Ethnic Identity and Mechanisms of Mental Health Service Engagement Among Young Adults with Serious Mental Illnesses

Abstract

Purpose

Young adults from minoritized racial and ethnic groups have lower rates of engagement in treatment for serious mental illnesses (SMI). Previous research suggests a relationship between ethnic identity development and engagement in mental health services, but it remains unclear how a sense of belonging and attachment to one’s racial and ethnic group influences participation in treatment among young adults with SMI.

Methods

Bivariate analyses and structural equation modeling (SEM) were used to examine whether ethnic identity was associated with treatment engagement (attendance and investment in treatment) and how ethnic identity might influence engagement through theoretical proximal mediators. Eighty-three young adults with SMI (95% from minoritized racial and ethnic groups) were recruited from four outpatient psychiatric rehabilitation programs and assessed at least 3 months after initiating services.

Results

Stronger ethnic identity was associated with greater investment in treatment but not with treatment attendance. The SEM analysis indicated that stronger ethnic identity may improve investment in treatment by enhancing hope (0.53, p < .05) and beliefs that mental health providers are credible (0.32, p < .05), and by increasing self-efficacy (−0.09, p < .05). Proximal mediators of engagement were associated with investment in treatment (hope and credibility, p < .05, and self-efficacy p = 0.055).

Conclusions

Findings provide preliminary evidence of an empirical and theoretical relationship between ethnic identity development and engagement in treatment among young adults with SMI. Assessment and strengthening of a young person’s ethnic identity may be a promising approach for improving their engagement in services and reducing inequities in their care.

Introduction

Young adults with serious mental illnesses (SMI; e.g., schizophrenia-spectrum and mood disorders) often face significant challenges as they transition to adulthood, such as lower educational attainment, unemployment, homelessness, incarceration, and increased risk of suicide [1–3]. Engaging in treatment consistently can help prevent negative outcomes during this developmental period. However, young adults with SMI are less likely to use mental health services compared to other age cohorts in the USA [4], particularly if they are from groups other than non-Latinx White [5].

Young adults from minoritized racial and ethnic groups are more likely to disengage from treatment [6, 7], with studies suggesting that those with lower socioeconomic status are at greater risk for premature termination of treatment for serious psychiatric conditions [8–11].

Barriers to consistent, effective mental health treatment among young adults include previous negative treatment experiences and mistrust of healthcare systems; cognitive and affective factors such as fear, stigma, and negative beliefs about treatment; and cost; and limited access to high-quality services and to services in languages other than English [12–17]. Minoritized young people also experience additional burdens to receiving appropriate care through interpersonal and structural discrimination that further engenders medical mistrust [ 18–20] and stigmatizing beliefs and attitudes toward seeking mental health treatment [21–23]. Young adults have also reported deterrents to ongoing engagement with treatment due to feeling that providers misinterpret them and their life experiences [14, 24]. Such perceptions of misalignment with providers may be intensified among minoritized young adults given that mental health providers from minoritized racial and ethnic groups are underrepresented, making up only 25% of U.S. providers [25, 26]. Even evidence-based programs that were designed to improve psychiatric and functional outcomes among young adults with SMI continue to see lower rates of treatment engagement (i.e., initiation and retention in care) among minoritized young people and their families and report limited capacity to address their cultural needs [27–31]. Research aimed at understanding how to make mental health services more responsive to and effective for young people with SMI from minoritized racial and ethnic groups is seriously needed to improve their service utilization and life outcomes. This study examined the relationship between ethnic identity development and treatment engagement processes among young adults with SMI in order to advance the evidence base for culturally responsive engagement approaches for this population.

Ethnic Identity as a Predictor of Young Adult Engagement with Mental Health Services

Ethnic identity processes are important to consider when examining contextual factors related to engagement in mental health care among young adults. Ethnic identity refers to one’s sense of self in terms of culture, race, language, and kinship [32, 33] and the quality of belonging and attachment to one’s ethnic group [34]. Ethnic identity, along with other aspects of one’s cultural context, shapes the way individuals perceive mental illness, treatment, and engagement in services [35–37]. Both culturally influenced beliefs (e.g., beliefs about mental illness etiology, medical mistrust, stigma) and cultural norms (e.g., religious, peer, and family norms for treatment seeking) affect help-seeking behaviors and treatment outcomes [36, 38, 39]. Recent research has shown that interventions that employ practices designed to explore ethnic identity and culture have been associated with improved engagement outcomes among youth in mental health services [40, 41]. However, just as cultural values and practices shared among a demographic group may vary in salience among individuals within that group, the extent to which individual young people have a sense of belonging and connection to their ethnic groups differs and may fluctuate over time and across social contexts [42, 43]. As such, interventions that are culturally adapted to improve engagement among young people from minoritized ethnic groups need to address these dynamic aspects of individual ethnic identity in order to be effective [41, 44].

Ethnic identity processes are of particular relevance in young adult services because identity formation is a key developmental task of the transition to adulthood, often involving intensive exploration of social identities that inform one’s sense of self and direction in life [45–47]. Key components of ethnic identity include self-identification as a group member; perceptions, behaviors, and feelings derived from one’s group membership; degree of ethnic knowledge; and involvement in ethnic social and cultural practices [48]. This is also a critical period where individuals are aging out of children’s health systems [49–51] and become independent decision-makers regarding their own treatment [52]. Decisions about whether to engage in mental health care are particularly precarious in early adulthood because symptom onset occurs around this time [53, 54], often leading to constraints on identity exploration [55] and thwarted developmental and life transitions [1, 2]. Developing a stronger sense of ethnic identity during adolescence and early adulthood has been associated with psychological well-being and protective effects on mental health, particularly among young people from racial and ethnic minoritized groups [56–58]. However, recent studies of young adults have linked ethnic identity with mental illness stigma [59, 60] and with less favorable attitudes toward professional mental health help-seeking [61, 62] among those from minoritized groups.

Very few studies have examined the association of ethnic identity with engagement in mental health services. Two studies found that among nationally representative samples of adults with any lifetime psychiatric disorder, U.S. Black and Hispanic individuals with a stronger ethnic identity were less likely to report psychiatric service use [63, 64], even after controlling for numerous factors associated with service utilization including insurance, income, symptom severity, and age at immigration (i.e., acculturation). In a large community sample of Black and White U.S. adults, Richman et al. [65] found that Black adults were less likely to use mental health services if they had high racial identity centrality (i.e., believing that their racial group is an important part of their self-image) than those with lower racial identity centrality. Conversely, higher racial identity regard (i.e., a person’s pride in or perceived positive judgement of their own race) was associated with more mental health care utilization among Black participants. Findings suggest that individuals may be more likely to seek care when they expect their racial group to be well-regarded and less likely to seek care when mental health services are not perceived as inclusive of or relevant to their racial identity. In a study comparing Black and White American adolescent girls, Yasui et al. [66] found that ethnic identity was related to mental health service utilization only for Black girls, and those with a stronger sense of belonging to their ethnic group were less likely to use mental health services. The authors suggested that Black girls who strongly identify with their ethnic group may find using mental health services to be incongruent with Black female cultural emphases on selfreliance and emotional regulation, and reflective of public stigma of mental illness observed in the Black community, which deters them from seeking services. These studies appear to capture cultural norms that have been shaped over time in response to an inequitable mental healthcare system, and systemic racism more generally. For instance, societal and medical discrimination may have left few alternatives to self-reliance for Black girls in need of mental health care. Overall, research suggests that minoritized individuals with stronger ethnic identities may be more intensely affected by culturally influenced norms and beliefs that present barriers to mental health service use.

Three recent studies examined relationships between ethnic identity and mental health service engagement among young adults. In a mixed methods study of Black and Latinx young adults who had experienced a mental health condition within the past 2 years, Moore et al. [67] found no differences in ethnic identity between participants who completed treatment and those that disengaged prematurely, although participants reported mental illness stigma and lack of culturally knowledgeable providers as barriers to initiating and continuing service use. Williams et al. [68] found that ethnic identity did not predict the use of professional mental health services after controlling for regional differences among Black young adults. The authors suggested that larger and closer-knit Southern U.S. Black communities offer more non-medical options for early intervention, such as churches and extended family, which may be more appealing and effective because they are aligned with cultural preferences for informal help. Similarly, Turner and Llamas [69] found no significant differences in ethnic identity among Latino/a college students with and without previous mental health service use. However, study results suggest that participant views about mental health services were influenced by their level of religious affiliation, with stronger religiosity associated with being less likely to use mental health services, which is consistent with culturally influenced mental disorder etiology (e.g., due to a moral/spiritual deficiency) and preferred spiritual healing practices among Latinx individuals [70, 71].

Taken together, these data imply that there are factors that impact the relationship between ethnic identity and engagement in mental health services among young adults, but there is a lack of research to precisely determine them. In addition, previous studies assessed relationships between ethnic identity and the most commonly reported behavioral dimensions of engagement with treatment, namely consulting with a mental health professional and treatment attendance [72]. However, treatment engagement is multidimensional, involving social, cognitive, and affective dimensions such as the degree of active involvement and investment in treatment that a person exhibits when they attend and the working alliance that they have with treatment providers [40, 73]. Examining belonging and attachment to one’s ethnic group in relation to these important aspects of treatment can yield vital evidence for how ethnic identity can influence engagement and introduce novel targets for intervention to increase service utilization among underserved and minoritized young adults. Yet, to our knowledge, no studies to date have included multiple dimensions of engagement when examining ethnic identity as a predictor of using mental health services. The current study will extend previous research to focus on the relationship between ethnic identity and two developmentally relevant dimensions of treatment engagement among young adults with SMI; attending services and investment in treatment.

Theory of Young Adult Service Use—Mechanisms of Engagement

The young adult framework of mental health service use [14] guides our investigation. The “young adult framework” integrates health behavior change, formal decision theory (e.g., Theory of Planned Behavior [74]), and elements of mental health service use theories to elucidate the main mechanisms of engagement identified in research with young adults with SMI [75]. Contextual factors at the social and community levels include important family and peer relationships, and the quality of interactions with providers. Other environmental factors include access issues such as availability of services, cost, transportation, and competing demands on one’s time (e.g., education, employment, care-giving). Individual-level factors that underlie mental health service utilization in young adults include (1) behavioral beliefs (perceived advantages of treatment, credibility of providers), (2) image management (perception of stigma surrounding service use), (3) hope (level of optimism), (4) self-efficacy (perceived ease and confidence in performing engagement behaviors, such as attendance and ongoing participation in treatment), (5) emotional reactions to treatment (e.g., fear), and (6) social norms (approval from important others) (see Fig. 1) [14]. The framework posits that in order to change the outcome of engagement in treatment, an intervention must first change one or more of these underlying mechanisms of action affecting the outcome. Recent research on engagement in mental health services has incorporated elements of experimental therapeutics from personalized medicine [76], which emphasizes specifying the processes by which intervention effects occur—what has been deemed the “science of how”—to better understand the mechanisms underlying service use behaviors [77, 78]. The young adult framework incorporates these principles and evolved from a study of young adults with SMI to provide the first developmentally specific service use framework delineating empirically-based mechanisms of engagement (Munson et al., 2012).

Fig. 1.

Young adult framework of mental health service use, including ethnic identity as a contextual factor

The mechanisms of the young adult framework were validated as proximal mediators to engagement in a prior clinical trial [79], and the framework has been applied in several other studies among young adults with SMI (see [51, 80]).

The Current Study

Our study’s purpose is to understand how a sense of belonging, pride, and involvement in one’s ethnic group might influence engagement in mental health services among young adults with SMI. The aforementioned studies have independently investigated associations among some relevant factors in young people and adults, and among individuals from minoritized ethnic and racial groups, but we did not find any studies that examined possible mechanisms of service use to assess the relationship between ethnic identity and engagement outcomes among young adults with SMI. As such, this study addressed two exploratory research questions: (1) What is the relationship between ethnic identity and engagement in mental health services (i.e., attendance and level of investment) among young adults with SMI; and (2) how does ethnic identity influence engagement in mental health services? Based on the small number of prior studies indicating that a stronger, more developed ethnic identity is associated with lower rates of mental health service utilization among large samples of adults living with a psychiatric disorder [63–65] and that overall differences in service use among studies of young adults might be indirectly related to participant ethnic identity [67–69], we hypothesized that ethnic identity would be negatively associated with engagement in our study. However, given that prior studies only measured help-seeking or attendance dimensions of engagement (e.g., “sought help from a professional in the past year” or “any lifetime service use”), we found insufficient empirical evidence to form consistent hypotheses concerning the relationship between ethnic identity, mechanisms of engagement, and level of investment in services among young adults. To investigate our research questions, we treated ethnic identity as a contextual factor that influences engagement, mediated by theoretical mechanisms identified by the young adult framework. Although it was not originally specified by the young adult framework, our review of prior research suggests that ethnic identity is a potentially important contextual factor for minoritized young adults. Our approach, described in detail below, developed an influence diagram of a subset of potential mediators to help explain how ethnic identity may impact engagement, ultimately using structural equation modeling (SEM) to examine the relative contribution of each mediator simultaneously.

Methods

Study Participants

Participants for the present study were part of a larger study conducted between 2018 and 2020 that recruited 121 young adults from four personalized recovery-oriented services (PROS) programs located in low-resourced communities in the Eastern United States. PROS programs serve adults with serious mental illnesses and offer therapeutic and skill-based groups, along with individual counseling, pharmacological services, and resource coordination for transportation, housing, employment, and other supports.

PROS program intake staff introduced the research study to potential participants. Young adults that were interested in participating contacted the onsite research staff, who explained the study and received written consent from all participants. Eligible young adults were within the 18–34 age range, consistent with the range defined by the US Census Bureau [81], and reflective of research indicating that key developmental milestones for young adults in the USA are occurring at older ages [82]. Additional eligibility criteria included (1) living with a serious mental illness, (2) enrolled in PROS or in the intake process, and (3) current or history of receipt of social services (e.g., Medicaid, supplemental nutrition assistance program) or involvement in juvenile justice or foster care systems. The present study uses data collected in the final assessment of the larger study that measured ethnic identity in 83 of the original 121 young adults. Further details on the larger study and data collection plan are described in the published study protocol [83]. The (The New York University Committee on Activities Involving Human Subjects approved all study protocols) University committee on the protection of human subjects approved all study protocols.

Measures

Demographics

Age, gender, ethnicity-race, and mental health diagnoses were self-reported. Ethnicity-race was categorized into four groups that matched participant self-reports: Black (non-Latino/a), Latino/a, multiethnic/multiracial (reporting more than one group), and White (non-Latino/a). Electronic health records were also used to assess primary psychiatric diagnoses. The Center for Epidemiological Studies–Depression Scale (CES-D; [84]) was used to measure current depression, with higher scores indicating the presence of greater symptomatology. Cronbach’s alpha for the CES-D with this sample was 0.88.

Engagement

The present study examines two distinct dimensions of engagement: (1) treatment attendance and (2) investment in treatment. Attendance was measured as the number of days that a young adult attended PROS over the past 2 weeks. Recommended attendance for PROS programs is 3 days per week. A modified version of the widely used Yatch-menoff engagement measure [85] was used as a multidimensional measure of engagement that centers social and cognitive dimensions. Investment in treatment is characterized as commitment to the helping process, active participation in services, and initiative in using help. The sum of 8 items measured investment; each rated on a 5-point disagree-agree scale (total score from 8 to 40). Sample items include “I am not just going through the motions. I’m really involved in working with PROS,” and “I really want to make use of the services PROS is offering me.” The Cronbach’s alpha for investment in treatment for this study was 0.83.

Ethnic Identity

Feelings of affirmation and belonging toward one’s ethnic group (e.g., “I have a clear sense of my ethnic background and what it means to me” and “I have a strong sense of belonging to my own ethnic group”) were measured by the 12-item Multigroup Ethnic Identity Measure (MEIM; [86]. Cronbach’s alpha for this study was 0.89. The first item of the scale asks participants to complete an open-ended statement: “in terms of ethnic group, I consider myself to be…” The following items are answered on a 4-point scale (strongly disagree–strongly agree; range = 12–48), with higher scores indicating a stronger, more developed ethnic identity. The total score was used for analysis. The 12-item MEIM has shown good reliability (0.67–0.91) in samples of Black, Latino/a, and White young adults [87, 88] and in adolescents of diverse ethnic and racial groups [86].

Mechanisms of Engagement

We measured six categories of empirically identified mediators of engagement in mental health care. Five categories (behavioral beliefs about perceived advantages/disadvantages of treatment (α = 0.85), emotional reactions to treatment (α = 0.85), social norms (α = 0.78), image management/stigma (α = 0.83), and self-efficacy (α = 0.76)) relied on strong psychometric approaches tied to decision theory constructs [89], using five-point disagree-agree scales. The psychometric properties of these measures were tested in previous studies with racially and ethnically minoritized young adults with mental health conditions and proved favorable in terms of alpha coefficients and low correlations with social desirability bias [75, 79]. Examples include the following: “Continuing my treatment will provide me with non-judgmental support” (perceived advantages); “When I think about continuing my treatment it makes me anxious” (emotional reactions); “If others who are important to me found out that I continue with my treatment, they would see me as crazy” (image management/stigma—higher scores indicate lower stigma); “How would your father feel about you continuing treatment for your issues at this time in your life?” (social norms—these were measured with a 6-point approval/disapproval scale and then averaged across 8 referents); “It would be difficult for me to continue my treatment sessions” (self-efficacy—lower scores indicate higher self-efficacy). We also assessed perceptions of the credibility of providers (e.g., “I feel like my current provider truly wants what is best for me” and “My current provider is honest with me”) (α = 0.90). Finally, we used a standardized measure of “generalized hope” [90] that had 12 items, each rated on an 8-point Likert scale (e.g., “I can think of many ways to get the things in life that are important to me”) (α = 0.75).

Analytic Strategy

Data collection for this study was originally powered to detect medium effect sizes. However, the pandemic occurred during data collection and a reduced sample size resulted (N = 83). This reduced the planned effect size sensitivity of the data to an ability to detect with a power of 0.80 population effect sizes for a predictor of approximately 7% unique explained variance or greater in a multiple regression context.

Preliminary analyses were conducted in SPSS 28.0 [91]. In addition to frequencies, bivariate correlation analyses were conducted to assess the relationships between ethnic identity, mediators, and engagement (see Table 1). Due to small sample size, the investigative team used theory and bivariate analyses to select the most meaningful subset of mediators, namely, those found to be significant at the bivariate level. Structural equation modeling (SEM) was completed using Mplus 8.0 [92] to obtain estimates of path coefficients linking ethnic identity scores to both treatment engagement and mediators. The analysis also included a baseline measure of engagement in the young adults’ last treatment experience as a covariate to control for unmeasured confounds. We used full information maximum likelihood [93] to treat missing data and robust estimation (Huber-White estimation) to address non-normality. The SEM model was just identified because of the inclusion of correlated disturbances for the mediators. Correlated disturbances between two variables are called for if there is an unmeasured omitted variable from the model that influences each of the variables in question.

Table 1.

Descriptives and correlations between primary study variables

Variables	EI	Eng. (Inv.)	Eng. (Att.)	BB TX	BB Cred.	IM stigma	Norms	Hope	SE	ER
Ethnic identity	1.0	.48***	.15	.17	.45***	.15	.15	.30**	−.28**	−.03
Engagement (investment)	.48***	1.0	.26*	.66***	.78***	.45***	.38***	.50***	−.46***	.413***
Engagement (attendance)	.15	.26*	1.0	.16	.21	.31**	−.03	.20	−.10	−.19
Range	12–48	8–40	1–5	0–52	8–40	0–44	0–6	12–96	3–15	0–28
Mean	36.96	34.27	3.20	41.08	35.63	31.23	4.90	65.33	4.76	19.72
SD	7.47	5.47	1.36	9.13	5.47	8.42	1.02	13.11	2.33	7.42

BB TX, perceived advantages/disadvantages of treatment; BB Cred., credibility of providers; IM stigma, image management/stigma; Norms, social norms; Hope; SE, self-efficacy; ER, emotional reactions to treatment

^*p < .05,

^**p < .01,

^***p < .001

Results

Sample Characteristics and Descriptive Statistics

Participants ranged in age from 18 to 34 (M = 27.17, SD = 3.38), with most between ages 23–30 (n = 58, 69.8%), and the majority of participants were from minoritized ethnic-racial groups (n = 79, 95.1%). Thirty-seven participants were Black (44.5%), 29 were Latino/a (34.9%), 13 were multiracial/multiethnic (15.6%), and four were White (4.8%). Most were male (n = 53, 63.8%) and reported a schizophrenia spectrum disorder as their primary diagnosis (n = 55, 66.2%). Twenty-eight participants reported mood disorders as their primary diagnosis (33.7%), and most participants (n = 61, 74.3%) had depression symptomology at risk for clinical depression at the time of the interview (CES-D ≥ 16).

The sample consists of individuals who receive assistance from Medicaid, Medicare, and/or Social Security Disability Insurance due to low income and psychiatric disability. Participants also reported receipt of at least one federally-funded safety-net service (e.g., Supplemental Nutrition Assistance Program, foster care, juvenile justice, and/or special education). Thirty-four participants (40.9%) reported having experienced homelessness. Most of the sample reported previous inpatient hospitalization (n = 75, 90.3%), and 44.5% reported at least one suicide attempt (n = 37). Forty participants (48%) completed high school or a GED, 22 (26%) reported having completed at least some post-secondary education, and 21 (20.7%) reported not completing high school. Seven participants (8.4%) were currently employed, and six (7.2%) were currently in school at the time of the study.

Preliminary Analyses

Descriptive statistics and bivariate correlations among core study variables are shown in Table 1. Ethnic identity was significantly correlated with the level of investment in mental health services (r = 0.48, p < .001). However, ethnic identity was not correlated with the number of days that the young adult attended PROS (r = 0.15, p = .20), so our analyses focus on potential mediators of the level of investment. Behavioral beliefs about treatment, emotional reactions to treatment, social norms, and stigma, were not significantly correlated with ethnic identity (p > .05). Credibility of providers, hope, and self-efficacy were significantly associated with ethnic identity and thus were selected for inclusion in a multivariate structural equation model predicting level of investment in treatment. In addition, ethnic identity scores were highest among Black (M = 3.21, SD = 0.62), Latino/a (M = 3.10, SD = 0.61), and multiracial (M = 2.79, SD = 0.61) participants compared to White participants (M = 2.60, SD = 0.41), but the mean differences were statistically non-significant (p > .05).

Structural Model

SEM was applied to the model shown in Fig. 1. A stronger, more developed ethnic identity was associated with improvement in the investment dimension of engagement (correlation = 0.46, p < .05). The SEM analysis was consistent with a model suggesting that a stronger ethnic identity is associated with change in three mechanisms of engagement: hope (0.53, p < .05), self-efficacy (−0.09, p < .05), and credibility of the treatment providers (0.32, p < .05) (Fig. 2). Further, these theoretical mechanisms of engagement were associated with engagement (hope and credibility, p < .05, and self-efficacy p = 0.055). The SEM analysis was consistent with the proposition that stronger ethnic identity improves engagement in treatment, specifically one’s investment in treatment, by enhancing one’s optimism and self-efficacy, and increasing one’s confidence in mental health providers.

Fig. 2.

Results of the structural model

Discussion

The current study is the first to examine the relationship between ethnic identity processes and two distinct dimensions of engagement with mental health services, attendance and investment, among young adults with SMI. The study considered theoretically grounded, proximal indicators of service use specific to young adults that might act as mechanisms through which ethnic identity influences treatment engagement. Results suggest that a stronger, more developed ethnic identity might promote psychological commitment and involvement in treatment during early adulthood. Findings highlight how the development of one’s ethnic identity and attachment to an ethnic group may be important for engagement through supporting a sense of optimism and agency in treatment and forming positive relationships with providers. Results have implications for service delivery among young adults with SMI and clarify individual and interpersonal engagement mechanisms relevant to young people from minoritized racial and ethnic groups that can inform future research and interventions designed to improve their treatment experiences and overall health outcomes.

A stronger ethnic identity was associated with increases in the investment dimension of engagement in our multivariate model but was not significantly related to the attendance dimension. This finding partially supports our hypothesis and aligns with prior studies of young adults in which ethnic identity was not associated with mental health service use behaviors [67–69]. The relationship between ethnic identity and treatment attendance, however, was found in prior research among samples including older age groups of adults [63, 64] and younger adolescents [66], suggesting that there may be circumstances unique to young adulthood that account for the difference. For instance, Munson et al. [94] found that session attendance was not a significant predictor of self-reported recovery among young adults with SMI when controlling for investment in treatment. Consistent with multidimensional concepts of engagement, simply attending services more frequently may not be sufficient to attain recovery goals that matter to young adults and may not be a strong single predictor of improved outcomes among this age group.

Active investment and commitment to treatment are likely needed to overcome common barriers to service use among young adults such as limited access, adverse experiences, and negative beliefs about treatment, particularly at this developmental stage when they are gaining greater autonomy and taking on more responsibility for managing their own healthcare. Young adulthood is a time when individuals are moving from school-based, child welfare or family-initiated mental health services into self-initiated treatment, where they will need to be motivated by their own recovery goals to sustain involvement. But young adults, especially those from minoritized groups, often have to endure a prolonged process of trial and error as they attempt to locate affordable and available providers that can understand and work with the challenges that they are facing [67]. It is easy to imagine that this struggle would negatively impact the frequency of session attendance, even among young adults that feel highly committed to receiving help with their mental health. Our findings indicate that ethnic identity development might offer psychological strengths that reinforce young adults’ ability to invest in their own treatment and can help them persist when they encounter barriers to service utilization.

Ethnic identity seemed to influence treatment investment through positive associations with hope and self-efficacy, two theoretical mechanisms of service use posited by the young adult framework [14] that likely reflect the psychological strengths among study participants that are important for engagement. Our results appear to align with prior research on psychological well-being among White, Black, and Latinx youth, in which a stronger sense of belonging to one’s ethnic group, along with positive attitudes about that group, have been associated with optimism [33, 86], positive future orientation [95], and mastery [96, 97]. Among our study participants, ethnic identity was associated with hope, conceptualized as general cognitive factors indicating the ability to determine and generate ways to achieve a desired goal [90]. A few prior studies have found a positive association between ethnic identity and hope among diverse groups of adolescents and young adults, linking stronger ethnic identity to increased positive goal-seeking and problemsolving behaviors [98–100].

Concurrently, ethnic identity was associated with reporting greater perceived ease and confidence in performing engagement behaviors among participants. This parallels studies that have demonstrated predictive relationships between ethnic identity and other types of self-efficacy, such as career decision-making [88] and psychological empowerment [101] among young adults and adolescents from minoritized groups. In these studies, stronger ethnic identity was associated with greater confidence in accomplishing important life tasks and having greater perceived levels of control within their environment. Overall, such studies have suggested that individuals with a positive ethnic group affiliation are less likely to internalize negative stereotypes about their group that could undermine their confidence in their capacity to persevere in under-resourced environments and to mobilize needed supports to achieve their goals.

Ethnic identity also appeared to influence treatment investment through a positive association with provider credibility. This finding is notable given prior research that shows distrust and perceived misalliance with providers to be significant barriers to mental health service engagement among young adults, particularly those from minoritized groups [15, 24]. Limited prior research has explored how ethnic identity development informs the process of therapeutic interactions in mental health care, indicating that associations between ethnic identity and perceptions of provider competency vary and often involve multiple factors (e.g., racial/ethnic concordance, racial/ethnic identity development of both patients and providers) [102, 103]. In the current study, it is possible that results reflect important characteristics of the providers working with participants. PROS programs attended by study participants reported that 67% of their clinical staff was composed of providers from Black and Latinx racial-ethnic groups, with 40% able to provide services in Spanish. This is considerably higher than nationally reported estimates [25, 26], suggesting that study participants (95% were Black, Latino/a, or multiracial) were more likely to be working with providers with a similar racial or ethnic background to their own. Some research on racial/ethnic concordance and psychosocial treatment outcomes has shown that patients from minoritized groups tend to perceive providers of their own ethnicity more favorably, but being matched with a provider based on only one aspect of identity does not guarantee that there will be shared understanding or greater trust [104].

Another important characteristic of PROS is the multidisciplinary staff that includes peer specialists, who are individuals with lived experience of having a mental health condition and of being a service user. There is evidence from prior research on peer providers in mental health services for young adults to suggest that racial/ethnic concordance contributes to the formation of trusting relationships and increased rapport among clients and their peer providers [105]. This also appears consistent with research showing that young adults from marginalized groups tend to prefer informal supports for help with mental health [68, 69], as peer specialists often utilize more informal approaches to working with clients and may be perceived as more culturally competent. The presence of providers who were from similar ethnic backgrounds, including some who had a mental illness, may have signaled to young adult participants in our study that providers were better equipped to understand their experiences and to appreciate and affirm their ethnic identities. Our findings seem to support research suggesting individuals with stronger ethnic identity are more likely to engage when services are perceived as inclusive of or relevant to their ethnic identity [65].

Ethnic identity was not associated with several proposed mediators of engagement among participants in the current study. As such, our findings did not support studies that have previously linked ethnic identity to beliefs [61, 62], emotions [69], social norms [66], or stigma [59] toward professional mental health service use among those from minoritized U.S. racial and ethnic groups. However, research has shown that stronger ethnic identity may involve greater identification with traditional ethnic values and social customs among young adults [106] that can differ widely between ethnic groups and also among ethnic group members. Results in prior research vary regarding the relationship between an individual’s adherence to ethnic values and norms and their attitudes about mental health treatment [60], suggesting that the influence of strong ethnic identity may be dependent on factors specific to one’s cultural group. In other words, strong adherence to ethnic group values and practices may or may not be associated with a proximal mediator of engagement, depending on what those values and practices are. As such, cultural diversity among participants may have made it more difficult to detect significant associations between levels of ethnic identity and several of the mediators proposed by the young adult framework.

It is also possible that ethnic identity may have been related to engagement through structural or societal-level factors (e.g., healthcare system and policies) [17] that were not adequately captured by the individual-level mediators of service use examined by the current study. For instance, racism has been identified as a macrolevel factor that can influence stigma associated with seeking mental health treatment [107], while a more developed ethnic identity has been shown to offer protective effects against racism (e.g., psychological well-being and social support) [58, 108] that may decrease self-stigma associated with service use among young adults from minoritized groups [60]. Research examining relationships among ethnic identity, discrimination, and service utilization have shown mixed results [65, 66], but more focused research in this area may help to clarify the roles of community-level factors and intersectional oppression (e.g., racism and ableism) in the relationship between ethnic identity and engagement among young adults of color with SMI.

Limitations

There are important limitations to consider in this research. Notably, the tested SEM model does not fully control for unmeasured confounds nor does it adjust for measurement error, and the study had a small sample, which may have limited power to detect mediators that are relatively weaker in magnitude, so conclusions must be limited accordingly. The small sample also limited the examination of more complex SEM models. However, none of our statistical tests are necessarily problematic for our sample size, and there is sufficient effect size sensitivity for the results to provide an important jumping-off point for future research. Both engagement and ethnic identity are constructs that fluctuate over time among young people [14, 42]. This study is but one cross-sectional snapshot of that phenomena in one urban city in the USA from which causal inferences cannot be made. As such, replication with longitudinal, multi-site data is needed to increase confidence in the results. Finally, this study focused on salient individual and interpersonal level factors that can contribute to mental health disparities among racially/ethnically minoritized young adults with SMI. While our study presents preliminary evidence to address a gap in the current research, it has limitations regarding societal-level factors that may influence the broader relationship between ethnic identity development and young adult engagement in mental health care.

Implications

Limitations notwithstanding, the current study has useful implications for clinical services that should be evaluated in future research on mental health care for minoritized young adults. Results suggest that providers might promote a young adult’s investment in mental health care by affirming the young adult’s positive feelings about their ethnic group membership. Negative stereotypes and the internalization of oppressive beliefs can also be explored and challenged to strengthen ethnic identity. Combining these practices with a specific emphasis on offering hope and positive expectations for recovery [109], exploring perceived barriers to recovery, and generating strategies for overcoming barriers as they enter adult services may decrease the likelihood of disengagement among young adults with SMI.

Although they are not universally employed with young adults, strategies such as cultural formulation [36] and cultural acknowledgement [110] that elicit discussion of an individual’s ethnic identity in order to address cultural barriers and/or incorporate cultural preferences and values into treatment are already evidence-supported engagement practices [40, 41]. In addition, services that assist young adults in developing their ethnic identities, through active participation with affirming providers and linkage to other ethnic identity supports (e.g., affinity groups, community organizations) [111], may encourage them to consider a broader set of recovery and life outcomes. Finally, increased representation of the client population’s racial and ethnic backgrounds, and lived experience of mental illness among providers, along with improved provider knowledge on the role of ethnic identity, culture, and distrust in young adult engagement may be essential components in culturally responsive engagement approaches for young adults.

Conclusion

The present study suggests that ethnic identity matters for young adult mental health care. It could be important for service providers to understand how young people feel about their ethnic backgrounds, and whether they feel a sense of belonging and positive attachment to their ethnic communities. In taking the time to appreciate this aspect of young adult identity development, providers can attain a better sense of what potential barriers and facilitators to recovery may be present and can use that insight to enhance the impact of their work with individual young adults. This study also indicates that research, policy, and practice experts need to pay much closer attention to the concept of ethnic identity moving forward, as it has implications for strength-based approaches to supporting mental health among racially and ethnically minoritized youth but has received little attention in the field of mental health services.

Data Availability

The data that support the findings of this study are available from the corresponding author upon reasonable request, in accordance with our data sharing plan.