Adverse Childhood Events in American Indian/Alaska Native Populations

Abstract

Introduction:

Adverse childhood experiences are linked to deleterious outcomes in adulthood. Certain populations have been shown to be more vulnerable to adversity in childhood than others. Despite these findings, research in this area lacks an empirical investigation that examines adverse childhood experiences among American Indian and Alaska Native populations using large, nationally representative data. As such, the authors have compiled what they believe is the largest empirical investigation of adverse childhood experiences among American Indian and Alaska Native individuals to date.

Methods:

Data were collected from the Behavioral Risk Factor Surveillance System from 34 states (2009–2017), whereby all individuals self-report as American Indian and Alaska Native (N=3,894). Adverse childhood experience scores were calculated and further stratified by sex, age, household income, education, employment status, sexual orientation, Census region, and state. In addition, frequencies and prevalence of each adverse childhood experience domain (stratified by the same categories) were calculated. Analysis was conducted in 2019.

Results:

The average adverse childhood experience score among American Indians and Alaska Natives was 2.32, higher than those of individuals identifying as White (1.53), Black (1.66), and Hispanic (1.63). Female participants had a higher average adverse childhood experience score than male participants (2.52 vs 2.12). Generally, younger individuals and those with lower incomes reported higher adverse childhood experience scores, whereas those with higher educational attainment reported lower scores.

Conclusions:

Compared with the few studies among American Indian and Alaska Native populations that have used either smaller samples or nontraditional adverse childhood experience data (i.e., asking parents about their children’s experiences), these results present overall higher adverse childhood experience averages than previously published studies. Nevertheless, aligning with other research on adverse childhood experiences, female individuals, younger adults, and sexual minorities reported higher adverse childhood experiences scores than other categories in their respective demographics.

INTRODUCTION

Cognition, mental and physical well-being, and lifelong health can be negatively affected by childhood events.^1–3 Adverse childhood experiences (ACEs) are a measure of adversity in childhood assessed by 8 domains, including emotional abuse, physical abuse, sexual abuse, intimate partner violence, household substance use, household mental illness, parental separation/divorce, and household member incarceration.^2,4 Individuals recounting a high number of ACEs are at a higher risk for deleterious health markers such as heart and pulmonary diseases, lung cancer, metabolic issues, inflammation, and liver diseases.^3,5–8 In addition, psychological well-being is impacted by ACEs because empirical investigations show statistically significant associations with depression, anxiety, mood disorders, and suicide.^1,9–11 Indeed, a recently released report from the Centers for Disease Control and Prevention (CDC) found that ACEs were linked to a higher risk of dying from 5 of the top 10 leading causes of death.¹² Nevertheless, demographic diversity within these studies is limited, with researchers calling for updated prevalence rates regarding ACEs by demographic and geographic characteristics and region.¹³ Although efforts have been made to update ACE data by demographics, smaller racial/ethnic minority populations are often overlooked or grouped into other or all other races/ethnicities category.¹³

Investigations regarding ACE scores largely neglect American Indian and Alaska Native (AI/AN) individuals when stratified by race.^13,14 AI/AN individuals may be at an increased risk of ACEs owing to many historical and contemporary social, political, and economic factors. ACE questions are specifically an individualistic way to measure ACEs; however, historical trauma (collective trauma experienced by a group of people who share an identity, affiliation, or circumstance across generations¹⁵) and cultural trauma (when members of a group collectively are subjected to a horrendous event(s) that leads to permanent trauma negatively impacting future identity and culture in fundamental ways¹⁶) in AI/AN populations contribute to disparities.¹⁷ ACEs in AI/AN populations may be associated with intergenerational experiences and trauma such as the genocide of AI/AN individuals, abuse from the boarding school system, interruption of traditional practices, and centuries of colonialism.¹⁸

Most studies that investigate ACEs in AI/AN populations used smaller sample sizes and were confined by geographic location.^19–21 One study reported 2011 data from a sample of 1,660 AI/AN individuals in an effort to assess ACEs.²² Although this was an important foundational study, these researchers assessed ACE scores through parents’ knowledge of their children’s experiences. Parents’ recount of their children’s ACE scores might introduce proxy bias of nonself reported behavior. ACE questions include parents’ self-reported responses to adverse conditions in the household (e.g., Seen parents hit, kick, slap, punch, or beat each other up?); thus, a parent’s recount of a child’s ACEs may be subject to bias particularly because studies show that parents might be more likely to downplay adverse events within their household, such as interpersonal violence.^23–25 Although ACE scores are cumulative through the age of 18 years, the ACE scores measured in this study were derived from children aged 0–17 years.²² Thus, ACE scores among younger children in their study may be underestimated because they had not experienced the full expanse of childhood and adolescence.

In 2009, CDC first gave the option for states to add ACEs questions as a part of their annual Behavioral Risk Factor Surveillance Survey (BRFSS)—a national survey of health behaviors/indicators.²⁶ From BRFSS, Merrick and colleagues¹³ compiled the most comprehensive ACEs data to date, acquiring ACEs data from >140,000 individuals in 25 states from 2015 to 2017. Findings showed that the prevalence of ACEs varied by race/ethnicity, with the populations experiencing the highest proportions of ≥4 ACEs as AI/AN (28.3%), followed by Black (17.7%), Hispanic (15.8%), White (15.0%), and Asian (8.6%) populations. A follow-up study by Merrick et al.²⁷ did report on AI/AN populations using a sample of 838; however, this was only a summary of ACEs frequencies (i.e., 0, 1, 2–3, or ≥4 ACEs) without addressing specific ACE domains or total ACE scores.

Many demographic and regional-level characteristics (e.g., states with higher levels of poverty or lower overall SES) may correlate with ACEs in AI/AN communities, but no reports investigated these factors as they relate to ACE prevalence. Given the very high ACE prevalence in AI/AN populations, the purpose of this paper is to provide updated socioeconomic and regional data of specific ACE domains as reported by AI/AN respondents. As such, the research questions are 2-fold. First, what are the overall frequencies of ACEs for AI/AN individuals delineated by demographic variables? Next, how do AI/AN respondents compare with normative data from other racial/ethnic groups on reports of ACEs?

METHODS

Study Sample

Data were acquired from CDC’s BRFSS from 2009 to 2017—an annual national survey that collects data through both cellular and landline telephone interviews among adults. The BRFSS is divided into 3 modules: (1) core modules, which are sets of survey questions consistently administered to all states and territories developed by CDC; (2) optional modules, which consist of CDC developed questions that states have the option to include in their questionnaire; and (3) state-added questions, which include customizable questions developed by each state coordinator or community stakeholders in partnership with each state. Core modules are publicly available, whereas optional modules and state-added questions must be individually requested from each state. ACEs module is an optional module. As of 2018, a total of 41 states had collected ACE data. Several states collected data in 2018 for the first time; however, these data are typically not available until at least 2 years after collection.

Measures

The ACEs module consists of 11 questions collapsed into 8 domains exploring the association of adverse experiences before age 18 years on health with well-being in adulthood.²⁸ The responses for each domain were dichotomized (yes/no) and summed, resulting in a range of 0–8 (higher scores indicating higher exposure to adversity). Ford et al.⁴ offer an in-depth description of the BRFSSCDC–ACEs module, a factorial structure, and calculated ACE scores.

In total, 38 states collected ACEs data from 2011 to 2017 (Washington, District of Columbia, and other U.S. territories did not collect ACEs data). Of these, 3 states declined to share data for various reasons (e.g., stopped giving data because of lack of personnel available to appropriate data, data collection is privately funded and not shared with the public, committee declined to have data included in the study owing to small percentages of data on AI/AN individuals who could potentially have identifiable characteristics, or only collected data on some ACE domains but not on others), with 1 additional state being unresponsive, resulting in a final state count of 34 and a final sample size of 3,894 self-identified AI/ANs.

Following similar procedures as Merrick et al.,¹³ states that included the optional ACE modules were contacted to establish data use agreements. After data were acquired, the ACEs data were merged from each state (along with their corresponding demographic variables) from 2009 to 2017 (Table 1). The BRFSS includes survey weights (also included in the acquired data) and were used to adjust the data to conform to population parameters provided by CDC (weights based on age, race/ethnicity, sex, and geographic region).

Table 1.

Number of American Indian/Alaska Native BRFSS Participants by Year and State, 2009‒2017 (N=3,894)

Year/State	n
2009
New Mexico	12
2011
Maine	37
Minnesota	71
Montana	541
Nebraska	21
Vermont	48
Washington	130
2014
Florida	16
North Carolina	41
West Virginia	34
2015
Alaska	521
California	56
Kentucky	58
Ohio	46
Texas	86
2016
Arizona	215
Arkansas	68
Georgia	36
Hawaii	11
Louisiana	28
Michigan	40
New York	60
Oklahoma	176
Pennsylvania	30
South Carolina	69
2017
Connecticut	37
Illinois	15
Iowa	18
Nevada	57
Oregon	16
South Dakota	1,080
Tennessee	40
Virginia	73
Wisconsin	107

BRFSS, Behavioral Risk Factor Surveillance System.

This study expands on the study by Merrick and colleagues¹³ in several ways. First, data were acquired from 10 additional states compared with those by Merrick et al.¹³ (a 42% increase) (Table 1). This study also provides ACE prevalence by region and individual states. Second, this study limits data inclusion to a single year per state. Merrick and colleagues¹³ used several years of data for a single state (e.g., Iowa 2011/2012/2013), whereas the current ACE data only allow for the most recent year for each state. Using several years of data from the same state may inflate results from that region or include duplicate data (i.e., individuals could have taken the survey in multiple years and subsequently counted multiple times in the data, although this is unlikely).²⁹

Statistical Analysis

First, frequency statistics were computed for the sample. Estimated weighted frequencies were further stratified by age, annual household income, employment status, educational attainment, sexual orientation, and 9 geographic regions (classified by the U.S. Census Bureau). Next, frequencies were conducted by each of the 8 domains (further stratified by the same aforementioned sociodemographic variables) and presented with corresponding 95% CIs through bias-corrected bootstrapping (1,000 samples). For comparisons with other races/ethnicities within each ACE domain, the authors used significance testing from chi-square analyses. An ANOVA was conducted to compare mean ACE scores among races/ethnicities (both are found in Table 2). Analyses were conducted in 2019 using SPSS, version 24.

Table 2.

Comparison of White and AI/AN ACE Prevalence in the U.S., 2009‒2017 (N=3,894)

Variable	AI/AN %/M (95% CI)	White %/M (95% CI)	Black %/M (95% CI)	Hispanic %/M (95% CI)	All races/ethnicities %/M (95% CI)	AI/AN significant difference
Emotional	43.1 (42.1, 44.4)	33.4 (33.3, 33.6)	28.8 (28.5, 29.2)	33.7 (33.4, 34.0)	33.4 (33.0, 33.9)	a>b,c,d
Physical	27.2 (26.3, 28.0)	15.8 (15.7, 15.9)	12.1 (11.8, 12.3)	23.8 (23.6, 24.0)	17.5 (17.1, 17.8)	a>b,c,d
Sexual	17.6 (16.8, 18.4)	11.0 (10.9, 11.1)	11.8 (11.5, 12.0)	11.2 (11.0, 11.4)	11.3 (11.0, 11.6)	a>b,c,d
Intimate partner violence	28.5 (27.6, 29.5)	15.5 (15.4, 15.6)	21.0 (20.8, 21.2)	22.5 (22.2, 22.7)	17.7 (17.4, 18.1)	a>b,c,d
Household substance abuse	40.9 (40.0, 41.8)	26.8 (26.7, 26.9)	25.4 (25.1, 25.7)	26.6 (26.3, 26.8)	26.8 (26.4, 7.2)	a>b,c,d
Household mental illness	22.7 (21.8, 23.5)	17.9 (17.8, 18.0)	11.1 (10.9, 11.3)	11.0 (10.9, 11.2)	16.1 (15.8, 16.5)	a>b,c,d
Parental separation/divorce	41.6 (40.7, 42.6)	25.3 (25.1, 25.3)	44.0 (43.7, 44.3)	28.5 (28.2, 28.7)	28.2 (27.8, 28.6)	a>b,d
Family incarceration	17.5 (16.7, 18.2)	6.3 (6.2, 6.4)	14.1 (13.8, 14.3)	9.2 (9.1, 9.4)	8.0 (7.8, 8.3)	a>b,c,d
ACE score mean	2.32 (2.28, 2.37)	1.53 (1.52, 1.54)	1.66 (1.65, 1.67)	1.63 (1.62, 1.64)	1.56 (1.54, 1.57)	a>b,c,d

Note: Significant differences for χ2 were assessed at the p<0.001 level. AI/AN=a, White=b, Black=c, and Hispanic=d. ACE score mean differences were assessed by ANOVA.

ACE, adverse childhood experience; AI/AN, American Indian and Alaskan Native; M, mean.

RESULTS

Table 3 presents the weighted estimates across demographic variables (N=3,894). Half of the sample was female (50.5%). Generally, all age groups were represented, with the lowest percentage being of those aged 18–24 years (11.6%), and the largest group being of those aged 45–54 years (22.2%). Approximately 27.2% of the sample made >$50,000 a year and represents the largest income category, with most individuals having some college (34.4%). More than half of the sample was employed (51.3%), with the vast majority identifying as straight (92.7%). The West South Central region of the U.S. (Arkansas, Louisiana, Oklahoma, and Texas) contained the most individuals (22.4%), whereas New England had the least (1.2%).

Table 3.

Demographic Characteristics of American Indian/Alaskan Native BRFSS Respondents From 34 U.S. States, 2009‒2017 (N=3,894)

Characteristic	n	Wgt %	95% CI
Sex
Female	2,208	50.5	49.5, 51.3
Male	1,686	49.4	48.4, 50.7
Age group, years
18‒24	239	11.6	11.0, 12.4
25‒34	501	17.9	17.1, 18.8
35‒44	534	16.3	15.5, 17.0
45‒54	802	22.2	21.4, 23.0
55‒64	899	16.2	15.4, 16.9
>64	919	15.8	15.0, 16.5
Household income
<$15,000	870	21.3	20.4, 22.2
$15,000–$24,999	851	22.4	21.4, 23.4
$25,000–$34,999	433	13.1	12.4, 13.8
$35,000–$50,000	440	15.7	14.9, 16.6
>$50,000	773	27.4	26.5, 28.3
Educational attainment
Less than high school	651	20.7	19.8, 21.5
High school diploma/GED	1,260	31.8	30.9, 32.8
Some college	1,201	34.4	33.4, 35.4
College degree	765	13.1	12.5, 13.8
Employment status
Employed	1,743	51.3	50.2, 52.4
Unemployed	412	9.7	9.1, 10.3
Unable to work	598	16.7	15.9, 17.5
Other	1,091	22.3	21.5, 23.2
Sexual orientation
Straight	1,625	92.7	92.0, 93.5
Gay/lesbian	29	2.4	2.0, 2.8
Bisexual	37	4.9	4.3, 5.5
Census region
New England	122	1.2	1.0, 15
Middle Atlantic	90	6.4	5.9, 6.9
East North Central	208	9.7	9.0, 10.4
West North Central	1,206	3.7	3.3, 4.0
South Atlantic	269	18.5	17.7, 19.2
East South Central	98	7.1	6.6, 7.6
West South Central	358	22.4	21.6, 23.2
Mountain	825	16.1	15.3, 16.8
Pacific	718	15.0	14.3, 15.8

BRFSS, Behavioral Risk Factor Surveillance System; Wgt, weight.

Most individuals experienced ≥1 ACEs (72.1%). Approximately 27.9% had an ACE score of 0, followed by 20.7% (1 ACE), 13.8% (2 ACEs), 10.8% (3 ACEs), 9.3% (4 ACEs), 7.5% (5 ACEs), 5.3% (6 ACEs), 3.5% (7 ACEs), and 1.3% (all 8 ACEs; not shown in tables).

Table 4 presents ACE prevalence and mean ACE scores by demographic variables among the 8 ACE categories. Overall, the average mean ACE score was 2.32, with the single highest ACE domain prevalence being emotional abuse (43.1%), followed by parental separation/divorce (41.6%) and household substance abuse (40.9%). The mean ACE score was higher for female (2.52) than for male participants (2.12), with female AI/ANs having higher percentages of ACEs in all the domains except for incarcerated household member. Of note, female participants reported sexual abuse at more than twice the rate for male participants (26% vs 9%). With respect to age, those aged 25–34 years reported the highest mean ACE score (3.04), followed by the group aged 18–24 years (2.91). For all other age groups, ACE scores decreased with an increase in age (2.43 for age 35–44 years, 2.28 for age 45–54 years, 2.17 for age 55–64 years, and 1.18 for age ≥64 years).

Table 4.

Frequencies of ACE Types and Mean ACE Score by Sociodemographic Characteristics Among American Indian/Alaskan Native BRFSS Respondents From 34 U.S. States, 2009‒2017 (N=3,894)

	Emotional abuse	Physical abuse	Sexual abuse	Intimate partner violence	Household substance abuse	Household mental illness	Parental separation/ divorce	Incarcerated household member	ACE score
Characteristics	Wgt (95% CI)	Wgt (95% CI)	Wgt (95% CI)	Wgt (95% CI)	Wgt (95% CI)	Wgt (95% CI)	Wgt (95% CI)	Wgt (95% CI)	Mean (95% CI)
Total	43.1 (42.1, 44.4)	27.2 (26.3, 28.0)	17.6 (16.8, 18.4)	28.5 (27.6, 29.5)	40.9 (40.0, 41.8)	22.7 (21.8, 23.5)	41.6 (40.7, 42.6)	17.5 (16.7, 18.2)	2.32 (2.28, 2.36)
Sex
Male	40.5 (39.0, 42.0)	23.8 (22.5, 25.2)	9.0 (8.2, 9.8)	28.2 (26.9, 29.5)	39.6 (38.1, 41.0)	19.4 (18.3, 20.6)	38.6 (37.0, 40.1)	18.8 (17.7, 20.0)	2.12 (2.06, 2.18)
Female	45.8 (44.3, 47.3)	30.3 (28.9, 31.7)	26.2 (24.8, 27.5)	28.6 (27.3, 29.9)	42.3 (40.8, 43.9)	25.8 (24.5, 27.2)	44.6 (43.1, 46.2)	16.3 (15.2, 17.3)	2.52 (2.45, 2.59)
Age group, years
18‒24	49.1 (45.7, 52.4)	29.8 (27.1, 32.5)	11.9 (9.9, 13.9)	39.5 (36.7, 42.2)	40.5 (37.5, 43.5)	32.6 (29.8, 35.2)	60.2 (57.2, 63.2)	32.1 (29.4, 34.9)	2.91 (2.77, 3.04)
25‒34	57.4 (55.1, 59.8)	32.4 (30.1, 34.6)	18.6 (16.4, 20.6)	37.2 (35.0, 39.5)	51.6 (49.3, 54.1)	30.1 (27.9, 32.4)	55.3 (52.6, 57.5)	30.0 (27.9, 32.2)	3.04 (2.94, 3.15)
35‒44	45.6 (43.2, 48.1)	25.3 (23.0, 27.4)	21.3 (19.3, 23.2)	26.9 (24.7, 29.1)	41.9 (39.2, 44.3)	27.1 (25.0, 29.4)	43.1 (40.4, 45.8)	19.1 (17.0, 21.1)	2.43 (2.31, 2.55)
45‒54	38.2 (36.0, 40.1)	29.7 (27.8, 31.5)	18.6 (16.7, 20.7)	26.4 (24.4, 28.6)	41.8 (39.4, 44.0)	25.7 (23.8, 27.7)	40.5 (38.5, 42.4)	14.3 (12.7, 15.8)	2.28 (2.18, 2.38)
55‒64	49.5 (47.0, 52.1)	29.1 (27.1, 31.3)	18.6 (16.7, 20.7)	29.9 (27.5, 32.3)	43.6 (41.3, 46.3)	13.8 (12.1, 15.5)	29.7 (27.3, 32.1)	8.6 (7.2, 10.2)	2.17 (2.05, 2.28)
>64	20.0 (18.1, 22.0)	15.8 (14.1, 17.6)	9.8 (8.2, 11.5)	13.9 (12.3, 15.5)	24.0 (21.7, 26.2)	7.2 (5.9, 8.6)	25.2 (23.0, 27.5)	4.8 (3.8, 5.8)	1.18 (1.11, 1.26)
Household income
<$15,000	49.4 (47.0, 51.9)	28.7 (25.8, 31.9)	25.8 (24.0, 27.6)	36.5 (34.5, 38.3)	48.0 (45.1, 51.0)	26.4 (23.8, 29.0)	41.2 (37.6, 44.5)	20.7 (18.6, 23.3)	2.65 (2.55, 2.72)
$15,000–$24,999	44.2 (42.3, 46.1)	29.9 (27.1, 32.7)	17.3 (15.9, 18.8)	29.6 (26.6, 32.5)	44.0 (42.2, 45.9)	23.0 (21.2, 25.1)	46.4 (44.4, 48.3)	27.0 (23.7, 30.2)	2.54 (2.45, 2.63)
$25,000–$34,999	48.6 (46.1, 50.8)	28.3 (25.0, 32.0)	19.9 (16.7, 23.4)	31.4 (29.2, 33.4)	43.5 (39.4, 47.4)	26.5 (24.3, 28.7)	51.0 (47.2, 54.4)	21.3 (18.3, 24.3)	2.64 (2.48, 2.80)
$35,000–$50,000	33.4 (31.1, 35.4)	24.9 (23.1, 26.7)	12.2 (10.3, 14.3)	28.2 (24.7, 31.5)	34.7 (31.4, 37.8)	14.3 (12.7, 15.9)	33.4 (29.9, 37.1)	9.5 (7.6, 11.3)	1.86 (1.76, 1.96)
>$50,000	49.6 (47.8, 51.2)	29.9 (27.6, 32.4)	17.2 (15.2, 19.4)	25.6 (23.4, 28.0)	40.3 (38.6, 42.1)	24.5 (23.0, 25.9)	37.9 (36.1, 39.7)	12.4 (11.3, 13.5)	2.31 (2.20, 2.44)
Educational attainment
Less than high school	43.6 (41.2, 45.9)	25.4 (23.4, 27.5)	22.8 (21.0, 24.7)	37.6 (35.4, 39.9)	46.8 (44.4, 49.3)	26.2 (24.0, 28.4)	40.1 (37.9, 42.5)	21.8 (20.0, 23.6)	2.55 (2.43, 2.65)
High school diploma/GED	42.3 (40.6, 44.0)	29.0 (27.3, 30.8)	15.3 (14.0, 16.5)	28.0 (26.4, 29.9)	39.0 (37.3, 40.8)	18.2 (16.8, 19.7)	35.7 (34.0, 37.4)	18.0 (16.6, 19.4)	2.19 (2.11, 2.27)
Some college	44.3 (42.3, 46.3)	29.5 (27.8, 31.1)	18.9 (17.6, 20.2)	24.1 (22.6, 25.6)	43.8 (42.0, 45.5)	24.9 (23.5, 26.4)	50.1 (48.3, 51.8)	17.6 (16.3, 19.0)	2.47 (2.39, 2.55)
College graduate	45.9 (43.3, 48.3)	21.1 (18.8, 23.6)	13.2 (11.4, 15.3)	29.0 (26.4, 31.6)	31.4 (28.7, 34.1)	23.7 (21.5, 26.0)	30.2 (27.5, 32.8)	9.9 (8.2, 11.7)	2.00 (1.89, 2.13)
Employment status
Employed	42.4 (40.9, 43.8)	25.6 (24.4, 26.8)	12.0 (11.1, 12.9)	27.9 (26.6, 29.2)	40.0 (38.7, 41.5)	22.8 (21.7, 24.0)	41.8 (40.2, 43.3)	18.4 (17.0, 19.8)	2.24 (2.18, 2.31)
Unemployed	49.0 (45.8, 52.4)	36.6 (33.3, 40.0)	21.6 (18.6, 24.2)	36.1 (33.0, 39.4)	52.6 (49.5, 56.1)	30.1 (27.1, 33.2)	44.0 (40.7, 47.4)	22.1 (19.5, 24.8)	2.84 (2.67, 3.00)
Unable to work	58.0 (55.5, 60.4)	34.9 (32.5, 37.3)	37.6 (35.4, 40.2)	36.7 (34.1, 39.2)	53.8 (51.2, 56.3)	29.8 (27.6, 32.2)	57.9 (55.5, 60.5)	25.7 (23.4, 27.9)	3.23 (3.12, 3.35)
Other	32.5 (30.3, 34.9)	22.5 (20.6, 24.4)	13.9 (12.3, 15.4)	22.5 (20.7, 24.3)	30.5 (28.5, 32.5)	16.6 (15.0, 18.4)	27.3 (25.6, 29.2)	9.4 (8.2, 10.6)	1.71 (1.61, 1.81)
Sexual orientation
Straight	40.6 (39.2, 42.0)	27.4 (26.2, 28.6)	16.0 (15.0, 17.1)	25.9 (24.6, 27.2)	38.1 (36.7, 39.5)	20.2 (19.0, 21.4)	41.5 (40.1, 43.0)	17.5 (16.5, 18.6)	2.21 (2.15, 2.27)
Gay/lesbian	55.2 (45.7, 63.8)	50.8 (42.4, 59.3)	34.7 (26.3, 43.2)	54.1 (44.1, 63.1)	69.5 (61.0, 77.1)	40.7 (32.2, 50.0)	70.3 (61.9, 78.8)	34.7 (25.4, 44.1)	4.05 (3.66, 4.46)
Bisexual	65.3 (58.7, 71.5)	47.7 (41.5, 53.5)	38.6 (33.2, 44.8)	34.7 (28.4, 41.3)	58.4 (52.7, 64.6)	41.9 (35.5, 48.4)	42.6 (36.2, 49.4)	7.1 (3.7, 10.4)	3.22 (3.01, 3.44)
Census region
New England	31.8 (24.5, 40.0)	19.1 (12.7, 25.5)	17.3 (11.8, 22.7)	24.5 (17.3, 32.4)	24.3 (17.1, 33.3)	17.3 (10.9, 23.6)	25.7 (18.1, 33.3)	11.8 (7.3, 16.4)	1.69 (1.90, 2.54)
Middle Atlantic	28.6 (25.1, 32.2)	23.9 (20.4, 27.5)	5.3 (3.7, 7.2)	17.6 (14.7, 20.8)	39.7 (35.9, 43.2)	28.5 (24.8, 32.5)	47.8 (43.9, 52.0)	31.5 (27.3, 35.1)	2.22 (2.06, 2.39)
East North Central	55.5 (52.0, 59.2)	35.1 (32.0, 38.2)	21.6 (18.8, 24.4)	33.3 (30.4, 36.0)	38.6 (35.9, 41.5)	15.8 (13.6, 18.1)	61.3 (58.2, 64.4)	15.3 (13.0, 17.4)	2.69 (2.54, 2.84)
West North Central	42.8 (38.1, 47.5)	28.8 (24.5, 33.2)	16.3 (12.2, 20.3)	33.9 (28.4, 39.4)	46.9 (41.6, 51.9)	28.3 (23.4, 33.2)	45.0 (39.3, 50.3)	17.2 (13.1, 21.6)	2.51 (2.29, 2.71)
South Atlantic	34.9 (32.6, 37.3)	22.3 (20.5, 24.3)	13.4 (11.7, 15.3)	24.8 (22.8, 26.7)	35.7 (33.6, 37.9)	23.7 (21.5, 25.8)	34.2 (31.9, 36.6)	14.2 (12.5, 15.9)	1.96 (1.86, 2.07)
East South Central	45.3 (41.7, 49.0)	24.8 (21.6, 28.1)	26.4 (23.2, 29.6)	24.8 (21.2, 28.1)	42.2 (38.1, 46.3)	36.2 (32.4, 39.8)	46.1 (42.1, 50.2)	19.5 (16.8, 21.9)	2.64 (2.45, 2.85)
West South Central	43.3 (41.1, 45.6)	25.4 (23.6, 27.3)	18.3 (16.7, 20.0)	25.0 (23.0, 26.8)	38.5 (36.4, 40.6)	18.9 (17.3, 20.5)	39.9 (37.8, 42.0)	14.0 (12.4, 15.2)	2.17 (2.08, 2.27)
Mountain	52.0 (48.9, 54.7)	31.4 (29.0, 33.8)	15.5 (13.7, 17.4)	39.1 (36.8, 41.4)	49.0 (47.0, 52.4)	27.0 (24.9, 29.2)	40.6 (38.0, 42.9)	21.0 (18.9, 23.2)	2.61 (2.51, 2.72)
Pacific	44.0 (41.1, 47.2)	29.0 (26.6, 31.4)	21.9 (19.6, 24.1)	29.3 (26.8, 31.8)	42.7 (40.2, 45.1)	16.9 (14.9, 18.7)	37.1 (34.3, 39.8)	18.2 (12.2, 24.2)	2.34 (2.25, 2.45)

ACE, adverse childhood experience; BRFSS, Behavioral Risk Factor Surveillance System; Wgt, weight.

Those with an income of ≥$50,000 reported the lowest ACEs (2.31), whereas those making <$15,000 reported the highest number of total ACEs (2.65). Similar decreasing ACEs were found in educational attainment, with the highest mean ACE scores in those with less than a high school degree (2.55) and the lowest mean scores in those with college degrees (2.00). AI/AN individuals who identified as unable to work had the highest mean ACE score among the employment categories (3.23) compared with those in the other category (1.71), which included students, homemakers, and retirees. Regarding employment status, employed individuals reported lower ACE scores than reported by unemployed individuals (2.24 vs 2.84).

Among demographic stratification variables, AI/ANs who identified as gay/lesbian had the highest mean ACE score (4.05), with approximately 70% of individuals in this category experiencing household substance abuse and parental separation/divorce. Individuals identifying as bisexual had the second highest mean ACE score among all stratification variables (3.22), with 65.3% reporting emotional abuse. With respect to region, those living in the East North Central region (Wisconsin, Michigan, Illinois, Indiana, and Ohio) had the highest ACE score (2.69), and those living in New England had the lowest (1.69).

Table 2 presents the ACE frequencies for Black, White, Hispanic, and all races/ethnicities. Compared with other races/ethnicities and national averages, AI/AN individuals had a significantly higher prevalence of ACEs across 7 domains, with the 1 exception being Black individuals with slightly higher parental separation/divorce rates (44.0% vs 41.6%). AI/AN individuals also had significantly higher total mean ACE scores than all the other categories.

DISCUSSION

This study, to the authors’ knowledge, is the most geographically diverse and largest numerical sample of ACEs data among AI/AN individuals and provides an expanded investigation of ACEs exposure across 34 states. Similar to Merrick et al.,¹³ the current findings reveal that certain demographic characteristics have higher rates of childhood adversity than others. Aligning with other research on ACEs, female individuals and younger adults reported higher ACE scores than other categories in their respective demographics.^13,30 In addition to these differences, the largest demographic disparity of mean ACE scores was found among sexual orientation subgroups (gay/lesbian=4.05, bisexual=3.22, and heterosexual=2.21). Although the ratio of mean ACEs (i.e., gay/lesbian individuals reporting nearly twice the ACEs reported by heterosexual individuals) is similar to other studies in the general population,¹³ the present results found higher averages among all the 3 categories. Although theories about why sexual minorities have higher ACEs have been postulated, such as that childhood adversity may catalyze shifts in sexual orientation or that sexual minorities are more likely to recognize and thus report adverse events,³¹ the association of higher ACEs among sexual minorities remains unclear.

Compared with the few studies that have investigated ACE scores among AI/AN populations, these results differ considerably. For example, when compared with the largest AI/AN study that examined ACE scores (n=1,660) using parents’ reports of their children’s experiences, the current results show substantially higher frequencies of ACEs in all categories.²² This includes parental separation/divorce (42% vs 20%), household substance abuse (41% vs 8%), family member incarceration (18% vs 8%), household mental illness (23% vs 6%), and intimate partner violence (29% vs 9%). These stark differences in ACE score reporting could be due to biases associated with parental reporting of household adversity. Because parents are more likely to under-report adverse events concerning their children, parents can be an unreliable proxy to report ACEs. Compared with a study that assessed self-reported ACE scores of AI/AN adults residing in South Dakota (N=516), which more closely aligns with data collection in this study, ACEs scores in this study are higher in 5 of the 8 ACE domains and only marginally lower in the remaining 3 categories (e.g., this study’s findings of household mental illness were 23% vs 24%).²⁰ Generally, this suggests that the current body of literature on ACEs among AI/AN individuals has underestimated the true prevalence of ACEs in the population.

It is also worth mentioning a recent follow-up study conducted by Merrick and colleagues,²⁷ which included AI/AN populations (N=838). Although their study did not disaggregate ACE domains nor total ACE scores to AI/AN populations, they did provide ACE frequencies for AI/AN populations. Merrick et al.¹³ reported that 28.8% of people had an ACE score of 0, slightly more than that reported in this study (27.9%).

Although ACE scores of AI/AN individuals may be higher than those of any other racial/ethnic group, these disparities can possibly be attributed to historical and contemporary social, political, and economic factors. ACE questions are specifically an individualistic way to measure adverse childhood events; however, historical trauma and cultural trauma in AI/AN populations contribute to disparities.¹⁷ There is evidence to suggest that historical trauma, including the loss of land, language, and culture that occurred more than a century ago (combined with discrimination in the form of both macroaggressions and microaggressions), manifests itself in present-day individuals who identify as AI/AN.³² For instance, Brockie et al.³² found that AI/ANs with higher rates of historical loss had significantly higher rates of depression, substance use, and suicide (rates were significantly increased in those reporting discrimination). As such, it is possible that historical loss and associated symptomology contribute to an increased risk of ACEs.

This study has several implications for population based public health. In particular, this study comprises the most comprehensive published ACEs data set for AI/AN populations, which captures disparities across a broader geographic spectrum. The findings highlight that AI/AN populations are more at risk for adversity in childhood (especially than other races/ethnicities), which can deleteriously manifest throughout the lifespan. As such, CDC released a recent 2019 report on evidence-based strategies to mitigate and prevent ACEs.³³ ACE prevention programming tailored to specific races/ethnicities has shown promising results in Black and Hispanic communities^34,35; yet, little is published on prevention/intervention efforts in AI/AN communities. One such effort showing promising results has been historical-based trauma counseling for AI/AN individuals.³⁶ These programs should be expanded in an effort to increase the efficacy of addressing AI/AN adversity. Because most studies subsume proportionally smaller races/ethnicities such as AI/AN into larger categories, perhaps the substantially higher ACE prevalence that AI/AN endure largely goes unnoticed, thus eclipsing the need for prevention/intervention programming for AI/AN populations that other races/ethnicities have been afforded.

Limitations

This study should be considered in conjunction with several limitations. For example, BRFSS relies on self-reporting and thus may be limited by several factors, including memory recall, response biases, events happening before children form memories, as well as individuals with the highest ACE scores being decreased as a result of the increased risk of disease.³⁷ However, self-report is an improvement in previous studies that used parents as a child’s proxy when measuring ACEs, and previous investigations have supported the validity of self-reported adversity in childhood.³⁸ Furthermore, adversity in childhood is a set of complex, intricate, and multidimensional processes that the ACEs framework attempts to simplify, especially among AI/AN individuals. ACE questions are not culturally centered nor framed in a cultural context relevant to AI/AN individuals. In addition, there are 574 federally recognized tribes and 66 additional state-recognized tribes in 36 states across different regions of the U.S.³⁹ Each tribe is unique with diverse cultural values and practices. These nationally representative BRFSS data do not indicate tribal affiliation and thus do not capture important differences between each community inherent in studying this highly diverse population.

CONCLUSIONS

Given that research indicates the role of historical and cultural traumas related to health disparities and psychosocial functioning,¹⁵ future research is needed in several areas regarding ACEs and AI/AN populations. First, cultural adaptations of ACE questions are largely nonexistent. AI/AN populations often experience remnants of historical trauma that have enduring consequences for community members,^32,40 a salient yet overlooked element when assessing childhood experiences. Framing childhood trauma in a cultural context could yield unique traumas for AI/AN populations not present in other populations. Next, future research should further explore the differences in ACEs by tribal affiliation such that prevention efforts can be culturally tailored to the AI/AN population and, more specifically, each individual tribe. Lastly, there has been little effort to assess ACEs among other indigenous populations worldwide. This study may catalyze other investigations into other indigenous and native populations, such as the First Nations in Canada, and further how those populations compare with AI/AN populations.