To prepare for this Discussion, read Knight et al.s (2014) study from

Research Article

Cultural Adaptation of Interventions in Real Practice Settings

Flavio F. Marsiglia1 and Jamie M. Booth2

Abstract This article provides an overview of some common challenges and opportunities related to cultural adaptation of behavioral interventions. Cultural adaptation is presented as a necessary action to ponder when considering the adoption of an evidence-based intervention with ethnic and other minority groups. It proposes a roadmap to choose existing interventions and a specific approach to evaluate prevention and treatment interventions for cultural relevancy. An approach to conducting cultural adaptations is proposed, followed by an outline of a cultural adaptation protocol. A case study is presented, and lessons learned are shared as well as recommendations for culturally grounded social work practice.

Keywords evidence-based practice, literature

Culture influences the way in which individuals see themselves

and their environment at every level of the ecological system

(Greene & Lee, 2002). Cultural groups are living organisms

with members exhibiting different levels of identification with

their common culture and are impacted by other intersecting

identities. Because culture is fluid and ever changing, the process

of cultural adaptation is complex and dynamic. Social work and

other helping professions have attempted over time to integrate

culture of origin into the interventions applied with ethnic

minorities and other vulnerable communities in the United

States and globally (Sue, Arredondo, & McDavis, 1992). In

an ever-changing cultural landscape, there is a renewed need

to examine social work education and the interventions social

workers implement with cultural diverse communities.

Culturally competent social work practice is well established

in the profession and it is rooted in core social work practice

principles (i.e., client centered and strengths based). It strives

to work within a client’s cultural context to address risks and

protective factors. Cultural competency is a social work ethical

mandate and has the potential for increasing the effectiveness

of interventions by integrating the clients’ unique cultural assets

(Jani, Ortiz, & Aranda, 2008). Culturally competent or culturally

grounded social work incorporates culturally based values,

norms, and diverse ways of knowing (Kumpfer, Alvarado,

Smith, & Bellamy, 2002; Morano & Bravo, 2002).

Despite the awareness about the importance of implementing

culturally competent approaches, practitioners often struggle

with how to integrate the client’s worldview and the application

of evidence-based practices (EBPs). When selecting and

implementing social work interventions, practitioners often

continue to unconsciously place themselves at the center of

the provider–consumer relationship. Being unaware of their

power in the relationship and undervaluing the clients per-

spective in the selection of EBPs tends to result in a type

of social work practice that is culturally incompetent and

nonefficacious (Kirmayer, 2012). This ineffectiveness can

be experienced and interpreted by practitioners in several

ways. In instances when clients do not conform to the content

and format of existing interventions, they are easily labeled as

being resistant to treatment (Lee, 2010). In other cases, when

clients fail to adapt to a given intervention that does not feel

comfortable to them, the relationship is terminated or the

client simply does not return to services. Thus, terms such

as noncompliance and nonadherence may hide deeper issues

related to cultural mismatch or a lack of cultural competency

in the part of the practitioner.

Culturally grounded social work challenges practitioners to

see themselves as the other and to recognize that the responsi-

bility of cultural adaptation resides not solely on the clients but

involves everyone in the relationship (Marsiglia & Kulis,

2009). In order to do this, practitioners need to have access

to interventions or tools that are consistent with the culturally

grounded approach. A culturally grounded approach starts with

assessing the appropriateness of existing evidence-based inter-

ventions and adapting when necessary, so that they are more

1 Southwest Interdisciplinary Research Center (SIRC), School of Social Work,

Arizona State University, Phoenix, AZ, USA 2 School of Social Work, University of Pittsburgh, Pittsburgh, PA, USA

Corresponding Author:

Jamie M. Booth, School of Social Work, University of Pittsburgh, 2117

Cathedral of Learning, 4200 Fifth Avenue, Pittsburgh, PA 15260, USA.

Email: [email protected]

Research on Social Work Practice 2015, Vol. 25(4) 423-432 ª The Author(s) 2014 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/1049731514535989 rsw.sagepub.com

relevant and engaging to clients from diverse cultural back-

grounds, without compromising their effectiveness. This process

of assessment, refinement, and adaptation of interventions will

lead to a more equitable and productive helping relationship.

The ecological systems approach provides a structure for

understanding the importance of cultural adaptation in social

work practice. Situated on the outer level (macro level) of

the ecological system, culture frames the norms, values, and

behaviors that operate on every other level: individual beliefs

and behaviors (micro level), family customs and communica-

tion patterns (mezzo level), and how that individual perceives

and interacts with the larger structures (exo level), such as

the school system or local law enforcement (Szapocznik &

Coatsworth, 1999). In this approach, the relationships between

individuals, institutions, and the larger cultural context within

the ecological framework are bidirectional, creating a dynamic

and rapidly evolving system (Bronfenbrenner, 1977; Gitterman,

2009). The bidirectional nature of relationships is an important

concept to consider when discussing the cultural adaptation

of social work interventions for two reasons: (1) regardless

of the setting, in social work practice, the clients and the

social workers engage in work partnerships in which both par-

ties must adapt to achieve a point of mutual understanding and

communication and (2) culture is in constant flux, as individ-

uals interact with actors and institutions which either maintain

or shift cultural norms and values over time.

Although culturally tailoring prevention and treatment

approaches to fit every individual may not be feasible, cultu-

rally grounded social work may require the adaptation of

existing interventions when necessary while maintaining the

fidelity or scientific merit of the original evidence-based

intervention (Sanders, 2000). This article discusses the need

for cultural adaptation, presents a model of adaptation from

an ecological perspective, and reviews the adaptations con-

ducted by the Southwest Interdisciplinary Research Center

(SICR) as a case study. The recommendations section con-

nects the premises of this article with the existing literature

on cultural adaptation and identifies some specific unresolved

challenges that need to be addressed in future research.

Empirically Supported Interventions (ESIs) in Social Work Practice

EBP has become the gold standard in social work practice and

involve the ‘‘conscientious’’ and ‘‘judicious’’ application of

the best research available in practice (Sackett, 1997, p. 2).

It is commonly believed that utilizing EBP simply requires the

practitioner to locate interventions that have been rigorously

tested using scientific methods, implement them, and evaluate

their effect; however, EBP acknowledges the role of individ-

uals and relationships in this process. EBP requires the inte-

gration of evidence and scientific methods with practice

wisdom, the worldview of the practitioner, and the client’s

perspectives and values (Howard, McMillen, & Pollio, 2003;

Regehr, Stern, & Shlonsky, 2007). The clinician’s judgment and

the client’s perspective are not only utilized in the selection of

the EBP intervention; they are also influential in how the inter-

vention is applied within the context of the clinical interaction

(Straus & McAlister, 2000). Achieving a balance between both

the client and the practitioner’s perspective in the application of

ESIs is essential for bridging the gap between research and prac-

tice (Howard et al., 2003). However, the inclusion of the clini-

cian’s judgment and the client’s history potentially muddles

the scientific merit of the intervention being implemented. This

is the fundamental tension and challenge when implementing

EBP and a key reason why the gap between research and prac-

tice exists (Regehr et al., 2007).

The attraction of EBP is clear; locating and potentially

utilizing empirically tested treatment and prevention inter-

ventions allow social workers to feel more confident that they

will achieve the desired outcomes and provide clients with

the best possible treatment, thereby fulfilling their ethical

responsibility (Gilgun, 2005). Despite this clear rationale, the

utilization of EBP is limited (Mullen & Bacon, 2006) and

when it is applied, research-supported interventions may not

be implemented in the manner the authors of the intervention

intended.

This lack of treatment fidelity when implementing EBP

may be due to practitioner’s awareness that the evidence

generated by randomized control trials (RCTs) may not be

applicable to the diverse needs of their clients or adequately

address the complexity of the clients’ life (Webb, 2001;

Witkin, 1998). Practitioners have natural tendency to adapt

interventions to better fit their clients (Kumpfer et al.,

2002). Some adaptations are made consciously, but others are

made quickly during the course of implementation and based

on clinical judgment (Bridge, Massie, & Mills, 2008; Castro,

Barrera, & Martinez, 2004). ESIs, however, can only be

expected to achieve the same results as those observed when

originally tested, if they are implemented with fidelity or

strict adherence to the program structure, content, and dosage

(Dumas, Lynch, Laughlin, Phillips Smith, & Prinz, 2001;

Solomon, Card, & Malow, 2006). Although adaptations are

typically made in response to a perceived need, when they

are not done systematically, based on evidence and with the

core elements of the intervention preserved, the efficacy that

was previously achieved in the more controlled environment

may not be replicated (Kumpfer et al., 2002). Informal adap-

tation has the potential for compromising the integrity of

the original intervention, thus negating the value of the accu-

mulated evidence that supports the intervention’s effective-

ness. This tension between fidelity and fit has generated a

need for strategies to create fit while insuring fidelity.

Cultural Adaptation

The primacy of scientific rigor over cultural congruence may

be a limitation in applying ESIs and a standard that should not

be maintained in culturally competent social work practice.

When working with real communities, both must be satisfied

to the highest degree possible (Regehr et al., 2007). One solu-

tion to tension between using culturally relevant practices and

424 Research on Social Work Practice 25(4)

ESIs is locating interventions that have been designed for and

tested with a given cultural group. However, the limited avail-

ability of culturally specific interventions with strong empiri-

cal support may create barriers to this approach. Despite the

progress that has been made to date, most ESIs are developed

for and tested with middle-class White Americans, with the

assumption that evidence of efficacy with this group can be

transferred to nonmajority cultures, which may or may not

be the case (Kumpfer et al., 2002).

For example, a prevention intervention with Latino parents

found that assimilated, highly educated Latino parents were

responsive to the prevention interventions presented to them,

while immigrant parents with less education were less likely

to benefit (Dumka, Lopez, & Jacobs-Carter, 2002). This high-

lights the differential effects of an intervention based on culture

as well as a clear need for a more culturally relevant interven-

tion for immigrant parents. Despite a clear need for adaptation

in some circumstances, there is a strong risk of compromising

the effectiveness of the ESI when unstructured cultural adapta-

tions are implemented in response to perceived cultural incon-

gruence (Kirk & Reid, 2002; Kumpfer & Kaftarian, 2000;

Miller, Wilbourne, & Hettema, 2003; Solomon et al., 2006).

For that reason, when culturally and contextually specific inter-

ventions exist with strong evidence, it is certainly preferable to

select that intervention; however, in the absence of an ESI

designed and tested for the population being served, adaptation

may be a more viable and cost-effective option for scientifi-

cally merging a client’s cultural perspectives/values and the

ESI (Howard et al., 2003; Steiker et al., 2008). Systematically

adapting an intervention may increase the odds that the treat-

ment will achieve similar results than those found in more

controlled environments by minimizing the amount of sponta-

neous adaptations that the practitioner feels that they must

make to communicate within the client cultural frame

(Ferrer-Wreder, Sundell, & Mansoory, 2012).

Cultural adaptation may not only preserve the ESI’s effi-

cacy but also enhance the results attained in clinical trials

(Kelly et al., 2000). Culturally adapted interventions have the

potential to improve both client engagement in treatment and

outcomes and might be indicated when either rates fall below

what could be expected based on previous evidence (Lau,

2006). In an evaluation of a culturally adapted version of

the Strengthening Families intervention, there was a 40% increase in program retention in the culturally adapted version

of the intervention (Kumpfer et al., 2002). Although outcomes

were not found to be significantly better in the adapted version

of the intervention, the increase in retention is a significant

improvement. Improving retention expands the intervention’s

potential to reach and impact individuals who would not

typically remain in treatment. Despite the lack of difference

in outcomes in the Strengthening Families intervention, some

evidence has emerged that culturally adapted interventions

not only increase retention but are also more effective. In a

recent meta-analysis, culturally adapted treatments had a

greater impact than standard treatments, produced better out-

comes, and were most successful when they were culturally

tailored to a single ethnic minority group (Smith, Domenech

Rodrı́guez, & Bernal, 2010).

Adapting interventions in partnership with communities also

enhances the community’s commitment to the implementation

and the chances that the program will be sustained overtime

(Castro et al., 2004). For example, efforts to adapt HIV pre-

vention programs by modifying the messages and protocols

in order for them to sound and feel natural or familiar intellec-

tually and emotionally to individuals, families, groups, and

communities have improved the communities’ receptiveness,

retention, outcomes, and overall satisfaction, in addition to

retaining high levels of fidelity (Kirby, 2002; Raj, Amaro,

& Reed, 2001; Wilson & Miller, 2003).

Finally, cultural adaptation is advantageous because it

allows the social worker to address culturally specific risk

factors and build on identified protective factors. In the case

of Latino families, differential rates of acculturation between

parents and youth appear to be a risk factor for substance use

and delinquency among youth, indicating that family-based

interventions may be the most culturally relevant intervention

(Martinez, 2006). In addition to a source of risk, cultural

norms that place a high value on family loyalty are protective

factors against a variety of negative outcomes (German,

Gonzales, & Dumka, 2009; Marsiglia, Nagoshi, Parsai, &

Castro, 2012). Identifying risk and protective factors unique

to a community and addressing these within an intervention

have the potential to increase the efficacy of the intervention.

The importance of EBP and culturally competent practice

has created tension in the field of social work. Evidence

has landed support to both claims: (1) interventions are more

effective when implemented with fidelity (Durlak & DuPre,

2008) and (2) interventions are more effective when they are

culturally adapted because they ensure a good fit (Jani et al.,

2008). These different perspectives highlight the tension in

the field between implementing manualized interventions

exactly as they were written versus to adjusting them to fit the

targeted population or community (Norcross, Beutler, &

Levant, 2006). Although this debate is far from resolved, the-

ories of adaptation have been developed that allow the

researcher/practitioner to adjust the fit without compromising

the integrity of the intervention (Ferrer-Wreder et al., 2012).

If the cultural adaptation is done systematically, it has the

potential for maximizing the benefit of the fit, as well as the

benefit of the ESI, thus providing a strategy that addresses

many of the concerns surrounding EBP’s applicability in

social work practice (Castro et al., 2004).

An Emerging Roadmap for Cultural Adaptation

Cultural adaptation is an emerging science that aims at

addressing these challenges and opportunities to enhance the

effectiveness of interventions by grounding them in the lived

experience of the participants. Strategies and processes to sys-

tematically adapt interventions while insuring a more optimal

cultural fit without compromising the integrity of scientific

merit have been proposed and are beginning to be tested

Marsiglia and Booth 425

(La Roche & Christopher, 2009). The first step in all adaptation

models is determining that the cultural adaptation of an interven-

tion should be perused. Adaptation of an ESI is indicated when

(1) a client’s engagement in services falls below what is

expected, (2) expected outcomes are not achieved, and (3) iden-

tified culturally specific risks and/or protective factors need to

be incorporated into the intervention (Barrera & Castro, 2006).

Once the determination is made to conduct an adaptation,

there are a variety of models that one could follow all of which

fall into two categories: content and process (Ferrer-Wreder

et al., 2012). Although most current adaptation models have

merged the discussions regarding the content that should be

modified and process by which this modification takes place,

it is useful to consider them separately.

Content models identify an array of domains that may be

crucial to address when conducting an adaptation. The ecolo-

gical validity model, for example, focuses on eight dimensions

of culture: language, persons, metaphors, content, concepts,

goals, methods, and social context (Bernal, Jiménez-Chafey,

& Domenech Rodrı́guez, 2009). The cultural sensitivity model,

also a content model, identifies two distinct content areas: deep

culture, which includes aspects of culture such as thought pat-

terns, value systems, and norms, and surface culture, which refers

to elements, such as language, food, and customs (Resnicow,

Soler, Braithwaite, Ahluwailia, & Butler, 2000). Proponents of

the cultural sensitivity model argue that both aspects of culture

should be assessed and potentially addressed if areas of conflict

or incongruence between the culture and the intervention are

identified (Resnicow et al., 2000). Surface adaptations allow the

participants to identify with the messages, potentially enhancing

engagement; while, deep culture adaptations ensure that the

outcomes are impacted (Resnicow et al., 2000).

Castro, Barrera, and Martinez (2004) and Castro, Barrera,

and Steiker, 2010 have proposed a content model that identifies

a set of specific dimensions—at the surface and deep levels—

that are essential to consider in the adaptation process: cogni-

tive, affective, and environmental. Cognitive adaptations are

considered when participants cannot understand the content

that is being presented due to language barriers or the use of

information that is not relevant in an individual’s cultural

frame. Vignettes given by the original intervention, for exam-

ple, may not be relevant to the participants or may be offensive

due to spiritual or religious taboos. The content may create a

negative reaction from the participants which in turn may block

their ability to hear and integrate the message. It is that content

that needs to be modified while the core elements of the inter-

vention are respected. Affective-motivational adaptations are

indicated when program messages are contrary to cultural

norms and values, creating a resistance to change within the

individual (Castro, Rawson, & Obert, 2001). Environmental

factors (later referred to as relevance) make sure that the con-

tents and structure are applicable to the participants in their

daily lived experience (Castro et al., 2010).

While content models of adaptation tell adaptors where to

look for cultural mismatch, process models provide a frame-

work for making systematic assessments of cultural match,

adjustments to the original intervention, and tests of the adap-

tations effectiveness. At a minimum adaption process, models

follow two systematic steps: (1) identifying mismatches

between the original intervention and the client’s culture and

(2) testing/evaluating changes that have been made to rectify

these disparities (Ferrer-Wreder et al., 2012).

Most process models of adaptation begin with building a

partnership or coalition with members of targeted community

(Castro et al., 2010; Harris et al., 2001; Wingood & DiCle-

mente, 2008). Sometimes the ESI that will be adapted is

selected at this stage; however, more information is often gath-

ered about the targeted population before selecting the inter-

vention that would provide the best fit (Kumpfer, Pinyuchon,

Teixeriade de Melo, & Whiteside, 2008; Mckleroy et al.,

2006; Wingood & DiClemente, 2008). Whether the interven-

tion has yet to be selected, extensive formative research is con-

ducted to assess the etiology of the social problem that is the

target of the intervention, possible population-specific risks

and protective factors, and measurement equivalence to insure

and accurate evaluation of intervention outcomes (Harris et al.,

2001). Some information about the target community may be

gained by reviewing relevant literature; however, interviews,

focus groups, and surveys are also used to collect primary data

about the social and cultural context that may impact the out-

come of the intervention or conflict with the program’s mes-

sages/implementation strategies.

At this point in the process, some adaptation models recom-

mend making changes based on the formative research

(Domenech-Rodriguez & Wieling, 2004; Harris et al., 2001), while

others suggest implementing the intervention with minimal

changes and assessing the need for further adaption. In an innova-

tive approach, the Planned Intervention Adaptation model suggests

making significant changes to one version of the intervention

while making minimal changes to another and implementing them

both simultaneously to test the differential effects (Castro et al.,

2010; Ferrer-Wreder et al., 2012; Kumpfer et al., 2008).

Regardless of the level of adaptation, the modified inter-

vention is pilot tested and based on the outcomes subsequent

adaptations are made (Ferrer-Wreder et al., 2012). Once a

final adaptation has been made, further testing takes place

in effectiveness trials. Across all theories of adaptation, the

process is iterative with refinements made to the intervention

at every stage based on the evidence generated in the prior

stage (Domenech-Rodriguez & Wieling, 2004). Regardless

of the depth of changes made, the adapted intervention must

be rigorously tested to ensure that the effects of the original

ESI are preserved after changes have been made.

Case Study: Adaptations of Keepin’it REAL (KiR), the Southwest Interdisciplinary Research Center (SIRC) Approach

Over the past 10 years of health disparities research, the SIRC

has developed a process of cultural adaptation that includes

most of the elements outlined previously. The specific

426 Research on Social Work Practice 25(4)

adaptation model utilized at SIRC is an expanded version

of the Barrera and Castro (2006) model as illustrated by

Figure 1.

KiR is the flagship empirically supported treatment SIRC

(Marsiglia & Hecht, 2005). KiR is a manualized school-

based substance abuse prevention program for middle school

students. It was designed to (a) increase drug resistance skills

among middle school students, (b) promote antisubstance use

norms and attitudes, and (c) develop effective drug resistance

and communication skills (Gosin, Dustman, Drapeau, &

Harthun, 2003). It was created and evaluated in Arizona

through many years of community-based research funded by

the National Institutes on Drug Abuse of the National Insti-

tutes of Health. It is a model program listed under Substance

Abuse and Mental Health Services Administration’s National

Registry of Evidence-Based Programs and Practices. There is

strong evidence about the efficacy of the intervention with

middle school Mexican American students (Marsiglia, Kulis,

Wagstaff, Elek, & Dran, 2005), however the community-

identified need to reach out to younger students and to stu-

dents of other ethnic groups generated a set of adaptation

efforts summarized in Figure 2.

As Figure 2 illustrates, KiR was adapted for fifth-grade stu-

dents (Harthun, Dustman, Reeves, Marsiglia, & Hecht, 2009)

following the SIRC adaptation model and an RCT was con-

ducted to test whether the effects of the intervention increased

by intervening earlier (fifth grade vs. seventh grade). Students

who received the intervention in both the fifth and seventh

grade were no different in their self-reported use of alcohol

and other drugs than students who received the intervention

only on the seventh grade (Marsiglia, Kulis, Yabiku, Nieri,

& Coleman, 2011). This effort did no yield the expected

results but provided evidence from a developmental perspec-

tive that starting earlier was not cost effective.

The second adaptation presented in Figure 2 was also

community-generated and supported from the evidence gath-

ered during the initial RCT of KiR. Urban American Indian

(AI) youth were not benefiting from KiR as much as other

children (Dixon et al., 2007). Following the principles of

community-based participatory research, a steering group,

including leaders from the local urban AI community and

school district personnel in charge of AI programs, was

formed to guide the adaptation process. In addition to enga-

ging community members and setting up a structure to ensure

a collaborative partnership, before beginning the adaptation

process, formative information was collected by consulting

the literature to identify culturally specific risks and protec-

tive factors and focus groups. Focus groups were conducted

with both Native American adults and youth to explore cultu-

rally specific drug resistance strategies that were frequently

applied by urban Native American youth (Kulis & Brown,

2011; Kulis, Dustman, Brown, & Martinez, 2013).

Based on this information, collected in conjunction with

four Native American curriculum development experts, KiR

was adapted, and while maintaining its core elements, the

content and structure were changed to be more culturally rel-

evant to Native American youth (Kulis et al., 2013). Changes

to the curriculum included (1) new drug resistant strategies

that were identified by the AI youth as being more culturally

relevant to them, (2) lesson plans designed to teach strategies

in a more culturally relevant way, (3) more comprehensiv