Determinants of psychological distress among Latina young adults during initial years in the United States

Abstract

Objectives:

This longitudinal study investigated how theorized cultural-, interpersonal/familial-, and individual-level experiences during initial years after immigrating to the United States may relate to changes in psychological distress over time among young adult Latina women in Miami-Dade County, Florida, United States (U.S.).

Methods:

Using latent growth curve modeling, we examined trajectories of psychological distress reported by 530 Latina women (aged 18–23 years) during their initial years in the U.S. The sample’s mean length of time in the United States since immigration was 11.52 months (SD = 9.94) at baseline assessment. Trajectories of distress were measured over a three-year time period. We investigated whether cultural- (immersion to dominant society/assimilation, immersion to ethnic society/enculturation, acculturative stress, marianismo), interpersonal/familial- (trust and communication with parents, peers, partner), and individual- (immigration authorization status, education level, relationship status) level determinants experienced soon after immigration predicted changes in participants’ distress over a subsequent two-year time period.

Results:

At baseline assessment, participants’ psychological distress was associated with higher levels of acculturative stress (β = .63, p < .001). Over time, the average level of women’s distress significantly declined (b = −.10, p = .001). Steeper declines in distress over time were reported by women who reported more baseline acculturative stress (β = −.80, p < .001) and less baseline trust and communication with their parents, peers, and partners (β = .52, p = .002).

Conclusions:

Findings inform theoretical and empirical knowledge bases concerning determinants of psychological distress during initial years after immigration among foreign-born young Latina women in the U.S. The study provides a longitudinal examination of psychological distress during the rarely studied time period of initial years in the U.S. Results elucidate the substantive influence of acculturative stress and interpersonal/familial relationship dynamics soon after immigration. Mental health interventions should attend to these modifiable factors during young Latina women’s early years in the U.S. to reduce psychological distress and promote well-being soon after immigration and subsequent years.

In response to Latinx individuals comprising the most rapidly growing ethnoracial community in the United States (U.S.; Peña, Lowe, and Ríos-Vargas 2023), psychology and allied professions study the prevalence, variability, and persistence of psychological distress among Latina women of multiple Latinx origins and contexts (e.g., Alegría et al. 2007, 2008; De Oliveira et al. 2017; Jones, Briones-Zamora, and Underwood 2024). When examined in the aggregate, national samples of foreign-born Latina women consistently indicate higher rates of psychological distress than men (Wassertheil-Smoller et al. 2014). Young adult Latina women also have increased susceptibility to psychological distress (e.g., depression, anxiety) than older women (Fortuna et al. 2007; Ornelas et al. 2023). Considering this, the purpose of the present study was to contribute to theoretical and empirical knowledge bases concerning psychological distress and its determinants among foreign-born, young Latina women in the U.S. Specifically, we studied psychological distress rates reported annually by Latina women (ages 18-23) over a three-year time period during their initial years in Miami-Dade County, Florida. In addition, we studied baseline cultural, interpersonal/familial, and individual factors as predictors of psychological distress over time because our primary aim was to examine whether and how experiences during participants’ initial 36 months in the in the United States (i.e., time of baseline assessment) may influence subsequent trajectories of psychological distress. By focusing on these early experiences, we sought to identify potentially modifiable factors that may shape long-term mental health outcomes. These findings contribute to the literature by suggesting that intervention efforts should target such factors during young Latina women’s early years in the U.S., with the goal of preventing psychological distress and promoting well-being both in the immediate post-immigration period and in subsequent years.

Social ecological theoretical framework

Our research questions and hypotheses were guided by social ecological theoretical framework (Bronfenbrenner 1986, 1994) because it is (a) inclusive of social and cultural determinants of health and (b) facilitates the study of the influence of co-occurring determinants on psychological distress experienced during young adult Latina women’s initial years in the U.S. Social ecological theory has explained the interdependence between health and social contexts of ethnoracial communities (Crawford 2020; Oh et al. 2023). More specifically, psychological distress is posited to result from an interplay of cultural-, interpersonal/familial-, individual-level factors that are nested within broader sociocultural structures to influence young Latina women’s mental health soon after immigration (Alegría, Álvarez, and DiMarzio 2017; Ayón, Marsiglia, and Bermudez-Parsai 2010). By obtaining self-reports of young adult Latina women soon after immigration, we attempted to advance empirical knowledge concerning the theorized cultural, interpersonal/familial, and individual factors that may affect their mental health during this distinct time period.

Cultural determinants of psychological distress

Acculturation is a multidimensional process involving interactions between heritage-cultural and receiving-cultural practices, values, and identifications during a person’s experience of a new culture (Driscoll and Torres 2020; Schwartz et al. 2010). Acculturation can entail changes in a person’s behaviors (e.g., language use), values (e.g., collectivism, individualism), and cultural identification (e.g., ethnic identity), among other factors (Alegría, Álvarez, and DiMarzio 2017; Torres, Driscoll, and Voell 2012). Multiple acculturation strategies are theorized, including (a) assimilation, (b) enculturation, (c) integration/biculturalism, and (d) marginalization (Berry 2005; Schwartz et al. 2010). Assimilation occurs when individuals adopt U.S. mainstream culture and abandon their cultural heritage, whereas enculturation refers to the retention of cultural heritage and the rejection of U.S. mainstream culture. Integration, or biculturalism, occurs when individuals adopt U.S. mainstream culture, while also maintaining their cultural heritage. Lastly, marginalization refers to the rejection of both U.S. mainstream culture and heritage culture. In terms of our guiding social ecological framework, acculturation can be conceptualized as primarily a macrosystem-level cultural adaptation process that is enacted within everyday interpersonal contexts (microsystems) and shaped by linkages across settings (mesosystem) over time (Bronfenbrenner 1986, 1994). We studied Latina adult’s perceptions of this construct after arrival to the U.S. and subsequent associations with psychological distress.

Acculturation strategies are dynamic and have varied health implications based on individual, social, and contextual factors (Alegría, Álvarez, and DiMarzio 2017; Chua et al. 2024; Ferguson, Causadias, and Simenec 2023). It is possible for one acculturation process to be beneficial in one context but not another, and for processes to vary between and within people over time depending on context (Schwartz et al. 2015). For example, assimilation is linked to increased psychological distress and substance use for Latina adult immigrants of different nationalities (Hernandez et al. 2019; López-Tamayo and Jason 2021). Alternatively, some proxies of assimilation (e.g., English language proficiency and preference) may be protective against psychological distress among some foreign-born Latina women when they facilitate access to community resources and employment opportunities (Bekteshi and Bellamy 2024; Valencia-Garcia et al. 2012). In a cross-sectional study examining different profiles of acculturation among a national sample of Latinx and Asian immigrants, Latinx adults in the assimilated and bicultural profiles reported better mental health than peers, potentially due to higher socioeconomic advantages (Bulut and Gayman 2016). Conversely, retaining and acquiring traditional Latinx cultural strengths, such as endorsing collectivistic values, maintaining social and familial supports, and pride in one’s ethnic identity may not only protect newly immigrated Latina adults from psychological distress (Cariello, Perrin, and Morlett-Paredes 2020; Sánchez, Vandewater, and Hamilton 2019), but also facilitate thriving after immigration (Martinez et al. 2021).

Acculturative stress or the stress experienced in response to the demands associated with acculturation (Arbona et al. 2010), is also a critical factor to consider. Acculturative stress has been conceptualized as arising from multiple domains such as social stress (e.g., difficulties with interpersonal belonging and acceptance after immigration); attitudinal stress (e.g., internal conflicts about one’s identity and cultural values due to acculturation); familial stress (e.g., family strain, conflict, and misunderstanding due to differential acculturation); and environmental stress (e.g., structural and contextual barriers such as discrimination, language barriers, and economic injustices) (Mena, Padilla, and Maldonado 1987). This operationalization has been supported by young first-generation and second-generation Mexican immigrant women reporting that their feelings of acculturative stress were heightened by unauthorized immigration legal status, language barriers, poverty, and discrimination (Viruell-Fuentes 2007). Acculturative stress also is associated with mental health, such as depression, alcohol use, and psychological distress (Bekteshi 2024; Verdaguer et al. 2023). Acculturative stress can be understood as a microsystem-level stress response that women in this study are likely to report from stressors arising from differing cultural settings experienced soon after their immigration (mesosystem) and structural-cultural differences in values, attitudes, ideologies from macrosystemic forces, particularly during their developmental transitions as recent immigrants and young adults (chronosystem) (Bronfenbrenner 1986, 1994).

Marianismo is a Latinx cultural value that shapes traditional feminine gender norms (Castillo et al. 2010). Marianismo beliefs include being family-focused, self-sacrificing, submissive, chaste, modest, and a source of familial and spiritual strength (Castillo et al. 2010; Sánchez, Vandewater, and Hamilton 2019). Marianismo is a complex, multidimensional construct that cross-sectional research suggests may impact Latina young adults’ mental health in their early post-immigration years. Both positive (e.g., spiritual strength) and negative (e.g., submissiveness) dynamics of marianismo influence mental health in distinct and opposing ways (Cano et al. 2020; Ertl et al. 2024; Nuñez et al. 2016). Cross-sectional studies suggest that increased endorsement of spiritual pillar marianismo beliefs (i.e., agreement with the expectation that Latina women should embody deep spirituality and serve as the family’s spiritual and moral anchor) is protective as it is related to less depression among samples of mostly Mexican immigrant women (Cano et al. 2020; Nuñez et al. 2016). Another study linking mental health and marianismo (e.g., Ertl et al. 2024), found that agreeing with beliefs about the importance for Latina women to be virtuous and chaste were linked to less substance use among young Latina college students who were predominantly U.S. born second-generation immigrants. In contrast, increased endorsement of the beliefs that Latina women should be subordinate and self-silencing to maintain harmony in relationships are positively associated with distress and depression, (Cano et al. 2020; Nuñez et al. 2016). It is posited that rejecting some types of traditional gender role beliefs may increase psychological distress among young Latina women due to an incongruence and dissonance between women’s own beliefs and marianismo expectations (Nuñez et al. 2016). Moreover, the heavy burden placed on Latina women who endorse a need to ensure their family’s functioning and be the primary source of strength while also be self-sacrificing, submissive and subordinate may also contribute to psychological distress. Based on social ecological theory, the dimensions of marianismo can be conceptualized as a macrosystem-level cultural gender ideology that is internalized at the individual-level (i.e., the level assessed in the present study) and enacted within daily microsystem relationships and shaped by microsystemic interactions (mesosystem) over time (Bronfenbrenner 1986, 1994).

Interpersonal/familial determinant of psychological distress

Women’s perceptions of supportive interpersonal and familial relationships, characterized by mutual trust and healthy communication at the microsystemic level, are likely to bolster young Latina immigrants’ mental health (Rios Casas et al. 2020). Indicators of supportive familial and interpersonal relationships have been found to inversely relate with psychological distress and stressors among Latinx immigrants (Rios Casas et al. 2020; Torres et al. 2016). Specifically, Rio Casas et al.’s (2020) qualitative, cross-sectional study revealed that familial, peer, and transnational social relationships helped alleviate immigration-related stressors (e.g., social isolation) among Latina immigrants of mostly Mexican descent, who had lived in the U.S. an average of 10 years. Moreover, Ryan et al.’s (2021) cross-sectional study found that positive social relationships were related to decreased anxiety symptoms among a community-based sample of Latina women of mostly Mexican descent, who had been living in the U.S. for over a decade. Additionally, Torres et al. ‘s (2016) cross-sectional study using a national foreign-born Latinx sample found that the supportive act of transnational remittance sending was associated with lower levels of psychological distress among Cuban immigrants. In contrast, when familial and interpersonal relationships are characterized by high conflict or absence, young Latina women may be vulnerable to acculturative stress, social isolation, and distress (Bostean and Gillespie 2018; Ryan et al. 2021). High levels of familial conflict was associated with increased acculturative stress among a nationally representative sample of foreign-born Latinx individuals of mostly Mexican descent (Bostean and Gillespie 2018). Additionally, Ryan et al. (2021) found that social isolation was related to increased depressive symptoms, suggesting that separation from supportive family and friends in one’s home country may increase distress after immigration.

Individual determinants of psychological distress

Because recently immigrated young Latina women are a heterogenous group, we examined several individual-level, microsystemic determinants of psychological distress. They include socioeconomic status, relationship status, and immigration authorization status. Self-reported indicators of socioeconomic status are typically inversely associated with psychological distress due to greater access to resources that alleviate distress (McSorley and Bacong 2023). For example, McSorley and Bacong’s (2023) cross-sectional study found that unemployment, decreased income, and low educational attainment were associated with increased levels of moderate and severe psychological distress among a national sample of disaggregated Latinx subgroups (Mexican, Central/South American, Puerto Rican, Cuban, Dominican). Moreover, Darghouth, Brody, and Alegría (2015) found that being separated or divorced was associated with increased reports of psychological distress compared to being married among Latina women in a national sample of aggregate Latinx individuals of mostly Mexican descent.

Psychological distress also varies depending on young Latina women’s immigration authorization status. Latina women residing in the U.S. without authorization are at risk for psychological distress because of increased risk of discrimination and violence (Navarro Flores et al. 2023). Navarro Flores et al. (2023) found that cumulative experiences of immigration-related trauma were associated with increased anxiety and depressive symptoms among a sample of Latinx individuals with unauthorized immigration status, predominantly women, residing in the U.S. for 10–20 years or more. Women without authorization also may experience more distress due to living in unsafe neighborhoods, actual or threatened separation from family due to deportation, social isolation, and loss of social status and legal rights due to anti-immigration public policies (Garcini et al. 2017, 2021; Verdaguer et al. 2023). Garcini et al.’s (2017) cross-sectional study found that post-migration difficulties (e.g., strained finances, discrimination) and being an emerging adult (18-25 years old) were associated with increased risk for depressive, anxiety, substance use disorders. Verdaguer et al.’s (2023) cross-sectional study also found that avoidance of seeking help due to fear of immigration officials was associated with increased psychological distress among a sample of Mexican immigrants, mostly women with a mean age of 43.

Present study

A critical next step of the existing literature is the examination of how Latina women’s perception of these multisystemic factors are associated with changes in their reports of psychological distress over time. Latina immigrants who have lived in the U.S. for an extended time period have been suggested to be more likely to experience psychological distress compared to more recently immigrated Latina women (Alegría, Álvarez, and DiMarzio 2017). Hence, we expected reports of psychological distress to increase during young Latina women’s initial three years in the U.S.

Specifically, we investigated whether women’s perceived social ecological factors reported soon after immigration would predict changes in psychological distress during women’s subsequent years in the U.S. At the cultural level, we hypothesized that women reporting more dominant society immersion (reflecting the aforementioned acculturation strategy of assimilation), less ethnic society immersion (reflecting the acculturation strategy of enculturation), and more acculturative stress at baseline assessment would indicate increases in psychological distress over time. Furthermore, based on the empirical and conceptual literature, we expected that women who endorsed marianismo beliefs of (a) spiritual pillar and (b) virtuous and chaste during the baseline assessment would experience less of an increase in psychological distress over time. We anticipated that women endorsing marianismo beliefs of (c) family pillar, (d) self-silencing to maintain harmony, and (e) subordinate to others during the baseline assessment would experience more of an increase in psychological distress over time. At the interpersonal/familial level, we hypothesized that women endorsing more trust and communication with parents, peers, and partners at baseline assessment would experience less of an increase in psychological distress over time. At the individual/microsystemic level, we hypothesized that women reporting (a) higher education and childhood socioeconomic status and (b) being married/partnered at baseline assessment would report less of an increase in psychological distress over time. Finally, we hypothesized that women with an unauthorized immigration legal status would experience more of an increase in psychological distress over time compared to authorized peers.

Materials and methods

Procedure

Data for the present study were drawn from a larger parent study of health determinants among recently immigrated young Latina women in Miami-Dade County, Florida, U.S. (see Sheehan et al. 2016 for a description of larger study). The present study focused specifically on psychological distress and its social and cultural determinants. The institutional review board of a large Southeastern public university approved the study. Eligibility criteria included: (a) identifying as a Latina woman, (b) being between ages 18 and 23 years, (c) immigrating to the U.S. from a Latin American country within 36 months of assessment, and (d) intending to stay in the U.S. for at least four years. The study consisted of three annual assessments: baseline (April 2013-April 2014), second (April 2014-April 2015), and third (April 2015-April 2016).

Recruitment

Respondent-driven sampling (Salganik and Heckathorn 2004) methods were used. Participants were recruited through announcements in neighborhoods, community agencies, and health fairs. Assessment interviews were conducted in Spanish by one of four research assistants who identified as foreign-born Latina women. Interviews occurred in a location chosen by participants or in university offices.

Participants

The baseline sample consisted of 530 young Latina women, 18–23 years (M = 20.81, SD = 1.80). Participants lived in the U.S. for an average of 11.52 months (SD = 9.94). Seventeen percent reported unauthorized immigration status. Approximately 57.2% had a high school education, 26.8% had some college, 4.3% earned a bachelor’s degree, and 1.1% held a graduate or professional degree. The majority of participants identified as Cuban (34%), followed by Colombian (11.1%), Nicaraguan (7.5%), Honduran (6.4%), Peruvian (5.6%), Mexican (5.1%), Venezuelan (4.9%), Ecuadorian (3.8%), Panamanian (3.6%), and Dominican (3.0%).

Measures

Measures used were either previously validated in Spanish (Brief Symptom Inventory; the Inventory of Parent and Peer Attachment) or translated to Spanish for the present study (Stephenson Multigroup Acculturation Scale; Social, Attitudinal, Familial, Environment Acculturative Stress Scale; Marianismo Beliefs Scale). Measures underwent a process of translation and back translation, modified direct translation, and checks for semantic and conceptual equivalence to ensure accurate translation from English to Spanish (Behling and Law 2000).

Psychological distress during initial years in the U.S.

Psychological distress was measured at each of the three annual assessments by the Global Severity Index (GSI) of the Brief Symptom Inventory (BSI; Derogatis 1993). The BSI is a 53-item self-report inventory that measures the extent to which individuals have been bothered in the past seven days by various symptoms using a 5-point Likert-type scale (0 = Not at all, 4 = Extremely). The BSI assesses nine symptom dimensions: Somatization, Obsession-Compulsion, Interpersonal Sensitivity, Depression, Anxiety, Hostility, Phobic Anxiety, Paranoid Ideation, and Psychoticism. The GSI is calculated using the sums for the nine symptom dimensions plus the four additional items not included in any of the dimension scores and divided by the total number of items to which the individual responded. If no items were skipped, the GSI score is the mean for all 53 items. The GSI score indicates overall level of psychological distress. Higher GSI scores indicate more problematic functioning due to distress. The English and Spanish version of the GSI have demonstrated evidence of reliability and validity (Derogatis and Fitzpatrick 2004; Ruipérez et al. 2001). The present study’s sample yielded Cronbach’s alpha estimates of .98, .99, and .98 for the GSI at each timepoint, respectively.

Cultural determinants of psychological distress

Assimilation and enculturation.

At baseline assessment, assimilation and enculturation levels were assessed by the Stephenson Multigroup Acculturation Scale (SMAS; Stephenson 2000) by Dominant Society Immersion and Ethnic Society Immersion subscales, respectively. The SMAS is a 32-item measure in which participants respond on a 4-point Likert-type scale (1 = False, 2 = Partly false, 3 = Partly true, 4 = True). Subscale scores were calculated by averaging across item responses. Subscales have evidenced acceptable reliability and validity (Ford and Kelly 2005). In the present study, the Ethnic Society Immersion (α = .93) and Dominant Society Immersion subscales demonstrated adequate internal consistency (α = .91).

Acculturative stress.

The Social, Attitudinal, Familial, Environment Acculturative Stress Scale (SAFE; Mena, Padilla, and Maldonado 1987) measured acculturative stress at baseline assessment. The SAFE is a 24-item measure that uses a 5-point Likert-type scale format (1 = Not at all, 5 = Extremely stressful). SAFE total scores were calculated by averaging across item responses. More acculturative stress is indicated by higher SAFE total scores. The SAFE has yielded psychometric support (Fuertes and Westbrook 1996; Hovey 2000). In this study, the internal consistency estimate for the SAFE total score was strong (α = .97).

Marianismo.

The 24-item Marianismo Beliefs Scale (MBS; Castillo et al. 2010) assessed the extent to which participants endorsed traditional gender role beliefs at baseline assessment. The MBS consists of five subscales: Family Pillar, Virtuous and Chaste, Subordinate to Others, Self-Silencing to Maintain Harmony, and Spiritual Pillar. Items are rated on a 4-point Likert-type scale (1 = Strongly disagree, 4 = Strongly agree). Subscale scores were calculated by averaging across items per subscale. Adequate psychometric evidence has been reported (Castillo et al. 2010). The present sample yielded adequate internal consistency estimates for the five subscales (α ranging from .88 to .97).

Interpersonal/familial determinants of psychological distress

Trust and communication with parents, peers, and partners.

During baseline assessment, participants completed trust and communication subscales of the Inventory of Parent and Peer Attachment (IPPA; Armsden and Greenberg 1987) to measure the quality of relationships between participants and their parents, best friend, and partner. Twenty-five questions from the Trust and Communication subscales were asked per relationship type using a 5-point Likert-type scale (1 = Never or almost never true, 5 = Always or almost always true). Items were averaged for each relationship type to arrive at a total score. Adequate reliability was reported for this measure in a sample of Latina women (De La Rosa et al. 2010). In this study, calculated total scores demonstrated support for strong internal consistency reliability (α = .95).

Individual determinants of psychological distress

Demographic variables.

During baseline assessment, participants provided their education level, perceived childhood socioeconomic status (When you were growing up, compared to other families, did you feel that your family was: 1 = Poorer than most, 2 = About the same as most, 3 = Richer than most), immigration authorization status (0 = Unauthorized immigration status, 1 = Authorized immigration status), and relationship status (0 = Single, 1 = Married/Partnered).

Attrition and missing data

Of the initial 530 participants, 508 completed the second assessment, and 505 completed the third (95% retention). To assess for potential sampling bias caused by attrition, we evaluated whether retained participants differed from non-retained participants on immigration authorization status, perceived childhood socioeconomic status, education level, relationship status, acculturative stress, enculturation, assimilation, marianismo, and psychological distress at baseline. No significant differences between retained and non-retained participants were observed.

The highest percentage of missing scale-level data for each variable was 1.1% (n = 6), which is below thresholds recommended to avoid biased results (Schlomer, Bauman, and Card 2010). Next, latent growth curve model (LGM) analyses were completed in Mplus 8.0 (Muthén and Muthén 1998–2017) using maximum likelihood estimation with robust standard errors (MLR) so that all available data were included (Enders 2011). The MLR estimator is more robust to data nonnormality observed than the regular maximum likelihood estimation method when missing data are present (Muthén and Muthén 1998–2017).

Analytic plan

Preliminary data analyses included (a) estimating descriptive statistics across time points and (b) examining distribution properties of continuous variables using suggested cutoffs of absolute values of 2.3 for skewness and kurtosis (Lei and Lomax 2005). A positively skewed distribution was observed for GSI scores only; therefore, a non-parametric test was used in aforementioned attrition analyses to detect potential sampling bias.

We next calculated LGMs to test (a) whether participants’ average level of psychological distress changed over time, and (b) whether hypothesized baseline predictors were associated with individual differences in average change of distress over time. First, participants’ distress trajectories were calculated via an unconditional LGM, with intercept and slope factors, for the GSI. Next, a conditional LGM was estimated by accounting for hypothesized baseline predictors of individual differences in average change of distress over time. To control for possible inflated type I error rates in the analyses of predictors of change, we applied the Benjamini-Hochberg false discovery rate procedure (Benjamini and Hochberg 1995) to calculate a corrected overall critical p value of .004. All estimated LGMs were evaluated according to standard structural equation model fit criteria: comparative fit index (CFI) values of ≥ .95 and root-mean-square error of approximation (RMSEA) and standardized root-mean-square residual (SRMR) values ≤ .05 represent excellent fit (Hu and Bentler 1999), with .90 representing the lower bound for acceptable CFI value, and .08 representing the upper bound for an acceptable RMSEA or SRMR value (Kline 2023).

Results

Descriptive statistics for all variables per time point are provided in Table 1. A bivariate correlation matrix containing all variables is presented in Table 2. Two Marianismo Belief Scale subscales (Subordinate to Others and Self-Silencing to Maintain Harmony) were found to highly correlate (r = .88). Thus, in accordance with recommendations to avoid effects of multicollinearity (Tabachnick and Fidell 2019), the subscales were combined by calculating an average of the two subscale scores for each participant. The new composite variable (Subordinate/Self-Silencing) yielded a Cronbach’s alpha of .97.

Table 1.

Descriptive statistics for study variables.

Variables	Assessment time point	Mean or %	Standard deviation
Brief symptom inventory global severity index
	T1	0.34	0.69
	T2	0.20	0.56
	T3	0.13	0.44
Individual determinants
Immigration Authorization Statusa	T1	83% authorized status	n/a
Relationship Statusb	T1	14.2% married/partnered	n/a
Education Levelc	T1	2.28	0.76
Perceived Childhood Socioeconomic Statusd	T1	2.00	0.48
Cultural determinants
Marianismo / Family Pillar	T1	3.27	0.65
Marianismo / Virtuous & Chaste	T1	2.98	0.71
Marianismo / Subordinate	T1	2.41	0.85
Marianismo / Self-Silencing	T1	2.21	0.92
Marianismo / Spiritual Pillar	T1	2.86	0.77
Acculturative Stress	T1	1.44	1.32
Enculturation / Ethnic Society Immersion	T1	3.25	0.63
Assimilation / Dominant Society Immersion	T1	2.59	0.71
Interpersonal/Familial Determinant
Trust and Communication with Parents, Peers, & Partners	T1	3.55	0.63

Note: T1 = First, baseline annual assessment, T2 = Second annual assessment, T3 = Third annual assessment.

^a0 = Unauthorized, 1 = Authorized.

^b0 = Unmarried/partnered, 1 = Married/partnered.

^c1 = Less than high school, 2 = Completed high school, 3 = Some college, 4 = Bachelor’s degree, or 5 = Postgraduate.

^d1 = Poorer than most, 2 = About the same as others, 3 = Richer than most; n/a = not applicable.

Table 2.

Descriptive statistics and correlations of study variables.

Variables	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
1. BSI – Global Severity Index (T1)	1
2. BSI – Global Severity Index (T2)	.06	1
3. BSI – Global Severity Index (T3)	.15**	.11*	1
4. Immigration Authorization Status (T1)a	.01	.08	.05	1
5. Marital/Relationship Status (T1) b	.04	.02	−.02	−.01	1
6. Education Level (T1)c	−.03	.01	.02	.04	−.02	1
7. Perceived Childhood Socioeconomic Status (T1)d	.06	−.03	−.01	.13**	.02	.05	1
8. Marianismo – Family Pillar (T1)	.05	−.01	.09*	−.07	−.03	.01	.03	1
9. Marianismo – Virtuous & Chaste (T1)	.03	.04	.05	−.10*	−.09*	−.06	.01	.69**	1
10. Marianismo – Subordinate (T1)	.01	.07	.02	−.13**	−.10*	−.05	−.04	.33**	.63**	1
11. Marianismo – Self-Silencing (T1)	−.01	.08	.02	−.18**	−.10*	−.03	−.08	.27**	.52**	.88**	1
12. Marianismo – Spiritual Pillar (T1)	−.02	.03	.04	−.14**	.08	.03	−.03	.63**	.66**	.60**	.57**	1
13. Acculturative Stress (T1)	.27**	−.03	.03	.17**	.05	−.06	.02	.23**	.01	−.33**	−.44**	−.12**	1
14. Enculturation / Ethnic Society immersion (T1)	.01	−.09	.01	.12**	.23**	−.04	.01	.32**	.10*	−.19**	−.28**	.06	.39**	1
15. Assimilation / Dominant Society Immersion (T1)	.03	.01	.06	.09*	−.07	.06	.06	.15**	.23**	.24**	.25**	.15**	−.12**	.32**	i
16. Trust and Communication with Parents, Peers, & Partners (T1)	−.06	−.12**	.07	.16**	.09*	.03	.06	.19**	.02	−.23**	−.32**	−.04	.30**	.48**	−.03

Note: (N = 530).

^**p < .001.

^*p < .05. BSI = Brief Symptom Inventory; n/a – not applicable. T1 = baseline assessment timepoint, T2 = 2nd annual assesment timepoint, T3 = 3^rd annual assessment timepoint.

^a0 = Unauthorized, 1 = Authorized.

^b0 = Unmarried/partnered, 1 = Married/partnered.

^c1 = Less than high school, 2 = Completed high school, 3 = Some college, 4 = Bachelor’s degree, or 5 = Postgraduate.

^d1 = Poorer than most, 2 = About the same as others, 3 = Richer than most.

Unconditional latent growth curve model for psychological distress

The unconditional latent growth curve model for the GSI yielded an acceptable SRMR (.02), an unacceptable CFI (.80), and an acceptable RMSEA (.04, 90% CI [0.00, 0.13]). Accordingly, we interpret the unconditional model as showing mixed fit and interpret findings cautiously. Although the CFI was lower than conventional standards, methodological work has noted that incremental fit indices such as the CFI may be difficult to interpret in latent growth curve models because they depend on the specification of the null model, which is not always the most appropriate comparison model for growth analyses (Preacher et al. 2008; Widaman and Thompson 2003).

The average intercept of the GSI was 0.32 (SE = 0.03, p < .001). The intercept variance was not significant, indicating little evidence of between-person variability in initial GSI levels (ψ < .01, SE = 0.05, p = .95). Slope estimates further indicated that, on average, GSI scores declined over time (b = −0.10, SE = 0.02, p < .001). The slope variance was also not significant, suggesting limited evidence of between-person variability in rates of change (ψ = 0.02, SE = 0.04, p = .54). Nonetheless, predictors of change were examined in a subsequent conditional model (see next section). This decision was guided by methodological recommendations indicating that tests of variance components are often underpowered, particularly in models with few time points, and that non-significant variance estimates therefore should not be interpreted as evidence of no meaningful heterogeneity (Curran, Obeidat, and Losardo 2010). More broadly, latent growth models are often extended to conditional models to test whether theoretically derived covariates account for differences in trajectories, regardless of the statistical significance of unconditional variance parameters (Bollen and Curran 2006; Duncan, Duncan, and Strycker 2006; Preacher et al. 2008). This rationale is particularly applicable to the present community-based sample, in which relatively low levels of psychological distress may have limited variability and attenuated slope variance estimates despite potentially meaningful individual differences in change over time.

Conditional latent growth curve model for psychological distress

The conditional model for the GSI yielded an acceptable SRMR (.04) and RMSEA (.04, 90% CI [0.00, 0.06]), but an unacceptable CFI (.81). Thus, we interpret the conditional model as demonstrating mixed fit and qualify conclusions accordingly. Consistent with the unconditional model, the relatively low CFI is interpreted cautiously because incremental fit indices such as the CFI may be difficult to interpret in latent growth curve models due to their dependence on the specification of the baseline/null model. The average intercept of the GSI in the conditional model was 0.74 (SE = 0.25, p = .003). Slope estimates in the conditional model also indicated that, on average, GSI scores declined over time (b = −0.43, SE = 0.15, p = .003).

As presented in Table 3, intercept estimates revealed that women reporting higher acculturative stress indicated higher GSI scores at baseline (β = 0.63, SE = 0.11, p < .001). Women experiencing higher acculturative stress at baseline tended to report greater declines in distress over time (β = −0.80, SE = 0.10, p < .001). Unexpectedly, women reporting more trust and communication with family, peers, and partners at baseline assessment reported less of a decline in GSI scores over time compared to their peers (β = 0.52, SE = 0.17, p = .002).

Table 3.

Results of conditional growth curve models testing determinants of psychological distress.

Parameter	Brief symptom inventory – Global severity index Standardized estimates (β)	Standard Error (SE)	p value
Individual determinants
Immigration Authorization Status a → Intercept	−.16	.09	.06
Immigration Authorization Status a → Slope	.09	.06	.12
Relationship Status b → Intercept	−.05	.06	.45
Relationship Status b → Slope	.02	.05	.70
Education Level c → Intercept	−.03	.10	.73
Education Level c → Slope	.06	.17	.74
Childhood Socioeconomic Status d → Intercept	.07	.24	.76
Childhood Socioeconomic Status d → Slope	−.14	.35	.70
Cultural determinants
Marianismo/Family Pillar → Intercept	−.06	.26	.81
Marianismo/Family Pillar → Slope	.45	.36	.21
Marianismo/Virtuous Chaste → Intercept	−.02	.17	.91
Marianismo/Virtuous Chaste → Slope	−.02	.26	.93
Marianismo/Subordinate-Self-Silencing → Intercept	.16	.21	.48
Marianismo/Subordinate-Self-Silencing → Slope	−.20	.30	.50
Marianismo/Spiritual Pillar → Intercept	−.17	.16	.31
Marianismo/Spiritual Pillar → Slope	.13	.25	.60
Acculturative Stress → Intercept	.63	.11	<.001
Acculturative Stress → Slope	−.80	.10	<.001
Enculturation /Ethnic Society Immersion → Intercept	−.23	.24	.34
Enculturation/Ethnic Society Immersion → Slope	−.01	.30	.97
Assimilation/Dominant Society Immersion → Intercept	.21	.17	.22
Assimilation/Dominant Society Immersion → Slope	−.06	.25	.80
Interpersonal/Familial determinant
Trust and Communication with Parents, Peers, & Partner → Intercept	−.31	.13	.02
Trust and Communication with Parents, Peers, & Partner → Slope	.52	.17	.002

^a0 = Unauthorized, 1 = Authorized.

^b0 = Unmarried/partnered, 1 = Married/partnered.

^c1 = Less than high school, 2 = Completed high school, 3 = Some college, 4 = Bachelor’s degree, or 5 = Postgraduate.

^d1 = Poorer than most, 2 = About the same as others, 3 = Richer than most.

Discussion

We examined changes in, and social ecological theory-based predictors of, psychological distress among foreign-born young Latina women during their initial years in Miami-Dade County, Florida. We investigated early immigration/baseline cultural, interpersonal/familial, and individual factors as predictors of psychological distress over time because our primary aim was to examine whether and how experiences during participants’ initial 1 to 36 months in the United States (i.e., range of baseline assessment time in U.S.) may influence subsequent trajectories of psychological distress. Women experiencing heightened acculturative stress also reported increased psychological distress at baseline assessment. During the subsequent three years, women’s psychological distress, on average, declined over time. Women indicating higher acculturative stress and less trust and communication with parents, peers, and partners at baseline assessment reported greater declines in distress over time. These findings highlight the importance of considering the time period soon after immigration as a potentially key contextual factor when studying mental health determinants among young adult, Latina immigrants.

Trajectories of psychological distress

Contrary to findings from past studies suggesting that Latina women may experience more distress the longer they reside in the U.S. (Alegría et al. 2008), young Latina women’s average level of psychological distress declined during their initial few years in the U.S. Participants’ psychological distress not only declined over time, but the highest levels of distress were reported at baseline assessment, soon after immigrating to the U.S. While our findings illustrated how young Latina women are at relatively higher risk for more distress during initial time period after immigration, declines in distress in subsequent years underscore a need to recognize and further study Latina women’s coping strategies and resilience. For instance, the reduction of distress observed in this sample is similar to a prior study of larger scope that examined trajectories of depressive symptoms over 13 years among disaggregated Latinx individuals by ethnicity, from youth into adulthood (Estrada-Martínez, Lee, and Shapiro 2019). They reported that decreases in depression occurred over time among first-generation immigrant Mexicans, Cubans, and Puerto Ricans into young adulthood. However, the decline was nonlinear in that depressive symptoms decreased, then increased, and later decreased over the 13-year trajectories for all Latinx subgroups. This nonlinear trajectory pattern was attributed to evolving social ecological factors (Estrada-Martínez, Lee, and Shapiro 2019), such as English language competency, discrimination, intra-group marginalization, contentment with decision to move to U.S., and sense of belonging, among others (Bekteshi 2024; Estrada-Martínez, Lee, and Shapiro 2019; Thornhill et al. 2022). Future research is encouraged to further investigate determinants over a longer time period and variability in coping and resilience with regard to psychological distress.

Determinants of change in psychological distress

The cultural factor of acculturative stress explained variability in women’s distress at baseline as suggested by past research (Lozano et al. 2024; Verdaguer et al. 2023). Lozano et al. ’s (2024) cross-sectional study similarly found that experiences of acculturative stress were related to elevated psychological distress among a sample of aggregate Latinx immigrant and non-immigrant parents, predominantly female-identified (63%). Verdaguer et al.’s (2023) cross-sectional study found that specific acculturative stress indicators (e.g., feelings of guilt related to leaving family or friends in home country, English language difficulties) were related to psychological distress among Mexican immigrants who were predominantly women (70%). In the present study, however, women experiencing more acculturative stress at baseline also experienced greater declines in distress. This result seems to be due to (a) the highest levels of distress and acculturative stress occurring at baseline and (b) the absence of bivariate associations between baseline acculturative stress and psychological distress at the second and third assessments. That is, baseline acculturative stress was predictive of declines in distress because the women who were reporting the most stress and distress at baseline also experienced the greatest declines psychological distress as they resiliently adjusted to the acute stress of immigration. It also is important to note some potential degree of regression to the mean in psychological distress over time may also explain why elevated baseline levels of acculturative stress unexpectedly predicted declines in psychological distress over time (Barnett, van der Pols, and Dobson 2005).

The declines in psychological distress observed in this sample may also have been influenced by immigration to a very distinct context of reception in the U.S.: Miami-Dade County, Florida. As we describe further in the below, participants may have experienced fewer social and environmental stressors than they would have if they immigrated to other contexts of reception (Salas-Wright et al. 2021). Nevertheless, the positive relations found between baseline enculturation and acculturative stress, as well as inverse relations between baseline assimilation and acculturative stress, suggest the participants reported attitudinal and familial acculturative stressors was substantive and related with acculturation processes as would be expected. Hence, findings do suggest the need for early interventions such as during the initial weeks and months after immigration to the U.S. to prevent the heightened acculturative stressors and psychological distress observed during this time period.

We also found that women reporting higher levels of interpersonal/familial trust and communication at baseline unexpectedly experienced less of a decline in distress than peers. Over time, supportive home country relationships may have lost their theorized protective influence against distress (Ryan et al. 2021). Gradual loss of connection with home country social support over time may induce feelings of guilt and loss (Verdaguer et al. 2023). Alternatively, this finding may reflect the notion that closer relationships with family, peers, and significant others and may introduce additional stressors, such as caring for others and witnessing loved one’s distress (Alegría, Álvarez, and DiMarzio 2017, 2022). Hence, research investigating changes in types and levels of microsystemic relationships over time during the immigration process are needed.

Limitations, constraints on generality, and implications for future studies

The contributions of this study should be considered alongside its limitations and implications for further study. Expansion of this study’s design is needed to assess the external validity of results while considering different contexts of reception and other unmeasured predictors of mental health. The present sample was community-based and non-clinical. It was also drawn from Miami-Dade County using respondent-driven, convenience sampling. Future studies could improve upon this one by examining psychological distress and its correlates among clinical samples with higher levels of distress. Also, if future studies use respondent-driven sampling, we recommend ensuring that participants are reasonably independent of one another to avoid violating assumptions of analyses. Moreover, the non-significant effects may vary in different study settings with more inclusive samples (Estrada-Martínez, Lee, and Shapiro 2019). For instance, although attempts were made to recruit participants from various origins, Mexican, Dominican, and Puerto Rican women were underrepresented, while Cuban and Colombian women were highly represented (as they are in Miami-Dade County, Florida). In addition, Miami-Dade County is a Latinx ethnic enclave, particularly for individuals of Cuban descent. Ethnic enclaves promote increased social integration with larger and more diverse social networks (Viruell-Fuentes et al. 2013); thus, the study’s setting may be protective against acculturative stress and psychological distress. However, research has noted that aggregate Latinx samples residing in neighborhoods with larger Latinx immigrant communities may be at increased risk for anxiety (Alegría, Molina, and Chen 2014), possibility due to disadvantaged economic and social conditions and other contextual forces that impact Latinx health (e.g., racial profiling; Alegría, Molina, and Chen 2014). Hence, future multi-site, studies of distress are needed focusing on experiences of Latinx subgroups residing in diverse neighborhoods in terms of ethnoracial composition, as distress levels are likely to vary among groups (Barragán et al. 2020) and by type of neighborhood characteristics (Alegría, Molina, and Chen 2014). Furthermore, the present study did not examine all acculturation processes, such as bicultural identity and marginalization. Inclusion of these dimensions may have provided more nuanced understandings of acculturation processes and distress over time. Adopting a bicultural identity (i.e., integrating one’s cultural heritage with U.S. mainstream culture), for example, facilitates healthier adjustment to the U.S. across studies with Latinx samples, mostly composed of only or predominant Mexican samples (87 out 152 studies; Safa and Umaña-Taylor 2021). Next, because data collection was conducted between 2013-2016, the external validity of findings may be affected by more recent macrosystemic, social ecological influences affecting distress for newly immigrated young Latina women in the U.S., such as increased anti-Latinx immigrant policies and recent changes in Latinx immigration patterns. Many more persons from Southern and Central American countries (e.g., Venezuela and Guatemala) sought asylum from increased political turmoil, crime, and poverty in their countries of origin after the conclusion of this study (Vazquez et al. 2022). Finally, scholars have called for studies of the roles that gender, race, and ethnicity play on the post-migration adjustment and health of Latinx in the U.S. (Capielo Rosario et al. 2021; Valdez and Tran 2019). Therefore, research should consider how factors such as skin color and ethnoracial identification may also influence the post-migration health of young Latina immigrants across varied contexts of reception (Valdez and Tran 2019).

Overall, to inform and promote the mental health of young Latina women, we encourage future researchers to study the importance of the initial years in the U.S. after immigration. Future research is encouraged to integrate additional health determinants (e.g., discrimination, access to mental health care) into longitudinal studies. Culturally-informed interventions also are encouraged soon after immigration to safeguard the mental health foreign-born young Latina women. Early interventions may be especially effective because of the detrimental effects of acculturative stressors soon after immigration.