Family-Centered Prevention Effects on the Association Between Racial Discrimination and Mental Health in Black Adolescents

Key Points Question Is participation in a family-centered prevention program, designed to enhance caregiving practices, associated with protection of Black adolescents from the effects of racial discrimination on their mental health? Findings This secondary analyses of data from 2 randomized clinical trials found that participation in a family-centered prevention program was associated with protection of Black adolescents from the effects of racial discrimination on conduct problems (in both trials) and on depression/anxiety symptoms (in 1 trial). These associations were partially explained by intervention-induced changes in protective parenting. Meaning These findings suggest that family-centered prevention programs reduce the effects of racial discrimination on subsequent increases in mental health problems among Black adolescents.


Introduction
Racial discrimination includes routine experiences with disrespect and treatment connoting that one is inferior. It is pervasive in Black youth's everyday lives. 1,2 Children and adolescents who encounter racial discrimination are at risk for poor mental health outcomes, such as hopelessness, conduct problems, drug use, and depression. 3 Importantly, not all Black youth exposed to racial discrimination go on to develop mental health problems. 3 This observation raises a fundamental question for pediatric scientists studying resilience and for clinical researchers developing applications: what enables some Black youth to remain mentally healthy despite experiences with discrimination? Research suggests that receipt of protective caregiving, defined as caregiving that includes high levels of emotional support and warmth, involvement, bidirectional communication, and cooperative problem solving, may mitigate mental health problems associated with discrimination. [4][5][6] However, causal inferences cannot be made on the basis of existing observational studies. In this study, we used secondary data from 2 randomized clinical trials of family-centered interventions to investigate the protective effects of caregiving practices. The Strong African American-Teen (SAAF-T) 7 program was designed for youth aged 14 to 16 years, and the Adults in the Making (AIM) 8 program was designed for youth aged 17 to 18 years. Both programs were developed to prevent mental health problems and substance use by enhancing protective caregiving. Previous research with the SAAF-T and AIM trials focused on establishing their efficacy and durability in preventing substance use and mental health problems. 9 Both interventions have also been shown to mitigate the effects of life stress on Black adolescents by increasing protective caregiving, and both programs have demonstrated stress-buffering capacities for a range of psychosocial outcomes. 9 However, these interventions' effects on youth's coping with discrimination have not been investigated.
Consistent with prior research that has documented larger intervention effects for those at higher risk, 10 we hypothesized that SAAF-T and AIM would produce the greatest mental health benefits for adolescents who reported more frequent discrimination at baseline. Specifically, we hypothesized that adolescents assigned to the SAAF-T or AIM intervention group who reported frequent discrimination at baseline would evince fewer subsequent increases in mental health problems than would adolescents assigned to a control group. Finally, we explored the possibility that SAAF-T and AIM effects in reducing the impact of discrimination on youth's mental health were attributable to improved protective caregiving.

Study Design
We conducted an unplanned secondary analysis of data from 2 community-based randomized clinical trials, SAAF-T and AIM. Both trials were unblinded, parallel studies. In each trial, families were randomly allocated on a 1:1 ratio to intervention or control groups by the study statistician. Additional trial details are provided in the trial protocol and statistical analysis plan in Supplement 1.

SAAF-T
Informed by a power analysis (Statistical Analysis Plan in Supplement 1), the SAAF-T trial included Black families from rural Georgia recruited from 2006 to 2007. In each family, an adolescent who was aged 16 years at recruitment and the adolescent's primary caregiver (in most families, the biological mother) provided data. Schools in 6 counties provided lists of 10th-grade students, from which participants were selected randomly. Eligible youth were in the 10th grade and self-reported African American or Black race/ethnicity. Youth or caregivers with developmental disabilities or psychiatric illnesses that prevented them from completing data collection or participating in the intervention were excluded. Figure 1 presents the flow of participants at each stage of the trial. American or Black race/ethnicity. Youth or caregivers with developmental disabilities or psychiatric illnesses that prevented them from completing data collection or participating in the intervention were excluded. Figure 1 presents the flow of participants at each stage of the trial.

Intervention Implementation
SAAF-T consisted of 5 consecutive weekly, 2-hour sessions held at community facilities, with separate skill-building curricula for youth and primary caregivers. 9 Caregivers were taught emotional and instrumental support, limit setting, adaptive racial socialization, and methods for communicating about sex and alcohol use. Youth learned the importance of abiding by household rules, setting goals for the future and making plans to attain them, and strategies for resisting substance use. Each meeting included a 1-hour session when youth and parents met separately, followed by a 1-hour parent-youth session during which participants practiced the skills they had learned in their separate sessions.
AIM provided support for developmentally appropriate caregiving before adolescents leave high school and assume occupational and continuing educational roles. Youth and their primary caregivers attended 6 consecutive weekly meetings and, as in SAAF-T, separate parent and youth skill-building curricula during each 2-hour session. Primary caregivers were taught protective emotional and instrumental support, occupational and educational mentoring, cooperative problem solving, and communication skills. Youth were taught how to make plans to meet their goals, to identify people in their communities who could help them with goal attainment, and to formulate self-care strategies.

Data Collection Procedures
All data were collected in participants' homes using standardized protocols. Interviews were conducted privately, with no other family members present or able to overhear the conversation.

Racial Discrimination
At baseline, adolescents in both trials completed the Schedule of Racist Events, 11 which has been used frequently in longitudinal studies with adolescents. 3 Items assessed the frequency with which adolescents encountered discriminatory events, such as racial slurs, physical threats, and false accusations (SAAF-T: α = .90; AIM: α = .86).

Adolescent Mental Health
For measures of adolescent mental health in SAAF-T, adolescents responded to the Conduct Problems Scale from the National Youth Survey. 12 Internal consistency analyses were not conducted because this instrument yields count data. In AIM, conduct problems were measured using the parents' reports on the Child Behavioral Checklist. 13 Responses to the Aggression and Rule Breaking subscales were summed to form an indicator of conduct problems ( baseline, α = .85; long-term follow-up, α = .89). In SAAF-T, depressive symptoms were assessed using adolescents' reports on the α = .76).

Socioeconomic Risk Index
A socioeconomic risk index with 6 dichotomous variables was used as a control in the data analyses.
A score of 1 was assigned to each of the following: family poverty (based on federal guidelines), caregiver unemployment, receipt of Temporary Assistance for Needy Families, caregiver single parenthood, caregiver education level less than high school graduation, and caregiver-reported inadequacy of family income. The scores were summed to form an index ranging from 0 to 6, with higher scores indicating more socioeconomic risk.

Intervention Status and Sex
Intervention participants were coded 1, and control participants were coded 0. Boys and men were coded 1, and girls and women were coded 0.

Statistical Analysis
Intent-to-treat analyses conducted with Mplus statistical software version 8. Study hypotheses were tested with regression models in which mental health outcomes were estimated from baseline levels of mental health variables, discrimination, intervention condition, and the discrimination × intervention condition interaction term. Sex and socioeconomic risk were controlled in all analyses. Linear regression was used except for conduct problems in SAAF-T, where Poisson regression better fit the data.
Intervention-induced changes in protective caregiving were examined with regression-based moderated mediation models. 18 First, we determined whether participation in SAAF-T or AIM improved protective caregiving for youth who frequently encountered discrimination (path A).
Second, we calculated regression coefficients reflecting the associations between improvements in protective caregiving and mental health outcomes (path B). Moderated mediation was examined using data from steps 1 and 2. To do this, path A × path B regression coefficients were calculated.
P values were 2-sided, and statistical significance was set at .05. Data were analyzed from June to August 2020.

Sample Equivalence at Baseline and Attrition
Baseline equivalence was established on socioeconomic risk, participant sex, protective caregiving, and mental health outcomes ( Table 1). Evaluation of baseline equivalence on study variables for participants who did or did not provide follow-up data × prevention group assignment revealed no significant main effects or interaction effects for any covariate or study variable.

Discrimination, Intervention Participation, and Adolescent Mental Health
Consistent with our hypotheses, significant discrimination × prevention status interactions emerged for 3 of 4 outcomes: conduct problems in the SAAF-T trial ( Table 2), and conduct problems and depression or anxiety symptoms in the AIM trial ( Table 3) No differences emerged between adolescents in the intervention or control groups when exposure to discrimination was low (SAAF-T conduct problems: incident risk ratio, 1.433 [95% CI, 0.910 to  Figure 2 presents mental health outcomes at lower (1-SD below the mean) and higher (1-SD above the mean) levels of discrimination.
Youth sex was not associated with moderation of any of the results.

Intervention-Induced Increases in Protective Caregiving and SAAF-T and AIM Effects on Mental Health Outcomes
We explored the hypothesis that the SAAF-T and AIM prevention effects for youth who encountered frequent discrimination were associated with improvements in protective caregiving. After controlling for youth sex, family socioeconomic risk, and baseline levels of protective caregiving and

Discussion
We analyzed data from 2 randomized clinical trials, including participants with similar demographic characteristics, that tested family-centered interventions. The programs were structured similarly, but were designed for different age groups (early high school vs high school seniors) and featured distinct content. The findings suggest that family-centered prevention attenuated associations between racial discrimination and subsequent increases in mental health problems. Youth who received the SAAF-T or AIM intervention who also encountered increased racial discrimination had significantly fewer conduct problems 1.5 to 2 years after the intervention than did youth in the control groups. Also, AIM participants who received the intervention and who experienced racial discrimination more frequently exhibited fewer increases in depressive or anxiety symptoms than did participants in the control group. However, this pattern did not hold for SAAF-T participants and depressive symptoms. This may be because the AIM participants were older than the SAAF-T participants, so AIM participants may have been at increased risk for exposure to racial discrimination than were the younger SAAF-T participants. Therefore, AIM participants may have benefited more from the increases in protective caregiving they received in the course of the prevention trial than did the participants in SAAF-T.
Moderated-mediation analyses were consistent with a scenario in which both SAAF-T and AIM reduced symptoms of mental health problems, in part, by enhancing the protective caregiving that youth received. The mediation analyses suggested that more supportive parents may be better able to establish strategies that enhance their children's emotion regulation for coping with racial discrimination. This in turn may have reduced the physiological and psychological effects of racial discrimination that can influence mental health. 19,20 To our knowledge, this is the first study to show that family-centered prevention was associated with buffering the effects of racial discrimination on adolescents' mental health. It is notable that nearly all of the findings in this study were reproduced across the 2 randomized trials, thus increasing confidence in the results.
These results suggest that the buffering influences of enhanced protective caregiving on mental health outcomes are likely to be robust. In that regard, the results also converge with findings in the pediatric literature suggesting that a significant proportion of youth develop resilience to the mental and physical health consequences of adversity if they receive protective caregiving. 21  findings suggest that SAAF-T, AIM, and perhaps other interventions that focus on strengthening protective caregiving could help to forestall or attenuate some of the mental health problems that racial discrimination can produce. Of particular relevance to pediatric clinical practice, efficacious family-centered prevention programs designed to enhance protective caregiving are available for Black preadolescents, adolescents, and youth who are about to transition to young adulthood. 9 Participation in these programs has been associated with stress-buffering effects on the development of self-control and on reductions in drug use, obesity, cytokine levels, epigenetic aging, and, in this study, mental health problems following discrimination. 9

Limitations
This study has some limitations, including a lack of data on the effectiveness of SAAF-T and AIM outside of the areas in the rural southern United States where they were developed. Their efficacy in urban settings must be evaluated. AIM used parent report for both mediation and outcome variables, thus potentially increasing associations owing to common method bias. The trials concluded 10 years ago, and the historical context is evolving rapidly; additional replication with more recent trials is needed. Because the trials examined the durability of intervention effects, we were not able to provide the intervention to control participants. It is important that future research uses designs to provide control group participants with the opportunity to experience the intervention, particularly when they are from historically underrepresented and exploited populations. The availability of such designs in prevention science is currently limited.

Conclusions
The findings of this secondary analysis of 2 randomized clinical trials underscore the susceptibility of adolescents who frequently encounter racial discrimination to elevated mental health symptoms.
Secondary data analyses of 2 randomized trials suggest that supportive parenting may offset these mental health risks.