How do Providers Serving American Indians and Alaska Natives with Substance Abuse Problems Define Evidence-Based Treatment?

INTRODUCTION

American Indians and Alaska Natives (AI/ANs) experience higher rates of alcohol and drug abuse and alcohol-specific mortality (Beauvais, Jumper-Thurman, & Burnside, 2008; Gilman, et al., 2008; U.S. Department of Health and Human Services, 2009), and less access to effective treatments for disorders from substance abuse, than persons in the general U.S. population (Castor, et al., 2006; McFarland, Gabriel, Bigelow, & Walker, 2006; Rieckmann, et al., 2012). While the inventory of evidence-based treatments (EBTs) for substance abuse problems is growing (Fals-Stewart & Lam, 2008; Longabaugh, Wirtz, Gulliver, & Davidson, 2009; Petry, et al., 2005; Substance Abuse and Mental Health Services Administration, 2008), there have been no systematic assessments of the knowledge of and attitudes towards EBTs among clinicians and clinical administrators working in treatment programs serving AI/AN communities. This limits our understanding of how providers in programs serving AI/ANs with problematic substance use perceive and use EBTs.

In contrast, there is a large and growing literature on EBT use in substance abuse treatment programs more generally. Two systematic reviews (Walters, Matson et al., 2005, Garner, 2009) described the literature relative to how EBTs are used within “mainstream” substance abuse treatment programs. Most of these investigations explored adoption (Fals-Stewart & Birchler, 2001; Knudsen, Ducharme, & Roman, 2007; Koch et al., 2006) and attitudes (Rieckmann et al., 2007; Bride, Abraham, & Roman, 2010; Henggeler, et al., 2008) about EBTs. Research on attitudes is more commonly focused on pharmacologic EBTs, but those examining the psychosocial treatments generally found more positive attitudes relative to the pharmacologic treatments (McGovern et al., 2004; Willenbring et al., 2004).

Studies exploring how EBTs generalize to substance abuse treatment for AI/AN populations are few. Gone’s study (2011) of how traditional practices are incorporated into AI/AN programs found only rare usage of EBTs. Miranda et al. (2005) described several adapted interventions that were shown effective in AI/AN populations, there is very little in the literature regarding EBT (either culturally adapted or not) effectiveness in programs serving AI/AN populations.

The dissemination and implementation of EBTs in programs for treatment of substance abuse are a national concern and the unique history of AI/ANs and the programs that serve them suggests the dynamics impacting this process may be different than those for treatment programs more broadly (Novins, et al., 2011). For example, structural barriers such as inadequate funding and a workforce that is not trained to deliver EBTs may limit the ability of these programs to identify, implement, and maintain their use (McLellan, Carise, & Kleber, 2003; McGovern, Xie, Segal, Siembab, & Drake, 2006; Humphreys & McLellan, 2011; Knudsen, Abraham, & Oser, 2011). This difficulty may be compounded by a lingering distrust of approaches associated with western standards of care among clinicians and their AI/AN clients that can be traced to the negative effects of colonization (Gone, 2008), the many instances of research abuse perpetrated by western institutions (Carpio, 2004; Mello & Wolf, 2010), and preferences for drawing on indigenous rather than biomedical healing traditions (Calabrese, 2008). These commentaries suggest that EBTs may represent a poor fit with AI/AN providers’ and communities’ cultural values and world views. We also have no clear guidance on how to assess cultural appropriateness in selecting an EBT for use in a given community or how to adapt interventions to address these concerns without compromising their therapeutic benefits (Gone & Calf Looking, 2011; Griner & Smith, 2006).

Because of these unique dynamics and concerns, it is particularly important that treatment programs serving AI/AN communities be able to assess the evidence base for specific interventions so that they make informed decisions about the types of services they should offer. The ability of these programs to assess the evidence base is unknown.

Aarons, Hurlburt, and Horwitz (2011), in their Exploration, Preparation, Implementation, and Sustainment (EPIS) implementation framework, have described the initial phase of the EBT dissemination and implementation process as one of exploration. In this phase, it is particularly important that service providers have the ability to assess the evidence base of available interventions and consider the fit of an EBT with their context. The ability to correctly define EBT using words found in standard definitions of the term (e.g., evidence, effectiveness, research, etc.) may elucidate providers’ awareness, knowledge, and even their attitudes about EBTs. While there are other aspects of the exploration process both in the inner context of a given organization or treatment program (e.g., ability to “scan” for potential approaches to improve services; Aarons, et al., 2011) and outer context of the service system (e.g., readily available information about innovative practices, inter-organizational networks for sharing ideas about innovative approaches, funding for improving services, availability of innovations that are acceptable and appropriate for the community served and feasible given the program’s resources; Aarons, et al., 2011; Kataoka, Novins, & Santiago, 2010), the ability to assess the evidence base for potential interventions is a core component of the exploration process.

In this paper we report the results of a qualitative analysis of the knowledge of and attitudes towards EBTs among clinicians and clinical administrators working for a sample of substance abuse treatment programs serving AI/AN communities. While recognizing that the broad regional and tribal diversity in these communities impacts any conclusions drawn, this work provides initial insights into the exploration process in these programs, and takes a first step in addressing the outlined gaps in knowledge and understanding of EBT use in AI/AN populations.

METHOD

Data for these analyses came from the program case study phase of the National Survey on Evidence-Based Practices. This project focused on how treatment programs serving AI/ANs with heavy or problem substance use design, implement, and assess their services; how they incorporate culture, evidence-based concepts, and healing techniques into the services they provide; and how they perceive EBTs. An Advisory Board supported this project. Members included administrators, service providers, evaluators from the AI/AN behavioral health services community, and researchers with expertise in AI/AN substance abuse treatment and dissemination and implementation research.

Settings and Participants

The study Advisory Board identified programs based on their reputations for innovative clinical services and to ensure adequate representation of the geographic, cultural, and reservation/rural/urban diversity of AI/AN communities. The criteria for innovative clinical services included a) combining cultural and biomedical treatment approaches, b) incorporating evidence-based practices into treatment, and c) receiving competitive federal funding for service implementation and expansion.

Eighteen treatment programs serving AI/AN communities were invited to participate in this study. All invitations were accepted. The 18 programs were clustered within three tribes, one tribal consortium, and three urban organizations from seven Indian Health Service regions and seven states. Ten of the programs were located on reservations, three were located in non-reservation rural areas, and five were located in urban areas. These programs were operated by tribes or tribal non-profit organizations, but all received some Indian Health Service funding to support their services. Seventy-six service participants (21 individuals in clinical administrative positions and 55 front-line clinicians) at these 18 programs participated in the project. A total of 21 key informant interviews and 10 focus groups were conducted. We emphasized participant and organizational confidentiality in order to maximize participation, facilitate candid responses, and minimize participant discomfort. Such discomfort is not uncommon in AI/AN communities due to prior research abuses and the stigma of mental illness (Gone & Alcantara, 2007; Norton & Manson, 1996). Accordingly, we did not collect individual participants’ demographic information and instead focused on the key topics of interest for this study.

Measures

The Advisory Board developed semi-structured interview and focus group guides to facilitate open-ended conversations about the community the program serves, the services provided and how they were developed, the challenges of service provision, and the participants’ awareness of and experience with a selected set of EBTs. The interview and focus group guides included open-ended “stem” questions designed to steer the conversation through key topics of interest to the research team. Most stem questions also included “probes” to explore specific issues if they were not spontaneously identified by the interview/focus group participants (e.g., whether the program felt pressure to provide specific treatments because of accreditation or reimbursement requirements). Copies of these guides may be found online at http://www.ucdenver.edu/caianh/ebp.

Data Collection

Data collection took place from August 2009 through July 2010. The Colorado Multiple Institutional Review Board reviewed and approved all procedures, which also underwent local review processes that always included an administrative review and sometimes included review by a research review committee or a formal institutional review board. At each participating program, we conducted one or more key informant interviews with program staff in clinical administrative positions, as well as a focus group with staff in front-line clinical positions. After reviewing a complete description of the study, each participant completed written informed consent. Focus group size ranged from three to thirteen participants and averaged eight. Because programs from a single tribe, tribal consortium, or urban organizations often participated together, focus groups usually included representatives from multiple treatment programs. This arrangement required more effort from the moderator to collect information about each program, but it allowed for richer conversation regarding these issues, especially given the small size of many of these programs. Key informant interviews lasted 60 minutes and focus groups lasted 90 minutes. Program directors had the option of distributing interview/focus group guides prior to the research team’s visit. The principal investigator conducted all interviews, which were recorded, and detailed notes were taken as a backup.

Analysis

Recordings were transcribed prior to analysis. One recording was of poor quality and required using the notes that were taken during the session by a research assistant. These data were imported into nVivo 9.0 for data management and to facilitate the analysis process.

Our goal in these analyses was to examine participants’ conceptualizations and attitudes towards EBTs. We therefore focused our coding on the following questions that asked participants to define EBTs and describe their knowledge, use, and assessment of the specific treatments listed in Tables 1 and 2: How would you define evidence-based treatment? Which of these treatments have you heard of? What do you know about [specific treatments participant reported having heard of]? What do you know about [with probes exploring experience with and attitudes towards these treatments]?

Table 1.

Key definition-related themes, percentage of interviews/focus groups in which theme/subtheme was mentioned, and examples of a definition of evidence-based treatment from 21 interviews and 10 focus groups with clinicians serving American Indian and Alaska Native populations

Definition-related themes and subthemes	Referenced in Interviews (n=21)	Referenced in Focus Groups (n=10)	Exemplary Quote
	n (%a)	n (%b)
Contains words found in formal definition
Effective	9 (42.9)	8 (80.0)	That’s treatment that’s been researched and been proven to be effective.
Research	10 (47.6)	6 (60.0)	Evidence-based treatment is empirically tested techniques that show a statistically significant improvement.
Manualized	3 (14.3)	0	…something that has been researched and proven effective and then manualized to put other programs together to use.
Replication	0	1 (10.0)	it took two peer review journal articles in order for it to be accepted for consideration as a best practice.
Contains words related to formal definition
Outcome related	4 (19.0)	3 (30.0)	Treatment being able to be measured with a measurable outcome.
Formalized	4 (19.0)	4 (40.0)	Well, because it’s something that somebody else developed and goes out there and teaches people and provides training on and, you know, it’s like a documented program.
Best Practice	0	2 (20.0)	Or just using best practices so that you have outcomes of more wellness.
Don’t know definition	3 (14.3)	2 (20.0)	Evidence-based what?

^aPercentage refers to the percent of Key Informants who referenced the theme

^bPercentage refers to the percent of Focus Groups in which the theme was referenced

Table 2.

Key attitude-related themes, percentage of interviews/focus groups in which theme/subtheme was mentioned, and examples of a definition of evidence-based treatment from 21 interviews and 10 focus groups with clinicians serving American Indian and Alaska Native populations

Attitude-related themes and subthemes	Referenced in Interviews (n=21)	Referenced in Focus Groups (n=10)	Exemplary Quote
	n (%a)	n (%b)
Not culturally relevant	6 (28.6)	3 (30.0)	…my only concern about Evidence-Based practices (is) that they’re not typically researched using minority populations. They’re almost exclusively Caucasian. So I don’t think there’s recognition of cultural differences in clients and in populations.
External mandate	2 (9.5)	2 (20.0)	It had to be out of that book and that’s how we had to do it.
Western standard	3 (14.3)	2 (20.0)	They want to dictate how we should be doing it, you know, from their vaulted ivory tower down there. I just…Oh God, it makes me mad.
Client-centered	2 (9.5)	3 (30.0)	Asking a client what they feel they need.
Resource Drain	3 (14.3)	0	Evidence-based is all of that stuff that we just don’t have the money or the staff to do.
Maintaining sobriety	0	2 (20.0)	We can show that after sixty days, ninety days, one year … for a fact that we know this person has maintained sobriety.

^aPercentage refers to the percent of Key Informants who referenced the theme

^bPercentage refers to the percent of Focus Groups in which the theme was referenced

Coding Procedures

For our analysis of the definition of an EBT, we coded participant responses relative to their match with the following formal definition: “any practice that has been established as effective through scientific research…” (Drake, et al., 2001). We also examined responses for components covered in more detailed definitions of EBTs, such as the importance of manual use, testing in a clinical trials paradigm, and hierarchies of evidence (e.g., clinical trials representing a higher level of evidence than expert opinion; Devereaux & Yusuf, 2005; Guyatt, et al., 2008). The a priori coding scheme we developed helped us examine whether participants identified the central components of the formal definition of EBT (e.g., effective, scientific research) as well as details provided in more elaborative definitions (e.g., formalized, outcome-related, use of manuals).

Applying grounded theory procedures (Strauss & Corbin, 1990), we also developed codes that were not directly related to the formal definition (e.g., concerns about cultural appropriateness, concerns that EBT use would interfere with client-centered care) but were indicative of participants’ attitudes towards these treatments. An initial review of the transcripts generated a tentative list of themes; we expanded and modified the coding framework in subsequent reviews. After confirming major themes and subthemes, we identified exemplary quotations that captured these ideas. Theme and subtheme prevalence was also noted.

JD and LM performed the initial coding, which was reviewed with DN to establish consensus.

RESULTS

Tables 1 and 2 list common themes and subthemes coded in key informant and focus group transcripts and provide explanatory quotations for each. All respondent answers to the question, how would you define evidence-based treatment? were coded to the listed themes and subthemes. Approximately 2.5 themes were endorsed per source.

Definition-related themes

Two key terms found in the standard definition of EBT, “effective” and “research,” were referenced in the majority of focus group transcripts (effective: 80% and research: 60%) and in nearly half of the key informant interviews (effective: 43% and research: 48%). One respondent succinctly defined EBTs by noting, “Pretty simply it’s the treatment techniques that have been researched and have proven to be effective in a population that’s been monitored.”

Additional words used that are synonyms or components of formal definitions were also mentioned. These include “empirical” (Focus Groups (FG): 40%; Key Informants (KI): 0%), “data” (FG: 0%; KI: 19%), and “[it] works” (FG: 60%; KI: 19%). “[It’s] treatment that’s worked, commented one participant, adding, “that shows proof that it’s worked, and that’s been researched and shown to be successful in this particular type of treatment.” Nineteen percent of sources were unable to define an EBT. One example from a respondent unsure about the term was, “You know, I’ve heard it, I’ve seen it on the covers of the book. But, no, [I don’t know what it means].”

We also identified themes consistent with expanded definitions of EBTs. That EBTs are outcome related was referenced in 30% of focus group and 19% of interview transcripts. One source described this by noting, “[it] is finding a method of intervention or treatment that has an impact on a person’s ability to change their behavior.” Formalized interventions were also mentioned in 40% of focus group and 19% of interview transcripts. One focus group participant described the formal nature of an EBT this way: “Then, someone’s going to write a curriculum, manualize it, get it all nice and beautiful, and then, they’re going to implement it and see if it works. And, if it works, then it’ll be an evidence-based practice and maybe it’ll be [listed in] NREPP [National Registry of Evidence-based Programs and Practices] …”

Terms such as “replication” and “use of a manual” were less frequently noted. In this analysis, 14% of key informants mentioned a manual, while the term was not mentioned in any of the focus group transcripts. Conversely, in 10% of the focus group transcripts, the term “replication” was referenced, while 0% of key informants referenced this term. Two other important higher-level terms in the definition of EBT, “randomization” and “hierarchy of evidence,” were not mentioned in either format.

Attitude-related themes

Positive, neutral, and negative attitudes towards EBTs were present across sources. Most of the positive comments were related to the effectiveness of certain EBTs. And while some respondents noted that certain EBTs were more translatable to AI/AN client needs, the most often mentioned attitudinal-related theme was concern about cultural relevance of EBTs to treat AI/ANs. This theme was present in 30% of focus group transcripts and 29% of interviews. One key informant described his concern by saying,

I think it was developed by folks that have been out in the field for a while or haven’t really experienced culturalized populations because it’s like saying, “Sam had evidence-based treatment in Alaska so if it works for them Alaskan Natives it’s going to work for the Hopi down in Arizona.” And it’s not true. Evidence-based just means that they have found a certain treatment approach or philosophy that helps with a certain population and it’s not true for all populations.

Among the 14% (KI)-20% (FG) who mentioned the influence of western ways in EBTs, each reference highlighted the western influence as a negative characteristic. For example, one respondent commented that the definition of an EBT is, “Western standards of numbers and success and failure rates are evidence-based. Yeah, that’s what evidence-based is – a few years of a study over time with numbers and success and failure rates, that’s what it is – western, western.”

External mandates (e.g., grant requirements, funding agency rules) to use EBTs were also mentioned in 20% of focus group and 10% of interview transcripts. These mandates were almost universally perceived negatively. Indeed, participants who brought up this concern consistently noted that external forces such as these may create barriers to implementing EBTs, as described below by one participant:

But, then, to try and force it on us and to say, “In addition to the fact that this is a heartbreaking job, we’re going to have you do fifty pages of paperwork every time you do an intake because the State tells us we have to, and now what we want you to do is evidence-based practice, too,” on top of that. And to have it forced on us, I guess it’s a bit of a parallel process with the client, and I just really want to say, “Thank you, but no thanks.”

Participants also reported that they often modify EBTs to meet the needs of clients in Indian Country. An example was related by one clinician: “You know, we see what they want, what they need and how we can support them on it. And, to me, that’s gonna be evidence-based because it’s working for that person.” For some participants, the perceived incongruities between individualized care and manualized treatments such as EBTs can create tension among the clinical staff and with clients. Participants found this conflict particularly challenging, since almost all EBTs have not been tested with AI/AN people. This challenge was described by one interviewee who said, “It’s really hard to find anything that is Native-specific or part of the culture program that’s going to support it. They’re just looking for data and there isn’t data out there…”

Familiarity with and use of EBTs

Respondents were asked to comment on familiarity and usage of the following EBTs: 12-step facilitation, Behavioral Couples Therapy, Cognitive Behavioral Therapy, Community Reinforcement and Family Training, Contingency Management, Matrix Model, Motivational Interviewing, Relapse Prevention Therapy, and groups of medications for comorbid psychiatric conditions, relapse prevention, and withdrawal. The top three EBTs that respondents reported familiarity with were Motivational Interviewing (MI), mentioned by 100% of both focus groups and key informant interviews, Matrix Model (MM) (FG: 100%; KI: 91%), and Cognitive Behavioral Therapy (CBT) (FG: 80%; KI: 62%). Sixty percent of focus groups and 48% of key informants reported using MI with their clients. Twelve-Step therapies were reported in use by half of focus groups and just over one-quarter of key informants, while reported use of CBT dropped to 30% in focus groups and 23% in key informant interviews. Familiarity and use of Multisystemic Therapy (MST), Community Reinforcement and Family Training (CRFT), and Behavioral Couples Therapy (BCT) were lowest among the listed EBTs with usage reported as low as 0% for MST in both focus groups and interviews, to just 5% and 10% for CRFT and BCT, respectively.

DISCUSSION

The findings presented here suggest that there is little consensus around perspectives regarding the relevance of EBTs for treatment in programs serving AI/AN communities where heavy or problem use of substances is common. And although some in our sample reported favorable attitudes about certain EBTs, in many cases, perceptions of EBTs seen as lacking cultural relevance or evidence specific to certain populations were reported. These factors may limit programs’ or clinicians’ motivation to fully evaluate the EBTs for potential implementation. Such perspectives may compound previously described structural barriers that limit adoption and implementation of EBTs in AI/AN substance abuse treatment programs (McLellan, et al., 2003; McGovern, et al., 2006; Humphreys & McLellan, 2011; Knudsen, et al., 2011). There is evidence, however, that some EBTs are effective across cultures, with little or no adaptation. Damashek, Bard, & Hecht (2012) found the a home-based EBT to reduce child neglect was seen by American Indian clients as more culturally appropriate compared to home-based services as usual. Hunter, Paddock, Zhou, Watkins, and Hepner, (2013) found that treatment effects of group cognitive behavioral therapy for depression in addiction treatment did not vary by race/ethnicity. Other studies examining treatment outcomes with the Adolescent Community Reinforcement Approach for alcohol and drug abuse (Godley, Hedges, & Hunter (2011), alcohol and marijuana use (Gil, Wagner, & Tubman, 2004), and cocaine dependence (Milligan, Nich, & Caroll, 2004) found no significant differences in the effectiveness of behavioral therapies for clients from different racial or ethnic backgrounds. Results from a study of adolescents enrolled in treatment for alcohol and other substance use problems utilizing the Community Reinforcement approach suggest that this EBT was equally effective across racial groups in increasing numbers of days abstinent from alcohol and other drugs (Godley, et al., 2011). Finally, Barry, Sullivan, and Petry’s (2009) study of contingency management for cocaine dependence was found equally effective for African Americans, Hispanics, and Whites.

Research in implementation science suggests the quality of the EBT implementation process is as important as the EBT itself (Aarons, et al., 2011; Fixsen, Blase, Naoom, & Wallace, 2009). More specifically, an effective treatment that is well implemented relative to human, infrastructural, and financial resources has a much greater chance of success than either a good treatment that is poorly implemented or a well-implemented but ineffective treatment. Moreover, without a constructive organizational culture and the support of both program administrators and clinical staff, mandated use of EBTs may not be well accepted or implemented (Aarons & Sawitzky, 2006; Glisson, Schoenwald, et al., 2010; Rieckmann, Bergmann, & Rasplica, 2011). Even if efficacy of the EBT is proven, if front-line clinicians do not consider it relevant to their practice and clients, there will likely be resistant to implementation (McGovern, Fox, Xie, & Drake, 2004).

Our findings are consistent with these and other investigations describing attitudes towards EBTs, especially those related to behavioral health, that are fraught with concerns about the complexity of effective implementation (Bride et al., 2012; Jack et al., 2011). Like these studies, those interviewed in this effort reported barriers to gaining the capacity to effectively implement EBTs within their unique cultural settings. In fact, some of the factors found to be instrumental in effective implementation (ongoing organizational support, provider training and networking, and regular feed; back regarding case progress) (Aarons, Fettes, et al., 2009; Aarons, Sommerfeld, & Walrath-Greene, 2009; Borntrager & Lyon, 2014; Kershner, Flynn, et al., 2014) may be more difficult to achieve in programs, such as those serving AI/AN populations, where lack of funding, geographic and professional isolation, and staff burnout/turnover are prevalent (Legha, et al., 2014).

Some resources can help to address different conceptualizations of EBTs. For example, the National Registry of Evidence-based Practices and Programs (2012) provides information on a number of characteristics of EBTs (e.g., analysis of the evidence for EBT effectiveness, reconciliation of required resources for implementation and ongoing sustainability relative to program capacity). However, such resources may fall short in identifying cultural fit or adaptability of EBTs for specific populations. To support AI/AN programs in assessing EBTs, it may be necessary to build a technical assistance infrastructure specific to treatment programs serving AI/AN communities that can facilitate the expansion of knowledge in these programs and assist in the exploration of potential EBTs, preparation for EBT implementation, and subsequent implementation and ongoing sustainment. Recent efforts by the Addiction Technology Transfer Centers (www.attcnetwork.org/regcenters/index_nfa_americanindian.asp) are consistent with such an approach, but may not be adequate to address the significant informational and infrastructural issues identified as challenging in addiction treatment for AI/AN people. Indeed, the development of a technical assistance infrastructure could address the cultural relevance of EBTs and provide resources to enhance programs’ ability to recruit and train qualified staff, and to develop relevant quality assurance and evaluation infrastructure to assess need for adaptation, adherence to EBT protocols, and assessment of clinically and culturally relevant client outcomes.

Finally, the results from this study suggest that there is some mistrust of EBTs in these programs and that efforts to enhance their use will need to squarely address concerns about the cultural appropriateness of EBTs, acceptable approaches to adaptation, and how EBTs can be utilized while still facilitating individualized treatment approaches. Indeed, the core concerns of the western/biomedical roots of EBTs suggest that open discussions of this process will be critical to improving the reach of key EBTs and the quality of their implementation. Program directors must grapple with important questions, such as whether they utilize EBTs, even if they have been shown to be effective in mainstream populations only, or embark on expensive and lengthy trials in AI/AN communities to ensure that these EBTs are culturally appropriate and effective. A full understanding of what constitutes an EBT may assist directors, and their programs, to better wrestle with such challenging decisions. However, our findings suggest that narrow definitions of EBTs that rely only on research evidence do little to enhance their broader acceptance. The use of tripartite EBT definitions that include research evidence; clinical judgment/expertise; and consumer choice, preference, and culture would likely go a long way towards allaying concerns regarding the imposition of practices based only on research with a limited population (Institute of Medicine, 2001; American Psychological Association Presidential Task Force on Evidence-Based Practice, 2006). In addition, the integration of cultural adaptation with implementation science holds promise for making the adoption and implementation of EBTs more relevant for different communities (Cabassa & Baumann, 2013).

More inclusive approaches to exploration and implementation could also be helpful in finding relevance and fit of EBTs for AI/AN communities. The use of collaborative approaches to intervention development and to dissemination and implementation of EBTs can facilitate researchers gaining a greater understanding of how to adapt or tailor interventions for specific populations. For example, community-based participatory research has great potential to have the culture and values of AI/AN communities drive research questions, research methods, and interpretation of data (Minkler & Wallerstein, 2010). Community-participatory partnered research also supports values of shared leadership, shared power and control, and shared fiscal resources in improving care for communities (Jones & Wells, 2007). Other process models that consider system, organization, community, provider, and client needs help to identify and adapt systems and organizational process and EBTs for a given community (Aarons, et al., 2012).

Some limitations of this study should be noted. First, broad heterogeneity exists among AI/AN populations, communities, and substance abuse treatment programs. Regional and tribal characteristics, geographic location, urbanicity, and patterns of substance use (Beals, et al., 2003; Etz, et al., 2012) differ widely among different AI/AN groups, limiting any generalizations we might make about providers who treat AI/ANs with substance use problems relative to their definition of EBTs. Second, because we selected programs that had reputations for providing innovative services, our respondents may not fully represent programs offering services to AI/ANs with substance use problems. Third, because we did not collect demographic data from respondents, we were unable to explore how characteristics such as ethnicity or years of experience in the addictions field may influence providers’ definitions of EBT. Finally, without comparable studies of treatment programs for substance abuse serving non-AI/AN communities, it is difficult to determine whether these findings, particularly those regarding the depth of knowledge of EBTs, are specific to programs serving AI/AN programs or whether this is a concern in treatment programs more broadly. However, our finding pointing to apprehensions regarding the cultural appropriateness of EBTs is highly consistent with prior commentaries and is likely a concern that is particularly salient in programs serving AI/AN communities. EBTs hold the promise for improving treatment services in AI/AN communities, but without concerted efforts to enhance these programs’ capacities to consider, assess, implement, and sustain the use of these interventions, the ability of these high-need communities to fully benefit from their culturally appropriate implementation will be limited.