Seeking Models and Methods for Cultural Adaptation of Interventions: Commentary on the Special Section

Early in the literature on developing and adapting psychosocial interventions for Hispanic groups, Rogler, Malgady, Costantino, and Blumenthal (1987) suggested three approaches. The first one was to make traditional treatments more accessible to Hispanics. Another was to select available therapies according to their proximity to Hispanic cultures and apply them to this population group. And a third was to extract elements from Hispanic culture to modify traditional treatments. These approaches are not mutually exclusive. In fact, they may work together very effectively in the selection and use of empirically supported therapies for Latinos. After all, there is a growing literature on adapting and tailoring therapies to diverse Latino groups in different parts of the U.S. and across the life span. The collective efforts in cultural-adaptation research have been to improve accessibility of tested interventions for Latinos and use critical Hispanic cultural elements to enhance the utility of these treatments. In many ways, Rogler et al.'s classification of approaches is perhaps more relevant today than when the paper first appeared. As someone who began in the mental health field when little if any attention was paid to race and ethnicity—and, when included, race and ethnicity were often framed as sources of pathology rather than as sources of strengths—I am impressed by the progress we have made in placing culture at its rightful place for ensuring treatment use and success. But I am fully aware that we still have a good distance to go to incorporate cultural factors stemming from a common ethnicity to make interventions relevant and effective with Hispanics. Our progress may have been limited by the lack of empirical substantiation of the specific features that reflect Hispanic cultures. Many cultural concepts that are associated with Latinos, such as respeto, familismo, marianismo, and personalismo, have appeared in the literature (and reappeared as they do in the papers in this section) with such regularity that they are now found in the common lexicon of researchers and clinicians working with Latinos. The general operating assumption has been that Latinos will be better engaged and served if we observe these cultural features when we design and implement our interventions.

This assumption, however, has not stood up to critical scrutiny. In a recent review of the literature on Latino cultural values, Melanie Domenech Rodriguez, Ana Baumann, Audrey Schwartz, and I are finding that there is some definitional consensus that has gradually coalesced around the cultural values. Still, many of these concepts have not been operationalized or operationalized with sufficient uniformity that we can comfortably say they can be or have been used in intervention-adaptation research. Dumka, Lopez, and Carter (2002), for example, found in their review of 21 parenting interventions for Latino families that most studies identified one or more pan-Latino values but few citations or other information were presented “to substantiate the validity of these values” (p. 215). This is certainly a cause of concern and should steel our determination to get a better handle on culture and how culture fits in the adaptation and implementation of empirically derived treatments and interventions.

Connecting Culture and Behaviors

The papers in this special section of Cognitive and Behavioral Practice are moving in the right direction. They display the work of our colleagues who are producing knowledge using ethnographic methods with Hispanics from different origins and heritage, different age groups, and dispersed geography (i.e., Indiana, New York, Massachusetts, Texas) and operationally defining cultural values, beliefs, behaviors, and practices.

The authors of this special section draw on the lived experience of youth and parents and on what Dumka et al. (2002) term “locationally contextualized knowledge” (p. 205). The authors of this group of papers are reaching for answers to questions that will advance our knowledge, such as Which cultural constructs are important for intervention research for specific gender, age groups, and situations? How do we operationalize the cultural features for the specific intervention and target group? How do we implant the cultural values and practices in the therapy process?

Our ultimate goal is to make concepts and approaches for adapting treatments for Latinos and other ethnic and cultural groups easily accessible to clinicians and researchers. Emerging from these papers are both conceptual and methodological approaches to intervention-adaptation for cultural relevance. We see in them the move from empirically validating the cultural values, beliefs, and practices towards identifying the behaviors that are associated with the cultural constructs. In this way we can guide clinicians and researchers on how cultural sensitivity appears in the interaction between Latino adults, between parents, between parents and children, and among family members in an extended family. Furthermore, we can heighten the cultural competence of therapists in their interactions with parents, children, and families. Barker, Cook, and Borrego (2010; this issue) provide insights into how values might appear behaviorally in interventions that target parenting behavior with young children. Dumas et al. (2010; this issue) highlight childrearing as both a cultural and educational task that parents undertake and the essential need to integrate parents' input in the intervention. Calzada (2010; this issue) documents the use of ethnographic methods with parents of preschoolers and gets at the operationalization of the familismo and respeto. In this manner, she explores two common aspects of familismo: the attitudinal (e.g., beliefs about the centrality of obligation to family, about emotional and physical closeness, about protecting the family) and the behavioral (e.g., sharing finances, living arrangements, daily activities, childrearing). Similarly, Calzada reaches for behavioral descriptions of respeto by asking parents, “What does it mean?” The parents' answers to the question yield in four domains for respeto: children's obedience and conformity to authority, particularly parents; children's deference, decorum, and public behavior.

Cardemil et al. (2010; this issue) shift away from parenting interventions to focus instead on adapting an intervention with a group of Latinas of childbearing age who show elevated levels of depression. No doubt their parenting skills are or will be strained under the conditions of their depression. Cardemil et al. draw on the urban context of a group of women representing different Hispanic groups and view cultural adaptation not as part of the active ingredient of the treatment but rather an enhancement that raises treatment uptake, acceptability, and buy-in. The adaptations resonate with what women believe and the worldviews that they bring from their cultures.

Common across these papers is the inclusion of both surface and deep structure dimensions for cultural adaptation (Resnicow, Baranowski, Ahluwalia, & Braithwaite, 1999). Surface structure modifications of interventions are those that show congruence with superficial characteristics of the Latino groups served (e.g., language, music, and other cultural symbols), and that enhance the receptivity, acceptability, and commitment to completing an intervention. One might think of these as face validity issues that are not themselves part of therapeutic action but that set the conditions for participants to want to attend and continue attending, follow-through on activities, and accept the proposed intervention. Deep structure modifications are those that have a textured appreciation and understanding of ethnic, cultural, social, environmental, behavioral, and psychological features that are inherent in a group. Thus, for example, knowing parents' attitudes about how a child should behave toward parents, grandparents, and other adults (e.g., respeto, obedience, conformity) and recognizing that parents prefer roles that are active as they teach their children to be bien educado (well mannered, well-behaved) are cultural beliefs and values that cannot be easily changed. They thus necessitate that the means by which the intervention is delivered be altered. Barker et al. (2010; this issue) provide considerable evidence for how we might look at cultural factors. More than what we can get from face validity, the deeply ingrained parenting beliefs and behaviors that we hear from parents are essential not just to inform the intervention but to motivate and engage parents in change efforts that they accept and believe as important to their children and families.

Deep structure issues are seen in the discussion of discipline. Discipline is identified by a consultant in Dumas et al.'s (2010; this issue) study as “the biggest topic parents want to talk about.” It is driven home by the feedback Dumas et al. get from mothers of Head Start–age children who indicated a preference for physical punishment in disciplining their children. But these mothers also understood the different expectations in the U.S. and were willing to subscribe to changes in their practices. It then becomes incumbent on the adaptation of the intervention to recognize that this long-held attitude and belief, and the corresponding behaviors (i.e., spanking), will require serious consideration. Cultural sensitivity would have the therapist recognize and validate the importance of this approach to the mothers. A culturally competent therapist would then help mothers find new ways of disciplining—by introducing and reinforcing those components of the intervention that provide parents with new options. Cultural competence then also includes repeating and adapting these components until parents feel comfortable and have internalized new alternatives and approaches to physical punishment.

Whose Behavior Should We Be Changing?

The special-section papers connect cultural beliefs and behaviors for the purpose of making interventions more relevant. Parents in Calzada's groups made it quite clear that ignoring their children's unacceptable or inappropriate behavior was not something that they believed would be effective. Parental reactions such as ignoring or giving children time-out seemed to run counter to their notions about childrearing. “I don't think that would work,” one of Calzada's parents said about ignoring her son's behavior. “As a parent, you need to teach your child what is right and what is wrong. You just can't walk away” (Calzada, 2010 -this issue, page 172). In her words, this parent, like many others, indicates that she subscribes to a culturally based belief that parenting is an active, not a passive, process.

Getting the child to obey, conform to the family and social good, defer to parents, grandparents, and other adults, and maintain public decorum that will not embarrass them or shame their family are considered aspects of active parenting process. The socialization of children to observe familismo and respeto, as stated mothers who participated in both Calzada's (2010; this issue) project and Dumas et al.'s (2010; this issue) project, points also to the importance in many Hispanic cultures of the family's good, not just the individual's good. It is for this reason that mothers told Dumas and his colleagues that they did not like particularly setting up child-centered routines and preferred physical affection as a form of praise over verbal praise or tangible rewards. Active parenting means a larger role for mothers and fathers, and adult family members, not a lesser one.

This discussion raises the question, Whose behavior it is it that we are trying to change when we adapt interventions? Sure, most interventions are intended to change how the recipient of the therapy behaves under specific and general conditions. Yet, there seems to be a point where the themes across them converge. That point is that cultural adaptation may be less a matter of changing the therapy and more of changing the therapist's behavior. Cardemil et al. (2010; this issue) seem to agree in that our approach is not to modify the intervention itself but rather that we are more likely to adapt the means and symbols of the intervention. If we believe, as many do, that there is an active ingredient that the therapies hold, usually in the nature of the therapeutic activities, their dosage, sequencing, and intensity, then it is the manner in which the therapists deliver the intervention that is what we change. Effectiveness may then be enhanced by how therapists alter their behavior, the language, symbols, and activities of the intervention. These changes then can create the many conditions that propel us to adapt interventions, so that interventions are perceived as preferable, accessible, and acceptable to the target group.

From the Interpersonal Level to Program Level

The approaches taken by Calzada, Cardemil et al., and Dumas et al. do not just stop at understanding culture in Hispanic families; they also consider the overall program within which an intervention is delivered. Calzada addresses adaptation at the program level by looking at the modality of the intervention (i.e., individual, family, group), its content, clinical process, and assessment. Cardemil et al. and Dumas et al. approach their interventions and populations with a similar emphasis on the program in which the intervention is embedded. Like Calzada, Cardemil et al. focus on design of the program, content, delivery, and providers. On the delivery of the intervention and the providers who provide the service, Cardemil et al. move to deeper layers of cultural adaptation by creating conditions in which the value of personalismo is integrated into the delivery and the therapists are trained and monitored so that their behavior, not just the intervention, is made more congruent. With more experience, Cardemil's young therapists will be prepared to advance the field of adaptation.

Dumas et al. are partially spurred into action by the urgent appeal of a local social service director for a Spanish-language, culturally informed intervention that already existed in the community. What makes the work so impressive is that the demand came from the community, not from the researcher. The recommendations that Dumas et al. make that clinicians and researchers consult widely with members of the target population reflects a search for local, specific contextual data, much as Calzada and Cardemil et al. do in deriving information from the persons for whom the intervention was intended.

Future Directions

The field of intervention, translational and implementation research needs more, not less, attention to adaptation. Adaptations can be based on key characteristics of the populations we serve, matters such as culture, race, gender, ethnicity, socio-economic status, immigration or migration histories, episodic or chronic financial hardship, homeless, and food insecurity. But adaptation science in the real world cannot simply take one of these characteristics; it has to take salient or the most pressing clusters of issues. Furthermore, we cannot, as Miranda, Nakamura, and Bernal (2003) point out, “test all treatments with every subpopulation” (p. 478) and in effect go through adaptation research each time an intervention is introduced to a particular community. We have to adopt more economical approaches to make extant interventions accessible to Latino and other populations.

What we need—and these papers show us some of the ways—are models and methods for adapting and delivering interventions that providers can implement without going through the intensive scientific approaches that researchers must undertake. After all, these interventions are provided to Latinos every day across our service systems. How can we expect to test each adaptation first before it is delivered? We would all agree that adapting interventions such that they are contextualized by acculturation, nationality, recency of immigration, geographic location of community, generational status, or any other set of variables can go a long way to transforming cultural competence from a lofty abstraction into a practical reality. Each of the four papers in this special section touches on this important goal. As we amass more papers like these and disseminate them widely, the findings reported by our colleagues about the values, definitions, and adaptations that were generated in their projects can be stored for use again in other interventions. Dissemination with similar populations helps others avoid having to “reinvent the wheel” each time. This stored knowledge would be useful to researchers studying other age groups. These papers can be points of departure for others. After all, not all values and beliefs are equally salient or necessary in every intervention or with every age group.

Our search has to take us to useful, applicable models for adaptation. What we can offer the field going forward are decision trees, algorithms, calculus, or other conceptual models that clinicians in the proverbial trenches of social and health service delivery can use as they tailor empirically supported therapies, like those described by Calzada, Cardemil et al., and Dumas et al., to the populations they serve. From such adaptation experiences as reported in these papers and those that exist in the literature, the field can begin to generate general principles for adapting treatments to different groups.

With the need for adaptation comes the need for methodologies for adapting. Models are one step in this direction. Each step in a conceptual model or algorithm, however, involves a decision (“What should we do?”) and an action (“We will do this”). This implies determining at what point the content of the intervention is changed and at what point it is rather a matter of altering the interventionist's behavioral interaction with clients. It seems worthwhile, too, for the field to provide the methods (e.g., tool kits) for converting decisions into actual adaptations that can be implemented, possibly tailored and “personalized,” at the client level.

Furthermore, as we continue our search for models and methods, we need to think carefully about the sectors of services in which we operate and in which we adapt and implement interventions. The selection of interventions for Hispanics, or any other population group, and how we adapt the interventions are dictated by whether we are working out of a mental health clinic, a residential treatment center under the auspices of the child welfare authority, a public school, a primary care center, or juvenile justice or law enforcement facility. It is conceivable that different cultural values might play unique parts in the adaptation based on who the population is and under what service conditions.

Conclusion

There is no doubt in my mind that our field must continue to pursue work that brings a higher level of conceptual and methodological uniformity and empirical validation to intervention development for Latinos and other groups. Methods for tailoring and adapting empirically supported treatments for specific populations are needed. And it is not just the characteristics of the populations that will determine what we need to modify. Rather, it is also the factors in the treatment that are amenable to adaptation that we need to carefully identify. And we must do this while retaining the critical aspects of the therapeutic action. We can look to future research that systematically identifies ways to tailor care consistently for optimal efficacy for diverse clients.