Vulnerable Latina/oFootnote 1 immigrant populations have been historically exposed to considerable adversity in the USA, recently exacerbated by the emergence of nationalist and anti-immigration movements (McHugh, 2018). During the Trump administration, aggressive deportation policies resulted in widespread trauma and fear among Latina/o immigrant communities, with particularly harmful impacts for children and youth (UNICEF, 2018). Thus, scholars have expressed the need for applied research initiatives aimed at ameliorating the deleterious impacts of cumulative immigration-related stress and trauma experienced by low-income Latina/o immigrants (Cardoso et al., 2018). This call for action is urgent, particularly when considering that vulnerable Latina/o immigrant families are very likely to be exposed to chronic stressors such as parents having multiple jobs characterized by low salaries and strenuous working conditions, lack of access to basic health and mental health services, language barriers, social isolation, discrimination, and distrust of formal health and mental health institutions (Garcini et al., 2015).

This context of adversity also has a negative impact on the quality and consistency of parenting practices of low-income Latino/a immigrant caregivers, which can lead to adverse consequences for children and youth (Cardoso et al., 2018). As an alternative, culturally adapted parent training (PT) interventions can enhance the protective parenting practices of immigrant parents (Gonzales et al., 2017). However, although meta-analytic data indicate that PT interventions constitute the gold standard for promoting effective parenting skills and preventing youth externalizing behaviors (Michelson et al., 2013), the dissemination of culturally adapted evidence-based PT interventions continues to be notoriously limited in low-income US Latina/o immigrant communities (Garcia-Huidobro et al., 2018).

Meta-analytic research on culturally relevant PT interventions indicates that Latina/o immigrant families benefit from this type of intervention as demonstrated by participants’ improvements on parenting practices, as well as children and youth adaptive behaviors (van Mourik et al., 2017). However, important gaps of knowledge exist with regards to the specific ways in which these interventions can enhance protective factors among Latina/o immigrant families. According to existing research, PT interventions are likely to strengthen specific Latina/o cultural protective factors such as the cultural value of familismo, as demonstrated by increased family cohesion and parental emotional involvement among families exposed to PT interventions. Similarly, enhanced limit-setting and supervision are likely to reinforce the cultural value of respeto, which refers to the importance of showing respect to parents and people in general (Marsiglia et al., 2019).

Addressing Discrimination and Cultural Conflicts in Prevention Parenting Interventions

Low-income Latina/o immigrant families are very likely to experience multiple forms of discrimination (McHugh, 2018). This understanding must inform the cultural adaptation of evidence-based PT interventions targeting Latina/o immigrant populations, particularly because experiencing discrimination is associated with multiple deleterious consequences ranging from poor health and mental health outcomes to reduced likelihood to engage in formal mental health services due to fear of discrimination (Ayón, 2015; Park et al., 2018). Furthermore, Latina/o youth in immigrant families who have experienced direct instances of discrimination are at an increased risk for drug use, chronic anxiety, depression, and rule-breaking behaviors (Schwartz et al., 2015). Similarly, youth who have witnessed or are aware of instances of discrimination experienced by their immigrant parents are more likely to report lower self-esteem and increased mood disorders than youth without exposure to these stressors (Espinoza et al., 2016). Based on these issues, scholars continue to emphasize the high need to overtly address various forms of discrimination in prevention interventions (Gonzales et al., 2017; Unger, 2015).

Latina/o immigrant families may also experience cultural conflicts, particularly if parents and youth embrace contrasting cultural identities and cultural values (Schwartz et al., 2015). Although parent–youth cultural conflicts vary among Latina/o immigrant families, prevention scholars agree on the premise that promoting biculturalism in immigrant families constitutes a relevant family protective factor (Schwartz et al., 2015).

Culturally Adapted Parenting Interventions

Cultural adaptation refers to the modification of evidence-based treatments to ensure their cultural and contextual relevance when delivered to diverse populations (Bernal et al., 2009). Despite the existence of model culturally relevant parenting prevention interventions for Latina/os (Garcia-Huidobro et al., 2019; Gonzales et al., 2017; Hernandez Robles et al., 2018; Martinez & Eddy, 2005; Marsiglia et al., 2019; van Mourik et al., 2017; Prado et al., 2007), the dissemination of culturally adapted PT prevention interventions across US low-income Latina/o immigrant communities remains extremely limited (Garcia-Huidobro et al., 2018). Furthermore, there is a scarcity of rigorous studies aimed at examining the impact of culturally adapted PT interventions that overtly address salient immigration-related stressors, such as various forms of discrimination experienced by Latina/o immigrants (Cardoso et al., 2018).

GenerationPMTO©

GenerationPMTO is a progenitor of parent training clinical and preventive family interventions. Originally developed by Gerald Patterson in the mid-1960s, the program has evolved over the course of more than five decades. The Social Interaction Learning Model underlying the program specifies parents as their children’s most important teachers. Thus, the intervention empowers parents by teaching them effective parenting strategies (Forgatch et al., 2009). The core content of GenerationPMTO is similar to that of other parenting programs (e.g., positive contingencies for behavior). However, the intervention’s unique focus on promoting specific clinical process skills and a strong emphasis on active teaching strategies constitutes a difference from other parenting interventions. These unique intervention characteristics were informed by years of observation-based research conducted by GenerationPMTO researchers, focused on the study of resistance during clinical intervention and analysis of parent–child parenting interactions (Chamberlain et al., 1984). Furthermore, the parent-based focus of the intervention does not require active participation from youth, which considerably increases the implementation feasibility of the intervention (Kuntsche & Kuntsche, 2016).

Cultural Adaptations of GenerationPMTO

The first cultural adaptation of GenerationPMTO for Latina/o immigrant families was conducted by Domenech Rodríguez et al. (2011). The adapted intervention was titled “CAPAS: Criando con Amor, Promoviendo Armonía y Superación” (Raising Children with Love, Promoting Harmony and Self-Improvement). The intervention was adapted for Latino/a families with young children (ages 4–11) and consisted exclusively of the core GenerationPMTO parenting components.

A subsequent study led by Parra-Cardona et al. (2017) empirically tested the impact of differential cultural adaptation. Specifically, the CAPAS intervention was compared against a CAPAS-Enhanced intervention. Both adapted interventions included all core GenerationPMTO parenting components. However, CAPAS-Enhanced also included components focused on overtly addressing immigration-related challenges, strategies to cope with discrimination, and promotion of family biculturalism. Participant families in the randomized controlled trial were low-income Latina/o immigrant families with one target child (ages 4–12). Six-month follow-up findings indicated that when compared to CAPAS, the CAPAS-Enhanced intervention was associated with the highest improvements on child internalizing and externalizing behaviors (Parra-Cardona et al., 2017). Qualitative data indicated that the incremental effects were associated with parents being able to refine their parenting practices by identifying the cumulative impacts of immigration-related stressors such as discrimination, as well as learning strategies to manage cultural conflicts in the family (Parra-Cardona et al., 2016).

The Current Study

The primary objective of this investigation was to empirically examine the implementation feasibility and initial efficacy of a version of CAPAS-Enhanced for low-income Latina/o immigrant families with adolescents, ages 12–14. The adapted intervention is known as CAPAS-Youth and integrates all the core GenerationPMTO parenting components, adapted for families with adolescents. CAPAS-Youth also includes sessions focused on identifying the impact of immigration-related challenges on parenting practices, strategies to address immigration-related stressors, and promotion of family biculturalism. Case management services are offered to all families to help them address a variety of immigration-related contextual challenges. The intervention was delivered on a weekly basis in a group format.

Key adaptations were conducted to increase the likelihood of sustaining the intervention in low-resource Latina/o immigrant communities. Thus, in contrast to the original prevention version of the GenerationPMTO intervention consisting of 14 sessions (Forgatch & DeGarmo, 1999), CAPAS-Youth consisted of 9 sessions. The intervention team also included members of the target community, which increases the likelihood of long-term sustainment.

Method

Overview of Research Design

Participant parents were randomly assigned to one of two conditions: (a) CAPAS-Youth or (b) wait-list control (WLC) group. Participants in the WLC group did not receive the parenting intervention until after (T2) assessments were completed for all parents in both intervention conditions. Case management services were offered to all families upon enrollment in the study, with personalized objectives according to the needs of each individual family. Due to the intense context of adversity experienced by the target Latina/o immigrant community, the research design consisted of an exploratory pre–post design. This determination was made given the considerable risk for high attrition related to contextual factors as families reported frequent relocations due to fear of widespread activities by immigration enforcement authorities in the area. Although we recognize that conclusive effects of prevention interventions can only be demonstrated with distal follow-up measurements (Gottfredson et al., 2015), a pre–post design was selected to prioritize examining implementation feasibility and initial efficacy of the intervention within this context of adversity. Thus, measurements were completed at baseline (T1) and intervention completion (T2).

The guiding research hypotheses were as follows: (a) Compared to parents in WLC at T2, parents assigned to CAPAS-Youth would report statistically significantly higher scores on all parenting components, with the exception of limit-setting, (b) at T2, the overall retention rate for both study conditions would be higher than 60%, (c) CAPAS-Youth parents would report high levels of satisfaction with the intervention according to session satisfaction aggregate scores, and (d) compared to participants in WLC at T2, CAPAS-Youth caregivers would report significant reductions on youth internalizing and externalizing behaviors. Furthermore, (e) we did not expect a significant difference between intervention conditions on discipline practices at T2 given the null findings reported on this variable at posttest in the original GenerationPMTO prevention trial (Forgatch & DeGarmo, 1999). In such a prevention study, positive limit-setting intervention effects were initially identified at 6-month follow-up measurement (Forgatch & DeGarmo, 1999). This hypothesis is also supported by meta-analytical research indicating that “sleeper” or delayed treatment effects have been reported in parent training trials (van Aar et al., 2017). Finally, (f) we hypothesized that mothers in both intervention conditions would report significant reductions related to extrafamilial immigration-related stress.

Participants and Sampling Procedures

Latina/o parents were recruited at community events, churches, and Latina/o serving organizations. A screening recruitment protocol was implemented according to the prevention focus of this study. Specifically, one focal youth (FY) per family was recruited if the following key inclusionary criteria were met: (a) FY was between the ages of 12 and 14 and (b) mothers’ reports of FY behavior problems were classified as mild-to-moderate according to Bird et al. (2001) youth behavior screening measure. The age criterion for youth was determined due to the importance of early prevention intervention (Forgatch et al., 2009). Eligibility criteria for parents included (a) self-identified as foreign-born Latino/a or Hispanic, (b) Spanish speaking, and (c) experiencing economic disadvantage, according to parental reports of financial difficulties. All study procedures were reviewed and approved by the Institutional Review Board prior to recruitment. Non-eligible families were offered the option of services by local mental health providers. Table 1 presents the participants’ demographic characteristics.

Table 1 Participants’ demographic information
Full size table

Randomization

From initial recruitment, 134 families were screened for eligibility with a 34% overall engagement of fathers across intervention conditions. Due to their limited exposure to the core components of the intervention, fathers were not included in efficacy analyses. Specifically, only seven fathers attended the six sessions that were required to classify them as having sufficient exposure to the core parenting components of the intervention. However, none of these fathers attended the review sessions that were critical to ensure that parents correctly incorporated and applied the PMTO parenting skills.

Enrolled mothers were randomly assigned to CAPAS-Youth or the WLC group. Allocation to intervention conditions was computer-generated and completed by the lead data analyst. Blocking was based on the sex of the adolescent (i.e., male, female). The data analyst did not have any contact with families, data collectors, or interventionists at any stage of the study. Neither participants nor research personnel was blinded to the condition assignment. The description of the flow of participants through the study is presented in Fig. 1.

Fig. 1
figure 1

CONSORT flowchart of participants

Full size image

Intervention Procedures

Parents were invited to participate in a 9-week parenting prevention program delivered in a group format, 2 h in length per session, with 8–12 parents and two facilitators. Caregivers in the intervention group received a mid-week call to troubleshoot home practice assignments. Families assigned to WLC were offered the intervention after all post-intervention assessments (T2) for both intervention conditions were completed for each cohort of the study. Participants were offered the parenting intervention at no cost and were compensated $30 USD for completion of baseline assessments (T1) and $40 for completion of post-intervention assessments (T2).

CAPAS-Youth

The CAPAS-Youth intervention differs from the CAPAS-Enhanced and other PT interventions because principles of racial socialization were adapted according to experiences of discrimination reported by participants. Furthermore, conversations were elicited from parents aimed at helping them refine communication strategies with their children to help them process experiences of discrimination, as well as to prevent youth from internalizing distressing emotions associated with these experiences.

An additional characteristic of CAPAS-Youth was its method of delivery. Specifically, GenerationPMTO is characterized by the frequent use of role-plays to help parents master new parenting practices. Thus, role-plays of common familial, cultural conflicts were enacted throughout the intervention to help parents address cultural challenges with their adolescent children, as well as to promote family biculturalism. Role-plays were also enacted according to specific stressors reported in real time by parents. Specifically, caregivers reported in parenting sessions instances of racial discrimination experienced in a variety of settings. Such reports were used to engage parents in mindfulness exercises to help them regulate distressing emotions, followed by role-plays of parenting situations once parents reported increased control over distressing feelings.

With regards to curriculum content, session 1 provides a framework of the parenting intervention according to the context of immigration, including discussion of immigration-related challenges and brainstorming of coping strategies related to immigration stressors. In session 2, the construct of biculturalism is introduced, including skills to promote biculturalism and prevent parent–youth cultural conflicts. The biculturalism components were informed according to principles of existing prevention interventions aimed at promoting biculturalism with Latino/a immigrant families (Martinez & Eddy, 2005; Smokowski et al., 2009).

Sessions 3 through 9 consisted of the core GenerationPMTO parenting components informed by immigration and biculturalism frameworks. Issues of discrimination were addressed by implementing three main strategies. First, in session 1, discrimination was overtly identified as a common stressor experienced by Latina/o immigrant families. A brief reflection was offered to parents focused on highlighting the critical contributions by Latina/o immigrants to the development of the USA as a nation, while also identifying the various forms of exploitation commonly experienced by this population.

Second, the curriculum was delivered by informing role-plays according to instances of discrimination reported by parents. Thus, we utilized mindfulness techniques to help parents regulate feelings associated with experiences of discrimination, prior to introducing them to the practice of new parenting skills. Finally, case management services were tailored to offer parents alternatives to cope with specific contextual stressors impacting their lives. For example, case managers worked with parents to find alternative employment opportunities if parents reported experiencing intense discrimination and exploitation in their workplace. The intensity of case management services varied according to the contextual stressors reported by each participating family.

Interventionists and Fidelity

The intervention teams consisted of a clinical social worker and a community leader affiliated with the largest faith-based organization serving Latina/o immigrant families in southwest Detroit. These interventionists were Latina/o, fully bilingual, and residents of the target community. Both interventionists have extensive training in GenerationPMTO with six years of experience delivering the intervention to Latino/a families with young children.

The interventionists community leaders had a unique role throughout intervention delivery. Specifically, in addition to sharing equal responsibility with the clinical social workers for the delivery of curriculum content, community leaders consistently provided a unique sense of connection to participants by highlighting how daily contextual challenges associated with their local community (e.g., gang violence, raids by immigration authorities) were likely to impact the participants’ lives and parenting practices.

To ensure fidelity and adherence to the core GenerationPMTO components, the lead author co-delivered the parenting groups for the initial two waves of the study. The lead author is a Latino immigrant, fully bilingual, and a certified GenerationPMTO specialist and trainer. Starting with the third wave of parenting groups, the lead author alternated between providing live supervision and coaching sessions. Supervision procedures followed fidelity guidelines as delineated in the Fidelity of Implementation (FIMP) Rating System (Forgatch et al., 2005), which evaluates adherence to the core GenerationPMTO components. Furthermore, the lead author periodically reviewed the cultural tailoring of the adapted intervention with the leadership team of Implementation Science International, Inc. (ISII), the independent organization that coordinates the dissemination of GenerationPMTO.

Assessment Procedures and Measures

Data were gathered through quantitative assessments implemented at baseline (T1) and intervention completion (T2). Assessments were completed at the participants’ homes or at a site of their choice (e.g., local church). For present analyses, only mothers completed all assessment measures, selecting Spanish as their language of preference.

Session Satisfaction

Mothers completed a weekly session satisfaction questionnaire (Forgatch et al., 2005). The questionnaire consists of a 10-item Likert-type scale (1 = lowest satisfaction, 5 = highest satisfaction). Internal consistency coefficients of the satisfaction measure across all waves of parenting groups ranged from α = .95 to α = .98.

Parenting

Five parenting measures were administered to assess specific parenting skills. For all scales, individual items were averaged to produce a mean score with higher scores indicating higher levels of functional parenting practices. All parenting measures have been reported to have linguistic appropriateness and content validity with Spanish-speaking Latina/o populations (Domenech Rodríguez et al., 2007; Martinez & Eddy, 2005; Smith et al., 2014).

Family Problem Solving

This 13-item measure was adapted from the scale developed by Domenech Rodríguez et al. (2007). Scale items measure family problem-solving behaviors (e.g., when solving a problem, all family members propose potential solutions). The measure is answered using a 5-point Likert-type scale (1= never to 5 = always). Cronbach’s alpha was α = .95.

Skills Encouragement

This revised measure consists of 19 items to assess parental support as youth develop new skills (e.g., you help your child to accomplish a difficult task), as well as active forms of parental encouragement (e.g., you verbally acknowledge your child’s capabilities; Martinez & Eddy, 2005). The measure is subdivided into two subscales (parental support and autonomy, parental reinforcement). Both subscales use a 5-point Likert-type scale ranging from 1= never to 5 = always. Cronbach’s alphas were acceptable for the total scale (α = .91), as well as parental support subscale α = .87 and parental reinforcement α = .81.

Limit-setting

Positive discipline was measured with four items of the Family Check Up Caregiver Questionnaire, aimed at examining noncoercive limit-setting strategies in response to youth misbehavior and parent–youth conflictual situations (e.g., you spoke calmly with your child when you were upset with him/her; Smith et al., 2014). The measure used a 5-point Likert-type scale ranging from 0 = never to 4 = very often. Cronbach’s alpha was α = .75.

Free Time Monitoring and Supervision

The measure consists of 15 items of the Family Check Up Caregiver Questionnaire, aimed at assessing parental supervision of free time and activities with friends (Smith et al., 2014). The scale used a 5-point Likert-type scale ranging from 0 = never to 4 = very often. Cronbach’s alpha for the scale was α = .87.

Parenting Warmth

Four items from the Family Check Up Caregiver Questionnaire were used to evaluate the level of the parent–youth emotional connection (Smith et al., 2014). The items measure the level of emotional closeness as indicated by youths’ behaviors (e.g., if upset, my child seeks comfort from me) and parental emotional involvement (e.g., it is easy to be in tune with what my child is feeling). The scale used a 5-point Likert-type scale ranging from 1= definitely not to 5 = definitely. Cronbach’s alpha was α = .85.

Extrafamilial Immigration-related Stress

Caregivers completed the extrafamilial immigration-related stress subscale of the Hispanic Stress Inventory-Immigrant Abbreviated version (HIS-I; Cavazos-Rehg et al., 2006). Selected items measured contextual stressors (e.g., I have been forced to accept low paying jobs). The level of stress is measured on a 5-point Likert scale ranging from 1 = not at all stressful to 5 = extremely stressful. The validity and reliability of this measure has been established with Latina/o, Spanish-speaking immigrant populations (Cavazos-Rehg et al., 2006). Cronbach’s alpha was α = .74.

Acculturation

Parental and youth acculturations were measured using the American subscale from the Bicultural Involvement Questionnaire (BIQ; Szapocznik et al., 1980). The American subscale is a six-item self-report with items measuring affinity with the English language and the US culture (e.g., How much do you enjoy going to places with an American orientation?), answered using a 5-point Likert-type scale ranging from 1 = not at all comfortable to 5 = very comfortable. It has been validated and used as a measure with both first-generation Mexican immigrant parents (Flores et al., 2004) as well as with second-generation Mexican youth (Guo et al., 2012). Cronbach’s alpha was α = .93.

Youth Internalizing and Externalizing Behaviors

Two revised scales from the Revised Behavior Problem Checklist (RBPC; Quay & Peterson, 1996) were used in this study. The Conduct Disorder Scale measures parental perceptions of child externalizing behavioral. The Anxiety-Withdrawal scale measures internalizing disorders such as fearfulness and anxiety. Items describe behaviors on a 3-point scale: 0 = not a problem, 1 = minor problem, and 2 = serious problem. Scale scores are summed with higher scores indicating more problematic behavior. Cronbach’s alphas were α = .85 (internalizing) and α = .93 (externalizing).

Data Analytic Strategy

We first conducted descriptive analyses to examine feasibility outcomes and data properties prior to conducting efficacy analyses. Next, we evaluated efficacy outcomes according to a series of mixed factorial ANCOVAs, aimed at examining group (CAPAS-Youth vs. WLC) by time (baseline vs. intervention completion) interaction effects of the intervention. All models were tested by controlling for the effects of youths’ binary sex (i.e., female, male) and maternal and youth acculturation, with analyses utilizing Bonferroni’s adjustment procedure. Missing data were less than 5% and missing completely at random, Little’s MCAR χ2 (590) = 262.60, p =.89. Furthermore, normality tests on all scales did not indicate significant skewness or kurtosis. None of the study outcomes significantly differed at baseline between the study conditions.

SPSS v.25 software was used for data analysis. Effect sizes (η2) with 95% confidence intervals were computed for each intervention effect (Feingold, 2009). Effect sizes are interpreted according to the standard benchmarks proposed by Cohen (1988), indicating small (η2 = .01–.05), medium (η2 = .06–.14), and large effects (η2 > .14). Because the same primary hypothesis was tested with 10 separate outcomes, we report corrected p values according to false discovery rate by implementing the Benjamini–Hochberg procedure (Benjamini & Hochberg, 1995). This analytical step was conducted to increase the rigor of analyses.

Results

Intervention Feasibility Outcomes

Of the 79 mothers randomly assigned to CAPAS-Youth (n = 41) and wait-list control condition (n = 38), 68 mothers completed T2 assessments, for an overall 86.1% retention rate (CAPAS-Youth = 85.3%; WLC = 86.8%). Of the mothers allocated to CAPAS-Youth that began the intervention, 94.5% completed six or more sessions, including at least one of the final review sessions (i.e., sessions 8–9). Completion of sessions by sex of participants is reported in Fig. 1. Session satisfaction data were aggregated from weekly ratings of parents’ satisfaction with the intervention. On a 1–5 satisfaction scale (5 = highest), satisfaction ratings across all waves ranged from 4.66 to 4.78 (SD = 0.74). Average satisfaction across cohorts was 4.71 (SD = 0.74) (Table 2).

Table 2 Means and standard deviations for mother parenting practices, immigration-related stress, and youth problem behaviors
Full size table

Intervention Outcomes

Means and standard deviations for mother parenting practices, immigration related stress and youth problematic behaviors are presented in Table 2. In general, results shown in Table 3 indicated that eight of the ten outcomes were statistically significant and in the hypothesized direction. Among the significant intent to treat contrasts, the intervention exhibited large effect sizes (η2 > .14). According to MANOVA analysis, both intervention conditions were equivalent, with the robust omnibus test (i.e., Pillai’s trace) discarding differences for at least one outcome variable between treatment conditions, V = .13, F (9, 50) = .83, p = .59. Each domain is discussed below.

Table 3 Analysis of covariance, estimated marginal means, and 95% confidence intervals of effect sizes
Full size table

Parenting Outcomes

Significant group by time intervention effects were detected for all parenting outcomes except for limit-setting. ANCOVA intervention effects for parenting outcomes are displayed in Table 3. Differences indicated that mothers assigned to CAPAS-Youth showed significant improvements at T2 compared to WLC. The effect sizes for all parenting measures were large (ranging from η2 = .21 to η2 = .32), with parental reinforcement presenting the largest effect size. Furthermore, the 95% confidence intervals for effect sizes did not contain zero, adding certainty to the effects found between the treatment and WLC groups.

Youth Outcomes

Significant group by time differences were identified, indicating decreased youth internalizing and externalizing problem behaviors at T2 in CAPAS-Youth intervention participants relative to controls (see Table 3). Effect sizes for both outcomes were large, as indicated by youth internalizing η2 = .25 and externalizing behaviors η2 = .24.

Extrafamilial Immigration-related Stress

ANCOVA intervention effects for extrafamilial immigration-related stress are displayed in Table 3. As expected, because case management services were offered to all families, no significant group by time intervention effects was detected. Furthermore, significant within-group reductions in the reported level of stress were identified between T1 and T2. Effect sizes were in the medium range for mothers in both intervention conditions, where CAPAS-Youth t (27) = 2.29, p = .03 (g = .48) and WLC t (29) = 2.39, p = .02 (g = .42).

False Discovery Rate

ANCOVA analyses were implemented using the Bonferroni adjustment procedure. To increase the rigor of analyses, we used correction for false discovery rate according to the Benjamini and Hochberg (1995) method, with an alpha level = .05. The resulting p values are included in Table 3, indicating that even with a false discovery rate of 5%, all of our effects remain significant.

Discussion

Feasibility Outcome

The retention rate of mothers in this study exceeded original expectations, despite the intense context of adversity experienced by participants. We attribute this result to having highly skillful and culturally sensitive research and intervention delivery teams, with a thorough awareness of the unique challenges faced by the target population. In addition, qualitative findings indicated that mothers considered that an overt focus on immigration-related challenges and biculturalism helped them become aware of the ways in which these stressors negatively impacted their parenting practices (Parra-Cardona et al., 2019).

Parenting and Immigration-related Stress Outcomes

Parenting Outcomes

When compared to WLC, CAPAS-Youth mothers reported significant improvements on four of the five GenerationPMTO parenting practices, with effect sizes in concordance with published parameters of efficacious PT prevention interventions (Michelson et al., 2013). Of particular interest are the large effect sizes for parenting outcomes. These findings indicate that strong intervention effects are associated with mothers receiving sufficient intervention dosage and coaching to help them implement the parenting skills at home.

Due to the exploratory nature of this study, our measurements consisted exclusively of pre- and posttests. Thus, null limit-setting intervention effects may be associated with the lack of distal measurements. In fact, we hypothesized null findings on limit-setting at posttest based on previously reported null posttest findings on this variable in the original GenerationPMTO prevention trial (Forgatch & DeGarmo, 1999). Furthermore, according to meta-analytical research, “sleeper” or delayed treatment effects are possible in PT prevention trials and may be associated with a variety of factors, such as parents needing additional time to master parenting practices or experiencing challenges to implement parenting practices in the midst of considerable contextual stressors (van Aar et al., 2017). In this study, the majority of participants were impacted by continuous and aggressive deportation activities enacted by the Trump administration. It was common for parents to share in parenting sessions stories of relatives or friends who were deported by immigration authorities. Thus, as suggested by van Aar et al. (2017), contextual challenges may have had a critical role in potential limit-setting sleeper effects, particularly because mothers may have experienced challenges in the implementation of emotionally charged discipline practices while coping with considerable contextual adversity.

Despite limit-setting null findings, current results are promising and highlight the relevance of culturally adapting PT interventions for Latina/o populations according to contextual and cultural factors of relevance to the target population (Garcia-Huidobro et al., 2018). However, it is also critical to highlight that according to standards of efficacy established by SPR, efficacy can only be claimed if there is “at least one long-term follow-up at an appropriate interval beyond the end of the intervention” (Gottfredson et al., 2015).

Significant within-group reductions in levels of extrafamilial stress for both conditions appear to indicate the relevance of advocacy approaches to help families address immigration-related stressors. This finding is highly promising as mothers in both groups reported significant reductions in perceived immigration-related stress, with no significant differences between groups. However, this hypothesis remains to be replicated in future research by conducting a more systematic evaluation of this intervention component.

Youth Outcomes

Study findings related to youth outcomes appear to support our hypotheses with regards to the positive impact of the intervention on youth outcomes. However, definitive conclusions on youth outcomes can only be claimed by relying on multiple sources of information to reduce potential bias associated with parental self-reports (e.g., teachers’ ratings of youth behavior). Notwithstanding this limitation, current findings are in line with previous studies confirming the positive impact that nurturing parenting practices have on adolescents’ well-being (Marsiglia et al., 2019).

Results also confirm the relevance of parent-based interventions that do not require the active participation of youth in prevention interventions (Kuntsche & Kuntsche, 2016). Finally, we consider it imperative to interpret current findings according to complementary qualitative data. Specifically, mothers in this study provided in-depth narratives describing the ways in which immigration-related stressors negatively impacted their parenting practices, as well as the mental health of their adolescent children. For example, in at least two cases identified during intervention delivery, mothers reported their adolescent children experiencing suicidal ideation resulting from instances of bullying by school peers who mocked them due to their parents’ undocumented status (Parra-Cardona et al., 2019). Based on the intense nature of these immigration-related stressors, current results indicate the critical protective effect associated with consistent and nurturing parenting practices (Marsiglia et al., 2019).

Study Limitations

Current findings should be considered in light of study limitations. First, outcome data consisted of parental self-reports, which carry the risk of reporting bias. In addition, the small sample size in this investigation did not allow for the identification of small intervention effects. Likewise, the lack of distal measurements did not allow for the identification of long-term intervention effects. Furthermore, the lack of systematic measurement of case management services does not allow for definitive conclusions on the impact of this intervention component. Finally, the low engagement rate of fathers considerably contrasted with our previous prevention study in which 84% of participating families had two caregivers (mother, father), with an overall retention rate of fathers of 85%. We attribute this contrast in father engagement to implementation limitations. Specifically, in this investigation, we had to complete study activities within a 12-month period. This represented a significant conflict for fathers who expressed interest to participate but were unable as the months in which the parenting program was offered were also months that fathers identified as in high demand for seasonal immigrant labor.

Conclusion

Notwithstanding study limitations, current findings are promising and justify the high need to continue to implement parenting prevention studies for low-income Latina/o immigrant families according to rigorous science, while also ensuring high cultural and contextual relevance. As a guiding principle, such efforts must be consistently grounded in the extraordinary strengths embraced by Latina/o immigrant families exposed to intense adversity.