A PROGRAM OF RESEARCH WITH HISPANIC AND AFRICAN AMERICAN FAMILIES: THREE DECADES OF INTERVENTION DEVELOPMENT AND TESTING INFLUENCED BY THE CHANGING CULUTURAL CONTEXT OF MIAMI

Abstract

In this article we summarize work with poor, inner-city Hispanic and African American families conducted at the University of Miami Center for Family Studies. We elucidate ways in which this research program has paralleled the treatment development paradigm and has been responsive to changes in local demographics. Specific cultural issues pertaining to Hispanics and African Americans are discussed in light of treatment development and implementation. Future directions and challenges for working with poor, inner-city minority families are addressed.

Since the early 1970s, the University of Miami Center for Family Studies has been designing, implementing, and refining family-based interventions for minority families in the Miami area. From the beginning, our work has focused on family-based treatment for Hispanic immigrant families with drug-using and conduct-problem adolescents. Although we have continued to conduct further research, training, and dissemination efforts in this area, we have also broadened our program of research to include African Americans as well as Hispanics, prevention as well as treatment programs, and ecological as well as within-family interventions.

Although we use the term “Hispanic” often in describing our work, there are drawbacks to such pan-ethnic terms. The term “Hispanic” (as well as the similar term “Latino/a”) is both a unifier and a source of irritation among individuals of Spanish-speaking descent. The term “Hispanic” was originally coined by the U.S. Census Bureau to replace the previous term “Spanish speaking, Spanish surnamed” (Padilla, 2003). The pan-ethnic term arose from the fact that individuals from 21 different countries all spoke the same language and had the same or similar names. However, the terms “Hispanic” and “Latino/a” obscure many important distinctions among the various Spanish-speaking nationalities. As Garvin (2000, cited in Rumbaut & Portes, 2001, p. 2) asks, “What do Cuban blacks have in common with Anglo-Argentines or Mayan-speaking Indians from Guatemala?” Mexicans, Puerto Ricans, Cubans, and individuals of other Spanish-speaking nationalities have vastly different histories in the United States, different racial and socioeconomic profiles, and different cultural preferences.

The success of our work has been brought about through four essential components: (a) collaborating with the Hispanic and African American communities; (b) designing and implementing culturally and clinically targeted interventions; (c) conducting rigorous research to test the impact of interventions; and (d) using our clinical and research findings as a foundation for continuous improvement and refinement of interventions. Becoming part of and collaborating with the ethnic communities has helped us to build trust and to establish ourselves as partners.

STRATEGIES FOR ENGAGING THE COMMUNITY

Our approach to engaging the community has been multilevel. Community can be viewed as comprised of different components. At the broadest level, there are the larger Hispanic and African American communities in Miami. Within each of these larger communities, there are a number of subcommunities, such as the provider community, collaborators drawn from local neighborhoods, and individual families. We engage the community at each of these levels using structural-strategic engagement interventions, drawn from our treatment approach, such as validating, respecting, and tracking the organization’s established power structure and establishing collaborative relationships with leaders within each component of the community.

Engaging the Larger Community

With regard to the larger community, over the 30 years of its existence, the Center for Family Studies has maintained a high profile in the Hispanic community by (a) establishing and maintaining close working relationships with leaders in the local Hispanic community, (b) participating in a broad range of Hispanic community activities, (c) participating in radio and television programming, and (d) having the work of the Center occasionally featured in newspaper articles. More recently, we have taken similar steps with the expansion of our intervention research into the African American community. We have worked closely with African American professional groups in South Florida, have appeared on radio programs, and have had articles about our work written in African American newspapers.

Perhaps our most effective way to engage the larger Hispanic and African American communities has been our respect for these cultures. This respect for culture includes approaching the communities in culturally syntonic ways. For example, in all of our intervention studies, we have used therapists, counselors, facilitators, outreach workers, and often project leaders that are of the same race and/or ethnicity as the client population. Moreover, we have always recruited assessors and other front-line staff members from the population and often the same neighborhoods in which clients live. This has greatly facilitated our understanding of the community as well as given us a familiar and comfortable face in the community.

In keeping with a focus on joining with the client community, a word of caution is in order regarding work with Hispanics. As a result of differences among Hispanic nationalities, therapists should take care in generalizing their experiences with families from one Spanish-speaking nationality to those from other nationalities. Even in cities such as Miami, where Cuban American culture predominates, Spanish speakers of other nationalities should be queried about their own ethnic identity, how they wish to be referred to, and what cultural norms they would like the therapist to follow. Moreover, it should be noted that Hispanic immigrant adults, especially older adults, are often not fluent in English (Mutchler & Brallier, 1999). Therefore, to avoid undue alliances with children or adolescents and insufficient alliances with parents, family therapists working with Hispanic families should be fluent in Spanish and should be familiar with the culture from which client families have originated.

Engaging the Provider Community

Throughout our 30-year history, we have been consistently involved with the provider community through a range of community activities specific to providers. In our efficacy studies, we often refer participants assigned to control conditions to local drug abuse treatment providers. In our on-going effectiveness studies, we have been funded by the National Institute on Drug Abuse to collaborate with local drug abuse treatment providers to design and administer state-of-the-science treatments. Through collaboration with local providers in both our efficacy and effectiveness research, we have become well known within the drug abuse treatment provider community. Similarly, in our work with HIV-seropositive individuals, we have recruited participants from local HIV clinics and referred control-condition participants to local treatment providers when necessary. Moreover, we have participated in the community network of drug abuse prevention and treatment, as well as HIV providers, and have led conferences and workshops for the provider community. Thus, we have become a partner in the network of prevention and treatment services for the range of populations served through our research.

Engaging School and Juvenile Justice Systems for Adolescent Interventions

Our community partnerships also include the public school system. Most of our research with adolescents has depended heavily on the Miami-Dade County Public School system, from which many participants have been recruited. The Center has always had a strong relationship with the public school system’s central office, regional offices, and local schools at which our interventions are conducted. In our prevention studies, we have relied heavily on school officials either to provide access to participants in the case of selected interventions, or to refer participants to us in the case of indicated interventions. In our treatment programs for drug-using and conduct-problem adolescents, we have also worked closely with the public school system’s drug abuse prevention counselors to secure referrals for our studies.

Our treatment research has also relied heavily on our strong relationship with the Public Defender’s office as well as the community’s juvenile drug abuse assessment and recovery facilities. The juvenile justice and drug abuse assessment facilities have provided us with many referrals. Securing referrals from these systems has required that we provide evidence that our intervention programs are efficacious in decreasing adolescent drug use and associated problem behaviors and maintain a consistently strong relationship with the juvenile courts and with the adolescent drug abuse treatment agencies. Ultimately, however, it is our reputation for excellent intervention services that sustains the school, juvenile justice, and adolescent drug treatment systems’ interest in collaborating with us.

Working with Healthcare and Justice Systems in Engaging Women with HIV and Drug Abuse Diagnoses

In our work with women with HIV, we have collaborated with special immunology OB/GYN clinics. Similarly, in working with postpartum women with drug abuse disorders, we have collaborated with doctors and nurses working in the delivery unit for drug-affected children in the county’s primary indigent-serving hospital. In addition, we developed a partnership with the local district office responsible for the management of dependency cases, as well as with the dependency judges themselves.

Engaging Individual Families

All of our interventions are comprised of both transcultural and culturally specific ingredients. Transculturally, our success in bringing hard-to-reach families into therapy has stemmed from our recognition that patterns of family interactions interfere with bringing the entire family into treatment. Consequently, in our treatment work, we have developed specialized family engagement interventions that approach the family according to its own ways of behaving. Further, in the spirit of our family-based work, we work with family members to help the target participants (i.e., with adolescents, the whole family is brought into treatment; with drug-addicted mothers, the mothers are brought into drug abuse treatment with the help of their families). In our prevention work, we have learned to distinguish between the family-process obstacles to engaging families into preventive interventions on the one hand, and the ecosystemic stressors that limit the range of options for poor, inner-city families on the other. These stressors and limited options have resulted in a loss of community support for Hispanic immigrant and African American families.

In terms of culturally specific ingredients, part of the reason for the success of our intervention programs for Hispanic and African American families may be due to the fact that family-based interventions match the value orientations of these communities. Both the African American and Hispanic cultures tend to value families, social interactions, respect, leadership and collaboration. Moreover, also syntonic with Hispanic and African American cultures, all of our interventions seek to develop relationships with families in which we provide leadership for change as well as develop a therapeutic system that is collaborative in nature. In the spirit of collaboration, our counselors are taught to be respectful of all family members.

The specific role of family differs somewhat between Hispanic and African American cultures. Hispanics tend to place a high degree of emphasis on family, to value family support, and to use family as their primary reference group (Sabogal, Marin, Otero-Sabogal, Marin, & Perez-Stable, 1987). The structure of African American families consists of a close network of blood and nonblood relatives including parents, grandparents, uncles, aunts, older siblings, ministers, and elders who operate inside and outside of the family’s home (Nobles, Goddard, Cavil, & George, 1987). Our interventions with Hispanics are, therefore, designed to restore the emphasis on family that may be lost as Hispanic adolescents and parents acculturate. In our interventions with African Americans, we target extended family and nonblood kin as well as nuclear family members.

In the last 10 years, in accordance with increasing disruption in inner-city communities and the subsequent breakdown of the family in inner-city neighborhoods, our services have become increasingly multisystemic and home based. As a result, we have modified our service models to adapt to the changing nature of the family and community contexts that we serve. It is important to note that the overwhelming majority of the families with whom we work are from poor communities. However, it is important to distinguish the influence of poverty from membership within an ethnic or racial group. This notion challenges the common practice of using terms like racial minority and inner city as codes for low income Hispanics or African Americans (Lott, 2002). We believe it is essential to separate the culture of poverty from the culture of Hispanics or African Americans living in the inner city. This is especially important with regard to African Americans, who have historically been marginalized in poor communities (Moore-Hines & Boyd-Franklin, 1996).

HISTORICAL CONTEXT OF INTERVENTION WORK AT THE CENTER FOR FAMILY STUDIES

The history of our work can be put into the context of the National Institute on Drug Abuse (NIDA) three-stage approach to treatment development (cf. Rounsaville, Carroll, & Onken, 2001). Stage Ia and Ib refer to treatment refinement and feasibility studies, respectively; Stage II refers to randomized clinical trials evaluating the efficacy of treatments in controlled laboratory settings; and Stage III refers to real-world effectiveness studies conducted to transport the treatment to community providers. Our work with a specific population or problem would often begin with focus groups or in-depth interviews or surveys with community members and providers, or with studies to understand the nature of problems to be later targeted through interventions. Using information gathered from these basic science methods, we would design or adapt an intervention model for use with the specific population or problem. Corresponding to Stage Ia of the NIDA treatment development model, we would identify intervention strategies that would be syntonic with the cultural and ecological circumstances underlying the group for which the intervention was intended. Based on the strategies identified, we would write a treatment guide for the intervention model, or adapt an existing treatment manual for use with our specific population. For example, our development of Brief Strategic Family Therapy (BSFT; Szapocznik & Kurtines, 1989) began with the integration of Minuchin’s (1974) Structural Family Therapy with Haley’s (1976) Strategic Family Therapy. Similarly, our development of Structural Ecosystems Therapy with adolescents (Robbins, Schwartz, & Szapocznik, 2004) began with the adoption of the home-based component of Multisystemic Therapy (Henggeler, Schoenwald, Borduin, Rowland, & Cunningham, 1998). Corresponding to Stage Ia, we would apply the intervention to a small number of cases, recalibrate the intervention guidelines, and move on to a Stage Ib (small randomized pilot) study. In our Stage-Ib studies, we pilot-tested the intervention on a small sample and modified the intervention manual as necessary based on clinical experience, measures of intervention outcome, and feasibility and acceptability for the target population. Provided that our pilot-test analyses suggested that the treatment model was efficacious, we would launch a more formal and larger Stage-II randomized trial of the manualized intervention developed in Stage I. Only the intervention in which we have worked the longest, BSFT, is currently being tested in a Stage-III effectiveness trial (in which the intervention is implemented by community providers).

Our work, which is described in Table 1, has followed three general trajectories, each of which is summarized in the sections that follow. First, we have designed and implemented culturally syntonic, family-based interventions to treat drug abuse and other problem behaviors in Hispanic adolescents. Second, we have designed and implemented culturally syntonic, family-based interventions to prevent drug abuse and other problem behaviors in Hispanic adolescents. Third, we have designed and implemented culturally syntonic interventions for use with African American adult women with HIV and drug abuse problems and their families. We have also examined the efficacy of some of our Hispanic-adolescent treatment and prevention interventions for use with poor, inner-city African Americans. Our prevention and treatment research with adolescents began as within-family and has expanded to include ecodevelopmental (ecosystemic, developmental, and social-interactional) principles. Our work with African American adult women has been solely family-ecodevelopmental.

Table 1.

Outcome Studies for Center for Family Studies Intervention Programs

Organizing Studies by Ethnicity
Interventions with Hispanics	Interventions with African Americans
Brief Strategic Family Therapy (BSFT)(Szapocznik et al., 1989, Santisteban et al., 1996, 1997)	BSFT (Santisteban et al., 1997)
One Person BSFT (Szapocznik, Kurtines, Foote, Perez–Vidal, & Hervis, 1983, 1986)	Engaging Moms (Dakof et al., 2003)
BSFT Engagement (Coatsworth, Santisteban, McBride, & Szapocznik, 2001; Santisteban et al., 1996; Szapocznik et al., 1988)	SET – African American Women (Szapocznik et al., 2004)
Bicultural Effectiveness Training (Szapocznik, Santisteban, et al., 1986)	SET – HIV-seropositive Women in Recoveryb (Feaster, 2003)
Family Effectiveness Training (Szapocznik, Santisteban, Rio, Perez-Vidal, Santisteban, & Kurtines, 1989)
Structural Ecosystems Therapy (SET) – Adolescentb (Robbins et al., 2004) preparation)	SET – Adolescentb (Robbins et al., 2004)
Familias Unidas (Pantin, Coatsworth, et al., 2003)	Familias Unidasa (unpublished)
Familias Unidas – HIV Preventionb (Pantin et al., 2003)
Organizing Studies by Focus
Within-Family Interventions	Family-Ecodevelopmental Interventions
BSFT (Szapocznik et al., 1989, Santisteban et al., 1996, 1997)	Engaging Moms (Dakof et al., 2003)
One Person BSFT (Szapocznik, Kurtines, Foote, Perez-Vidal, & Hervis, 1983, 1986)	SET – African American Women (Szapocznik et al., 2003)
BSFT Engagement (Coatsworth, Santisteban, McBride, & Szapocznik, 2001; Santisteban et al., 1996; Szapocznik et al., 1988)	SET – HIV-seropositive Women in Recoveryb (Feaster, 2003)
Bicultural Effectiveness Training (Szapocznik, Santisteban, et al., 1986)	SET – Adolescentb
Family Effectiveness Training (Szapocznik, Santisteban, Rio, Perez-Vidal, & Kurtines, 1989)	Familias Unidas (Pantin et al., 2003)
Structural Ecosystems Therapy (SET) – Adolescentb (Robbins et al., 2004)	Familias Unidas – HIV Preventionb Szapocznik & Pantin, 2000b)

Note: Results indicated positive intervention effects unless otherwise noted.

^aProgram produced iatrogenic effects.

^bStudy in process; findings are not included in this article.

Cultural and Contextual Influences on Treatment Development

Brief Strategic Family Therapy (BSFT) was the first intervention model developed at the Center for Family Studies. BSFT was developed as a family-based intervention that was responsive to the findings of a cross-cultural survey given to Cuban immigrants and White Americans in Miami (Szapocznik, Scopetta, Arnalde, & Kurtines, 1978). The survey was developed to gauge the value orientation of the Miami Cuban population, with the goal of developing a treatment model to match this value orientation. Respondents surveyed indicated that the Cubans, when compared with White Americans, valued hierarchical relationships (compared to individualistic), were present-oriented (compared to past or future), and valued a doing, rather than a being, orientation. Given the emphasis on familism in Hispanic culture and a hierarchical (rather than individualistic) value orientation among Cubans (Santisteban, Muir-Malcolm, Mitrani, & Szapocznik, 2002), we assumed that the intervention would need to be family based. As a result, BSFT emphasizes the family as the primary agent of change vis-à-vis adolescent symptomatology. Respecting the family hierarchy, including restoring parental leadership, and directive, present-oriented therapist interventions are of primary importance in BSFT (Szapocznik, Scopetta, & King, 1978).

In the 1970s, most Hispanic families came to treatment for adolescent behavior problems accompanied by strong parent–adolescent cultural conflicts (Szapocznik, Scopetta, Kurtines, & Arnalde, 1978). Although it is natural for younger individuals to adapt more quickly to a new host environment, it appeared that the adolescent’s natural striving for autonomy was exacerbated by acculturation to individualistic behaviors and values, while the parents’natural striving to maintain family integrity was exacerbated by a strong adherence to hierarchical values. One of the earliest targets of BSFT was to correct maladaptive family interactions that accompany exaggerated acculturation/intergenerational gaps between Hispanic immigrant parents and their adolescents by assisting the family in developing conflict resolution skills, increasing understanding and respect for each generation’s values, and restoring the parents’ leadership in the family (Szapocznik, Scopetta, & King, 1978).

To develop a treatment model that would be syntonic with the predominant Cuban value orientation, Szapocznik and colleagues (see Szapocznik & Kurtines, 1989) sought to integrate the structural family therapy tradition (e.g., Minuchin, Rosman, & Baker, 1978) with strategic approaches to therapy (e.g., Haley, 1976). This integration of structural and strategic approaches found in BSFT formed the basis for all of the intervention programs designed at the Center for Family Studies. The structural aspect of the approach holds that problem behavior is embedded in maladaptive family functioning and restructuring those maladaptive family interactions most related to the symptom is likely to produce reductions in problem behavior. The strategic aspect of the approach holds that problematic behavior can be addressed by focusing strategically on specific aspects of maladaptive family interaction that are most amenable to intervention and that are most central to the adolescent’s symptomatology. The structural-strategic approach represents the most transcultural aspect of our work, in that we have successfully used the approach with families from Hispanic and African American cultural backgrounds, and with adolescent as well as adult target participants.

As we have noted elsewhere (e.g., Szapocznik, Hervis, & Schwartz, 2003; Szapocznik & Williams, 2000), the structural-strategic approach guides the process by which we implement our interventions. The content of intervention activities is guided by the principles of the participants’ cultural background, symptoms, and their situation (e.g., poverty, HIV infection, immigration status). For example, in our recent work with Cuban and Central American immigrant families, we attend to the importance of immigration and acculturation in these families’ lives, as well as to the fact that many of these families have left members behind in their native countries. These important aspects of cultural context are used to understand the family’s problems and the adolescent’s maladaptive behavior and to design a strategy to correct the within-family and social ecosystemic mechanisms that help to maintain the adolescent’s symptoms.

INTERVENTIONS DESIGNED FOR HISPANICS

Within-Family Interventions

Brief Strategic Family Therapy (BSFT)

Brief Strategic Family Therapy (Szapocznik, Hervis, & Schwartz, 2003; Szapocznik & Kurtines, 1989) was designed to match the value orientations reported by early Cuban immigrants in Miami (Szapocznik, Scopetta, Arnalde, & Kurtines, 1978). However, the features of the model that were built on what we thought were exclusively Cuban values appear to have certain transcultural qualities. BSFT’s major therapeutic techniques fall into three major categories: Joining, diagnosing and restructuring. The therapist joins the family by initially supporting the family structure, by tracking its patterns of interactions, by reflecting the family’s style, affect, activity, and mood, and by encouraging the family to behave or interact in its characteristic fashion. Family problems are diagnosed in the areas of family organization (including power distribution), boundaries, developmental appropriateness, identified patienthood, and conflict resolution. To assist in structural diagnosis (Szapocznik & Kurtines, 1989) and in assessment of changes in family structure (i.e., interactional patterns), our team developed the Structural Family Systems Ratings to organize the information obtained when families are asked to perform three standardized tasks (Szapocznik et al., 1991).

Restructuring refers to the strategies that the therapist uses in changing maladaptive interactional patterns by capitalizing on existing adaptive interactions. These strategies are implemented to correct only those maladaptive interactional patterns that are most related to the adolescent’s presenting problems (e.g., drug abuse, delinquent behavior). For example, a mother’s habit of hiding her son’s drug abuse from the father may have to be changed in such a way that the parents collaborate around matters related to their son’s drug abuse. Although the parents’ partnership in relation to managing the adolescent needs to be strengthened, the parents’ relationship as a couple may not have to be addressed.

The efficacy of BSFT in reducing behavior problems and drug abuse has been tested in two randomized, controlled NIDA Stage-II studies. Szapocznik et al. (1988) randomly assigned behavior-problem and emotional-problem Cuban preadolescent boys to BSFT, individual psychodynamic child therapy, or a recreational placebo-control condition. Outcome analyses indicated that BSFT was as effective as individual psychodynamic child therapy, and more efficacious than the recreational control, in reducing children’s behavioral and emotional problems, and in maintaining these reductions at 1-year posttermination. However, at 1-year follow-up, BSFT was associated with a significant improvement in observer-rated family functioning, whereas individual psychodynamic child therapy was associated with a significant deterioration in observer-rated family functioning.

In a second study, Santisteban et al. (2003) randomly assigned Hispanic (one-half Cuban and one-half from other Hispanic countries) adolescents with behavior problems to receive either BSFT or group counseling. Group counseling was modeled after an intervention widely used in the community with youth with moderate behavior problems. Brief Strategic Family Therapy was significantly more efficacious in reducing conduct problems, associations with antisocial peers, and marijuana use, and in improving observer-rated family functioning.

We have also explored the extent to which BSFT can be used with African American as well as Hispanic adolescents with behavior problems. In an uncontrolled demonstration study examining the suitability of BSFT for adolescents from both ethnic groups, Santisteban et al. (1997) assessed conduct problems, delinquency in the company of peers, and observer-rated family functioning before and after BSFT treatment. Although BSFT significantly reduced association with antisocial peers and improved family functioning for both Hispanics and African Americans, BSFT was significantly more powerful in reducing association with antisocial peers among African Americans than among Hispanics. Conversely, BSFT was significantly more powerful in improving family functioning among Hispanics than among African Americans.

One-Person BSFT

The prevailing belief in traditional family systems theory was that family therapy had to be delivered to the entire family conjointly. In response to the difficulties of engaging and retaining the entire families of drug-abusing adolescents in therapy, we explored whether it was possible to achieve the same goals of BSFT, as typically conducted in a conjoint fashion, if therapy was conceptualized systemically but conducted primarily through one person. One-Person BSFT (Szapocznik, Foote, Perez–Vidal, Hervis, & Kurtines, 1985) capitalizes on the systemic concept of complementarity, which suggests that when one family member changes, the rest of the family responds by either resisting the change or adapting to the new changes (Minuchin & Fishman, 1981). The goal of One-Person BSFT is to change the drug-abusing adolescent’s participation in maladaptive family interactions that include her or him. Either this causes the family to adapt by producing more adaptive interactions or, more frequently, it produces a crisis that brings the whole family to a session that can then be used to help the family adapt to the adolescent’s change.

To test the efficacy of One-Person BSFT, Hispanic drug-abusing adolescents were randomized to receive either one-person or conjoint (whole-family) BSFT (Szapocznik, Kurtines, Foote, Perez–Vidal, & Hervis, 1983, 1986). In the One-Person condition, therapists were required to work with the drug-abusing adolescent alone in therapy for at least 75% of the sessions, whereas in the conjoint condition therapists were required to work with the conjoint family for at least 75% of the sessions. Both forms of BSFT showed significant improvements in family functioning and significant decreases in adolescent drug use and other problem behaviors, and the efficacy of One-Person BSFT was not significantly different from that of conjoint BSFT.

Bicultural Effectiveness Training

Our first prevention program was based on our observation that families of Cuban adolescents with behavior problems were often characterized by acculturation discrepancies between parents and adolescents. Parents tended to be more oriented toward Hispanic culture, whereas adolescents tended to be more oriented toward American culture. This acculturation gap was often the source of conflict between parents and adolescents and of poor adolescent adjustment (Szapocznik, Kurtines, & Fernandez, 1980). As a result, we designed an intervention, Bicultural Effectiveness Training (BET; Szapocznik, Santisteban, Kurtines, Perez-Vidal, & Hervis, 1984), to achieve the same goals as BSFT through a psychoeducational intervention. The intervention was organized into 12 sessions, each of which covered a culturally laden theme, such as hierarchical versus individualistic ways of relating and conflict diffusion versus emergence with resolution. After a presentation of the cultural differences (content) on specific issues of family interactions (process), adolescents and parents were asked to role-play one another’s cultural orientation and to discuss and negotiate conflicts that emerged during the role-plays. The major reframe in BET is to change the family’s view of the conflict from intergenerational (i.e., parent–adolescent) to intercultural (i.e., American vs. Hispanic). This reframe is used to promote a family-wide view of differences as external to the family, and thereby to lower obstacles to effective problem solving and other restructuring required in the parent–adolescent relationship.

A small, randomized, controlled trial of BET, compared with BSFT, indicated that the program was effective in facilitating parents’ awareness of the American cultural system in which their adolescents were functioning and facilitating adolescents’ awareness of the Hispanic culture from which their families originated. Outcomes were similar between BET and BSFT (Szapocznik, Santisteban, et al., 1986). However, clinical observations suggested that BET appeared to be less stressful on families than was BSFT. Bicultural Effectiveness Training might therefore be recommended in lieu of BSFT for immigrant families manifesting intergenerational and intercultural conflict.

Family Effectiveness Training

We next designed an intervention to combine the preventive and psychoeducational focus of BET with the restructuring techniques used in BSFT in an indicated population of Hispanic preadolescents with behavior problems. Lectures derived from BET were used to stimulate discussions during which the family’s typical way of interacting was allowed to emerge. Brief Strategic Family Therapy restructuring techniques were then used to correct the maladaptive family interactions. This intervention, called Family Effectiveness Training (FET; Szapocznik, Santisteban, et al., 1989), was tested with children who were brought by their parents to the clinic with complaints about the children’s behavior, typically for misbehavior at home. A randomized clinical trial comparing FET to a no-contact control condition was conducted. Outcome analyses indicated that FET was significantly more efficacious than a no-intervention control condition in reducing children’s problem behaviors and improving family functioning, both risk/protective factors for future drug abuse and other problem behaviors (Szapocznik, Santisteban, Rio, Perez-Vidal, Kurtines, & Hervis, 1986).

Engagement into Treatment

Although family-based therapy has been shown to be one of the most effective treatment modalities for adolescent behavior problems and substance abuse (Ozechowski & Liddle, 2000; Williams, Chang, & Addiction Centre Adolescent Research Group, 2000), family-based therapy faces the difficult task of engaging and retaining multiple family members in treatment. We sought to utilize the basic principles of BSFT to devise an intervention to engage families of adolescents with behavior and drug problems into treatment. This work was based on our recognition that families were unsuccessful in entering treatment because of exactly the same patterns of family interactions that caused the symptoms in the youth. To engage a family in treatment, we recognized the need to identify the family’s pattern of interactions that was the obstacle to the family’s entering treatment, and thus we devised a set of interventions to overcome these problems. Within the Cuban immigrant population with whom we worked, we identified four such patterns of interactions. The most prevalent pattern was one in which the adolescent was the most powerful member of the family (Szapocznik, Hervis, & Schwartz, 2003). The parents were not able to bring the adolescent to treatment. If the therapist allies with the parents, she or he acquires their “powerlessness.” Therefore, the therapist joins with the powerful adolescent, who is able to bring the family into treatment. For example, the therapist might approach the adolescent on her or his own “turf” (e.g., in a park after school) and build an alliance with the youth. The therapist identifies what family outcome interests the adolescent, and agrees to help the adolescent accomplish some aspect of this goal. Once the family has been successfully engaged into treatment, the therapist will work to restructure the family so that the parent figures regain authority in the family and the children are subordinated to parental authority.

The effectiveness of BSFT Engagement was tested in three separate studies with Hispanic adolescents with behavior and drug problems and their families (Coatsworth, Santisteban, McBride, & Szapocznik, 2001; Santisteban et al., 1996; Szapocznik et al., 1988). In the first study (Szapocznik et al., 1988), Hispanic (mostly Cuban) families with drug abusing adolescents were randomly assigned to BSFT + engagement as usual (i.e., the control condition) and BSFT + BSFT Engagement (i.e., the experimental condition). In the engagement as usual condition, client families were approached in a way that resembled as closely as possible the kind of engagement that usually takes place in outpatient centers.

The effectiveness of BSFT Engagement was measured using two criteria, first engaging client families to attend the intake session and, second, retaining the client families to the completion of treatment. Retention was measured because it represents a full dose of therapy and is the ultimate goal of engagement interventions. The results of the study revealed that 93% of the families in the BSFT Engagement condition, compared with only 42% of the families in the engagement as usual condition, were engaged into treatment. Moreover, 75% of families in the BSFT Engagement condition completed treatment, compared with only 25% of families in the treatment as usual condition.

The second study (Santisteban et al., 1996) was a replication of the previous engagement study and it explored factors that might moderate the effectiveness of the engagement interventions. In addition, Santisteban et al. (1996) more stringently defined the success of engagement in terms of client families that attended the intake session and the first therapy session. Treatment retention was defined in the same way (i.e., completion of treatment). Participant families were randomly assigned to experimental and control conditions. In the BSFT Engagement condition, 81% of the families were successfully engaged, whereas in the control conditions, 60% of the families were successfully engaged. A major finding of this study was that the effectiveness of BSFT Engagement procedures was moderated by Hispanic nationality. Among the non-Cuban Hispanics (composed primarily of Nicaraguan, Colombian, and Puerto Rican families) assigned to the BSFT Engagement condition, the rate of engagement was high (93%) compared with the lower rate for Cubans assigned to this same condition (64%). There is evidence that, in Hispanic families, acculturation to American values and behaviors is associated with decreased familism (Santisteban, Coatsworth, Briones, & Szapocznik, 2004), and perhaps that the lower engagement rate found for Cubans was due to higher rates of acculturation in the Cuban families (Santisteban et al., 1996). It is possible that such families perceive less need for family involvement in adolescent drug abuse treatment. On the basis of this finding, aspects of Liddle’s (2003; Liddle & Schwartz, 2002) reconnection strategies have been incorporated into a modified version of BSFT Engagement.

This finding has implications for our distinction between transcultural and culturally specific aspects of interventions. The basic assumption about patterns of family interactions probably applies across cultures; however, the specific types of patterns that prevent families from engaging in treatment are probably culturally specific. Thus, we must attend to the specific patterns of interactions in each population and develop specific strategies to overcome them.

A third study (Coatsworth et al., 2001) tested the ability of BSFT + BSFT-Engagement to engage and retain adolescents and their families in comparison to a community control condition. An important aspect of this study was that the control condition was implemented by a community treatment agency and, as such, was less subject to the influence of the investigators. The Hispanic adolescents and families in this study were primarily Cuban or Nicaraguan. Findings in this study showed that BSFT Engagement had an 81% engagement rate, significantly higher than the 61% rate in the community control condition. Likewise, among those engaged, 71% of BSFT cases, compared to 42% in the community control condition, were retained to treatment completion.

Family-Ecodevelopmental Interventions

In the 30 years since BSFT was developed, a number of important social and cultural changes have occurred in the United States. First, the demographic profile of Miami’s Hispanic population has changed considerably. The vast majority of early Hispanic immigrants in Miami were Cuban. However, during the 1980s, immigrants began to come to Miami from Central and South America (Fernandez-Kelly & Curran, 2001). The result of this broadening immigration was a Hispanic population that was diverse in terms of racial background and immigration-related experiences. Although Hispanics in Miami are from diverse economic backgrounds, our work has always focused on poor, nonpaying clients.

Second, increases in crime rates, single parenthood, neighborhood disorganization, and other deleterious social conditions have amplified the number and quality of stressors on families and thus have increased risks for adolescent drug abuse and other problem behaviors. As the social conditions in poor neighborhoods continued to deteriorate, it became clear that modifying within-family processes might not be sufficient to address the multiple risk factors predisposing Hispanic immigrant adolescents to substance abuse and other problem behaviors. Influences on adolescent behavior and development originating from other sources, such as the peer network and the school system, would have to be addressed by improving family–school and family–peer relations and providing parents with skills to collaborate with these important systems. At the same time, the literature on risk and protective factors for adolescent drug abuse and related problem behaviors (Szapocznik & Coatsworth, 1999) as well as for unsafe sexual behaviors (Perrino, Gonzalez-Soldevilla, Pantin, & Szapocznik, 2000) increasingly suggested the important role of a complex set of social contexts.

During the 1990s, all of our research programs (adolescent treatment, adolescent prevention, and programs with African American adult women) began to incorporate ecodevelopmental principles. The specific interventions that resulted from this incorporation differed according to the cultural and developmental needs of the target population and according to the specific problems being addressed. Our intervention work shifted from an intrafamilial structural-strategic perspective to an ecodevelopmental perspective that combined structural-strategic, developmental, and extrafamilial interventions.

Ecodevelopmental theory (Pantin, Schwartz, Sullivan, Coatsworth, & Szapocznik, 2003; Szapocznik & Coatsworth, 1999) brings together an ecosystemic focus, a developmental focus, and a focus on social interactions. The ecosystemic focus is drawn from Bronfenbrenner (1979), who posited four levels of social context: Microsystem (e.g., family, peers, school), mesosystem (e.g., parental involvement in school, family’s contacts with health care providers), exosystem (e.g., a target adolescent’s parents’ workplace or support system), and macrosystem (i.e., social or cultural context).

The second element of ecodevelopmental theory is a developmental perspective that emphasizes individual development as a function not only of one’s current social context, but also of changing conditions in the social context over time. Hence, both person and context are viewed as evolving and changing across the lifespan, and these changes take place across various systems. For example, according to Dodge and Pettit (2003) and to Patterson, Reid, and Dishion (1992), the development of antisocial behavior may begin with the deleterious effects of difficult temperament on attachment relationships in infancy and evolves into poor self-regulation and rejection by prosocial peers in childhood, association with deviant peers and substance use in early adolescence, delinquency and criminal involvement in later adolescence, and finally antisocial personality in adulthood.

The third element of ecodevelopmental theory is social interactions. It is postulated that risk and protection are expressed in the patterns of relationships and direct transactions between individuals within and across the different contextual levels of the social ecology (Garbarino & Abramowitz, 1992; Szapocznik & Coatsworth, 1999). For example, the amount of social support that parents perceive from others is directly predictive of the supportiveness of their own parenting (Swick & Broadway, 1997), which in turn may affect the likelihood of adolescent drug abuse (Vazsonyi, Hibbert, & Snider, 2003).

Putting together these three factors, ecosystemic, developmental, and social-interactional, ecodevelopmental theory (Szapocznik & Coatsworth, 1999) views social context as influencing the child through social interactions, with the effects of one aspect of the social context affecting the child in ways that may later be manifested in a totally different aspect of context. For example, interactions occurring within the family may predispose adolescents to associate with specific types of peers.

Our family–ecodevelopmental treatment and prevention interventions focus on the microsystem and mesosystem levels. We also work extensively in the exosystem to increase social support for Hispanic immigrant parents. We work within the family microsystem in all of our interventions. With regard to the mesosystem, in our work with adolescents we work within the family–school, family–peer, and family–juvenile justice mesosystems, and in our work with drug abusing and HIV-seropositive adult women important mesosystems are the family-healthcare and the family-drug treatment mesosystem. With these women, self-help groups can be important microsystems as well that need to be created or strengthened. The macrosystem, although not amenable to direct intervention, serves as a cultural frame for much of our work. For example, the reality of life in urban America creates parenting difficulties for Hispanic immigrants from rural villages. These parents may be unprepared to handle the tasks (e.g., monitoring schoolwork, supervising peer relationships) involved in raising an adolescent in the United States (Pantin, Schwartz, et al., 2003). For another example, pervasive racial discrimination and lack of access to appropriate healthcare create difficulties for African American families with drug-abusing members in accessing the services that the symptomatic individual requires.

Structural Ecosystems Therapy

In our treatment work with Hispanic and African American adolescents, we (Robbins et al., 2004) have developed Structural Ecosystems Therapy (SET), an ecological extension of BSFT, to address the need for extrafamilial as well as within-family intervention. In SET, basic BSFT techniques developed for use within the family are utilized in working within the family–peer, family–school, and family–juvenile justice mesosystems. The intervention is currently being tested with a sample of Hispanic and African American adolescents presenting with delinquency and drug abuse.

Familias Unidas

In our early prevention work (BET and FET), we intervened mostly within the family microsystem. However, we sought to create a prevention program that would simultaneously reduce risk and increase protection at the multiple ecodevelopmental levels. Intervention activities proceed from the umbrella of parent support networks, where groups of parents meet with a facilitator to discuss the worlds of the child: Family, school, and peers. Home visits are also provided, during which parents have the opportunity to enact the intervention skills learned in session with their own adolescents in the family setting.

Within the family microsystem, activities are conducted to reinvest parents in their adolescents’ lives, to encourage effective behavior management techniques, and to increase parent–adolescent bonding. In the family–peer and family–school mesosystems, activities are conducted to encourage parents to manage their adolescents’ peer activities (e.g., organize and monitor peer activities; collaborate with peers’ parents) and to develop collaborative relationships with the school system. Parent support groups are used to connect Hispanic immigrant parents with one another and to ameliorate the social isolation that often accompanies moving to a new country with unfamiliar cultural practices (cf. Leon & Dziegielewski, 2000).

In an outcome study, Pantin, Coatsworth, et al. (2003) randomly assigned Hispanic adolescents to either Familias Unidas or a no-intervention control condition. Results indicated that Familias Unidas was significantly more efficacious than was the control condition in increasing parental investment and in decreasing adolescent behavior problems. The reduction in behavior problems within the Familias Unidas condition was most apparent during the summer vacation, when behavior problem scores (combined parent and adolescent report) in the control condition were nearly five times as high as those in the Familias Unidas condition. This finding is particularly important because poor minority families do not have the financial resources to provide structured activities for their adolescents during the summer (cf. Black & Krishnakumar, 1998). The parenting techniques, supervisory skills, and social support fostered through the intervention helped Hispanic immigrant parents to manage the large blocks of unstructured time that adolescents have during the summer vacation.

Although Familias Unidas was developed with Hispanic families, we decided to implement it with African American families to assess the generalizability of the intervention. African American adolescents and their families were randomly assigned to either Familias Unidas or to a no-treatment control. Outcome analyses revealed that parental investment, family unity, and family cohesion decreased in both the Familias Unidas and control conditions, with sharper decreases observed in the Familias Unidas condition. At present, we have not been able to explain these unexpected findings. Familias Unidas, an intervention designed specifically for Hispanic immigrant families, may not have been appropriate for, or compatible with, African American culture.

Future directions

Based on the results of Familias Unidas with Hispanic immigrant families, we have developed a briefer and more focused version of the intervention. In the streamlined version, the intervention is much more structured, and the number of group sessions and home visits is standardized. Given the growing importance of preventing sexual behavior and HIV infection among adolescents (Call et al., 2002), we have expanded the intervention focus to include HIV prevention as well as drug abuse and problem behavior prevention. The streamlined version of Familias Unidas combines the original module (i.e., strengthening the family, connecting the family to the adolescent’s peer and school worlds, and providing social support for parents) with a module focusing specifically on promoting parent–adolescent discussions about sexuality and about HIV (Krauss et al., 2000).

It is important to address racism and discrimination as barriers when working with African Americans. Failure to recognize such experiences will likely result in disengagement from the therapeutic work or, even if engagement is successful, failure to achieve desired outcomes. For example, in teaching parents about increasing involvement with the school system in Familias Unidas, we learned that African American parents were often ignored or their concerns minimized by school staff, and that these parents reported that their children often received harsher punishment (e.g., suspension versus detention) for the same offenses committed by children of other ethnic groups.

Future Directions for Work with Hispanic Families

A new direction taken by a team of investigators in our Center has focused on the utilization of Stage-I treatment development mechanisms to develop, manualize, and test new innovative treatments that address specific client characteristics among Hispanics (Santisteban, Muir, Mitrani, & Mena, 2004; Santisteban, Suarez-Morales, & Mena, in press). One such effort has focused on Stage-Ia and -Ib studies to develop and test an integrative family-based treatment addressing the specific needs of drug abusing adolescents with borderline personality disorder features (Santisteban et al., 2004). A second on-going Stage-I treatment development effort seeks to integrate family interventions, individually oriented interventions and psychoeducation modules into a “flexible manual” for the treatment of Hispanic drug abusing adolescents and their families (Santisteban et al., in press). The goal of this flexible manual is to ensure that the treatment package can be tailored to the specific life situations (e.g., blended families), culture-related stressors (e.g., immigration-related parent–child separations; acculturation stress), and other possible unique characteristics (e.g., traditional perspectives on drug/alcohol use and risky sexual behaviors). Having developed the flexible manual and procedures for selection of modules, the project is currently in the Stage-Ib phase.

INTERVENTIONS DESIGNED SPECIFICALLY FOR AFRICAN AMERICANS

Structural Ecosystems Therapy for HIV-seropositive African American Women

We began our work with poor, inner-city, African American women because of the unusually high HIV contraction rates among this population (Centers for Disease Control and Prevention, 2002a, 2002b, 2002c). We first conducted in-depth case studies of four pregnant, HIV-seropositive, inner-city, African American women (Shelton et al., 1993) to identify major areas of need and concern in these women’s lives. Among the primary themes that emerged from these case studies were daily hassles (e.g., child care [a family stressor], transportation) and the presence and importance of at least one supportive, involved, and concerned family member. The case studies highlighted the protective effect of family involvement against life hassles, depression, and distress.

Our next step was to conduct a larger descriptive study of African American women with and without HIV (Smith et al., 2001). When asked to list their primary sources of support, nearly all of the women (both HIV-seropositive and HIV-negative) included at least one family member. Compared with those without HIV, African American women with HIV reported significantly less perceived control over hassles and other life stressors and significantly higher levels of avoidant and support-seeking coping.

Smith et al. (2001) noted that both the women with and without HIV in this sample were significantly more anxious and depressed than the seropositive and seronegative gay men in the Sahs, Goetz, Reddy, and Rabkin (1994) sample. This placed the women with HIV at greater risk, because psychological distress has been associated with worsening of HIV symptoms and decreased quality of life in individuals with HIV (Cruess et al., 2000; Sowell et al., 1997). Moreover, using the same data set from which the Smith et al. (2001) study was published, Feaster and Szapocznik (2002) found that, among poor, inner-city, African American women, the mean levels of hassles across all family members accounted for more variance in distress than did the woman’s own stressors. Taken together, all of these findings support the use of a family-based intervention to reduce family-related hassles and to reduce distress in the HIV-seropositive woman.

Within our case and descriptive studies, we observed that poor, inner-city, African American women with HIV often experience a multitude of deleterious ecosystemic influences, such as isolation from the health care system, strained relationships with family members and friends, persistent residential mobility, and economic hardships (Jackson-Gilfort, Mitrani, & Szapocznik, 2000; Nelson, Mitrani, & Szapocznik, 2000). As a result, we decided that a family–ecodevelopmental intervention was most appropriate for this population.

In our interventions for African American adult women, we attempted to understand adult behavior as a function of social–ecological context, developmental transitions, and social interactions (Mitrani, Szapocznik, & Robinson-Batista, 2000). In our work with African American women with HIV, we aimed to (a) establish positive connections between the family and needed support systems; (b) reduce conflict between and among representatives of the woman’s social context (given that such conflict is a primary source of social stress); and (c) strengthen supportive relationships within the family (Mitrani et al., 2000). We have used SET, our ecological adaptation of BSFT, with African American adult women with HIV. Similar to the version of SET that we use with adolescents, the version that we use with African American women with HIV includes basic BSFT techniques, such as joining, diagnosing, and restructuring to correct maladaptive interactions, both within the family and between the family and other important support systems in the woman’s life (e.g., health care providers). Such connections are designed to help the HIV-seropositive woman “to skillfully and strategically interact with service systems and to capitalize on and enhance the tradition of strong social supports in African American families and communities” (Mitrani et al., 2000, p. 244). Moreover, SET aims to connect the various microsystems in the woman’s life (e.g., family, friends, health care providers) so that individuals within these microsystems can collaborate in supporting the woman.

The efficacy of SET was tested in a randomized controlled study. Szapocznik et al. (2004) randomly assigned African American women with HIV to SET, a Person-Centered Approach condition designed to control for common factors in psychotherapy or a no-contact, community control condition. At baseline, the sample had mean distress scores around the clinical threshold. Results indicated that SET reduced psychological distress significantly more than did the Person-Centered Approach or community control. Post hoc analyses indicated that SET was most efficacious in reducing psychological distress in those women who were closest to the clinical threshold for distress at baseline. In contrast, SET was not efficacious in reducing distress for the small number of women in the sample who were more highly distressed at baseline. Structural Ecosystems Therapy was associated with significantly greater reductions in family hassles compared with the Person-Centered Approach and community control. Post hoc analyses indicated that SET was associated with reductions in family hassles only in women reporting the highest amounts of family hassles at baseline. Structural Ecosystems Therapy was not efficacious in increasing family support.

Enrolling Mothers of Substance Exposed Infants into Drug Abuse Treatment

Our work with mothers of substance-exposed infants began with interviews conducted with pregnant and postpartum women in drug abuse treatment. Our initial intent was to develop a structural ecosystems intervention for these women. However, through our interviews with the women and with treatment program staff, we learned that drug-abusing, pregnant and postpartum women often desperately wanted to maintain custody their children. Most of these women eventually lost custody of their children in dependency court, because they failed to enter and complete drug abuse treatment. The women that we interviewed suggested that we should design an intervention to help other drug-abusing, pregnant and postpartum women to enter drug abuse treatment. Unless they enter and successfully complete drug abuse treatment, women who give birth to a drug-exposed child will have their parental rights terminated (Kovalesky, 2001).

Because of most mothers’ desire to maintain custody of their children, the birth of a child may provide a unique window of motivational opportunity to engage drug-abusing women in treatment. To facilitate engagement and retention of these African American mothers in drug abuse treatment, we created an adaptation of our BSFT engagement procedures called the Engaging Moms program. Engaging Moms procedures involved asking the woman to construct a genogram to identify those family or extended kin members who were most influential in her life. The therapist then asks the woman’s permission to contact each of these family members. Those family members whom the therapist perceives as possessing the majority of power in the family (often the woman’s partner, sibling, or parent) are then enlisted to help engage the woman into drug abuse treatment. Only non-drug-using family members are involved in these engagement efforts.

In family meetings, the woman’s family members (coached previously by the therapist) offer support to the woman but insist that they will withdraw their support unless she enters treatment. The family members are asked to reaffirm their care and concern for the woman and for her children, and to emphasize that the woman’s children may be taken away from her if she does not enter drug abuse treatment. An additional component in this intervention is the development of a close relationship between the therapist and the woman to facilitate the trust the woman needs to allow herself to be guided by the therapist.

The efficacy of the Engaging Moms program in enrolling substance-abusing, postpartum mothers in drug abuse treatment was tested in a randomized controlled study with a services-as-usual control condition (Dakof et al., 2003). Each of these mothers or their infants had tested positive for cocaine at the time of delivery. To facilitate retention in treatment in the Engaging Moms condition, participants who were successfully engaged into drug abuse treatment received study services (e.g., reminders to attend treatment sessions) for their first 4 weeks of treatment. Eighty-eight percent of mothers in the Engaging Moms condition were successfully enrolled in drug abuse treatment, compared with 46% in the services-as-usual condition. Moreover, 67% of mothers in the Engaging Moms condition were retained in treatment for at least 4 weeks, compared with 38% in the services-as-usual condition.

Further Challenges to be Addressed in Working with African American Families

As mentioned earlier, we found that SET was not efficacious with highly distressed HIV-seropositive African American women. To continue to improve the SET intervention, a module for highly distressed women has been developed and integrated in an on-going study (Feaster, 2003). This second-generation study tests the efficacy of SET in preventing drug abuse relapse and in promoting antiretroviral medication adherence among HIV-seropositive women who have recently completed drug abuse treatment.

The disruption of African American marital relationships has presented extraordinary challenges for the African American family (Moore-Hines & Boyd-Franklin, 1996; Sue & Sue, 1999). In African American families, fathers and father figures often do not live in the same home as mothers and children (McAdoo, 2002). Fathers or father figures should be invited to therapy regardless of whether they reside in the children’s home. Because fathers are often not involved in or invited to therapy with behavior-problem and drug-abusing adolescents (Duhig, Phares, & Birkeland, 2003), it is particularly easy for therapists to marginalize African American fathers. In general, mental health clinicians are accustomed to seeing women, more than men, particularly for the treatment of children and adolescents. It is also important to recognize that mainstream society often views African American men as aggressive and violent, and it is important that this mindset is not taken into family therapy interventions.

It is important to recognize that the African American families seen in our research studies have fathers or father figures present. These men can be overlooked if the assumption is to look for the traditional nuclear family. To avoid excluding a member of the family, in BSFT we propose meeting with the entire family in the first session. It is particularly damaging to therapy to meet with the father only after conducting several sessions with his wife, ex-wife, or current or former female partner. This leads to suspicion and feeling of blame when the father is finally invited to therapy. Moreover, to ensure that extended family or non-blood relatives are not excluded, we ask the target participant to identify the individuals who serve in family-related capacities, using the Protocol to Identify the Family (Pequegnat et al., 2001). This procedure inquires about individuals who live in or outside the home, help with housework, provide childcare, provide financial support, or serve as mentors. Individuals identified on this protocol are invited to therapy even if they are not related to the target participant by blood or marriage.

Addressing single parenthood and social isolation among African American mothers

There is a common perception of the strong African American woman who is often raising her family single handedly. For single African American mothers, especially those isolated in poor ghettoes, struggling to raise their families alone, it is important to acknowledge and encourage their strength and self-reliance. However, it is equally important to strengthen support systems for socially isolated African American women, as in our implementation of SET with HIV-seropositive African American women. African American culture tends to espouse a collectivist value orientation, suggesting that social isolation maybe particularly painful for these women.

Referring African American adolescents with behavior and drug problems to treatment rather than incarceration

Securing African American referrals from the juvenile justice system has been a particular challenge for our work. Unlike their Hispanic counterparts who enjoy majority status in Miami, African American youth with severe drug and behavior problems are more likely to be referred to juvenile or adult incarceration than to treatment. The lower rate of referral of African-American youth to mental health treatment prevents them from gaining access to beneficial interventions.

CONCLUSION

In this article, we have reviewed three decades of culturally and clinically targeted intervention development and implementation with poor, inner-city Hispanic and African American families. We have reviewed the historical and cultural circumstances confronting these ethnic groups in the Miami area. We have drawn on these circumstances in formulating our intervention programs and in suggesting future directions and challenges in working with these populations. As our work has been primarily with Hispanics and has only recently expanded to target African Americans, we have devoted considerable effort to engaging the African American community. Moreover, in light of our findings that interventions designed for Hispanic immigrants may not be efficacious with African Americans, we acknowledge the need to design culturally targeted intervention programs for African American families.

We have also capitalized on lessons learned from our Stage-II efficacy studies. For example, in our original work with BSFT, we found that engagement rates were surprisingly low. As a result, we developed specialized strategies to engage families into treatment. Similarly, in our clinical trial evaluating the efficacy of SET for African American women with HIV, we found that the intervention was not efficacious for women who were highly distressed entering treatment. We have used this finding to design an adaptation of SET (currently being tested) that incorporates strategies to manage high levels of distress (Feaster, 2003).

In summary, our work illustrates the value of developing culturally and clinically targeted interventions for ethnic minority families. It is important that intervention programs be designed to match the culture of the target population as well as its daily realities, such as living in poor and highly disrupted neighborhoods. However, it is especially important to distinguish the culture of a given minority group from the culture of the inner city or the culture of poverty. Doing so may help to maximize the extent to which minority group members feel valued, remain engaged in treatment, and find that the treatment meets their specific needs.