Culturally tailored substance use interventions for Indigenous people of North America: a systematic review

Abstract

Purpose–

The purpose of the current study is to conduct a systematic review of peer-reviewed work on culturally tailored interventions for alcohol and drug use in Indigenous adults in North America. Substance use has been reported as a health concern for many Indigenous communities. Indigenous groups experienced the highest drug overdose death rates in 2015, the largest percentage increase in the number of deaths over time from 1999 to 2015 compared to any other racial group. However, few Indigenous individuals report participating in treatment for alcohol or drug use, which may reflect the limited engagement that Indigenous groups have with treatment options that are accessible, effective and culturally integrative.

Design/methodology/approach–

Electronic searches were conducted from 2000 to April 21, 2021, using PsycINFO, Cumulative Index to Nursing and Allied Health Literature, MEDLINE and PubMed. Two reviewers classified abstracts for study inclusion, resulting in 18 studies.

Findings–

Most studies were conducted in the USA (89%). Interventions were largely implemented in Tribal/rural settings (61%), with a minority implemented in both Tribal and urban contexts (11%). Study samples ranged from 4 to 742 clients. Interventions were most often conducted in residential treatment settings (39%). Only one (6%) intervention focused on opioid use among Indigenous people. Most interventions addressed the use of both drugs and alcohol (72%), with only three (17%) interventions specifically intended to reduce alcohol use.

Originality/value–

The results of this research lend insight into the characteristics of culturally integrative treatment options for Indigenous groups and highlight the need for increased investment in research related to culturally tailored treatment across the diverse landscape of Indigenous populations.

Introduction

Indigenous populations represent a growing proportion of drug and alcohol fatalities in the USA and Canada. The Centers for Disease Control reported that Indigenous groups had the highest drug overdose death rates in 2015, and the largest percentage increase in the number of deaths over time from 1999 to 2015 compared to any other racial group in the USA (Mack, 2017). Opioid-related overdose deaths have increased by five times between 1999 and 2015 in Indigenous populations, equating to 22 deaths per 100,000 in metropolitan areas and almost 20 deaths for 100,000 in nonmetropolitan areas. Alcohol-related fatalities are also rising. In 2016, alcohol-related death rates were 113.2 and 58.8 deaths per 100,000 residents among Indigenous men and women, respectively, making Indigenous groups the highest age-standardized rate for alcohol fatalities (Mack, 2017). Indigenous groups located in Canada face similar risks to those located in the USA. Despite representing just 2.6% of the total population, Indigenous groups account for 10% of total drug overdose deaths across Canada (Government of Canada, 2021). Though Indigenous individuals in North America have the highest rates of abstinence from drugs and alcohol (Cunningham et al., 2016), they are still considered an at-risk population for substance-related death. This makes the treatment of substance use disorder (SUD) in this population of unique and paramount concern.

The prevalence of Indigenous SUD stems in large part from the intergenerational experiences of colonialism, namely, though historical trauma. Prominent scholars have defined historical trauma as the psychological effects stemming from the multitude of atrocities associated with colonization, including forced removal from Indigenous homelands, mandated boarding school attendance and the subsequent cultural oppression in these spaces, and the extreme levels of physical, sexual and spiritual violence that are still perpetuated today (Gone et al., 2019). Historical trauma has been identified as a prominent source of health disparity (Gone et al., 2019 for a review), and has been used as a lens to contextualize rates of drug and alcohol use disorder in Indigenous populations (Brave Heart, 2003).

Despite these experiences, Indigenous people are highly resilient and have been especially successful in using Indigenous systems of knowledge to address the behavioral manifestations of historical trauma. In particular, culturally tailored treatment regimens based in Indigenous systems of knowledge show promising evidence related to improved SUD outcomes (Brave Heart et al., 2011; Hartmann et al., 2019). However, health systems that service Indigenous groups can be poorly resourced (Kruse et al., 2022) and are sometimes located in areas that are geographically removed from the Indigenous populations they intend to serve (Indian Health Service, 2018). There is little catalogued information about the characteristics of culturally tailored programming; how these programs are distributed across tribal, rural and urban Indigenous groups; and what types of substances these interventions address.

Study objective

To our knowledge, there has been no systematic review of peer-reviewed studies that examine culturally tailored treatment regimens in the USA and Canada that assess the study quality. To contribute to the growing body of knowledge and understanding the treatment of SUD in Indigenous populations, the current study presents a systematic review of peer-reviewed work on culturally tailored interventions for alcohol and drug use in Indigenous adults in North America. Over the course of this work, we use the word Indigenous, a term that does not center arbitrary geo-political lines, to describe American Indian, Canadian Aboriginal, Native American and Alaska Native populations.

The articles that met inclusion criteria for this review were analyzed both in terms of the intervention itself (primary population, intervention type, substance use addressed), geographic location (country, region and tribal versus urban) and study design. We include a quality assessment for each of the relevant studies. The goal of this research is to explicate the current state of empirically based and culturally tailored intervention efforts to identify potential gaps in accessibility and improve the provision of treatment services.

Methods

This systematic review was guided by core tenets of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) (Page et al., 2021), which included creating a protocol that defined the rationale and objectives, inclusion and exclusion criteria, search methods and study selection and data collection and extraction guidelines.

Study inclusion and exclusion criteria

This review included studies that were explicitly culturally tailored for Indigenous people living in the USA or Canada and addressed substance misuse treatment, prevention and/or intervention. The word “Indigenous” is used here to describe American Indian, Native American and Alaskan Native populations in a way that does not center arbitrary geo-political lines. We define culturally tailored as any intervention that tailors its messaging, service delivery and context to address the unique strength and needs of an Indigenous community (Samuels et al., 2009). We examined studies that included adults aged 18 and older, who self-identified as American Indian and Alaska Native (AI/AN), Indigenous, Native Hawaiian, AK Native, American Indian or Native American. All behavioral, trauma-informed, community level and structural intervention modes (e.g. individual, couple, group, community, structural, social media) were included. Additionally, delivery formats of multiple types were included (e.g. in person, telephone, mobile app, computer-assisted) as well as multiple settings including community, residential, outpatient and criminal justice settings.

In regard to the peer-reviewed studies associated with interventions, we were also intentional in documenting study type. We included randomized controlled trials (RCTs), individual or cluster, quasi-experimental studies including quasi-randomized studies, non-RCTs, pre/posttest and outcome assessments (program evaluation), as well as qualitative studies and mixed methods studies. Only English-language articles were searched and included. We excluded studies that did not include Indigenous populations and studies that only examined pharmacologic treatments and interventions that were not culturally tailored.

Literature search

Electronic searches through PsycINFO, Cumulative Index to Nursing and Allied Health Literature (CINAHL), MEDLINE and PubMed were all conducted on April 19, 2021. We limited the search to include articles published in peer-reviewed journals from January 1, 2000 to April 19, 2021, which were available in English. Dissertations and theses were excluded. No additional limits or filters were used in the search. Search terms were divided into four search strings:

• AI/AN and Indigenous;

• culturally adapted;

• substance use/misuse; and

• drug treatment.

Search terms within each search string were connected with the “OR” Boolean operator. Search terms for each search string are as follows:

• “AI/AN” OR “Indigenous” OR “Native Hawaiian” OR “Alaska Native” OR “American Indian” OR “Native American”;

• “Culturally adapted” OR “community led” OR “traditional medicine” OR “culturally responsive” OR “ADAPT-ITT” OR “trauma informed” OR “historical trauma” OR “intergenerational trauma” OR “urban” OR “tribal” OR “reservation based” OR “cultural sensitivity” OR “culturally informed” OR “cultural integration” OR “community-based”;

• “Substance Use” OR “drug use” OR “inject” OR “intravenous” OR “drug abuse” OR “substance use” OR “problematic drug use” OR “substance misuse” OR “addiction” OR “dependence” OR “substance use disorder” OR “SUD” OR “illicit drug” OR “substance-related disorder” OR “illicit” OR “narcotics” OR “overdose” OR “poly-drug use” OR “poly-substance use” OR “multiple drug abuse” OR “drug overdose” OR “heroin” OR “opiate” OR “opium” OR “opioids” OR “diamorphine” OR “morphine” OR “methadone” OR “analgesics” OR “buprenorphine” OR “fentanyl” OR “cocaine” OR “crack” OR “stimulants” OR “amphetamine-type stimulants” OR “amphetamine” OR “methamphetamine” OR “ATS” OR “methamphetamine hydrochloride” OR “non-prescribed barbiturates” OR “ecstasy” OR “MDMA” OR “benzodiazepine use” OR “alcohol”; and

• “Drug treatment” OR “medication assisted treatment” OR “MAT” OR “medications for opioid use disorder” OR “MOUD” OR “drug treatment” OR “substance use disorder treatment” OR “methadone maintenance treatment” OR “buprenorphine” OR “naltrexone” OR “suboxone” OR “harm reduction” OR “residential treatment” OR “nonresidential treatment” OR “outpatient.”

After the initial search strings were conducted, they were combined with the following Boolean operators: string 1 AND string 2 AND string 3 OR string 4.

The study selection was supported by Covidence (Covidence Systematic Review Software, 2022), a primary screening and data extraction tool. After the searches were conducted, the raster image files were uploaded into Covidence that cross-referenced each article and removed duplicate articles. Both authors screened all articles independently using the inclusion/exclusion criteria and selected a rating of “yes,” “no,” or “maybe” in the Covidence (2022) program. Both authors discussed conflicts and came to a consensus based on the inclusion and exclusion criteria. In instances where abstracts were seemingly relevant or unclear (e.g. it was unclear the level of cultural adaptation or the age group included), the article was selected for full-text review. Both coauthors independently reviewed all relevant articles and selected “include” or “exclude” in Covidence. Similar to abstract review, the coauthors met to discuss conflicts in accordance with the exclusion and inclusion criteria.

Multiple papers that reported data from the same data set or project were merged and counted as the same study. If it was unclear whether the information was from the same study, the coauthors attempted to reach out to the authors for clarification. Coauthors also reviewed references of relevant articles, but this did not yield any additional studies. We developed a list of five prominent researchers who lead studies related to the inclusion and exclusion criteria of this systematic review. We reached out to these relevant researchers via phone and email to ask about interventions and studies that we may have missed. Three of the five researchers responded, which yielded two additional studies.

Data extraction

A standardized template for data extraction of key information was developed and completed in Covidence for each article. For studies that met the inclusion criteria during the full-text review stage, data extraction categories included publication characteristics (article title, year of publication, full citation), geographic characteristics (study location, region, urban/tribal), study aims, study design, description of population, substance use addressed, main outcomes, whether it was community-based, program type, inclusion/exclusion criteria, total number of participants and quality appraisal (discussed further in the following section).

Quality appraisal

Both authors independently reviewed and assigned quality scores to each of the articles using the Mixed Methods Appraisal Tool (MMAT) (Hong et al., 2018). The tool provides a framework to assess multiple study designs including qualitative research, quantitative RCTs, quantitative nonrandomized studies, quantitative descriptive studies and mixed methods studies. Each study design category asks five questions that signal study quality relative to the study design (e.g. “Is randomization appropriately performed?” and “Is the qualitative approach appropriate to answer the research question?”). Reviewers can select, “Yes,” “No” or “Can’t Tell.’ Study quality ratings from one to five are assigned relative to the number of “Yes” answers selected for the study. Studies with a bias rating of five indicate low overall bias, and a bias rating of one indicated a high overall bias. Bias ratings are as follows: 1 – critical bias, 2 – serious bias, 3 – moderate bias, 4 – adequate bias and 5 – low bias. Discrepancies were settled by review and consensus between both authors.

Data synthesis

After data extraction in Covidence, we downloaded the compiled extraction data in the form of a spreadsheet. Both authors independently reviewed the spreadsheet and made notes about information that needed further clarity. Together, both authors reviewed the spreadsheet ensuring that information was presented clearly and consistently across each data extraction variable. In an effort to streamline how this information was presented in Table 1, we combined first author, year of publication and study name (if applicable). Country, region and tribal or urban designations were combined and presented in the same column. Type of study, main outcomes/aim and study bias rating were presented together as were sample size and inclusion/exclusion criteria, and substance use addressed and program type. The final column present if the study was community-based.

Table 1.

Characteristics of included studies (n = 18)

First author (year) and study name, if applicable	Country; region; tribal or urban	Type of study; main outcomes/aim; Bias rating (1–5)	Sample size and inclusion criteria	Substance use addressed; program type	Community-based
Burduli etal. (2018) HONOR study	USA;Midwest;rural	RCT; Outcomes: Primary: Recent alcohol use (past three days), recent drug use (opioids, amphetamine, methamphetamine, cocaine, cannabis) Secondary: self-reported alcohol and drug use Bias rating: 5	114; Inclusion: (1) self-reported American Indian race; (2) seeking alcohol misuse or dependence and drug misuse or dependence treatment on a participating reservation; (3) age 18–65years; (4) Diagnostic and Statistical Manual, fourth edition diagnosis of current alcohol dependence; (5) current drug misuse, defined as using drugs without a prescription at least once in the past 30 days; (6) ability to read and speak English; and (7) ability to provide written informed consent. Exclusion: (1) significant risk of dangerous alcohol withdrawal or expression of concern by the participant, research project leader, and/or health-care provider about dangerous withdrawal; (2) Diagnostic and Statistical Manual, fourth edition diagnosis of drug dependence; (3) significant risk of dangerous drug withdrawal and/or selfreported or medically documented severe withdrawal from drugs in the six months before study entry; (4) any medical or psychiatric condition, such as organic brain disorder, dementia or psychotic disorder, that the research project leader determines would compromise safe study participation; and (5) receiving drugs under the direction of a physician for pain management or another medical condition for which drug abstinence is contraindicated	Alcohol, illicit substances; contingency management	Yes
Dickerson etal. (2012)Drum-Assisted Recovery Therapy for Native Americans (DARTNA)	USA; West; urban	Qualitative study - Focus Group; Aim: Develop DARTNA protocol to obtain pretest and effectiveness study. Bias rating: 5	8; Inclusion: 1. Meeting Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR) criteria for current or past alcohol or drug abuse or dependence, self-identifying as having at least one-quarter AI/AN heritage, being at least 18 years old, and reporting no psychiatric conditions that would preclude focus group participation; 2. providers included certified alcohol and drug counselors, social workers, counselors, psychologists or physicians with experience providing substance abuse services to AIs/ANs.Exclusion: None listed	Drugs and alcohol; Alcoholics Anonymous	Yes
Edwards (2003) Friendship House	USA; West; urban	Qualitative study - in-depth interview Aim: to understand and document the experience of substance abuse recovery from the perspective of the Native Americans in treatment at the Friendship House in San Francisco Bias rating: 5	12; Inclusion: Successfully completed the 90-day residential substance abuse treatment program. Exclusion: None listed	Multiple; residential	Yes
Gone and Calf Looking (2015) Pikuni Blackfeet Indian Culture Camp	USA; West; rural	NRE - Pilot study; Aim: Explore pilot offering of the culture camp primarily served as a demonstration of “proof of concept” for alternative Indigenous intervention. Bias rating: 4	4; Inclusion: Volunteered to participate. Exclusion: None listed	Multiple; residential	Yes
Gray etal. (2010) CBSBfor AI Women in the Southwest	USA; Southwest; tribal	RCT; Outcomes: frequency/quantity of alcohol consumption, confidence in ability to resist alcohol, depression, self-esteem Bias Rating: 3	268; Inclusion: Any woman ages 18 to 50 who were tribal members. Exclusion: None listed	Multiple; Culturally based health promotion intervention	Yes
Hanson and Pourier (2015)Oglala Sioux CHOICES	USA; Midwest; tribal	NRE - Feasibility study; Aim: Engagement, or the positive group atmosphere; conflict or any tension felt in the group; and avoiding, or perceived avoidance of personal responsibility or group work by the other members, fidelity Bias rating: 2	230; Inclusion: Had more days of drug use than alcohol use within the last 90 days, experienced an alcohol detoxification-associated seizure or loss of consciousness within the last 12 months, or had a medical or psychiatric illness that required hospitalization.Exclusion: None listed	Alcohol; motivational interviewing with feedback	Yes
Kanate etal. (2015) North Caribou Lake First Nation outpatient program	Canada; Ontario; tribal	NRE - Program evaluation; Outcomes: community-wide measures of wellness, number of criminal charges, addiction related medical evacuations, child protection agency cases, school attendance and attendance at community events. Bias rating: 3	140; Inclusion: All patients met the Diagnostic and Statistical Manual of Mental Disorders, 4th edition, criteria for substance dependence. Exclusion: None listed	Opioids; intensive outpatient	Yes
Kelley etal. (2017) Transitional Recovery and Culture Program (TRAC)	USA; West; tribal and urban	NRE - Retrospective review; Outcomes: government performance and results act (GPRA), alcohol use, binge drinking, illegal drug use, combined alcohol and drug use, depression, anxiety, suicide attempts, psychological and emotional impacts, voluntary selfhelp groups and social support. Bias rating: 4	224; no inclusion/exclusion criteria listed	Drugs and alcohol; peer recovery support	Yes
Kelley etal. (2018) Substance use prevention program for youth in the Rocky Mountains	USA; West; rural	NRE - Program evaluation; Outcomes: Lower substance use among American Indian youth, increase community readiness to support prevention, increase in the number of community members reach through culturally based prevention. Bias rating: 4	369; Inclusion: youth participating in at least one prevention activity, ages 12–25. Exclusion: None listed	Multiple; residential	Yes
Naquin etal. (2006) Therapeutic communities of care	USZ;Alaska;rural	Mixed methods - focus groups and quantitative; Outcomes: resident engagement with treatment process, advancement through treatment levels and subsequent social reintegration, continued sobriety. Bias rating: 4	187; no inclusion/exclusion criteria listed	Multiple;residential	Yes
Pearson etal. (2019) CPT in AI/AN women in the Pacific Northwest	USA; West; tribal and urban	RCT; Outcomes: PTSD symptoms, alcohol-related problems, alcohol use frequency, substance use disorder, total high-risk sexual behaviors, percentage of noncondom-protected vaginal/anal intercourse acts Bias rating: 5	73; Inclusion: (a) 0Being an AI/AN female ≥ 18years of age; (b) at least two days of heavy drinking in the past year or any use of illicit substances in the last three months; (c) willingness to abstain from substance use during therapy; (d) a minimum of subthreshold PTSD symptoms defined as meeting Criterion A and at least two of the B-E DSM- IVsymptom criteria; and (e) selfreported sexual activity in the past 12-months. Exclusion: (a) psychiatric medication changes or dose changes in the past two months; (b) presence of a psychotic disorder; (c) past 30-day suicide attempt, suicidal ideation with intent or plan or selfharm, (d) past three-month opioid use to ensure that participants are stable and are unlikely to need detoxification	Drugs and alcohol;cognitive processing therapy	Yes
Ray etal. (2020) Family Wellness Warriors Initiative (FWWI)	USA;Alaska;rural	NRE - Propensity score matching; Outcomes: total health-care system visits, substance use visits, somatic visits, emergency department visits Bias rating: 4	180; Inclusion: completed an FWWI training between 10/1/2012 and 11/06/2016 and to have been empaneled in the SCF health-care system at least six months before the training date. Exclusion: None listed	Drugs and alcohol; 5-day intensive training	Yes
Running Bear etal. (2017)Alcohol detoxification among Alaska Native People	USA;Alaska;tribal	NRE - Retrospective cohort study; Outcomes: detoxification treatment, acceptance of a referral to substance abuse treatment, substance abuse treatment entry. Bias rating: 5	383; Inclusion: adult patients admitted to the detoxification unit over a two-year time period.Exclusion: Participant was neither Alaska Native, nor American Indian or admitted for a primary diagnosis of drug withdrawal	Alcohol; residential	Yes
Saylors (2003)The Women’s Circle	USA; West; urban	Mixed methods-Quantitative (pre/post) and ethnography; Outcomes: substance use (alcohol, marijuana, nonprescription methadone, hallucinogens, uppers, downers, inhalants), violence, PTSD, depression Study Quality: N/A*	742; no inclusion/exclusion criteria listed	Multiple; residential	Yes
Tonigan etal. (2020) Culturally adapted AA	USA;Southwest;urban	NRE - Pre/post design; Outcomes:12-step participation questionnaire, health-care utilization, acculturation, enculturation, adverse alcohol-related consequences, alcohol dependency Bias rating: 5	61; Inclusion: Study inclusion criteria included (1) attended at least one AA meeting in the prior three months, (2) consumed alcohol in the prior three months, (3) selfidentified as an urban AI adult, (4) sixth-grade English reading level, (4) ability to provide documented informed consent, (5) age 18 years and older, and (6) total Alcohol Dependence score (ADS) ≥ 8. Exclusion: Inability to perform informed consent and pending legal convictions	Alcohol; Alcoholics Anonymous	No
van der Woerd etal. (2010)Namgis Treatment Center (NTC) Program	Canada;British Columbia;tribal	NRE - Program evaluation; Outcomes: client ratings of program sessions and staff, support for client after treatment, abstinence and relapse, clientrated well-being Bias Rating: 2	218; Inclusion: Clients who had been out of treatment at NTC for 3–37 months. Exclusion: None listed	Drugs and alcohol; Contingency management	Yes
Venner etal. (2021) MICRA for American Indians in a Southwest Tribe	USA;Southwest;tribal	RCT; Outcomes: percent days abstinent at 12 month follow-up Bias rating: 4	79; Inclusion: 18 years or older, an enrolled tribal member, diagnosed with a DSM-IV-TR SUD, and able to speak English fluently. Exclusion: None listed	Drugs and alcohol; intensive outpatient	Yes
Wright etal. (2011) HSOC	USA; West; urban	NRE - Program evaluation pre/post; Outcomes: substance in past 30 days, stress resulting from substance use in past 30 days, arrest/committed crime past 30 days, psychological symptoms, employment status, enrollment in school/training program. Bias Rating: 4	490; Inclusion: Aged 18 years or older, able to provide consent, able to participate in treatment, began treatment within the specified period, and completed both baseline and six-month follow-up interviews. Exclusion: None listed	Multiple kinds;residential	Yes

Notes:

^aWest (WA, OR, CA, ID, NV, MT, WY, UT, CO); Southwest (AZ, NM, TX, OK); Midwest (ND, SE, NE, KS, MN, IA, MO, WI, IL, IN, MI, OH); Pacific Northwest (WA, OR).

^bNRE = nonrandomized experiment.

^cRCT = randomized control trial.

^*Quality assessment could not be performed because there was no clearly defined research question (Hong et al., 2018); CBSB = cognitive-behavioral skills building; CPT = cognitive processing therapy; PTSD = post traumatic stress disorder; SCF = Southcentral Foundation; MICRA = motivational interviewing and the community reinforcement approach

Results

A total of 1,040 records were yielded from the PsycINFO, CINAHL, MEDLINE and PubMed databases. There were 1,033 merged studies when accounting for multiple data sets coming from the same study. Once we removed 393 duplicates, we screened the remaining 640 records for relevance using their titles and abstracts. During the abstract review, we excluded 534 papers that lacked clear relevance to the inclusion and exclusion criteria, leaving 106 full-text articles for review. We added two articles from the hand search and expert review, which brought the total full-text review articles to 108. Upon close full-text review of 108 articles, we excluded 90 studies that did not meet the inclusion and exclusion criteria. Article exclusion reasons are as follows: 55 had no intervention, 25 focused only a youth population, four had wrong outcomes, three were not culturally tailored, two did not include an Indigenous population and one study was outside of the USA or Canada which was out of the scope of this study. No articles appeared to meet inclusion criteria that were later excluded. We identified a total of 18 studies published from January 1, 2000, through April 19, 2021, that fit the exclusion criteria, from which we extracted data. See Figure 1 for the PRISMA diagram of the review process.

Figure 1.

Flow chart of the systematic review of culturally tailored interventions since 2000

Geographic characteristics

Of the 18 peer-reviewed interventions, the vast majority were located in the USA (n = 16; 88.89%). Eight (44.44%) interventions were located in the West region of the USA (Dickerson et al., 2012; Edwards, 2003; Gone and Calf Looking, 2015; Kelley et al., 2017, 2018; Pearson et al., 2019; Saylors, 2003; Wright et al., 2011), three (16.67%) were located in the Alaska (Naquin et al., 2006; Ray et al., 2020; Running Bear et al., 2014) and three (16.67%) were located in the Southwest (Gray et al., 2010; Tonigan et al., 2020; Venner et al., 2021). This means that of the 16 total interventions throughout the USA, only two (11.11%) were located on the Eastern side of the country – specifically in the Midwest (Burduli et al., 2018; Hanson and Pourier, 2015). Given the large presence that Indigenous groups have on the East Coast, this distribution is notably skewed.

Two (11.11%) of the 18 interventions were located in Canada. Of the two interventions, one was located in the province of British Columbia (van der Woerd et al., 2010) and one was located in Ontario (Kanate et al., 2015). The locations of these interventions take place in areas of Canada with the highest overdose fatalities (Government of Canada, 2021). Similar to the USA, there are glaring gaps in territories that have large Indigenous populations.

The majority (n = 11; 61.11%) of the interventions took place in strictly rural or tribal settings (Burduli et al., 2018; Gone and Calf Looking, 2015; Gray et al., 2010; Hanson and Pourier, 2015; Kanate et al., 2015; Kelley et al., 2018; Naquin et al., 2006; Ray et al., 2020; Running Bear et al., 2017; van der Woerd et al., 2010; Venner et al., 2021). Fewer (n = 5; 27.78%) took place strictly in urban contexts (Dickerson et al., 2012; Edwards, 2003; Saylors, 2003; Tonigan et al., 2020; Wright et al., 2011), and two (11.11%) interventions took place in both urban and rural contexts (Kelley et al., 2017; Pearson et al., 2019). Notably, all but one intervention that included urban contexts (n = 6) were located in the West in the USA (Dickerson et al., 2012; Edwards, 2003; Pearson et al., 2019; Saylors, 2003; Tonigan et al., 2020; Wright et al., 2011). Given that many Indigenous adults live in urban settings across the USA (Yuan et al., 2014) and Canada (Government of Canada, S. C., 2019), it appears that urban Indigenous populations are underserved relative to their counterparts located on reservations, villages or in surrounding rural areas.

Serviced population

The sample size of the studies ranged from 4 to 742, with the average number of participants being approximately 213 individuals. Though the majority (n = 13; 72.22%) of interventions serviced male and female participants (Burduli et al., 2018; Dickerson et al., 2012; Edwards, 2003; Kanate et al., 2015; Kelley et al., 2017, 2018; Naquin et al., 2006; Ray et al., 2020; Running Bear et al., 2017; Tonigan et al., 2020; van der Woerd et al., 2010; Venner et al., 2021; Wright et al., 2011), a few (n = 4; 22.22%) served women exclusively (Gray et al., 2010; Hanson and Pourier, 2015; Pearson et al., 2019; Saylors, 2003) and one (5.55%) serviced only male participants (Gone and Calf Looking, 2015). There was no indication that any intervention was gender-specific or serviced two-spirit or nonbinary populations. All programs serviced adults (as expected based on the exclusion criteria); however, only one study (5.55%) included youth (i.e. individuals under the age of 18) in addition to adults (Kelley et al., 2018).

Study type and outcomes measured

Study types fell within one of four categories, including nonrandomized experimental studies, randomized control studies, mixed method studies and qualitative studies, across the 19 peer-reviewed works. Nonrandomized experimental studies (n = 9; 52.94%) were the most common study designs represented in this systematic review. This included a feasibility study (n = 1; 5.55%) (Hanson and Pourier, 2015); program evaluations (n = 4; 22.22%) (Kanate et al., 2015; Kelley et al., 2018; van der Woerd et al., 2010; Wright et al., 2011); retrospective cohort studies or reviews (n = 2; 11.11%) (Kelley et al., 2017; Running Bear et al., 2017); and one each of a propensity score matched design (n = 1; 5.55%) (Ray et al., 2020) and a pre-/poststudy (n = 1; 5.55%) (Tonigan et al., 2020).

RCTs were the next most common type (n = 4; 22.22%) of study design (Burduli et al., 2018; Gray et al., 2010; Pearson et al., 2019; Venner et al., 2021). The two (11.11%) articles that used mixed methods study designs combined quantitative outcome measurement with focus groups (Naquin et al., 2006) and ethnography (Saylors, 2003). Finally, three (16.67%) articles used strictly qualitative methods including in-depth interviews (Edwards, 2003; Gone and Calf Looking, 2015) and focus groups (Saylors, 2003).

In regard to the measured outcomes of each intervention, all but four (77.78%) of the studies explicitly reported substance use outcomes. It may be important to note that two of those articles were qualitative in nature (Dickerson et al., 2012; Edwards, 2003) and sought to understand the experience of participants and two articles were pilot and feasibility studies (Gone and Calf Looking, 2015; Hanson and Pourier, 2015). Fewer articles (n = 5; 22.22%) measured mental health outcomes such as depression, anxiety and self-esteem (Gray et al., 2010; Kelley et al., 2017; Naquin et al., 2006; Pearson et al., 2019; Wright et al., 2011). Seven (38.89%) of the 18 studies measured spiritual or cultural outcomes of interventions, including community engagement, spirituality and connectedness to culture (Dickerson et al., 2012; Edwards, 2003; Gone and Calf Looking, 2015; Gray et al., 2010; Hanson and Pourier, 2015; Kelley et al., 2018; Saylors, 2003).

In sum, experimental studies without randomized treatment were the most common mode of intervention dissemination, followed by RCTs. Though RCT serves as the “gold standard” for evaluation, the ethics of treatment distribution complicate its use in Indigenous communities who have been excessively and inappropriately researched (Bessarab and Ng’andu, 2010). The studies presented primarily measured outcomes related to substance use, but rarely reported effects on spiritual dimensions of health despite the nature of the interventions. This result reinforces the need for researchers to acknowledge the contributions of culturally tailored treatment that fall outside of the bounds of Western conceptions of wellness.

Intervention setting and substances addressed

Interventions were most often conducted in residential treatment settings (n = 7; 38.89%) (Edwards, 2003; Gone and Calf Looking, 2015; Kelley et al., 2018; Naquin et al., 2006; Running Bear et al., 2017; Saylors, 2003; Wright et al., 2011). Two studies (11.11%) were conducted in intensive outpatient settings (Kanate et al., 2015; Venner et al., 2021), and two studies (11.11%) adapted Alcoholics Anonymous/Twelve Steps frameworks for Indigenous populations (Dickerson et al., 2012; Tonigan et al., 2020). Two studies (11.11%) applied cognitive behavioral therapies or motivational interviewing within their interventions (Hanson and Pourier, 2015; Pearson et al., 2019), and one study (5.55%) used contingency management techniques in their interventions (Burduli et al., 2018). Most interventions (n = 13; 72.22%) addressed the use of drugs and alcohol, with three (16.67%) interventions intended to specifically reduce only alcohol use (Hanson and Pourier, 2015; Running Bear et al., 2017; Tonigan et al., 2020). Only one intervention (5.55%) was specifically tailored to address opioid use among Indigenous people (Kanate et al., 2015). In sum, most interventions serviced both drug and alcohol use disorders, with a few specific to either alcohol or opioid use.

Cultural considerations

Though all 18 interventions were culturally tailored to Indigenous populations, all but three (83.33%) adhered to the principles of Community-Based Participatory Research (CBPR) to design the treatment (Burduli et al., 2018; Edwards, 2003; Tonigan et al., 2020). Integration of cultural considerations occurred through meetings with tribal elders or formal boards and councils made up of tribal representatives. Eleven (61.11%) of the interventions integrated cultural practices into the interventions, which included drum circles, sweat lodges, ceremony and developing traditional skills (Dickerson et al., 2012; Edwards, 2003; Gone and Calf Looking, 2015; Gray et al., 2010; Hanson and Pourier, 2015; Kanate et al., 2015; Kelley et al., 2017, 2018; Pearson et al., 2019; Saylors, 2003; Wright et al., 2011).

A large proportion of these interventions followed the considerations of CBPR, and many consulted with tribal elders, Indigenous community boards or relevant authorities to inform best practices in implementation. This level of integration with the surrounding community is critical to the success of treatment as it allows for increased cultural specificity, which may lead to better health and wellness outcomes (Wright et al., 2011) and increased trust between practitioners and community members (West et al., 2022).

Quality appraisal

Of the nine nonrandomized studies, three (33.33%) were rated a bias rating of 5 (low bias) (Ray et al., 2020; Running Bear et al., 2017; Tonigan et al., 2020). Three (33.33%) were rated a bias rating of 4 (adequate bias) (Kelley et al., 2017, 2018; Wright et al., 2011). One study (11.11%) had a bias rating of 3 (moderate bias) (Kanate et al., 2015). Finally, two studies (22.22%) were rated a bias rating of 2 (serious bias). It may be important to note that one article was a feasibility study (Hanson and Pourier, 2015) and one is a program evaluation (Wright et al., 2011), as these types of study designs are less appropriate for this particular quality assessment.

Of the four RCT studies, two (50%) were rated a bias rating of 5 (Burduli et al., 2018; Pearson et al., 2019), one (25%) was given a bias rating of 4 (Venner et al., 2021) and one (25%) was received a bias rating of 3 (Gray et al., 2010).

Of the two mixed methods articles, one received a bias rating of 4 (Naquin et al., 2006). We could not assess one mixed methods study because a research question was not clearly defined (Saylors, 2003). Per the guidance of the MMAT, quality should not be assessed for articles without clearly defined research questions. Despite the lack of a clearly defined research question, the article is included in this systematic review because it meets the goal of more clearly understanding what culturally tailored interventions are available. Of the three qualitative studies, two (66.67%) were rated a bias rating of 5 (Dickerson et al., 2012; Edwards, 2003) and one (33.33%) received a bias rating of 4 (Gone and Calf Looking, 2015).

In total, seven studies (38.89%) were rated a bias rating of 5 (Burduli et al., 2018; Dickerson et al., 2012; Edwards, 2003; Pearson et al., 2019; Ray et al., 2020; Running Bear et al., 2017; Tonigan et al., 2020); six studies (33.33%) were rated a bias rating of 4 (Gone and Calf Looking, 2015; Kelley et al., 2017, 2018; Venner et al., 2021; Wright et al., 2011); two studies (11.11%) were rated a bias rating of 3 (moderate bias) (Gray et al., 2010; Kanate et al., 2015); and two studies (11.11%) were rated a bias rating of 2 (serious bias) (Hanson and Pourier, 2015; van der Woerd et al., 2010). One study (Saylors, 2003) could not be assessed for quality because it lacked a clear research question.

Overall, the coauthors chose to include a variety of study designs to gain the fullest understanding of the current state of the literature. These studies not only reported quantitative findings, but contextualized findings qualitatively. In consideration of the quality appraisal of the studies, the coauthors did not exclude any articles because of high bias. Based on the objectives of the review, it was expected that none of the studies would be generalizable to Indigenous populations at large, and the wider US population. Overall, the RCT studies included had low bias ratings based on the MMAT quality appraisal. The qualitative studies were largely rated low or adequate bias. We still included the three qualitative studies because of their rich description of the setting, intervention and the use of participants’ voices. This type of description is important when discussing, developing and replicating culturally tailored and community-based interventions. Nonrandomized studies displayed the most variability in terms of bias, with quality ratings ranging from 2 (serious) to 5 (low). All nonrandomized studies were included because of the valuable insights the nature of some of the study designs program evaluations and pilot/feasibility studies provide to a systematically excluded and underserved population. Overall, this field of study would benefit from more mixed methods studies, which would provide a holistic understanding of the utility of substance use interventions.

Discussion

The purpose of this work was to document the current state of culturally tailored treatment of SUD in Indigenous communities through peer-reviewed studies. All studies identified were in North America, involved individuals above the age of 18 and were culturally tailored. Our literature search identified 18 studies of culturally tailored treatment for Indigenous groups that varied in their implementation, serviced population, geographic location and delivery of services.

Overall, this work highlighted several needs and considerations relevant to the treatment of SUD in Indigenous populations. First, there exists a profound lack of access to culturally relevant treatment resources for Indigenous communities outside of the Southwestern USA, and in urban settings. The development of policy and programming for SUD that disproportionately services white populations (Hansen and Netherland, 2016) has caused disparities in effective treatment protocols for nonwhite opioid users, who benefit most from treatment that is community-centered and culturally responsive (Lau, 2006). Because individuals who belong to racial and ethnic groups that have been minoritized have experienced increasing rates of SUD and death (for example, see Amiri et al., 2022), there is a need to invest resources in interventions that are developed and rooted in these communities. For Indigenous groups specifically, culturally responsive treatment is not just preferable – it is necessary to address experiences of historical trauma, decolonize traditional treatment regimens and allow tribal sovereignty in addressing the needs of their specific communities. This review highlights the scope of this need across Indigenous groups in the USA and Canada, though this need is ubiquitous for Indigenous populations globally.

Second, this review highlighted that few interventions were specific to either alcohol or opioid use. Given the dearth of programs available to Indigenous people, the diversity of treatment options to address an individual’s drug of choice is beneficial. However, past studies have suggested that SUD treatment is most effective when it is specialized around specific substances, especially across alcohol and drug use programs (Andersson et al., 2021). As such, increasing access to programs that specialize in the treatment of specific substances may improve the outcomes.

Third, this review illuminates important gaps in treatment provision as it pertains to Indigenous groups. The low number of peer-reviewed studies documenting the nature and results of culturally integrated interventions illustrates a pressing need for program expansion and evaluation. Further, the results illustrate that the majority of these programs were located in the American Southwest, with only two located across all of Canada. In keeping with the statistics regarding the state of alcohol- and substance-related fatalities, Indigenous groups located in Canada are expressly in need of resources to address the prevalence of SUD in recent years and could benefit from increasing program accessibility. In addition, most treatment sites were in tribal or rural areas, which is problematic considering most (approximately 71%) Indigenous people live in urban areas in the USA (Yuan et al., 2014). Statistics show that only one in four urban Indigenous people live within an Indian Health Services service area (Roubideaux, 2008; Whitesell et al., 2012), illustrating the extent to which this gap is felt by individuals living off the reservation.

Further, this review identifies the gaps in the way culturally relevant treatment programs are studied and assessed. Most studies used nonrandomized designs, creating some concern about effectively isolating treatment effects. Further, the majority of the studies contained in this review were quantitative in nature, and few used qualitative data or mixed methods. Qualitative data and analyses are especially necessary to center the voices of Indigenous people and to accurately capture the nature and complexity of SUD in Indigenous communities. Further, the majority of the outcomes measured in these studies are related to substance use. Future research should seek to integrate holistic conceptions of health into the assessment of treatment outcomes, including (but not limited to) feelings connection to culture, emotional and spiritual health and engagement in cultural practices.

Nevertheless, there were many encouraging elements related to the research of culturally relevant treatment in Indigenous communities that were highlighted in this research. The vast majority (83.33%) of studies identified in this review adhered to the principles of CBPR, meaning that Indigenous communities acted as equal partners in the research. The interventions assessed in these studies contained a variety of elements meant to connect individuals with SUD to their culture, including drum circles, sweat lodges, ceremony and developing traditional skills. Past research has called for tribal sovereignty in the development of effective substance abuse treatment to effectively imbue cultural perspectives and themes to best serve their own communities (Richer and Roddy, 2022). The 18 programs contained in this review serve as meaningful examples in culturally tailored treatment as they effectively center culture in treatment, highlight the importance of community and improve accessibility for Indigenous groups.

Ultimately, while this work focused on the characteristics of culturally tailored treatment in the USA and Canada, the findings of this research have applications to Indigenous populations more generally. Indigenous groups across the globe have experienced the impacts of colonization, historical trauma and racism (for example, in Guatemala and New Zealand, see Beristain et al., 2000; Pihama et al., 2014), and these experiences largely set the context for substance use (Brave Heart, 2003). Because cultural tailoring of treatment has been found to increase treatment engagement and retention across different minority populations (Hall, 2001) and leads to sustainable and long-lasting improvements in behavioral health (Lau, 2006), the expansion of access to culturally tailored treatment should aid Indigenous communities outside of the USA and Canada. In light of the benefits these programs produce, systematic factors (e.g. poor funding) act to limit resources for research and treatment for Indigenous populations globally (Valeggia and Snodgrass, 2015). The methodological examples from this review may inform work and research agendas for Indigenous populations located in other geographic areas.

Strengths and limitations

There were several limitations of note that may have impacted the conclusions drawn from this study. First and foremost, this work only included peer-reviewed studies on culturally tailored treatment programs. This meant that community-driven and Indigenous-led interventions that did not have a documented study component or assessed results were omitted. We recognize that this method inherently reinforces the belief and hierarchy of Western (academic) versus Indigenous (community) systems of knowledge. Despite this limitation, we continued with the study protocol as designed, as it allowed us to assess and compare the study quality and bias across peer-reviewed works explicitly. Future research should involve the exploration of nonacademic literature in the documentation of culturally tailored interventions, which can serve as a source of comparison for the current study.

A second limitation is the explicit focus on North American Indigenous populations rather than Indigenous communities globally. We are aware of the ways in which dividing our sample based on political nation states rather than Indigenous identity represents a larger issue related to the geopolitics of Indigeneity and colonialism. However, we are hopeful that focusing on these specific, well-resourced countries may inspire increased government investment in the care of Indigenous populations.

Conclusion

This work illuminated the state of research on, as well as the characteristics of, culturally tailored drug and alcohol treatment for Indigenous groups. Based on the findings of this systematic review, increased resources must be devoted to addressing the culturally specific needs of Indigenous groups with SUD, especially those located in urban areas outside of the American Southwest. Scholarship in this area can be bolstered by the inclusion of mixed methods evaluations of culturally tailored interventions, and the consideration of Indigenous conceptions of health in assessing the efficacy of treatment. The results of this research can be used to educate researchers, practitioners and policymakers on the needs of Indigenous communities, and ways to effectively measure intervention outcomes as they continue to become more prevalent and accessible to Indigenous communities.

Biographies

Ariel M.S. Richer is a former NIH T32 Pre-Doctoral Fellow at Columbia University School of Social Work working within the Social Intervention Group. She is the recipient of an F31 NRSA training and research grant through the National Institute of Minority Health Disparities. Her focus is on intimate partner violence and access to relevant services for black and Indigenous women who use drugs and are involved in the criminal-legal system. All of her work is founded in principles of community-based participatory research, as she works collaboratively with Indigenous and Native communities. Ariel is the cofounder and chief executive officer of Urban Indigenous Collective. She is a descendant of the Indigenous people of Trinidad and Tobago and Venezuela. She earned her Master of Science in Social Work, Social Enterprise Administration from Columbia University School of Social work in 2015 and is a Licensed Master Social Worker.

Ariel L. Roddy is an Assistant Professor in the Department of Sociology at the University of Utah. Her research focuses on the economic marginalization of and barriers to reentry for women of color in the justice system through an Indigenous feminist lens.