Strategies for Inclusivity of American Indian and Alaska Native Peoples in Behavior Therapy Research: Within-Group Diversity, Data, and Ethical Recommendations

THE DOMINANT narrative and depictions of American Indian and Alaska Native (AI/AN) communities are situated in the past, within fictionalized media representations, and posit that AI/AN peoples no longer exist. The reality is that AI/AN communities have and still do inhabit their traditional homelands, and tribal communities can be traced across the U.S. A constant is that most, if not all, tribal names reflect an interconnectivity and a recognition of humanity; these names hold variations that translate to “people,” the “real people,” or “human beings.” Identity reflects geography, history, ancestry, and an understanding of what it means to be alive and in communion with all that is alive. Likewise, the authors of this piece come to you humbly as academics, teachers, and learners, but also as the real people of the tribal lands of Turtle Island (i.e., the U.S.). The goal of this commentary is to (1) convey the richness, diversity, and inherent strengths of AI/AN peoples, which cannot and have not been extinguished by any force of humankind; (2) lay charge to the persistent and unrelenting impacts of historical trauma and oppression upheld by sociopolitical systems and institutions of settler-colonial dominance and evident in racism; (3) posit that behavioral therapy research, data, conclusions and applications therein are rarely specific to nor examined among AI/AN communities; and (4) hold that addressing issues with sampling, methodology, measurement, and data management are key to increasing appropriate research and data analyses specific to the AI/AN population, and thus, pertinent to research reconciliation and sovereignty. Moreover, we assert that the twofold process of decolonization and indigenization is crucial. Smith (2021) defines decolonization as “once viewed as the formal process of handing over the instruments of government, is now recognized as a long-term process involving the bureaucratic, cultural, linguistic and psychological divesting of colonial power” (p. 112). Whereas indigenization refers to the inclusion of Indigenous peoples, epistemologies, knowledge systems, worldviews in science, research, and policy to predominantly Western spaces and systems (Gaudry & Lorenz, 2018).

Tribal Diversity Across Turtle Island

AI/AN peoples are often viewed as a large, homogenized group, leading to the prevalent belief in a monolithic culture. However, this misconception cannot be further from reality. Across the U.S., there are currently 574 federally recognized tribes that represent distinct cultures, geographic locales, and 175 unique languages. Though there are similarities among AI/AN groups, the 574 federally recognized tribes reflect a great diversity of culture and self-governance of tribal citizens that exist among and across these communities (National Congress of American Indians [NCAI], 2020). The cultural distinctions span from the most remote village or reservation to the most populated cities, where the needs and behaviors of the people can differ greatly (Herron et al., 2021). In fact, while only 3% of the AI/AN population lives on reservations, it is estimated that over 70% live in urban areas, which indicates an outward migration from rural communities and/or reservations towards higher resourced centers (e.g., transportation, medical care, employment). Alarmingly, only 3% of AI/AN research is based within these urban centers. There could be many reasons for this, such as access and availability, proximity to Indian Health Services, real or perceived need, or limited knowledge of population distribution and what it means to identify as AI/AN. However, to unravel the discrepancies and disparities within AI/AN research, it is important to understand the historical context.

Historical Trauma and Postcolonial Distress

A discussion of historical trauma is pertinent to the psychosocial, ethical, and methodological issues pertaining to AI/AN communities and research. Historical trauma is defined as culturally based traumas intentionally inflicted on a cultural group systematically and systemically (e.g., genocide, forced relocation, residential boarding schools; Graziosi et al., 2021). However, historical trauma is not situated in history alone—it represents an ongoing process that has not been stopped or ceded (Kirmayer et al., 2014). Rather, postcolonial distress is characterized as (a) the initial colonial wounding perpetrated against AI/AN peoples through extermination, relocation, decimation, segregation, and assimilation; (b) the traumas that halted the natural course of change and stole critical components of culture (e.g., language, medicine people) while simultaneously enforcing settler-colonial ideals; (c) the colonization is continuous in sociopolitical systems, institutions, and policies, while also being cumulative and exacerbated by present stressors (e.g., racism, violence); and (d) the intergenerational transmission of these traumas and effects continues through physiology, environment, and/or social interaction until the cycle is broken.

Postcolonial distress is exacerbated by racism in all its forms: oppressive policies, stealing of tribal lands, inadequate resources, and limited access to healthcare, education, housing, employment, and healthy foods (Kirmayer et al., 2014; Whitesell et al., 2012). Further, the symptoms and sequelae include health disparities and disproportionate rates of homelessness, violence, suicide, and substance use disorders (SUDs; BigFoot & Schmidt, 2010; Lopez et al., 2021; Whitesell et al.). In fact, AI/AN peoples are two times more likely to experience posttraumatic stress disorder (PTSD) and have a higher 12-month and lifetime prevalence of PTSD in contrast to all other U.S. ethnoracial groups (Gameon & Skewes, 2020; Goldstein et al., 2016). In addition, AI/AN populations experience disproportionately high rates of SUDs; namely, AI/AN communities experience the highest rates of alcohol use disorder (AUD), binge drinking, alcohol-related consequences, and severity for lifetime prevalence and alcohol-related morbidity and mortality (Grant et al., 2015; Lopez et al., 2021; Whitesell et al., 2012). Moreover, while there are variations in rates by tribe and region, more research is needed to understand these rates to work with communities to improve, increase, and culturally tailor mental health treatments.

Culturally Congruent Behavior Therapies

Behavioral interventions have largely been normed and, thus, adapted for the general, predominantly White U.S. population (Novins et al., 2016). Even intervention studies that consider broader ethnoracial categories focus mainly on White, Black, and Latinx population, whereas the AI/AN sample is frequently lumped into the “other” category that often includes Asian, Native Hawaiian, Pacific Islander, and/or multiracial participants (Gartner et al., 2021; Novins et al., 2016). Cultural derived and adapted evidence-based treatments (EBTs) have been found to be efficacious and are well received among AI/AN communities, which is critical given the disparities experienced and the resultant treatment needs (Coser et al., 2021; Gameon & Skewes, 2020; Graziosi et al., 2021). In other words, EBTs have demonstrated cultural congruency, acceptability, and efficacy among AI/AN populations when they are based on cultural values and practices (e.g., sweat lodge, ceremony, traditional teachings; Rowan et al., 2014) and when they are culturally adapted (Venner et al., 2016). For example, incorporating cultural practices into dialectical behavior therapy (DBT) among AI/AN youth diagnosed with a SUD demonstrated that 96% had improved or recovered at discharge, as measured by internalizing and externalizing on the Youth Outcome Questionnaire–Self Report and evidenced by significance and large treatment effects (Beckstead et al., 2015). As another illustration, a clinical trial of a cognitive processing therapy (CPT) adaptation to treat PTSD, illicit drug and alcohol use, and high-risk sexual behavior among AI/AN adults, it was observed that the treatment group had large reductions in symptoms across all treatment targets (Pearson et al., 2019). Furthermore, evidence-based interventions that are culturally derived/adapted (i.e., based in and developed from cultural practices and worldviews within the target community) have demonstrated efficacy for the treatment of PTSD, AUD, SUDs, and other mental health diagnoses (BigFoot & Schmidt, 2010; Dickerson et al., 2018; Johnson et al., 2021). However, while behavioral EBTs for mental health and SUDs have demonstrated efficacy for AI/AN peoples, barriers related to underutilization of data, and lack of cultural interventions that increase access, efficacy, and retention, persist.

Identity Politics and Sampling

An array of factors that contribute to how and why AI/AN peoples do or do not self-identify has been found in U.S. Census data (Wood & Hays, 2014). In 2010, 5.6 million AI/AN peoples reported identifying as multiracial, and 2.9 million reported identifying solely as AI/AN (Connolly et al., 2019). AI/AN peoples are the only racialized group in the U.S. who must prove their identity through blood quantum and tribal enrollment, which has long been a tool of conquest and acquisition of tribal lands through systematic dilution of identity (Gartner et al., 2021). However, with the insurgence of ancestry testing (e.g., DNA) and long-lost Indigenous ancestry being in vogue, there has been an increase in people reporting mixed AI/AN heritage on census data, but not a comparable increase in tribal enrollment (Connolly et al.). Nonetheless, it is likely that those who have a matching between AI/AN identity and culture are more likely to be utilizing relevant services and resources pertinent to AI/AN specific mental health treatment and research. While the use of data for AI/AN alone might be appropriate for policy or program research that could impact funding and be specific to federally recognized tribal entities, AI/AN alone or in conjunction with another race could be more appropriate when considering more general cultural categories and identities related to behavioral intervention research (Connolly et al., 2019)—namely, those who are tribal members of tribes not federally recognized or decedents of enrolled members who have been raised in the culture and context of their tribal ancestry. Consideration of AI/AN heritage in conjunction with another ethnoracial category is an important means of understanding and validating the lived realities of the critically underrepresented Indigenous Black and mixed Indigenous American and Latinx populations. Moreover, while beyond the scope of this commentary, Native Hawaiian, Chamorro, and American Samoan communities are also considered Native American, albeit not federally recognized but as trust territories, and have experienced similar histories of colonization, historical trauma, and postcolonial distress (Leavitt et al., 2015; NCAI, 2020).

There are copious reasons as to why deindividuation of AI/AN identity occurs, such as invisibility and erasure; internalization of homogenized and pejorative representations (e.g., “savagery,” mascots); and the very real dangers of identifying as AI/AN (e.g., racism, oppression, violence), to name a few (Leavitt et al., 2015; Wood & Hays, 2014). However, government and medical officials tend to underreport AI/AN identity even when the person is tribally enrolled. This is due to multiple factors embedded in racist views of how an AI/AN person is expected to speak, look, and present (Wood & Hays). In some instances, tribal enrollment relative to birth is contrary to race on death certificates. In other words, a person is born AI/AN but dies non-Native. In the National Cancer Institute’s Surveillance, Epidemiology, and End Results (SEER) and the related National Longitudinal Mortality Study survey, 78% of those of AI/AN ancestry were placed into a different racial category, which was 57–77 percentage points higher than all other racial categories’ misclassification (Clegg et al., 2007). Administrative misclassification has been demonstrated in comparisons of survey data and clerical data within the Veterans Affairs (VA), prison system, and healthcare settings (Leavitt et al., 2015). Posited reasons for misclassification incongruous with self-report include unfamiliarity with tribal diversity, assumptions made about last names (e.g., Hernandez, Smith), and racist or biased views of phenotypes and what it means to be AI/AN (Leavitt et al., 2015; Wood & Hays).

Importantly, inclusion of AI/AN peoples in clinical research is critical to culturally congruent, tailored, and accessible mental health treatment for this underserved population. Notably, Hays and Iwamasa (2019) state that “the omission of ethnic and other cultural information is the rule rather than the exception in clinical and counseling research” (p. 5). Part of integrating AI/AN, multicultural, and diverse perspective into clinical theory and practice starts with the inclusion of those knowledge systems and very peoples at the data gathering, analytic, and reporting levels. In other words, sampling and cultural identity—and thus, cultural practices, knowledge, and worldview—are crucial to the fit of clinical interventions (Bernal & Domenech Rodriguez, 2012). While some strictly adhere to the fidelity of an EBT protocol in resistance to adaptations, research (BigFoot & Schmidt, 2010; Coser et al., 2021; Pearson et al., 2019; Venner et al., 2016) has observed that culturally derived and adapted treatments are efficacious, relevant, and reflect culturally responsive research and practice. Furthermore, meta-analyses have demonstrated that culturally adapted treatments are equally if not more effective that nonadapted treatments (Benish et al., 2011; Rathod et al., 2018; Smith et al., 2011).

Appropriate Methodology and Measurement

Methodology and measurement within AI/AN EBT research have the potential to be decolonizing and have been defined by an iterative process representative of the pan-Indigenous medicine wheel (Walls et al., 2019). Occurring across three levels of specificity, methodology is tailored for (a) the general and broad AI/AN community, (b) a population, region, or locale (e.g., rural, urban); or (c) a specific tribal group or community. A priori AI/AN conceptualizations and operationalization for the research specific to the level of tribal specificity are posited to increase validity (Lopez, 2020; Walls et al., 2019). Collaborative and participatory research is most analogous to Indigenous research approaches and recognizes tribal and data sovereignty and knowledge ownership. These partnerships also greatly impact the results and the ways in which the results are understood and shared.

Correspondingly, measurement equivalence is integral to making comparisons and obtaining appropriate data, just as cultural adaptation is essential for salient behavior therapies. For example, Motivational Interviewing (MI) has previously been critiqued for its usage with AI/AN peoples based on its development and norming with predominantly White U.S. populations (Venner at al., 2008; Venner et al., 2016); however, after cultural adaptations and equivocal norming, it has been found to be efficacious with AI/AN populations (Dickerson et al., 2018; Novins et al., 2016). Where larger, epidemiological datasets are considered, data weighting has been accepted as another way to contextualize data for typically small samples like AI/AN subsamples in contrast to other groups. However, to mitigate error and inaccuracies, it is suggested to only weight the data when appropriate (e.g., larger sample size, more than 100) and usage of fewer variables is indicated (Becker et al., 2021; Gartner et al., 2021; Urban Indian Health Institute [UIHI], 2020). Furthermore, a careful consideration by researchers of best practices in designing research, measurement strategies, and use and analyses of AI/AN data has the potential to effect larger changes across research and presents real-world impacts for AI/AN communities that have been historically under- and misrepresented.

The othering of AI/AN in datasets can be misleading, represent a systematic erasure of Indigenous communities, diminish the urgency to serve these populations, and act as an injustice to unique mental health and treatment needs of AI/AN peoples (Gone & Kirmayer, 2020). Furthermore, the lack of emphasis or outright exclusion of these populations in datasets directly erases the AI/AN participant community and thereby limits the resources that are developed and made available. The diversity that prevails in AI/AN populations is negated by the secular or ignored reporting in ethnicity-based reports. For datasets to best identify ways to serve Indigenous communities’ quality of health, the ethnicity-based data collected must be segregated by location so that tribal-specific data can identify the unique needs of different tribes. This cannot be achieved without first including the AI/AN data on all reports that concern ethnicity or race. The NCAI (2020) refers to AI/AN peoples in research as “the asterisk nation,” meaning that the population is not represented in the data or as a data point but rather is demarcated by an asterisk in reports. This is particularly poignant for AI/AN peoples, as national datasets are used to make policy decisions and allocate large funds to tribal communities for crucial resources.

Data disaggregation has been identified as a strategy for inclusion and increased participant numbers and, thus, increased power for analyses among the AI/AN subgroup (Becker et al., 2021; Lopez, 2020). By definition, “data disaggregation refers to the separation of compiled information into smaller units to elucidate underlying trends and patterns” (Pan American Health Organization [PAHO], 2021). For AI/AN peoples this would mean disaggregating race by tribal affiliation (e.g., Navajo Nation, Aztec, Nome Eskimo Community). According to the Fundamental Principles of Official Statistics, at a minimum, efforts should also be made to disaggregate by demographics, such as age, gender, sex, education, and geography/region. For example, data that disaggregates AI/AN people as a distinct ethnoracial group (as opposed to the typical categorization of “multiracial” or “other”) is more representative of the population and yields more contextualized data. Moreover, the disaggregation of the AI/AN category into specific tribal/regional groups helps further contextualize the nuance of tribal diversity by identifying the position and constraints defined by sociopolitical systems (e.g., sexual orientation, location, socioeconomic status) as it pertains to tribal/ regional differences.

Undoing Past Hurts of Research

There are long and painful histories of unethical research in AI/AN communities that have all but eroded trust between researchers and participants, healthcare systems and patients, and governments and constituents (Griffiths et al., 2021; Herron et al., 2021; Pacheco et al., 2013). However, ethical, inclusive, indigenized, and decolonial research practices in these communities have the power to build relationships, strengthen partnerships, instill trust, and, ultimately, begin to undo the history of the hurts of the past. Managing and stewarding the data in ways that are inclusive of tribal communities is a best-practice framework for conducting research and engaging with AI/AN data. Several recommendations related to research and analyses of AI/AN datasets are in Table 1 and based on existing literature (Becker et al., 2021; Gartner et al., 2021; Lopez, 2020; UIHI, 2020). This table lists recommendation and related strategies to consider whether sampling is large (e.g., epidemiological), medium (e.g., most clinical research), small (e.g., unable to achieve significance but able to report effect sizes), and if one is considering engaging the AI/AN community. Additionally, the table can be used generally to consider ways to enhance and improve research and/or clinical work among the AI/AN population. Last, for further readings and materials, please see Table 2.

Table 1.

Recommendations for Clinical Research and Datasets

Recommendation	Strategies
1. Consult with tribal communities and understand the data sovereignty protocols, guidelines, and practices in the community or region the research is being/was conducted.	Ideal for engaging the community for ethical research, initial introduction to recruitment, and/or increasing participation.
2. Be clear when communicating racial and ethnic identification selection procedures to ensure accurate reporting. It is recommended that participants be encouraged to select AI/AN alone and careful considerations be made about selecting multiple eth- noracial categories.	Selection can affect what is reported regardless of sample size and can impact future sampling procedures, techniques, and strategies.
3. Under no circumstances should participants be categorized without their knowledge, as this is impacted by implicit biases; AI/AN categorization should be by selfselection.	Never assume identity; always let participant/client self-identify to obtain the most accurate and contextual information.
4. Include culturally relevant variables to enhance understanding of the data and to make the data culturally salient, consistent, and of good quality. For example, collect data about tribal affiliation.	Regardless of target sample size, ask about tribe, tribal affiliation, reservation/village involvement, etc. Unless targeting a non-AI/AN sample, do not assume none will be AI/AN.
5. Conduct mixed methods research to obtain both quantitative and qualitative data that represents the lived realities of the AI/AN sample.	Both are valuable methods and are strengthened together, particularly when sample sizes are small (e.g., qualitative data can contextualize the sample).
6. Use culturally derived/culturally adapted measures for construct equivalence in the AI/AN sample (i.e., AI/AN measures, measures adapted for AI/AN peoples).	Using measures not otherwise used with the AI/AN sample can be limiting, not measuring the same construct, and be culturally inappropriate.
7. Count AI/AN peoples alone and in conjunction with other ethnoracial groups. Inclusion is the antidote to othering. Thus, resist grouping AI/AN into the multiracial or the other category.	Do not lump data into an “other” category; rather, report the subsample and what is known, even if subsample analyses cannot be done.
8. Use weighted sampling procedures if indicated and applicable for proportional comparisons and contextualization of the sample.	Use ranking, matching, propensity weighting, or a combination, as deemed appropriate for small, medium, or large sample sizes.
9. Share what is there. Small sample size and inability to achieve statistical significance is not a barrier for AI/AN communities who want to know and understand their data.	Sharing the sample data is critical to representation, current and future data usage, including metanalytic procedures, data pooling, and transparency of AI/AN data.
10. Disseminate based on the recommendations of tribal communities, advisor(s), and/or the tribal institutional review board(s).	Always consult with tribal review boards, community members, and/or advisor(s) to determine who owns the data, ow confidentiality is protected, and to guide all aspects of the research.

Table 2.

Recommended Reading and Resources for Ethical Research Practices With AI/AN Peoples

Readings	Websites	Videos
Best Practices for American Indian and Alaska Native Data Collection (Urban Indian Health Institute, 2020)	National Congress of American Indians https://www.ncai.org/policy-research-center/research-data/introduction-to-research	“Decolonizing Data: Strengthening Community Voices to Take Action for Our Missing Relatives” https://www.uihi.org/resources/decolonizing-data-strengthening-community-voices-to-take-action-for-our-missing-relatives/
Culturally Responsive Cognitive Behavior Therapy, Second Edition: Practice and Supervision (Hays & Iwamasa, 2019; Book)	National Council of Urban Indian Health https://ncuih.org/	“Ethical Frameworks for Research Collaboration with Indigenous Communities” https://elsihub.org/video/ethical-frame-works-research-collaboration-indigenous-communities
Indigenous Data, Indigenous Methodologies and Indigenous Data Sovereignty (Walter & Suina, 2019; Article)	Substance Abuse and Mental Health Service Administration https://www.samhsa.gov/tribal-affairs	“Supporting Ethical Research Involving American Indian/Alaska Native (AI/AN) Populations” https://videocast.nih.gov/watch=42314

Conclusion

Othering is not only an extension of historical trauma and violence, but a function of sociopolitical structures of racism and oppression that has no place in behavioral treatment or research. Future directions include concerted efforts to include the AI/AN populations in behavioral research; inclusive statistical strategies for analyses of small group data; intervention norming and cultural adaptations; and inclusion of AI/AN participation and voice in behavioral research for collective group visibility, belonging, and empowerment. The insufficient data-collecting and reporting of AI/AN communities is a pertinent example of how systems that carry historically harmful intended actions continue to jeopardize Indigenous peoples. The product of the faulty data–collecting and reporting of AI/AN communities is perpetual invisibility that fails to recognize the support that could otherwise be provided for the Indigenous people of Turtle Island. The disaggregation of data and inclusion of AI/AN subsamples would provide invaluable information about the efficacy of culturally derived and adapted EBTs. Expectantly, this increase in data clarity and availability could increase AI/AN focused and funded research. Furthermore, an indigenized, decolonial process could undo past hurts, promote reconciliation, and support tribal sovereignty.

Why Become Board Certified?

• “Gold standard” of professional practice

• ABPP is a “trusted credential” (i.e., psychologists have met their board’s specialty’s standards and competencies)

• Potential for increased financial compensation (VAs, PHS, DOD, some hospitals)

• Reflects a higher standard of expertise over a generalist license

• Enhances one’s qualifications as an expert witness; facilitates applying to insurance companies’ networks

• 40 CE credits after passing the exam

3 Steps to Board Certification:

1. Submission of educational/training materials.

2. Review of a practice sample or senior option (≥ 15 years of experience, there are alternatives for a practice sample).

3. Collegial exam.

• We encourage diversity of all types and we recognize diversity has breadth

• We offer a discounted application fee for graduate students, interns, and postdoctoral residents

• We conduct exams at APA and ABCT conferences, and other locations as possible (virtual and face-to face)

• We offer a free workshop at the ABCT conference and free mentoring.

• Online application: https://abpp.org/Applicant-Information/Specialty-Boards/Behvioral-Cognitive.aspx