Introduction

Twenty-eight publications generated by the Familias en Acción (FEA) community-based participatory action research (CBPR) program were reviewed to synthesize key findings and methods for those engaging in Latino youth violence prevention and research [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24, 25•, 26•, 27, 28]. FEA was established in 2005 as a community-academic research partnership with a Latino community in San Antonio, TX, USA and was committed to using CBPR to address youth violence. As a principal investigator of FEA, this review is influenced by personal knowledge and perspectives.

The FEA literature reflects research conducted between the summer of 2005 through the spring of 2020 under the umbrella of the FEA CPBR program. Eight publications were about the original FEA project [1,2,3,4,5,6,7,8], eight were about Familias SUAVE (Families on the Southside United Against a Violent Environment) [9,10,11,12,13,14,15,16], seven were about an initial Healthy Adolescent Initiative [17,18,19,20,21,22,23], three were about a second Healthy Adolescent Initiative [24, 25•, 26•], and two were about the Education, Life Choices, and Injury Risk among Adolescents and Young Adults study [27, 28].

Review findings are organized into five sections: (1) effects of interventions on violent behaviors and attitudes, (2) predictors of future violent behaviors, (3) cross-sectional associations with violent behaviors and attitudes, (4) respondent-driven sampling, and (5) community-based participatory action research.

Findings

Effects of Interventions on Violent Behaviors and Attitudes

The FEA literature includes prospective evaluations of the effects of interventions on violent behaviors and attitudes [5, 14, 15, 20, 25•]. Table 1 summarizes the interventions with positive program effects on violent behaviors and attitudes. Participants of three interventions reported significantly lower levels of violent behaviors [14, 25•]. Participants of one intervention, El Joven Noble, did not report lower levels of violent behaviors in two studies [5, 14, 15]. Participants of two interventions reported improvements in violence-related attitudes [5, 20].

Table 1 Interventions with positive program effects on violent behaviors and attitudes
Full size table

In a primary violence prevention study, participants of a Violence Prevention Program (VPP) reported fewer acts of violence outside of school at 12 months (p = 0.025), and participants of a Positive Youth Development Program (PYDP) reported fewer acts of violence in school at 6 months (p = 0.018) [25•]. The study used a prospective randomized controlled design to assess the effects of two programs that used CBPR to engage adolescents, young adults, and parents in the development and implementation of interventions. The VPP focused on four community-prioritized risk factors for violence: the initial perpetration of violence, exposure to community violence, mental health issues, and social injustice issues. The PYDP focused on four community-prioritized protective factors: succeeding in school, developing supportive adult relationships, developing supportive peer relationships, and developing a sense of purpose. The two interventions used 8-week internet-based programs and an in-person youth summit [25•].

In a secondary violence prevention study, high school students who participated in the Teen Medical Academy reported fewer acts of non-physical aggression (p < 0.001) and physical violence (p = 0.002) at 9 months [14]. In addition, participants of the Teen Medical Academy reported less intimate partner violence (p = 0.02). The study was conducted with students in a Disciplinary Alternative Education Program (DAEP) and used a prospective repeated measures quasi-experimental intervention/control design. El Joven Noble, a culturally tailored character development program developed in East Los Angeles, was the intervention program. The Teen Medical Academy, a health career promotion program developed in San Antonio, was the control program. Both programs consisted of eighteen 45-min sessions that were conducted twice a week. Participants opted out of their daily life skills class on Tuesdays and Thursdays [14].

Participants of El Joven Noble did not report lower levels of violent behaviors in two prospective studies [5, 14, 15]. Additional analyses from the previously mentioned study in a DAEP demonstrated that high school students who participated in El Joven Noble reported more acts of school (p = 0.02) and non-school (p = 0.003) violence at 9 months as compared to those who participated in the Teen Medical Academy [15]. In a primary violence prevention study, there were no El Joven Noble program effects on violent behaviors [unpublished data, 5]. The study used a prospective randomized controlled design and was implemented weekly with third to fifth graders during an existing afterschool program [5].

Participants of two interventions reported improvements in violence-related attitudes [5, 20]. Participants of the Familias en Acción Scholarship Program reported a more negative attitude toward the use of violence at 3 months (p = 0.01) [20]. The study used a prospective randomized controlled design to assess the effects of a program that used CBPR to engage adolescents, young adults, and parents in the development and implementation of the intervention. The scholarship program focused on three community-prioritized positive youth development strategies: increasing student academic achievement motivation, increasing parental educational support, and addressing perceived barriers to accomplishing educational goals [17, 20, 23]. The intervention was implemented weekly on a weekday evening for four consecutive weeks. In the previously mentioned study with third to fifth graders, there was an El Joven Noble program effect on non-violent conflict resolution self-efficacy at 3 months among higher-risk students (p < 0.05) [5].

Predictors of Future Violent Behaviors

The FEA literature includes longitudinal analyses that examined predictors of future violent behaviors [3, 11, 15, 18, 25•]. Table 2 displays variables positively and negatively associated with future violent behaviors. Previously engaging in an act of violence and exposure to community violence were robust significant predictors of future violence [15, 18, 25•]. Depression, alcohol use, and belief in aggression were also found to be predictors in one study [11, 15]. Three attitudes were found to have protective effects [3, 15].

Table 2 Variables positively and negatively associated with violence
Full size table

Previously engaging in an act of violence was the most significant predictor of future violence [15, 18, 25•]. Baseline perpetration of non-physical aggression, physical violence, intimate partner violence, school violence, and non-school violence were the strongest predictors of engaging in these acts at 3- and 9-month post-intervention in a secondary violence prevention study [15, 18]. Similarly, baseline perpetration of school violence and non-school violence were the strongest predictors of engaging in these acts at 6- and 12-month post-intervention in a primary violence prevention study [25•].

Exposure to community violence was also a predictor of future violence [11, 15, 18, 25•]. Exposure to community violence was a predictor of non-physical aggression, physical violence, intimate partner violence, school violence, and non-school violence at 3- and 9-month post-intervention in a secondary violence prevention study [11, 18, 25•]. Similarly, exposure to community violence was associated with perpetration of school violence and non-school violence at 6- and 12-month post-intervention in bivariate analyses of a primary violence prevention study [25•]. However, in this study, exposure to community violence was not found to be a major predictor in the multivariate analyses [25•].

Depression and alcohol use were positively associated with school violence and non-school violence at 3 and 9 months in a secondary prevention study with middle and high school students [15]. In the same study, belief in aggression was associated with non-physical aggression at 3 months [11].

Three attitudes were found to have future protective effects [3, 15]. Higher academic achievement motivation was negatively associated with school violence and non-school violence at 3 and 9 months in a secondary prevention study among middle and high school students [15]. Greater non-violent conflict resolution self-efficacy and negative attitudes toward gangs were associated with less perpetration of violence immediately post-intervention in a primary prevention study among elementary school students [3].

Cross-sectional Associations with Violent Behaviors and Attitudes

The FEA literature includes cross-sectional analyses of variables potentially associated with violent behaviors and attitudes [4, 10, 16, 18, 19, 22, 24]. Table 2 displays variables positively and negatively associated with violent behaviors and attitudes.

Exposure to community violence was positively associated with non-physical aggression, physical violence, intimate partner violence, hitting back when someone hit me first, and carrying a weapon [10, 19, 22]. Exposure to a firearm in the home was positively associated with school, non-school, and intimate partner violence [16].

Multiple variables were found to be negatively associated with violent behaviors [10, 18, 24]. Thoughtful decision-making, low parental support of violence, low exposure to community violence, low hyperactive–impulsive symptoms, and low inattentive symptoms were significant predictors of not engaging in violence [24]. Higher academic achievement motivation was associated with fewer violent behaviors [18]. Belief in familismo was associated with less non-physical aggression and physical violence [10].

Two variables were found to be positively associated with violent attitudes [10, 19]. Exposure to community violence was associated with violent attitudes [19], and belief in male privilege was associated with greater belief in aggression and greater acceptance of intimate partner violence [10].

Four variables were found to be negatively associated with violent attitudes [4, 10, 18]. Higher collective efficacy and academic achievement motivation were associated with negative attitudes toward violence [4, 18]. Belief in familismo was associated with greater belief in non-violent conflict resolution strategies [10]. Higher collective efficacy and better neighborhood conditions were associated with negative attitudes toward intimate partner violence [4].

Respondent-Driven Sampling

The FEA literature includes the use of respondent-driven sampling (RDS) [27, 28]. RDS was used to estimate the proportion of young people in the partnering Latino community who are at risk for homicide and to elicit existing cultural funds of knowledge on how to reduce homicide risk [27, 28]. RDS was used to ensure the inclusion of youth who do not attend school regularly and/or lack residential stability [27, 28]. Based on responses to the Violence Perpetration and Injury Scale, 29.4% (22.1–38.1%) of 14–21-year olds in this community are at risk of being the victim of a homicide [27]. A total of 46.6% (34.1–58.2%) believe that avoiding dangerous situations is a way to decrease the risk of getting shot or stabbed [28].

Community-Based Participatory Action Research

The FEA literature includes information about using CBPR to conduct Latino youth violence prevention and research. Information on goals, models, facilitators of success, lessons learned, and the importance of using CBPR are included.

A major goal of FEA was to uphold the ideals of CBPR while implementing rigorous methods of prospective randomized controlled trials [7, 15, 20, 22, 25•, 26•]. The goal was to facilitate the inclusion of locally relevant experiences, values, priorities, culture, and history into both the research process and the products [15, 17, 23, 25•]. Adolescents, young adults, and parents are engaged in pursuing funding, setting priorities, and designing, implementing, evaluating, and disseminating findings of their own prevention research programs [3,4,5,6,7, 10, 11, 13,14,15,16,17,18,19,20, 22,23,24, 25•, 27, 28].

The academic partners made a commitment to continually work toward achieving a community-controlled model of a community–academic research partnership [1, 7]. The model recognized that “The fluid reality of community partnerships requires a cyclical process of discovery that allows for and even encourages movement in unanticipated directions and also pays great attention to the cultural beliefs and practices of all partnership participants” [7]. The FEA CBPR program cycled multiple times through pre-engagement, engagement, community assessment, intervention design, implementation, evaluation, dissemination, sustainment, and community self-advocacy phases [3,4,5,6,7, 10, 11, 13,14,15,16,17,18,19,20, 22,23,24, 25•, 27, 28].

A major facilitator of the success of FEA was the reciprocal cultural competence of the original site principal investigator and the partnering Latino community [2]. When FEA began, she enlisted the help of key informants, was sensitive to prior local history, and demonstrated a commitment to respect the local culture and people. She established an environment of trust and quickly recognized that this community already had the collective experience and shared wisdom that would make the project successful. She struggled with, but successfully addressed culture conflict with genuineness, kindness, and a focus on how any action would be received and perceived by the community. The community demonstrated their own cultural competence. “Over the many years of dealing with people that have tried to manipulate and take advantage of them, this community has become very sophisticated and skilled at dealing with outsiders, including academic researchers. … They are cautious and selective as they make choices. They accept the responsibility for and are committed to nurturing, providing, and protecting a healthy future for their children” [2].

Public health students who engaged with FEA shared the following “lessons learned” [6]. Partnerships are a privilege, and developing them takes time. One needs to relinquish control, listen, and be emotionally engaged. Social interaction is the heart of monthly meetings, including food and celebration. Community is power; interconnectedness “provides the strength to endure, grow, and create change” [6]. In collaboration with local women, it was also learned that empowered Latinas have passion, give and receive genuinely, change negatives into positives, and make progress [8].

In the editorial “Why We Need Primary Youth Violence Prevention Through Community-Based Participatory Research,” external commentators discuss FEA and the importance of using CBPR for youth violence research [26•]. “CBPR is an approach to research that allows those most affected by a health issue to be equitably engaged in research. This can have a significant impact, particularly when addressing public health problems in which traditional approaches have had only limited success, and especially in which a clear health disparity exists, such as youth violence” [26•]. They comment on the negative impact of violence on young people and state that it is “important and urgent not only for researchers to advance the science but also for funders to see the clear and growing need to meaningfully support further research in this area” [26•].

Conclusions

The Familias en Acción CBPR program upheld the ideals of CBPR while implementing rigorous methods of prospective randomized controlled trials. The findings indicate that CBPR can be used successfully to engage a Latino community in youth violence prevention and research. Adolescents, young adults, and parents can, will, and should participate in the development, implementation, and evaluation of prevention programs.