PubMed Bauer , A. M. , Chen , C. N. , & Alegría , M . ( 2010 ). English language proficiency and mental health service use among Latino and Asian Americans with mental disorders . Medical Care , 48 , 1097 – 1104 . doi: 10.1097/MLR.0b013e3181f80749 Google Scholar CrossRef Search ADS PubMed Bebbington , P. E. , Dunn , G. , Jenkins , R. , Lewis , G. , Brugha , T. , Farrell , M. , & Meltzer , H . ( 1998 ). The influence of age and sex on the prevalence of depressive conditions: Report from the national survey of psychiatric morbidity . Psychological Medicine , 28 , 9 – 19 . doi:10.1017/s0033291797006077 Google Scholar CrossRef Search ADS PubMed Berry , J. W . ( 1997 ). Immigration, acculturation, and adaptation . Applied Psychology: An International Review , 46 , 5 – 68 . doi:10.1111/j.1464-0597.1997.tb01087.x Blazer , D. , Burchett , B. , Service , C. , & George , L. K . ( 1991 ). The association of age and depression among the elderly: An epidemiologic exploration . Journal of Gerontology , 46 , M210 – M215 . doi:10.1093/geronj/46.6.m210 Google Scholar CrossRef Search ADS PubMed Carreon , D. , & Noymer , A . ( 2011 ). Health-related quality of life in older adults: Testing the double jeopardy hypothesis . Journal of Aging Studies , 25 , 371 – 379 . doi:10.1016/j.jaging.2011.01.004 Google Scholar CrossRef Search ADS Cole , M. G. , & Dendukuri , N . ( 2003 ). Risk factors for depression among elderly community subjects: A systematic review and meta-analysis . The American Journal of Psychiatry , 160 , 1147 – 1156 . doi: 10.1176/appi.ajp.160.6.1147 Google Scholar CrossRef Search ADS PubMed Constantine , M. G. , Okazaki , S. , & Utsey , S. O . ( 2004 ). Self-concealment, social self-efficacy, acculturative stress, and depression in African, Asian, and Latin American international college students . The American Journal of Orthopsychiatry , 74 , 230 – 241 . doi: 10.1037/0002-9432.74.3.230 Google Scholar CrossRef Search ADS PubMed Council on Graduate Medical Education . ( 2016 ). Supporting Diversity in the Health Professions. Retrieved from http://www.hrsa.gov/advisorycommittees/bhpradvisory/cogme/Publications/diversityresourcepaper.pdf. Dannefer , D . ( 2003 ). Cumulative advantage/disadvantage and the life course: Cross-fertilizing age and social science theory . The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 58 , S327 – S337 . doi:10.1093/geronb/58.6.s327 Google Scholar CrossRef Search ADS Derose , K. P. , Escarce , J. J. , & Lurie , N . ( 2007 ). Immigrants and health care: Sources of vulnerability . Health Affairs (Project Hope) , 26 , 1258 – 1268 . doi: 10.1377/hlthaff.26.5.1258 Google Scholar CrossRef Search ADS PubMed Dowd , J. J. , & Bengtson , V. L . ( 1978 ). Aging in minority populations. An examination of the double jeopardy hypothesis . Journal of Gerontology , 33 , 427 – 436 . doi:10.1093/geronj/33.3.427 Google Scholar CrossRef Search ADS PubMed DuBard , C. A. , & Gizlice , Z . ( 2008 ). Language spoken and differences in health status, access to care, and receipt of preventive services among US Hispanics . American Journal of Public Health , 98 , 2021 – 2028 . doi: 10.2105/AJPH.2007.119008 Google Scholar CrossRef Search ADS PubMed Enders , C. K. , & Bandalos , D. L . ( 2001 ). The relative performance of full information maximum likelihood estimation for missing data in structural equation models . Structural Equation Modeling , 8 , 430 – 457 . doi:10.1207/s15328007sem0803_5 Google Scholar CrossRef Search ADS Ferraro , K. F. , & Farmer , M. M . ( 1996 ). Double jeopardy, aging as leveler, or persistent health inequality? A longitudinal analysis of white and black Americans . The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 51 , S319 – S328 . doi:10.1093/geronb/51b.6.s319 Google Scholar CrossRef Search ADS Flores , G . ( 2006 ). Language barriers to health care in the United States . The New England Journal of Medicine , 355 , 229 – 231 . doi: 10.1056/NEJMp058316 Google Scholar CrossRef Search ADS PubMed Gee , G. C. , & Ponce , N . ( 2010 ). Associations between racial discrimination, limited English proficiency, and health-related quality of life among 6 Asian ethnic groups in California . American Journal of Public Health , 100 , 888 – 895 . doi: 10.2105/AJPH.2009.178012 Google Scholar CrossRef Search ADS PubMed Gonzalez-Barrera , A. , & Krogstad , J. M . ( 2015 ). What we know about illegal immigration from Mexico . Washington, DC: Pew Research Center, 20 . Guarnaccia , P. J. , Good , B. J. , & Kleinman , A . ( 1990 ). A critical review of epidemiological studies of Puerto Rican mental health . The American Journal of Psychiatry , 147 , 1449 – 1456 . doi: 10.1176/ajp.147.11.1449 Google Scholar CrossRef Search ADS PubMed Humes , K. , Jones , N. A. , & Ramirez , R. R . ( 2011 ). Overview of race and Hispanic origin, 2010 . Washington, DC: US Department of Commerce, Economics and Statistics Administration, US Census Bureau . Jang , Y. , Haley , W. E. , Small , B. J. , & Mortimer , J. A . ( 2002 ). The role of mastery and social resources in the associations between disability and depression in later life . The Gerontologist , 42 , 807 – 813 . doi:10.1093/geront/42.6.807 Google Scholar CrossRef Search ADS PubMed Jang , Y. , Yoon , H. , Park , N. S. , & Chiriboga , D. A . ( 2016 ). Health vulnerability of immigrants with limited english proficiency: A study of older Korean Americans . Journal of the American Geriatrics Society , 64 , 1498 – 1502 . doi: 10.1111/jgs.14199 Google Scholar CrossRef Search ADS PubMed Kandula , N. R. , Lauderdale , D. S. , & Baker , D. W . ( 2007 ). Differences in self-reported health among Asians, Latinos, and non-Hispanic whites: The role of language and nativity . Annals of Epidemiology , 17 , 191 – 198 . doi: 10.1016/j.annepidem.2006.10.005 Google Scholar CrossRef Search ADS PubMed Kim , G. , Aguado Loi , C. X. , Chiriboga , D. A. , Jang , Y. , Parmelee , P. , & Allen , R. S . ( 2011 ). Limited English proficiency as a barrier to mental health service use: A study of Latino and Asian immigrants with psychiatric disorders . Journal of Psychiatric Research , 45 , 104 – 110 . doi: 10.1016/j.jpsychires.2010.04.031 Google Scholar CrossRef Search ADS PubMed Kim , G. , Chiriboga , D. A. , & Jang , Y . ( 2009 ). Cultural equivalence in depressive symptoms in older white, black, and Mexican-American adults . Journal of the American Geriatrics Society , 57 , 790 – 796 . doi:10.1111/j.1532-5415.2009.02188.x Google Scholar CrossRef Search ADS PubMed Kim , G. , DeCoster , J. , Bryant , A. N. , & Ford , K. L . ( 2016 ). Measurement equivalence of the k6 scale: The effects of race/ethnicity and language . Assessment , 23 , 758 – 768 . doi: 10.1177/1073191115599639 Google Scholar CrossRef Search ADS PubMed Kim , M. , Kwok , O. , Yoon , M. , Willson , V. , & Lai , M. H . ( 2016 ). Specification search for identifying the correct mean trajectory in polynomial latent growth models . The Journal of Experimental Education , 84 , 307 – 329 . doi:10.1080/00220973.2014.984831 Google Scholar CrossRef Search ADS Kim , G. , Worley , C. B. , Allen , R. S. , Vinson , L. , Crowther , M. R. , Parmelee , P. , & Chiriboga , D. A . ( 2011 ). Vulnerability of older Latino and Asian immigrants with limited English proficiency . Journal of the American Geriatrics Society , 59 , 1246 – 1252 . doi: 10.1111/j.1532-5415.2011.03483.x Google Scholar CrossRef Search ADS PubMed Krause , N. , & Goldenhar , L. M . ( 1992 ). Acculturation and psychological distress in three groups of elderly Hispanics . Journal of Gerontology , 47 , S279 – S288 . doi:10.1093/geronj/47.6.s279 Google Scholar CrossRef Search ADS PubMed Krogstad , J.M. & Gonzalez-Barrera , A . ( 2015 ). A majority of English-speaking hispanics in the U.S. are bilingual . Pew Research Center . Retrieved from http://www.pewresearch.org/fact-tank/2015/03/24/a-majority-of-english-speaking-hispanics-in-the-u-s-are-bilingual/. Krogstad , J.M. , Stepler , R. , & Lopez , M.H . ( 2016 ). English Proficiency on the Rise Among Latinos . Pew Research Center Hispanic Trends . Retrieved from http://www.pewhispanic.org/2015/05/12/english-proficiency-on-the-rise-among-latinos/. Lin , J. G. , & Yi , J. K . ( 1997 ). Asian international student’s adjustment: Issues and program suggestions . College Student Journal , 31 , 473 – 479 . Little , R.J.A. , & Rubin , D. B . ( 2002 ). Statistical analysis with missing data ( 2nd ed .). Hoboken, NJ : Wiley . Google Scholar CrossRef Search ADS Mui , A. C. , Kang , S. Y. , Kang , D. , & Domanski , M. D . ( 2007 ). English language proficiency and health-related quality of life among Chinese and Korean immigrant elders . Health & Social Work , 32 , 119 – 127 . doi:10.1093/hsw/32.2.119 Google Scholar CrossRef Search ADS PubMed Muthen , L. K. , & Muthén , B. O . ( 2010 ). Mplus user’s guide, v 6.1 . Los Angeles, CA : Muthen and Muthen, UCLA . Nguyen , H. , Rawana , J. S. , & Flora , D. B . ( 2011 ). Risk and protective predictors of trajectories of depressive symptoms among adolescents from immigrant backgrounds . Journal of Youth and Adolescence , 40 , 1544 – 1558 . doi: 10.1007/s10964-011-9636-8 Google Scholar CrossRef Search ADS PubMed Nwadiora , E. , & McAdoo , H . ( 1996 ). Acculturative stress among Amerasian refugees: Gender and racial differences . Adolescence , 31 , 477 – 487 . Google Scholar PubMed Pandya, C., Batalova J., & McHugh M. (2011). Limited English proficient individuals in the United States: Number, share, growth, and linguistic diversity. Washington, DC: Migration Policy Institute . Ponce , N. A. , Hays , R. D. , & Cunningham , W. E . ( 2006 ). Linguistic disparities in health care access and health status among older adults . Journal of General Internal Medicine , 21 , 786 – 791 . doi: 10.1111/j.1525-1497.2006.00491.x Google Scholar CrossRef Search ADS PubMed Preacher , K. J. , & Hayes , A. F . ( 2004 ). SPSS and SAS procedures for estimating indirect effects in simple mediation models . Behavior Research Methods, Instruments, & Computers: A Journal of the Psychonomic Society, Inc , 36 , 717 – 731 . Google Scholar CrossRef Search ADS PubMed Radloff , L. S . ( 1977 ). The CES-D scale: A self-report depression scale for research in the general population . Applied Psychological Measurement , 1 , 385 – 401 . doi:10.1177/014662167700100306 Google Scholar CrossRef Search ADS Sadule-Rios , N . ( 2012 ). A review of the literature about depression in late life among Hispanics in the United States . Issues in Mental Health Nursing , 33 , 458 – 468 . doi: 10.3109/01612840.2012.675415 Google Scholar CrossRef Search ADS PubMed Sentell , T. , & Braun , K. L . ( 2012 ). Low health literacy, limited English proficiency, and health status in Asians, Latinos, and other racial/ethnic groups in California . Journal of Health Communication , 17 ( Suppl 3 ), 82 – 99 . doi: 10.1080/10810730.2012.712621 Google Scholar CrossRef Search ADS PubMed Sentell , T. , Shumway , M. , & Snowden , L . ( 2007 ). Access to mental health treatment by English language proficiency and race/ethnicity . Journal of General Internal Medicine , 22 ( Suppl 2 ), 289 – 293 . doi: 10.1007/s11606-007-0345-7 Google Scholar CrossRef Search ADS PubMed Snowden , L. R. , Masland , M. , & Guerrero , R . ( 2007 ). Federal civil rights policy and mental health treatment access for persons with limited English proficiency . The American Psychologist , 62 , 109 – 117 . doi: 10.1037/0003-066X.62.2.109 Google Scholar CrossRef Search ADS PubMed Takeuchi , D. T. , Zane , N. , Hong , S. , Chae , D. H. , Gong , F. , Gee , G. C.,… Alegría , M . ( 2007 ). Immigration-related factors and mental disorders among Asian Americans . American Journal of Public Health , 97 , 84 – 90 . doi: 10.2105/AJPH.2006.088401 Google Scholar CrossRef Search ADS PubMed U.S. Census Bureau . Population 60 years and over in the United States: 2010 [on- line] . Retrieved March 6, 2017, from https://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?src=bkmk. U.S. Census Bureau . Selected social characteristics of the United States: 2005 [on-line] . Retrieved March 6, 2017, from http://factfinder.census.gov/servlet/ADPTable?_bm=y&geo_id=01000US&-qr_name=ACS_2005_EST_G00_DP2&-ds_name=&-redoLog=false&-format-. Zarit , S. H. , & Zarit , J. M . ( 1998 ). Mental disorders in older adults: Fundamentals of assessment and treatment . New York : Guilford Press . Zhang , W. , Hong , S. , Takeuchi , D. T. , & Mossakowski , K. N . ( 2012 ). Limited English proficiency and psychological distress among Latinos and Asian Americans . Social Science & Medicine (1982) , 75 , 1006 – 1014 . doi: 10.1016/j.socscimed.2012.05.012 Google Scholar CrossRef Search ADS PubMed Zong , J. , & Batalova , J . ( 2015 ). The limited English proficient population in the United States[on-line] . Retrieved March 6, 2017, from http://www.migrationpolicy.org/Article/Limited-English-Proficient-Population-United-States/. © The Author(s) 2018. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Gerontologist Oxford University Press

Limited English Proficiency and Trajectories of Depressive Symptoms Among Mexican American Older Adults

Loading next page...
 
/lp/ou_press/limited-english-proficiency-and-trajectories-of-depressive-symptoms-ws3FmNPoYM
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
0016-9013
eISSN
1758-5341
D.O.I.
10.1093/geront/gny032
Publisher site
See Article on Publisher Site

Abstract

Abstract Background and Objectives This study examined the effect of limited English proficiency (LEP) on trajectories of depressive symptoms among Mexican American older adults in the United States. Research Design and Methods The sample was drawn from Waves 1 to 6 (1993–2007) of the Hispanic Established Population for Epidemiological Studies of the Elderly (H-EPESE). A total of 2,945 Mexican American older adults were included in the analyses. A latent growth curve modeling was conducted. Results After adjusting for covariates, results show that Mexican American older adults with LEP had higher levels of depressive symptoms than those with English proficiency (EP) at baseline as well as over a 14-year period. Differential trajectories were observed between those with LEP and EP over time, indicating that those with LEP had a significantly steeper curve of depressive symptom trajectories over time than those with EP. Discussion and Implications These results suggest that LEP is a risk factor not merely for greater depressive symptomatology at each time point, but for an accelerated trajectory of depressive symptoms over time among Mexican American older adults. Overall, the findings emphasize the need to assist not only the informal support system of Mexican American older adults with LEP, but also the formal system. Depressive symptoms, Depression trajectories, Latent growth curve modeling, Limited English proficiency, Mexican American older adults The Hispanic population—defined as “people of Cuban, Mexican, Puerto Rican, South or Central American or other Spanish culture or origin regardless of race” (Humes, Jones, & Ramirez, 2011)—is not only the largest racial/ethnic minority group in the United States, but the population with the largest proportion of persons with limited English proficiency (LEP) (Zong & Batalova, 2015). According to the U.S. Census Bureau, 25.3 millions of Americans, or 8.6% of the U.S. population, speak English less than “very well” and are classified as having LEP (U.S. Census Bureau, 2005; Pandya, McHugh, & Batalova, 2011). The proportion of persons with LEP is estimated at 8.3% of immigrants aged 60 years or older in the United States (U.S. Census Bureau, 2010). The highest proportions of LEP are found in the Hispanic population (Zong & Batalova, 2015), and it is of particular relevance to this paper that across all racial/ethnic groups, older Hispanics are the group most likely to report LEP (Krogstad & Gonzalez-Barrera, 2015, Krogstad, Stepler, & Lopez, 2016). LEP’s impact is pervasive for immigrants living in the United States and other English-speaking countries. First, LEP has been reported as a risk factor for poor health conditions (Derose, Escarce, & Lurie, 2007; Dubard & Glizlice, 2008; Gee & Ponce, 2010; Jang, Yoon, Park, & Chiriboga, 2016; Kandura, Lauderdale, & Baker, 2007; Kim, Aguado Loi, et al., 2011; Kim, Worley, et al., 2011; Mui, Kang, Kang, & Domanski, 2007; Ponce, Hays, & Cunningham, 2006; Sentell & Braun, 2012). Individuals with LEP tend to have poorer physical health status (e.g., poorer self-rated health, poorer health-related quality of life), as well as poorer mental health status (e.g., greater depressive symptoms, greater diagnosis for major depression and anxiety, and greater psychological distress) compared to those who are proficient in English or native English speakers (Derose et al., 2007; Dubard & Glizlice, 2008; Gee & Ponce, 2010; Kandula et al., 2007; Kim, Worley, et al., 2011; Mui et al., 2007; Ponce et al., 2006; Sentell & Braun, 2012). Second, literature also suggests that LEP acts as a significant barrier to access to and utilization of physical and mental health care (Abe-Kim et al., 2007; Kim, Aguado Loi, et al., 2011; Kim, Worley, et al., 2011; Ponce et al., 2006; Sentell, Shumway, & Snowden, 2007). This is mainly due to health care settings’ large dependence on verbal communications required for diagnosis and treatment of illness (Dubard & Gizlice, 2008; Kim, Worley, et al., 2011; Sentell et al., 2007). Lastly and more related to the present study, LEP is frequently reported to be associated with greater depression as language barriers due to LEP could be a significant source of stress, which can exacerbate the acculturative stress among immigrants (Constantine, Okazaki, & Utsey, 2004; Lin & Yi, 1997; Nwadiora & McAdoo, 1996; Sadule-Rios, 2012). For instance, Krause & Goldenhar (1992) reported that older Mexican Americans with LEP tended to show higher risk for depressive symptoms, as compared to those with greater English proficiency (EP). Researchers suggest that LEP itself could be an important social status variable that can directly influence depressive symptoms (Berry, 1997; Zhang, Hong, Takeuchi, & Mossakowski, 2012). Specifically, the LEP status can create inconveniences and discomforts in daily lives in English-speaking countries, especially when communicating with and developing social networks with English-speaking individuals, as well as accessing useful information, which may lead to increased levels of stress (Zhang et al., 2012). Given LEP’s relation to depressive symptoms, researchers have further questioned LEP’s effects on longitudinal courses of depressive symptoms. For example, Nguyen, Rawana, and Flora (2011) examined the effect of immigration factors (i.e., host language proficiency and years since immigration) on the trajectories of depressive symptoms among immigrant adolescents in Canada. They discovered that proficiency in host language significantly influenced on the quadratic trajectories of depressive symptoms; adolescents who were proficient in host language showed steeper increases in depressive symptoms until they diverge at about mid-adolescence and exhibited higher rates of depressive symptoms in early adulthood, as compared to their counterparts with no proficiency in English. To our knowledge, there is no literature on the effect of EP on depressive symptoms among older adults from a longitudinal perspective; therefore, we will attempt to explore the longitudinal effect of LEP among older adults in the present study. The present study was guided by two theories to conceptualize the potentially cumulative disadvantageous effects of LEP: the cumulative disadvantage theory (Dannefer, 2003) and the double jeopardy theory (Dowd & Bengtson, 1978). The cumulative disadvantage theory describes how early advantage or disadvantage is critical to how cohorts become differentiated over time (Dannefer, 2003). Many health disparities researchers employed the cumulative disadvantage theory to examine longitudinal effects of risk factors on minority older adults’ health (e.g., Carreon & Noymer, 2011). In the case of the present sample, given the generally very low socioeconomic status of LEP Hispanic older adults, the majority of them have faced disadvantages across their live courses. The double jeopardy theory (Dowd & Bengtson, 1978) has also been widely used in minority aging research. According to the double jeopardy theory (Dowd & Bengtson, 1978), racial/ethnic minority older adults may suffer from the double disadvantage to health due to the interactive effects of age and race/ethnicity. In the case of the present sample, Mexican Americans have in general faced lives with greater disadvantages than those faced by non-Hispanic White Americans. As Mexican Americans reach later life and begin to experience many physical and social and behavioral problems associated with the aging process, this is yet another jeopardy, which supports the double jeopardy framework. Taken together the abovementioned theoretical frameworks, we view that racial/ethnic minority older adults with LEP may be in a disadvantaged position and this disadvantaged effect influences cumulatively and longitudinally compared to those who are proficient in English. The purpose of the present study is to assess the effect of LEP on trajectories of depressive symptoms among Mexican American older adults in the United States. Our overarching research questions are whether LEP is harmful for mental health and whether the cumulative disadvantage arising from the lack of language skills influences mental health among minority older adults. Based on prior literature, we hypothesized that (1) Mexican American older adults with LEP would have higher levels of depressive symptoms than those with EP at baseline as well as over time, and (2) Mexican American older adults with LEP would have a steeper curve of depressive symptom trajectories over time compared with those with EP. Methods Sample The sample was drawn from the first six waves of the Hispanic Established Population for Epidemiological Studies of the Elderly (Hispanic EPESE). The Hispanic EPESE is a multistage stratified probability study of community-dwelling Mexican Americans aged 65 years and over who reside in the five southwestern states that contain the majority of Mexican Americans: Texas, California, New Mexico, Colorado, and Arizona. The first wave of interviews (n = 3,050) was conducted between September 1993 and June 1994. Respondents were interviewed at a place of their choosing including their homes by trained interviewees and the interviews were conducted in English or Spanish depending upon respondents’ preference. Follow-up interviews were then conducted approximately every 2 to 3 years, with a supplemental sample of 902 respondents from the same cohorts as the original sample added at Wave 5 (2004–2005). All of the new respondents were aged 75 years or older, were identified using the same multistage sampling procedures and again were interviewed at a place of their choosing. For the present analyses, among a total of 3,952 respondents from Waves 1 to 6 (1993–2007), 2,945 Mexican Americans aged 65 years and older, who have a depression score at least at one time point for the first four waves of data were included in further analyses. As mentioned above, given that there was a large, new group of respondents (n = 902) added at Wave 5, we removed those newly added cases for our analysis but retained the original group of respondents only in the current study. Measures Depressive Symptoms Depressive symptoms were measured with the Center for Epidemiologic Studies Depression scale (CES-D), a 20-item self-administered questionnaire (Radloff, 1977). Respondents were asked to assess the frequency of depressive symptoms experienced during the past week, based on a 4-point scale with categories in the subsequent order: rarely or none of the time (0), some or a little of the time (1), much of the time (2), and most or all of the time (3). The total scores for 20 items potentially ranged from 0 to 60, with higher scores indicating more depressive symptoms. A score of 16 or higher is generally deemed as a sign of probable clinical depression (Zarit & Zarit, 1998). An internal consistency of the CES-D was reported as moderately high (Cronbach’s α = .88) in this Hispanic population. LEP Respondents were asked to respond how well they speak English using the following rating scale: 1 (not at all), 2 (not too well), 3 (pretty well), and 4 (very well). Based on previous studies (Kim, Aguado Loi, et al., 2011; Kim, Worley, et al., 2011; Ponce et al., 2006), LEP was determined if the responses were not at all or not too well and coded as 1 for further analysis (1 for LEP; 0 for non-LEP). Covariates Respondents’ sociodemographic background characteristics (i.e., age, gender [1 for female; 0 for male], marital status [1 for married; 0 for else including separated, divorced, widowed, and never married], educational attainment), nativity (1 for U.S. born; 0 for non-U.S. born), and interview language (1 for Spanish; 0 for English) were used as covariates. These variables were selected as previous studies reported significant correlations with depressive symptoms (Akhtar-Danesh & Landeen, 2007; Angel, Frisco, Angel, & Chiriboga, 2003; Bebbington et al., 1998; Blazer, Burchett, Service, & George, 1991; Cole & Dendukuri, 2003; Jang, Haley, Small, & Mortimer, 2002). Data Analysis The present study used the Full Information Maximum Likelihood estimation method (FIML) to include all available cases rather than deleting those with missing data as FIML estimation has shown superior performance for estimating the parameters compared to other traditional missing data analysis (e.g., listwise deletion, pairwise deletion) (Enders & Bandalos, 2001; Little & Rubin, 2002). To examine the longitudinal trajectories of depressive symptoms, we employed a latent growth curve model (Preacher & Hayes, 2004), which takes into account the between-individual difference as well as within-individual variance. In latent growth models, each growth related term is treated as a latent construct having the observed measures as indicators (see an example in Supplementary Figure S1). Mplus 7.1 (Muthen & Muthén, 2010) with ESTIMATOR=MLR was used to analyze the latent growth curve models by handling the missing data as well as adjusting for the standard errors given the non-normal data. To search for the best fitting growth trajectory to our data, we first analyzed an unconditional growth model, which has no predictors or covariates in the model. To search for the optimal growth trajectory, we then fit a series of four unconditional growth models: including intercept-only, linear growth, quadratic growth, and cubic growth. For the within-subject error variance structure, all possible variance-covariance components were freely estimated with no constraints (i.e., UN structure) given that UN structure is best suited in the search for the optimal growth trajectory (Kim, Kwok, Yoon, Willson, & Lai, 2016). The Likelihood Ratio Test, the Akaike Information Criteria (AIC), and the Bayesian Information Criteria (BIC) were then used to compare the competing models. If there is no significant improvement in LRT, the simpler model is considered to be a better fitting model over the more complex model. Likewise, a smaller AIC and BIC represents a better fitting model over the competing model. Results Background Characteristics of the Sample As summarized in Table 1, age at baseline ranged from 65 to 107 with an average age of 72.83 years (SD = 6.57 years). Over half were female and married. The majority had attained education less than high school (less than 11 years) with an average of 4.85 years of education. Over half (55.9%) reported that they were born in the United States, whereas 78% were interviewed in Spanish (22% in English). Regarding our variable of interests, over half (60.9%) appeared to have LEP by assessing their English-speaking ability as not at all or not too well. All demographic characteristics were significantly different between LEP and EP groups (all ps < .001). The missing data rate for longitudinal depressive symptom variables ranged from 4% for Wave 1 and 72% for Wave 6 (see Table 1). Table 1. Descriptive Statistics for the Demographic Characteristics (n = 2,945) Variable M (SD) or % Total sample LEP group EP group Age (years)** 72.83 (6.57) 73.22 (6.92) 72.28 (5.95) Female** 57.9% 60.3% 53.7% Educational Attainment**  Less than high school (0–11 years) 90.1% 98.3% 77.1%  High school (12 years) 6.8% 1.2% 15.8%  Less than college (13–15 years) 1.6% .1% 3.8%  College (16 years) 1.0% .2% 2.3%  More than college (>17 years) .5% .2% 1.0% Nativity**  U.S. Born 55.9% 41.1% 80.3%  Non-U.S. Born 44.0% 58.9% 19.7% Marital status**  Married 55.8% 52.6% 60.9%  Separated 3.0% 3.6% 1.9%  Divorced 4.7% 3.8% 6.3%  Widowed 30.8% 34.0% 25.9%  Never married 5.6% 6.0% 5.0% Interview language**  English 22.0% 8.2% 43.9%  Spanish 78.0% 91.8% 56.1% Depressive symptoms (CES-D)  Wave 1 (n = 2,823)** 9.94 (9.59) 10.60 (9.73) 8.75 (9.07)  Wave 2 (n = 2,241)* 7.10 (8.41) 7.47 (8.66) 6.58 (7.95)  Wave 3 (n = 1,813)** 8.44 (9.02) 8.93 (9.50) 7.56 (8.07)  Wave 4 (n = 1,587) 7.22 (7.78) 7.33 (7.87) 7.09 (7.72)  Wave 5 (n = 1,060)* 9.58 (8.94) 10.06 (9.45) 8.82 (8.07)  Wave 6 (n = 812) 9.35 (8.94) 9.94 (8.91) 8.30 (8.91) Variable M (SD) or % Total sample LEP group EP group Age (years)** 72.83 (6.57) 73.22 (6.92) 72.28 (5.95) Female** 57.9% 60.3% 53.7% Educational Attainment**  Less than high school (0–11 years) 90.1% 98.3% 77.1%  High school (12 years) 6.8% 1.2% 15.8%  Less than college (13–15 years) 1.6% .1% 3.8%  College (16 years) 1.0% .2% 2.3%  More than college (>17 years) .5% .2% 1.0% Nativity**  U.S. Born 55.9% 41.1% 80.3%  Non-U.S. Born 44.0% 58.9% 19.7% Marital status**  Married 55.8% 52.6% 60.9%  Separated 3.0% 3.6% 1.9%  Divorced 4.7% 3.8% 6.3%  Widowed 30.8% 34.0% 25.9%  Never married 5.6% 6.0% 5.0% Interview language**  English 22.0% 8.2% 43.9%  Spanish 78.0% 91.8% 56.1% Depressive symptoms (CES-D)  Wave 1 (n = 2,823)** 9.94 (9.59) 10.60 (9.73) 8.75 (9.07)  Wave 2 (n = 2,241)* 7.10 (8.41) 7.47 (8.66) 6.58 (7.95)  Wave 3 (n = 1,813)** 8.44 (9.02) 8.93 (9.50) 7.56 (8.07)  Wave 4 (n = 1,587) 7.22 (7.78) 7.33 (7.87) 7.09 (7.72)  Wave 5 (n = 1,060)* 9.58 (8.94) 10.06 (9.45) 8.82 (8.07)  Wave 6 (n = 812) 9.35 (8.94) 9.94 (8.91) 8.30 (8.91) Note: **p < .001, *p < .05 based on results from a chi-square test or t tests comparing between LEP and EP groups. CES-D = Center for Epidemiologic Studies Depression scale; EP = English proficiency; LEP = Limited English proficiency. View Large Table 1. Descriptive Statistics for the Demographic Characteristics (n = 2,945) Variable M (SD) or % Total sample LEP group EP group Age (years)** 72.83 (6.57) 73.22 (6.92) 72.28 (5.95) Female** 57.9% 60.3% 53.7% Educational Attainment**  Less than high school (0–11 years) 90.1% 98.3% 77.1%  High school (12 years) 6.8% 1.2% 15.8%  Less than college (13–15 years) 1.6% .1% 3.8%  College (16 years) 1.0% .2% 2.3%  More than college (>17 years) .5% .2% 1.0% Nativity**  U.S. Born 55.9% 41.1% 80.3%  Non-U.S. Born 44.0% 58.9% 19.7% Marital status**  Married 55.8% 52.6% 60.9%  Separated 3.0% 3.6% 1.9%  Divorced 4.7% 3.8% 6.3%  Widowed 30.8% 34.0% 25.9%  Never married 5.6% 6.0% 5.0% Interview language**  English 22.0% 8.2% 43.9%  Spanish 78.0% 91.8% 56.1% Depressive symptoms (CES-D)  Wave 1 (n = 2,823)** 9.94 (9.59) 10.60 (9.73) 8.75 (9.07)  Wave 2 (n = 2,241)* 7.10 (8.41) 7.47 (8.66) 6.58 (7.95)  Wave 3 (n = 1,813)** 8.44 (9.02) 8.93 (9.50) 7.56 (8.07)  Wave 4 (n = 1,587) 7.22 (7.78) 7.33 (7.87) 7.09 (7.72)  Wave 5 (n = 1,060)* 9.58 (8.94) 10.06 (9.45) 8.82 (8.07)  Wave 6 (n = 812) 9.35 (8.94) 9.94 (8.91) 8.30 (8.91) Variable M (SD) or % Total sample LEP group EP group Age (years)** 72.83 (6.57) 73.22 (6.92) 72.28 (5.95) Female** 57.9% 60.3% 53.7% Educational Attainment**  Less than high school (0–11 years) 90.1% 98.3% 77.1%  High school (12 years) 6.8% 1.2% 15.8%  Less than college (13–15 years) 1.6% .1% 3.8%  College (16 years) 1.0% .2% 2.3%  More than college (>17 years) .5% .2% 1.0% Nativity**  U.S. Born 55.9% 41.1% 80.3%  Non-U.S. Born 44.0% 58.9% 19.7% Marital status**  Married 55.8% 52.6% 60.9%  Separated 3.0% 3.6% 1.9%  Divorced 4.7% 3.8% 6.3%  Widowed 30.8% 34.0% 25.9%  Never married 5.6% 6.0% 5.0% Interview language**  English 22.0% 8.2% 43.9%  Spanish 78.0% 91.8% 56.1% Depressive symptoms (CES-D)  Wave 1 (n = 2,823)** 9.94 (9.59) 10.60 (9.73) 8.75 (9.07)  Wave 2 (n = 2,241)* 7.10 (8.41) 7.47 (8.66) 6.58 (7.95)  Wave 3 (n = 1,813)** 8.44 (9.02) 8.93 (9.50) 7.56 (8.07)  Wave 4 (n = 1,587) 7.22 (7.78) 7.33 (7.87) 7.09 (7.72)  Wave 5 (n = 1,060)* 9.58 (8.94) 10.06 (9.45) 8.82 (8.07)  Wave 6 (n = 812) 9.35 (8.94) 9.94 (8.91) 8.30 (8.91) Note: **p < .001, *p < .05 based on results from a chi-square test or t tests comparing between LEP and EP groups. CES-D = Center for Epidemiologic Studies Depression scale; EP = English proficiency; LEP = Limited English proficiency. View Large Results From Growth Curve Modeling Analysis Depressive symptom scores over six time points are displayed in the bottom rows of Table 1. The mean depressive symptom score at each wave shows that there is some fluctuation on depressive symptom scores over time. These changes in depressive symptoms over time provide justification for searching for the appropriate growth trajectory. As displayed in Table 2, different growth models were compared in order to find the best fitting model. Based on model fit statistics presented in Table 2, both quadratic and cubic growth models fit the best to the data to represent the growth trajectory of depressive symptoms across six waves. We have, therefore, plotted the two competing growth models (i.e., quadratic vs cubic growth model) to compare the overall trajectory and have found that they have almost equivalent growth patterns across times. Based on this equivalence, we selected the more parsimonious quadratic growth trajectory model for further analyses. After comparing the model having the interaction effect between the time and LEP to the model without the interaction, we selected the interaction model as our final model and interpreted the results based on the corresponding model. Table 2. Model Specification Model specification X2 df AIC BIC Intercept-only 278.176 5 73,212 73,344 Linear growth 274.069 4 73,210 73,348 Quadratic growth 139.284 3 73,078 73,221 Cubic growth 115.978 2 73,056 73,206 Quadratic growth with a predictor LEP and covariates 206.739 38 70,991 71,176 Quadratic growth with a moderator LEP and covariates 198.851 36 70,987 71,184 Model specification X2 df AIC BIC Intercept-only 278.176 5 73,212 73,344 Linear growth 274.069 4 73,210 73,348 Quadratic growth 139.284 3 73,078 73,221 Cubic growth 115.978 2 73,056 73,206 Quadratic growth with a predictor LEP and covariates 206.739 38 70,991 71,176 Quadratic growth with a moderator LEP and covariates 198.851 36 70,987 71,184 Note: AIC = Akaike Information Criteria; BIC = Bayesian Information Criteria; LEP = Limited English proficiency. View Large Table 2. Model Specification Model specification X2 df AIC BIC Intercept-only 278.176 5 73,212 73,344 Linear growth 274.069 4 73,210 73,348 Quadratic growth 139.284 3 73,078 73,221 Cubic growth 115.978 2 73,056 73,206 Quadratic growth with a predictor LEP and covariates 206.739 38 70,991 71,176 Quadratic growth with a moderator LEP and covariates 198.851 36 70,987 71,184 Model specification X2 df AIC BIC Intercept-only 278.176 5 73,212 73,344 Linear growth 274.069 4 73,210 73,348 Quadratic growth 139.284 3 73,078 73,221 Cubic growth 115.978 2 73,056 73,206 Quadratic growth with a predictor LEP and covariates 206.739 38 70,991 71,176 Quadratic growth with a moderator LEP and covariates 198.851 36 70,987 71,184 Note: AIC = Akaike Information Criteria; BIC = Bayesian Information Criteria; LEP = Limited English proficiency. View Large Table 3 summarizes the estimated parameters from the quadratic growth model with LEP as a moderator for the growth trajectories while controlling for covariates. The intercept coefficient (β00 = 8.70, p < .01) represents the average level of depression symptoms when all other variables are constrained to be zero. In other words, depressive symptoms at baseline averaged 8.70 for those who had LEP, were average aged, male, have no education, were born outside of the United States, not married, and were interviewed in Spanish. As shown in Table 3, Mexican American older adults with LEP demonstrated greater depressive symptoms than those with EP at baseline when controlling for background characteristics (β01 = 1.10, p = .01). By switching the time coding, we examined the difference on the level of depression symptoms between older adults with LEP and those with EP at each time point. While the significant difference diminished from the second wave to fifth wave, respondents with LEP significantly increased in their depressive affect at the sixth wave than their counterparts. Table 3. Parameter Estimates for Quadratic Growth Model for Depressive Symptoms Parameter Est. SE p-value Intercept β00 8.70 .51 .00 Time β10 −1.07 .25 .00 Time2 β20 .29 .05 .00 LEP β01 1.10 .41 .01 LEP × Time β11 −.88 .32 .01 LEP × Time2 β21 .19 .07 .01 Age (centered) β02 .11 .02 .00 Female β03 2.27 .27 .00 Educational Attainment β04 −.11 .04 .01 Born outside the United States β05 .70 .27 .01 Interviewed in Spanish β06 −1.00 .33 .00 Marital Status β07 −.48 .27 .08 Parameter Est. SE p-value Intercept β00 8.70 .51 .00 Time β10 −1.07 .25 .00 Time2 β20 .29 .05 .00 LEP β01 1.10 .41 .01 LEP × Time β11 −.88 .32 .01 LEP × Time2 β21 .19 .07 .01 Age (centered) β02 .11 .02 .00 Female β03 2.27 .27 .00 Educational Attainment β04 −.11 .04 .01 Born outside the United States β05 .70 .27 .01 Interviewed in Spanish β06 −1.00 .33 .00 Marital Status β07 −.48 .27 .08 Note: LEP = Limited English proficiency. View Large Table 3. Parameter Estimates for Quadratic Growth Model for Depressive Symptoms Parameter Est. SE p-value Intercept β00 8.70 .51 .00 Time β10 −1.07 .25 .00 Time2 β20 .29 .05 .00 LEP β01 1.10 .41 .01 LEP × Time β11 −.88 .32 .01 LEP × Time2 β21 .19 .07 .01 Age (centered) β02 .11 .02 .00 Female β03 2.27 .27 .00 Educational Attainment β04 −.11 .04 .01 Born outside the United States β05 .70 .27 .01 Interviewed in Spanish β06 −1.00 .33 .00 Marital Status β07 −.48 .27 .08 Parameter Est. SE p-value Intercept β00 8.70 .51 .00 Time β10 −1.07 .25 .00 Time2 β20 .29 .05 .00 LEP β01 1.10 .41 .01 LEP × Time β11 −.88 .32 .01 LEP × Time2 β21 .19 .07 .01 Age (centered) β02 .11 .02 .00 Female β03 2.27 .27 .00 Educational Attainment β04 −.11 .04 .01 Born outside the United States β05 .70 .27 .01 Interviewed in Spanish β06 −1.00 .33 .00 Marital Status β07 −.48 .27 .08 Note: LEP = Limited English proficiency. View Large The LEP status associated with both time factors (i.e., β11, β21) was also statistically significant, suggesting that there is a differential effect of LEP on growth trajectories of depressive symptoms. As hypothesized and displayed in Figure 1, those with LEP revealed a steeper curve of depressive symptom trajectory over time than those with EP. Figure 1. View largeDownload slide Quadratic growth trajectories of depressive symptoms over six waves by English proficiency. CES-D = Center for Epidemiologic Studies Depression scale; EP = English proficiency; LEP = Limited English proficiency. Figure 1. View largeDownload slide Quadratic growth trajectories of depressive symptoms over six waves by English proficiency. CES-D = Center for Epidemiologic Studies Depression scale; EP = English proficiency; LEP = Limited English proficiency. Among covariates adjusted, several demonstrated independent contributions. As shown in Table 3, age was significantly related with depressive symptom level (β02 = .11, p < .01), indicating that every year older was associated with a .11 increase in the level of depressive symptoms. Females exhibited higher levels of depressive symptoms than males on average at the baseline (β03 = 2.27, p < .01). Years of education were also significantly related with depressive symptoms (β04 = −.11, p < .01), meaning that every year increase in education was associated with a .11 decrease in the level of depression symptoms. Respondents who were born outside the United States showed significantly more depressive symptoms (β05 = .70, p < .01) while respondents interviewed in Spanish showed less depressive symptoms (β06 = 1.00, p < .01) at the baseline. Although marital status was not statistically significant at p < .05 level, it showed a marginally significant relationship with depressive symptoms (β07 = −.48, p = .08), indicating that married people reported less on their depressive symptoms at the baseline. Discussion Given the reported vulnerability of Hispanic older adults with LEP (Kim, Aguado Loi, et al., 2011; Kim, Worley, et al., 2011; Krause & Goldenhar, 1992) and LEP’s potentially harmful effect on mental health (Mui et al., 2007; Takeuchi et al., 2007), we attempted to explore the effect of LEP on longitudinal trajectories of depressive symptoms among Mexican American older adults over a 14-year period. To our knowledge, this was the first study examining the longitudinal effect of LEP on mental health among older Hispanics in the United States. As hypothesized, LEP was associated not merely with greater depressive symptoms, but with an accelerated trajectory of depressive symptoms among older Mexican Americans over 14 years. This finding also supports what would be expected on the basis of the double jeopardy and cumulative disadvantage theories. Our findings suggest the need for assistance for those with LEP in order to reduce depressive symptoms, which may also help reduce disparities arising from language barriers. The most intriguing finding was the differential growth trajectories of depressive symptoms between with LEP and EP groups over 14 years, especially showing a steeper curve for those with LEP. Despite the nonsignificant differences in depressive symptom levels from 2 to 5 waves, the overall growth trajectory curve of depressive symptoms was significantly steeper for those with LEP compared to those with EP. Given that those with LEP tend to have difficulty navigating the proper health care system and understanding health information and treatment (Flores, 2006; DuBard & Gizlice, 2008; Kim, Aguado Loi, et al., 2011), it seems natural to speculate that poor access to and quality of health care that LEP individuals often experience in health care settings may lead to their poor mental health status both cross-sectionally and longitudinally. Given the availability of variables on health care access and use in the Hispanic EPESE data set, future mediation analyses focusing on health care access and use as potential mediators should be helpful in terms of determining where to target interventions for Hispanic older adults. Given the shortage of bilingual providers and the lack of affordability of mental health care due to the lack of mental health parity in Medicare that had persisted during the study period, LEP older adults are even less likely to access mental health care (Bauer, Chen, & Alegría, 2010; Sentell et al., 2007; Snowden, Masland, & Guerrero, 2007). What we did not know until now was LEP’s long-term impact on poor mental health status through more than ten years of follow-up. Future research should investigate mechanisms deriving long-term vulnerability of LEP older adults. It is worth pointing out opposing growth trajectories of depressive symptoms by host language proficiency found in a previous study (Nguyen et al., 2011). The main difference in the present study and Nguyen and colleagues’ study (2011) was the target population: adults aged 65 years and older versus adolescents aged 12–23 years. Nguyen and colleagues (2011) reported that becoming fluent in the mainstream language (i.e., English or French) was associated with a risk for steeper increases in depressive symptoms among adolescents from immigrant backgrounds in Canada, which was a completely opposite direction observed in the present study. While Nguyen and colleagues (2011) speculated that becoming fluent in the mainstream language often accompanied with a loss of heritage language might affect the parent-child cohesion and closeness negatively, this was conflicting to many other studies including the current study conceptualizing LEP as a marker for less acculturation and therefore as a source of stress. The findings from the current study suggest the need that LEP’s effect should be investigated further across the life span. Findings from this study provide implications for research, clinical practice, and policy pertaining geriatric mental health disparities. Mental health professionals and clinicians working with older Mexican Americans with LEP should be aware of the risk that they tend to have higher levels of depressive symptoms in general, as well as long-term disadvantage to mental health status and high stress relating to communicating with others. It is worth noting regarding clinical relevance that mean depressive symptom scores for both LEP and EP groups were well below the level for probable depression. However, given that previous studies reported standard clinical cut-offs of the CES-D may not be good indicators of depression and significantly higher CES-D scores were found among Hispanic older adults than those of other racial/ethnic elderly groups (e.g., Guarnaccia, Good, & Kleinman, 1990; Kim, Chiriboga, & Jang, 2009), clinicians and researchers should be cautious when interpreting the CES-D scores. In addition, clinicians and researchers should recognize that the CES-D results may need to be interpreted differently cross-culturally. Direct interventions to increase English skills may not be effective for this geriatric minority population. Instead, increasing support from informal and formal network that can build a system to help LEP older adults should be needed. Policy-makers need to focus on educating and training health/mental health care providers in order to increase bilingual workforce, especially in health and mental health professional shortage areas, which may be an effective step for policy intervention. For example, student training fellowships that the Health Resources and Services Administration (HRSA) has awarded fellowships to multiple health professions institutions and students from underrepresented minority groups in order to enhance the workforce diversity of health care professions such as physicians, social workers, and psychologists (Council on Graduate Medical Education, 2016). Policy-makers can also consider developing formal programs to build or invigorate a formal support system for LEP patients. These supports might focus on day care, respite care, meals on wheels and similar home and community-based services, but would require Spanish-language proficiency on the part of the providers. Also, providers can engage someone from informal support system identified by the LEP patient as a translator to help to overcome LEP barriers during treatment. As an extension of treatment, providers can use someone from informal network to help motivate the patient to engage in health promotion, to do homework in between treatment sessions, to take the medications, and to keep the LEP patient coming back to therapy. The present study is not without limitations. First and foremost, given that this study solely focused on Mexican Americans, generalizability is limited. Future research should be extended to other Hispanic subgroups as well as other racial/ethnic minority populations. Second, there may be the heterogeneity issue even within the Mexican American elderly sample. Even if some potentially important covariates were controlled in our analyses, subgroup differences might have existed. Third, like many longitudinal studies, attrition over time may have affected the findings. Fourth, we were not able to control potentially stressful life events such as the death of a spouse that older adults often experience in later life. Lastly, given that measurement equivalence of our outcome measure (i.e., the CES-D) by EP has not yet been established, the CES-D may not have functioned equally across linguistic groups. Given that prior research suggests evidence of measurement bias of a psychological distress measure (K6) when language use is considered (Kim, Decoster, Bryant, & Ford, 2016), further research should test the potential bias of the CES-D by EP. In conclusion, the current longitudinal analysis revealed LEP as a risk factor for an elevated depressive symptom trajectory over a 14-year period in the Hispanic elderly population. Supporting the double jeopardy theory was that those with LEP exhibited greater depression at all waves than did those with EP. Supporting the cumulative disadvantage theory was that those with LEP had a steeper curve of depressive symptom trajectory than those with EP over time. What was particularly noteworthy in the present study was the cumulative disadvantage of LEP to mental health that Mexican American older adults had experienced over 14 years. Overall, the findings emphasize the need to assist the informal support system of Mexican American older adults with LEP, but the formal system may help reduce depressive symptoms arising from language barriers. Focusing on those experiencing language barriers may assist with the goal of improving the mental health status of vulnerable elderly populations. Supplementary Data Supplementary data are available at The Gerontologist online. Funding The Hispanic EPESE was supported by the National Institute on Aging (R01AG10939, PI: Markides, K.). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Conflict of Interest None reported. References Abe-Kim , J. , Takeuchi , D. T. , Hong , S. , Zane , N. , Sue , S. , Spencer , M. S. , & Alegría , M . ( 2007 ). Use of mental health-related services among immigrant and US-born Asian Americans: Results from the National Latino and Asian American Study . American Journal of Public Health , 97 , 91 – 98 . doi: 10.2105/AJPH.2006.098541 Google Scholar CrossRef Search ADS PubMed Akhtar-Danesh , N. , & Landeen , J . ( 2007 ). Relation between depression and sociodemographic factors . International Journal of Mental Health Systems , 1 , 4 . doi: 10.1186/1752-4458-1-4 Google Scholar CrossRef Search ADS PubMed Angel , R. J. , Frisco , M. , Angel , J. L. , & Chiriboga , D. A . ( 2003 ). Financial strain and health among elderly Mexican-origin individuals . Journal of Health and Social Behavior , 44 , 536 – 551 . doi:10.2307/1519798 Google Scholar CrossRef Search ADS PubMed Bauer , A. M. , Chen , C. N. , & Alegría , M . ( 2010 ). English language proficiency and mental health service use among Latino and Asian Americans with mental disorders . Medical Care , 48 , 1097 – 1104 . doi: 10.1097/MLR.0b013e3181f80749 Google Scholar CrossRef Search ADS PubMed Bebbington , P. E. , Dunn , G. , Jenkins , R. , Lewis , G. , Brugha , T. , Farrell , M. , & Meltzer , H . ( 1998 ). The influence of age and sex on the prevalence of depressive conditions: Report from the national survey of psychiatric morbidity . Psychological Medicine , 28 , 9 – 19 . doi:10.1017/s0033291797006077 Google Scholar CrossRef Search ADS PubMed Berry , J. W . ( 1997 ). Immigration, acculturation, and adaptation . Applied Psychology: An International Review , 46 , 5 – 68 . doi:10.1111/j.1464-0597.1997.tb01087.x Blazer , D. , Burchett , B. , Service , C. , & George , L. K . ( 1991 ). The association of age and depression among the elderly: An epidemiologic exploration . Journal of Gerontology , 46 , M210 – M215 . doi:10.1093/geronj/46.6.m210 Google Scholar CrossRef Search ADS PubMed Carreon , D. , & Noymer , A . ( 2011 ). Health-related quality of life in older adults: Testing the double jeopardy hypothesis . Journal of Aging Studies , 25 , 371 – 379 . doi:10.1016/j.jaging.2011.01.004 Google Scholar CrossRef Search ADS Cole , M. G. , & Dendukuri , N . ( 2003 ). Risk factors for depression among elderly community subjects: A systematic review and meta-analysis . The American Journal of Psychiatry , 160 , 1147 – 1156 . doi: 10.1176/appi.ajp.160.6.1147 Google Scholar CrossRef Search ADS PubMed Constantine , M. G. , Okazaki , S. , & Utsey , S. O . ( 2004 ). Self-concealment, social self-efficacy, acculturative stress, and depression in African, Asian, and Latin American international college students . The American Journal of Orthopsychiatry , 74 , 230 – 241 . doi: 10.1037/0002-9432.74.3.230 Google Scholar CrossRef Search ADS PubMed Council on Graduate Medical Education . ( 2016 ). Supporting Diversity in the Health Professions. Retrieved from http://www.hrsa.gov/advisorycommittees/bhpradvisory/cogme/Publications/diversityresourcepaper.pdf. Dannefer , D . ( 2003 ). Cumulative advantage/disadvantage and the life course: Cross-fertilizing age and social science theory . The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 58 , S327 – S337 . doi:10.1093/geronb/58.6.s327 Google Scholar CrossRef Search ADS Derose , K. P. , Escarce , J. J. , & Lurie , N . ( 2007 ). Immigrants and health care: Sources of vulnerability . Health Affairs (Project Hope) , 26 , 1258 – 1268 . doi: 10.1377/hlthaff.26.5.1258 Google Scholar CrossRef Search ADS PubMed Dowd , J. J. , & Bengtson , V. L . ( 1978 ). Aging in minority populations. An examination of the double jeopardy hypothesis . Journal of Gerontology , 33 , 427 – 436 . doi:10.1093/geronj/33.3.427 Google Scholar CrossRef Search ADS PubMed DuBard , C. A. , & Gizlice , Z . ( 2008 ). Language spoken and differences in health status, access to care, and receipt of preventive services among US Hispanics . American Journal of Public Health , 98 , 2021 – 2028 . doi: 10.2105/AJPH.2007.119008 Google Scholar CrossRef Search ADS PubMed Enders , C. K. , & Bandalos , D. L . ( 2001 ). The relative performance of full information maximum likelihood estimation for missing data in structural equation models . Structural Equation Modeling , 8 , 430 – 457 . doi:10.1207/s15328007sem0803_5 Google Scholar CrossRef Search ADS Ferraro , K. F. , & Farmer , M. M . ( 1996 ). Double jeopardy, aging as leveler, or persistent health inequality? A longitudinal analysis of white and black Americans . The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 51 , S319 – S328 . doi:10.1093/geronb/51b.6.s319 Google Scholar CrossRef Search ADS Flores , G . ( 2006 ). Language barriers to health care in the United States . The New England Journal of Medicine , 355 , 229 – 231 . doi: 10.1056/NEJMp058316 Google Scholar CrossRef Search ADS PubMed Gee , G. C. , & Ponce , N . ( 2010 ). Associations between racial discrimination, limited English proficiency, and health-related quality of life among 6 Asian ethnic groups in California . American Journal of Public Health , 100 , 888 – 895 . doi: 10.2105/AJPH.2009.178012 Google Scholar CrossRef Search ADS PubMed Gonzalez-Barrera , A. , & Krogstad , J. M . ( 2015 ). What we know about illegal immigration from Mexico . Washington, DC: Pew Research Center, 20 . Guarnaccia , P. J. , Good , B. J. , & Kleinman , A . ( 1990 ). A critical review of epidemiological studies of Puerto Rican mental health . The American Journal of Psychiatry , 147 , 1449 – 1456 . doi: 10.1176/ajp.147.11.1449 Google Scholar CrossRef Search ADS PubMed Humes , K. , Jones , N. A. , & Ramirez , R. R . ( 2011 ). Overview of race and Hispanic origin, 2010 . Washington, DC: US Department of Commerce, Economics and Statistics Administration, US Census Bureau . Jang , Y. , Haley , W. E. , Small , B. J. , & Mortimer , J. A . ( 2002 ). The role of mastery and social resources in the associations between disability and depression in later life . The Gerontologist , 42 , 807 – 813 . doi:10.1093/geront/42.6.807 Google Scholar CrossRef Search ADS PubMed Jang , Y. , Yoon , H. , Park , N. S. , & Chiriboga , D. A . ( 2016 ). Health vulnerability of immigrants with limited english proficiency: A study of older Korean Americans . Journal of the American Geriatrics Society , 64 , 1498 – 1502 . doi: 10.1111/jgs.14199 Google Scholar CrossRef Search ADS PubMed Kandula , N. R. , Lauderdale , D. S. , & Baker , D. W . ( 2007 ). Differences in self-reported health among Asians, Latinos, and non-Hispanic whites: The role of language and nativity . Annals of Epidemiology , 17 , 191 – 198 . doi: 10.1016/j.annepidem.2006.10.005 Google Scholar CrossRef Search ADS PubMed Kim , G. , Aguado Loi , C. X. , Chiriboga , D. A. , Jang , Y. , Parmelee , P. , & Allen , R. S . ( 2011 ). Limited English proficiency as a barrier to mental health service use: A study of Latino and Asian immigrants with psychiatric disorders . Journal of Psychiatric Research , 45 , 104 – 110 . doi: 10.1016/j.jpsychires.2010.04.031 Google Scholar CrossRef Search ADS PubMed Kim , G. , Chiriboga , D. A. , & Jang , Y . ( 2009 ). Cultural equivalence in depressive symptoms in older white, black, and Mexican-American adults . Journal of the American Geriatrics Society , 57 , 790 – 796 . doi:10.1111/j.1532-5415.2009.02188.x Google Scholar CrossRef Search ADS PubMed Kim , G. , DeCoster , J. , Bryant , A. N. , & Ford , K. L . ( 2016 ). Measurement equivalence of the k6 scale: The effects of race/ethnicity and language . Assessment , 23 , 758 – 768 . doi: 10.1177/1073191115599639 Google Scholar CrossRef Search ADS PubMed Kim , M. , Kwok , O. , Yoon , M. , Willson , V. , & Lai , M. H . ( 2016 ). Specification search for identifying the correct mean trajectory in polynomial latent growth models . The Journal of Experimental Education , 84 , 307 – 329 . doi:10.1080/00220973.2014.984831 Google Scholar CrossRef Search ADS Kim , G. , Worley , C. B. , Allen , R. S. , Vinson , L. , Crowther , M. R. , Parmelee , P. , & Chiriboga , D. A . ( 2011 ). Vulnerability of older Latino and Asian immigrants with limited English proficiency . Journal of the American Geriatrics Society , 59 , 1246 – 1252 . doi: 10.1111/j.1532-5415.2011.03483.x Google Scholar CrossRef Search ADS PubMed Krause , N. , & Goldenhar , L. M . ( 1992 ). Acculturation and psychological distress in three groups of elderly Hispanics . Journal of Gerontology , 47 , S279 – S288 . doi:10.1093/geronj/47.6.s279 Google Scholar CrossRef Search ADS PubMed Krogstad , J.M. & Gonzalez-Barrera , A . ( 2015 ). A majority of English-speaking hispanics in the U.S. are bilingual . Pew Research Center . Retrieved from http://www.pewresearch.org/fact-tank/2015/03/24/a-majority-of-english-speaking-hispanics-in-the-u-s-are-bilingual/. Krogstad , J.M. , Stepler , R. , & Lopez , M.H . ( 2016 ). English Proficiency on the Rise Among Latinos . Pew Research Center Hispanic Trends . Retrieved from http://www.pewhispanic.org/2015/05/12/english-proficiency-on-the-rise-among-latinos/. Lin , J. G. , & Yi , J. K . ( 1997 ). Asian international student’s adjustment: Issues and program suggestions . College Student Journal , 31 , 473 – 479 . Little , R.J.A. , & Rubin , D. B . ( 2002 ). Statistical analysis with missing data ( 2nd ed .). Hoboken, NJ : Wiley . Google Scholar CrossRef Search ADS Mui , A. C. , Kang , S. Y. , Kang , D. , & Domanski , M. D . ( 2007 ). English language proficiency and health-related quality of life among Chinese and Korean immigrant elders . Health & Social Work , 32 , 119 – 127 . doi:10.1093/hsw/32.2.119 Google Scholar CrossRef Search ADS PubMed Muthen , L. K. , & Muthén , B. O . ( 2010 ). Mplus user’s guide, v 6.1 . Los Angeles, CA : Muthen and Muthen, UCLA . Nguyen , H. , Rawana , J. S. , & Flora , D. B . ( 2011 ). Risk and protective predictors of trajectories of depressive symptoms among adolescents from immigrant backgrounds . Journal of Youth and Adolescence , 40 , 1544 – 1558 . doi: 10.1007/s10964-011-9636-8 Google Scholar CrossRef Search ADS PubMed Nwadiora , E. , & McAdoo , H . ( 1996 ). Acculturative stress among Amerasian refugees: Gender and racial differences . Adolescence , 31 , 477 – 487 . Google Scholar PubMed Pandya, C., Batalova J., & McHugh M. (2011). Limited English proficient individuals in the United States: Number, share, growth, and linguistic diversity. Washington, DC: Migration Policy Institute . Ponce , N. A. , Hays , R. D. , & Cunningham , W. E . ( 2006 ). Linguistic disparities in health care access and health status among older adults . Journal of General Internal Medicine , 21 , 786 – 791 . doi: 10.1111/j.1525-1497.2006.00491.x Google Scholar CrossRef Search ADS PubMed Preacher , K. J. , & Hayes , A. F . ( 2004 ). SPSS and SAS procedures for estimating indirect effects in simple mediation models . Behavior Research Methods, Instruments, & Computers: A Journal of the Psychonomic Society, Inc , 36 , 717 – 731 . Google Scholar CrossRef Search ADS PubMed Radloff , L. S . ( 1977 ). The CES-D scale: A self-report depression scale for research in the general population . Applied Psychological Measurement , 1 , 385 – 401 . doi:10.1177/014662167700100306 Google Scholar CrossRef Search ADS Sadule-Rios , N . ( 2012 ). A review of the literature about depression in late life among Hispanics in the United States . Issues in Mental Health Nursing , 33 , 458 – 468 . doi: 10.3109/01612840.2012.675415 Google Scholar CrossRef Search ADS PubMed Sentell , T. , & Braun , K. L . ( 2012 ). Low health literacy, limited English proficiency, and health status in Asians, Latinos, and other racial/ethnic groups in California . Journal of Health Communication , 17 ( Suppl 3 ), 82 – 99 . doi: 10.1080/10810730.2012.712621 Google Scholar CrossRef Search ADS PubMed Sentell , T. , Shumway , M. , & Snowden , L . ( 2007 ). Access to mental health treatment by English language proficiency and race/ethnicity . Journal of General Internal Medicine , 22 ( Suppl 2 ), 289 – 293 . doi: 10.1007/s11606-007-0345-7 Google Scholar CrossRef Search ADS PubMed Snowden , L. R. , Masland , M. , & Guerrero , R . ( 2007 ). Federal civil rights policy and mental health treatment access for persons with limited English proficiency . The American Psychologist , 62 , 109 – 117 . doi: 10.1037/0003-066X.62.2.109 Google Scholar CrossRef Search ADS PubMed Takeuchi , D. T. , Zane , N. , Hong , S. , Chae , D. H. , Gong , F. , Gee , G. C.,… Alegría , M . ( 2007 ). Immigration-related factors and mental disorders among Asian Americans . American Journal of Public Health , 97 , 84 – 90 . doi: 10.2105/AJPH.2006.088401 Google Scholar CrossRef Search ADS PubMed U.S. Census Bureau . Population 60 years and over in the United States: 2010 [on- line] . Retrieved March 6, 2017, from https://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?src=bkmk. U.S. Census Bureau . Selected social characteristics of the United States: 2005 [on-line] . Retrieved March 6, 2017, from http://factfinder.census.gov/servlet/ADPTable?_bm=y&geo_id=01000US&-qr_name=ACS_2005_EST_G00_DP2&-ds_name=&-redoLog=false&-format-. Zarit , S. H. , & Zarit , J. M . ( 1998 ). Mental disorders in older adults: Fundamentals of assessment and treatment . New York : Guilford Press . Zhang , W. , Hong , S. , Takeuchi , D. T. , & Mossakowski , K. N . ( 2012 ). Limited English proficiency and psychological distress among Latinos and Asian Americans . Social Science & Medicine (1982) , 75 , 1006 – 1014 . doi: 10.1016/j.socscimed.2012.05.012 Google Scholar CrossRef Search ADS PubMed Zong , J. , & Batalova , J . ( 2015 ). The limited English proficient population in the United States[on-line] . Retrieved March 6, 2017, from http://www.migrationpolicy.org/Article/Limited-English-Proficient-Population-United-States/. © The Author(s) 2018. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

The GerontologistOxford University Press

Published: Apr 23, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off