Abstract
Objective
The COVID-19 pandemic has disproportionately affected American Indian and Alaska Native (AI/AN) people, who experience a 3.2 times higher age-adjusted rate of hospitalization and nearly double the attributed deaths compared to non-Hispanic Whites. We examined pandemic effects on emotional health and substance use in urban AI/AN people.
Methods
From January–May 2021 we collected cross-sectional data from 642 patients seen at five health organizations serving primarily AI/AN people in urban settings. The outcomes are self-reported, cross-sectional changes in emotional health and substance use since pandemic onset. Exposures of interest include infection history, COVID-19 risk perception, pandemic-related life disruption, and feared effects on AI/AN culture. Poisson regression was used to model adjusted multivariate associations.
Results
Since pandemic onset, 46% of participants reported worsened emotional health; 20% reported increased substance use. Very or extremely disruptive pandemic experiences and increasing reported feared pandemic effects on culture were associated with worse pandemic emotional health [adjusted Prevalence Ratio 1.84; 95% CI 1.44, 2.35 and 1.11; 95% CI 1.03, 1.19], respectively. COVID-19 infection and risk perception were not associated with emotional health after adjustment for other factors. The primary exposures were not associated with change in substance use.
Conclusions
The COVID-19 pandemic has impacted the emotional health of urban AI/AN people. The finding that poor emotional health is associated with pandemic-related threats to AI/AN culture may signal a protective role for community and cultural resources. This warrants further study as exploratory analysis did not find hypothesized effect modification according to strength of affiliation with AI/AN culture.
Keywords: American Indian and Alaska Native people, COVID-19, Culture, Emotional health, Pandemic, Substance use
1. Introduction
The impacts of the COVID-19 pandemic have been widespread and severe. As of September 2022 in the United States, there have been over nine million cases of infection, five million hospital admissions, and one million deaths [1]. The COVID-19 pandemic has disproportionately affected American Indian and Alaska Native (AI/AN) communities, as evidenced by higher rates of morbidity and mortality compared to Non-Hispanic White (NHW) people. AI/AN people experienced 3.2 times higher age-adjusted rate of hospitalization and nearly double the attributed deaths compared to NHW people [2,3]. These inequities have gradually attenuated, reflecting AI/AN community vaccination successes, which are the highest rates in the US [4]. Beyond direct effects of infection, the impact of COVID-19 on AI/AN peoples' emotional health and substance use may exacerbate existing inequities. In 2020, 5.6% of AI/AN participants in the National Survey on Drug Use and Health reported past-year serious thoughts of suicide [5]. According to the Centers for Disease Control and Prevention statistics reporting system, in 2020 the rate of suicide for AI/AN people exceeded the US general population (age adjusted rate 14.53 vs 13.48 per 100,000, respectively) [6]. Likewise, from 2019 to 2020 drug overdose as a cause of death was highest among AI/AN people (30.5 per 100,000 for AI/AN people vs 26.2 for NHW people in 2019, increasing to 42.5 vs 33.1 in 2020) [7]. Historical trauma and decreased social support may be associated with COVID-19 related stress among AI/AN people [8], adding to the potentially adverse consequences for members of this population. However, the intrinsic strengths of AI/AN culture may reduce risk of worsening pandemic emotional health and substance use, as suggested by pre-pandemic evidence of traditional AI/AN cultural activities associated with positive mental health [9] and consideration of culture in strength-based approaches to modify substance use inequities among AI/AN people [10].
The COVID-19 pandemic has a distinct impact on emotional health and substance use in the general population. Beyond the direct harm conferred by poor emotional health, people with prior diagnoses of mental illness have a 65% increased risk of COVID-19 infection, even after controlling for other known contributors [11]. Therefore, increased likelihood of COVID-19 infection may exacerbate emotional health concerns, which may contribute to COVID-19 inequities among AI/AN people. Furthermore, a growing literature suggests COVID-19 related stressors contribute to increased substance use, with an uptick in emergency room positive screens for opiate and stimulant use during the pandemic [12,13]. Thus, identifying factors impacting COVID-19-related emotional health and substance use, in terms of risk as well as protection such as AI/AN cultural factors, is critical to informing actionable intervention for COVID-19 and its sequalae.
Despite the burden of COVID-19 in AI/AN communities, the extent of preexisting mental illness diagnoses and substance use [7,[14], [15], [16], [17], [18]], and inequities in social determinants of health [19,20], there is minimal literature on the impact of the pandemic on emotional health and substance use in AI/AN people, especially among the majority who live in urban areas [21]. To address this gap, we assessed factors such as perception of COVID-19 risk, actual infection, life impacts, and feared AI/AN cultural harms thought to be associated with worsening pandemic-related emotional health and substance use.
2. Methods
2.1. Study design
Community Organizations for Natives: COVID-19 Epidemiology, Research, Testing, and Services (CONCERTS) was designed to identify and remove barriers to COVID-19 testing among urban AI/AN people and to address health inequities related to COVID-19 in this population. As part of CONCERTS, we created and implemented a cross-sectional survey in partnership with five tribal health organizations serving primarily AI/AN people in urban settings to identify barriers, facilitators, attitudes, and factors related to COVID-19 testing and vaccination. CONCERTS community partners and investigators anticipated the importance of gathering a wide range of data that assess the impact of COVID-19 on AI/AN peoples' health. Following survey implementation, this study was proposed and approved by the CONCERTS team and community partners, allowing utilization of these data to explore the impact of the pandemic on the health of AI/AN peoples' emotional health and substance use.
The included tribal health organizations provide a range of services including outreach and referral, dental, primary, and behavioral health care, health promotion and disease prevention supported by additional state, federal, and local resources. The areas served by the tribal organizations participating in CONCERTS are Albuquerque, NM, Anchorage, AK, Denver, CO, Minneapolis-St. Paul, MN, and Wichita, KS. The active patient populations ranged from 1269 to 25,043 unique patients seen in 2019.
Patients eligible for inclusion in the study were seen at any of the five tribal primary care clinics within the year prior to the survey, were 18 years or older, AI/AN persons, and not diagnosed with dementia or other serious cognitive issues. To ensure we enrolled enough older adults at highest likelihood of contracting COVID-19, we stratified sampling by age (18–54 vs. 55 and over). Each clinic generated a list of eligible patients from their electronic health records and randomly sampled 200 in each age group. Clinic patients with an email address in their respective electronic health records were sent an invitation to participate in CONCERTS along with a link to an online REDCap survey. Patients without an email address were mailed invitations to participate. Patients received up to four emails or phone call reminders over a 14-day period. The goal was to enroll up to 150 participants per clinic. If this sample size was not achieved through the first round of emails, letters, and phone calls, we selected a new random sample of eligible patients from the clinic in question. Participants were enrolled from January to May 2021 and were compensated $100. Of 4603 eligible patients contacted for this study, 788 (17.1%) participants were enrolled and completed the survey. This study was approved by the Washington State University IRB (#18590), local review boards, and the Indian Health Service national IRB as appropriate.
2.2. Survey development and measures
The participant survey was designed to collect comprehensive patient-reported information on sociodemographic characteristics, co-morbid health conditions, individual and structural social determinants of health, access to health services, pandemic impact on physical and emotional health, socioeconomic and quality of life status, experience with and attitudes toward COVID-19, and receipt of and structural barriers to COVID-19 testing and vaccination. Survey questions were generally developed based on the NIH RADX-UP Common Data Elements and PhenX Toolkit. When necessary, questions were modified for cultural appropriateness (for example, when asking about insurance status, inclusion of population-specific coverage such as Indian Health Services), administration mode, or to better reflect study aims. After the survey was developed, providers and staff at each clinic provided feedback on the appropriateness and relevance of all survey questions to their clinic populations. The resulting survey has 164 questions, took about 40 min to complete and included a full study description with informed consent.
The primary outcomes were self-reported change in emotional health and substance use since the pandemic started in February 2020, assessed as the same/better/worse and same/no use/less/more, respectively. To focus on identification of factors associated with inequities, worsened emotional health was compared to same and better emotional health grouped together as reference. Substance use was compared with the group of interest, either increased or decreased use, with all remaining categories as reference; as detailed further in analysis, we conducted sensitivity analyses excluding participants who reported no substance use to support this decision. Four primary measures were used to assess domains related to COVID-19 pandemic experiences: COVID-19 infection status, perception of COVID-19 risk, life impacts, and feared effects on AI/AN culture. COVID-19 infection was assessed by a positive response to either “Did any test indicate that you had COVID-19?” and/or “Has a doctor, nurse or other health care professional ever told you that you were infected with or have COVID-19?” Perception questions included if the respondent felt “at risk of being infected” (dichotomous) and how concerned they were about becoming infected (Likert, not/a little/moderately/very/extremely). To summarize the concepts embodied by these questions, they were combined into an aggregate variable with the four categories defined by the combinations of at-risk vs not (yes/no, “I am at risk of being infected with COVID-19”), and very vs minimally concerned (“very” or “extremely” defining the former group, with the sample distribution approximating 50% for each group), to describe the presence and magnitude of perceived risk. COVID-19 life impacts were assessed by a global measure of “How disruptive has the COVID-19 pandemic been to your everyday life?” and dichotomized to “Not at all/A little/Somewhat disruptive” vs “Very/Extremely disruptive”, with the sample distribution approximating 50% for each. Six dichotomous cultural impact questions were reported descriptively in Table 1 and included feared difficulties or losses of Native languages, of elders and wisdom keepers, of participation in cultural practices, loss of tribal population, AI/AN stigma, and similarity to historical diseases. Feared cultural impacts were assessed via the aggregate number endorsed.
Table 1.
Participant characteristics among urban American Indians and Alaska Native (AI/AN) people Jan-May 2021 (N = 642).
| N=6421 | |
|---|---|
| Sociodemographic and health | |
| Age in years, mean (SD) | 42.8 (13.5) |
| Female | 60% |
| Education | |
| Less than high school | 7% |
| High school graduate or GED | 23% |
| Some college | 29% |
| Associate, occupational, technical, or vocational degree | 17% |
| College degree or higher | 24% |
| Marital status | |
| Married or member of an unmarried couple | 44% |
| Divorced, separated, widowed | 20% |
| Never married | 36% |
| Global self-rated health | |
| Excellent | 7% |
| Very good | 26% |
| Good | 45% |
| Fair/Poor | 22% |
| High cultural identity | 24% |
| Pandemic-related exposures | |
| Positive COVID-19 infection | 16% |
| COVID-19 perceptions | |
| Not at risk / Not concerned | 18% |
| Not at risk / Concerned | 10% |
| At risk / Not concerned | 33% |
| At risk / Concerned | 39% |
| Effect of pandemic on everyday life | |
| Not at all / A little / Somewhat disruptive | 47% |
| Very / Extremely disruptive | 53% |
| Count of 6 concerns about cultural impacts of pandemic, mean (SD) | 3.9 (1.9) |
| It will be harder to practice or learn our Native language | 48% |
| We will lose our elders and wisdom keepers | 81% |
| It will be harder to participate in our cultural practices | 79% |
| The population of my AI/AN tribe or group will be reduced | 76% |
| I or my AI/AN tribe or group is being stigmatized, or marked or seen in a negative way | 38% |
| It reminds me of how our ancestors were wiped out from similar diseases | 73% |
| Outcomes | |
| Overall emotional health since beginning of COVID-19 pandemic in February 2020 | |
| Improved or stayed the same | 54% |
| Gotten worse | 46% |
| Substance use since beginning of COVID-19 pandemic in February 2020 | |
| Increased | 20% |
| Stayed about the same | 27% |
| Do not use alcohol or drugs | 38% |
| Decreased | 15% |
Unweighted sample size, results weighted for sampling and non-response.
Patient-reported covariates included age, sex, marital status, educational attainment, global self-rated assessment of health (Likert, poor/fair/good/very good/excellent). Sex was selected as participants more frequently reported sex than gender, and review of sample responses indicated concordance between reported sex and gender. Cultural identity was assessed via reported degree of alignment with an AI/AN way of life (a lot vs somewhat, a little, not at all) [22]. We performed an exploratory analysis to assess effect modification by cultural identity, hypothesizing participants with greater alignment would be more negatively affected by from feared cultural impacts.
2.3. Statistical analysis
Mean, standard deviation, and frequency were used to describe sociodemographic characteristics and pandemic-related exposures and outcomes. Poisson regression was used to estimate prevalence ratio (PR) with 95% confidence interval (CI) for the association between COVID-19 pandemic experiences and changes in the dependent variables emotional health and substance use. One model was fit to examine factors related to worsened emotional health. Two models were fit related to change in substance use: one comparing participants who reported increased use to those who did not (same use, decreased use, no substance use) and one comparing participants who reported decreased substance use to those who did not (same use, increased use, no substance use). For each outcome comparison, we fit a set of three models containing progressively inclusive independent variables. The first model (1) examined the unadjusted bivariate association of each independent variable with the specific outcome. The second model (2) examined the association of each pandemic-related exposure adjusted for sociodemographic variables. The final model (3) included all pandemic-related exposures and sociodemographic variables. We also conducted sensitivity analyses for change in substance use employing the same method but excluding participants reporting no substance use from the analysis. Effect modification was formally assessed by fitting regression models that included model (3) independent variables, a main effect for cultural identity, and the product term for the pandemic-related exposure and cultural identity (low vs high). Separate models were fit for each effect modification analysis (four models for each outcome) but each model included main effect terms for all pandemic related exposures. Inferential results are presented as prevalence ratio (PR) with 95% confidence interval (CI). Poisson rather than logistic regression was selected because odds ratios are biased estimates of prevalence ratios when assessing non-rare outcomes [23]. All analyses incorporated inverse probability weights to account for age-based sample selection, nonresponse according to age and sex, and weighting by clinic. Weights were scaled so that each of the five participating clinics contributed equally to analyses, preventing a single clinic with a very large population from disproportionately influencing the analysis. Analyses were conducted using Stata 17.0 [24].
3. Results
Of the 788 participants who completed some portion of the CONCERTS survey, 642 (81%) with complete data for all study variables were included in analysis. After inverse probability weighting was applied, the mean age was 42.8 years, 60% were female, 24% had a college degree or higher, 44% were married, and 78% reported their global health as good or better (Table 1). A high degree, specified as “a lot”, of alignment with an AI/AN cultural identity was reported in 24% of respondents. COVID-19 related stressors were prevalent in the study: 16% reported COVID-19 infection; 39% described feeling at risk of COVID-19 infection and were concerned; 53% felt the pandemic was very or extremely disruptive; the mean number of feared cultural impacts were 3.9 of 6 possible. Relative to pandemic onset, 46% reported worsened emotional health, 20% reported increased substance use, and 15% reported decreased substance use.
3.1. Worsened emotional health
In the univariate emotional health models, increasing age was associated with lower prevalence of worsened emotional health, (PR 0.93; 95% CI 0.90, 0.97; p < 0.001) for a five-year increase in age (Table 2 ). Additionally, worse global self-rated health was associated with worsened emotional health, “fair/poor” vs “excellent” (PR 2.35; 95% CI 1.33, 4.14; p = 0.003), a somewhat imprecise result. Other covariates were not significantly associated with emotional health. Perception and concern of being at risk for COVID-19 infection was associated with worse emotional health, (PR 1.76; 95% CI 1.23, 2.51; p = 0.002) compared to those feeling not at risk and not concerned. Report of very/extremely disruptive pandemic experiences and more concerns about cultural impacts were associated with worse emotional health, (PR 2.18; 95% CI 1.70, 2.80; p < 0.001) and (PR 1.19; 95% CI 1.11, 1.27; p < 0.001), respectively. COVID-19 infection was associated with higher prevalence of worsened emotional health, but the 95% confidence interval included the null (PR 1.22; 95% CI 0.94, 1.56; p = 0.13).
Table 2.
Association of factors with overall worse emotional health since beginning of COVID-19 pandemic in February 2020, assessed among urban American Indians and Alaska Native people Jan-May 2021 (N=6421).
|
Model 12 |
Model 22 |
Model 32 |
|
|---|---|---|---|
| PR (95% CI) | PR (95% CI) | PR (95% CI) | |
| Sociodemographic and health | |||
| Age in years3 | 0.93 (0.90, 0.97)4 | 0.92 (0.89, 0.96)4 | 0.94 (0.91, 0.97)4 |
| Sex | |||
| Male | Reference | Reference | Reference |
| Female | 1.21 (0.95, 1.54) | 1.16 (0.93, 1.45) | 1.19 (0.96, 1.48) |
| Education | |||
| Less than high school | 1.45 (0.93, 2.25) | 1.35 (0.90, 2.01) | 1.32 (0.86, 2.02) |
| High school graduate or GED | Reference | Reference | Reference |
| Some college | 1.33 (0.95, 1.87) | 1.44 (1.04, 1.97)4 | 1.26 (0.92, 1.73) |
| Associate, occupational, technical, or vocational degree | 1.22 (0.82, 1.82) | 1.25 (0.85, 1.83) | 1.05 (0.72, 1.52) |
| College degree or higher | 1.37 (0.98, 1.93) | 1.57 (1.13, 2.17)4 | 1.22 (0.88, 1.70) |
| Marital status | |||
| Married or member of an unmarried couple | Reference | Reference | Reference |
| Divorced, separated, widowed | 0.99 (0.75, 1.32) | 1.02 (0.78, 1.34) | 0.98 (0.76, 1.27) |
| Never married | 1.13 (0.90, 1.44) | 1.02 (0.81, 1.30) | 1.08 (0.86, 1.35) |
| Global self-rated health | |||
| Excellent | Reference | Reference | Reference |
| Very good | 1.54 (0.85, 2.77) | 1.70 (0.94, 3.08) | 1.58 (0.93, 2.70) |
| Good | 1.94 (1.10, 3.40)4 | 2.25 (1.28, 3.97)4 | 1.79 (1.06, 3.03)4 |
| Fair/Poor | 2.35 (1.33, 4.14)4 | 2.95 (1.64, 5.29)4 | 2.32 (1.36, 3.98)4 |
| Pandemic-related exposures | |||
| COVID-19 infection | |||
| No | Reference | Reference | Reference |
| Yes | 1.22 (0.94, 1.56) | 1.18 (0.93, 1.50) | 1.12 (0.88, 1.42) |
| COVID-19 perceptions | |||
| Not at risk / Not concerned | Reference | Reference | Reference |
| Not at risk / Concerned | 1.11 (0.66, 1.85) | 1.13 (0.70, 1.83) | 0.90 (0.57, 1.41) |
| At risk / Not concerned | 1.19 (0.80, 1.76) | 1.17 (0.80, 1.69) | 1.08 (0.75, 1.56) |
| At risk / Concerned | 1.76 (1.23, 2.51)4 | 1.68 (1.19, 2.36)4 | 1.19 (0.83, 1.70) |
| Effect of pandemic on everyday life | |||
| Not at all / A little / Somewhat disruptive | Reference | Reference | Reference |
| Very / Extremely disruptive | 2.18 (1.70, 2.80)4 | 2.03 (1.59, 2.60)4 | 1.84 (1.44, 2.35)4 |
| Count of 6 concerns about cultural impacts of pandemic | 1.19 (1.11, 1.27)4 | 1.16 (1.08, 1.24)4 | 1.11 (1.03, 1.19)4 |
Unweighted sample size, results weighted for sampling and non-response.
Model 1 includes the indicated variable as the only independent variable; Model 2 includes indicated variable and all sociodemographic and health variables; Model 3 includes all sociodemographic and health variables and all pandemic-related exposures.
Prevalence ratio for a 5-year increase in age.
p < 0.05; PR = prevalence ratio; CI = confidence interval.
In adjusted models, the magnitude of the association between worsened emotional health and pandemic- related exposures was generally attenuated compared to the unadjusted models. The association with perception of COVID-19 risk was similar to unadjusted estimates when accounting for sociodemographic covariates only, but the effect size was reduced after adjusting for the other pandemic related exposures (PR 1.19; 95% CI 0.83, 1.70; p = 0.34) (Table 2). Both disruptive pandemic experiences and concerns about cultural impacts remained associated with worsened emotional health after adjustment for sociodemographics and other pandemic related exposures, though results were moderately attenuated (PR 1.84; 95% CI 1.44, 2.35; p < 0.001 and PR 1.11; 95% CI 1.03, 1.19; p = 0.005, respectively).
3.2. Change in substance use
There was little evidence for an association between any change in substance use, and the selected pandemic-related exposures. In unadjusted and adjusted models, increasing age was consistently associated with lower prevalence of increased substance use with all others as reference, (fully-adjusted PR 0.82; 95% CI 0.75, 0.89; p < 0.001) for a five-year increase in age (Table 3 ). COVID-19 infection was associated with higher prevalence of increased substance use vs all others (PR 1.21; 95% CI 0.75, 1.94; p = 0.44) and lower prevalence of decreased substance use vs all others (PR 0.84; 95% CI 0.43, 1.65; p = 0.62); however, both estimates are roughly compatible with associations on either side of the null. Results from sensitivity analyses excluding participants who reported they did not use substances from the reference group for both increase and decrease in substance use were largely unchanged (Appendix A). The most substantial difference was for the association between age and decreased substance use, which was null in the sensitivity analyses (PR 1.00; 95% CI 0.93, 1.08; p = 0.91).
Table 3.
Association of factors with increased or decreased substance use since beginning of COVID-19 pandemic in February 2020, assessed among urban American Indians and Alaska Native people Jan-May 2021 (N=6421).
| Increased substance use |
Decreased substance use |
|||||
|---|---|---|---|---|---|---|
| Model 12 |
Model 22 |
Model 32 |
Model 12 |
Model 22 |
Model 32 |
|
| PR (95% CI) | PR (95% CI) | PR (95% CI) | PR (95% CI) | PR (95% CI) | PR (95% CI) | |
| Sociodemographic and health | ||||||
| Age in years3 | 0.81 (0.75, 0.88)4 | 0.81 (0.74, 0.88)4 | 0.82 (0.75, 0.89)4 | 0.94 (0.88, 1.01) | 0.93 (0.86, 0.99)4 | 0.93 (0.86, 1.00) |
| Sex | ||||||
| Male | Reference | Reference | Reference | Reference | Reference | Reference |
| Female | 0.73 (0.48, 1.12) | 0.73 (0.49, 1.07) | 0.73 (0.49, 1.08) | 1.05 (0.66, 1.67) | 0.96 (0.61, 1.52) | 0.96 (0.61, 1.50) |
| Education | ||||||
| Less than high school | 0.66 (0.24, 1.83) | 0.66 (0.25, 1.74) | 0.66 (0.25, 1.71) | 0.57 (0.11, 2.89) | 0.56 (0.11, 2.92) | 0.52 (0.10, 2.81) |
| High school graduate or GED | Reference | Reference | Reference | Reference | Reference | Reference |
| Some college | 0.90 (0.49, 1.66) | 0.91 (0.51, 1.62) | 0.91 (0.49, 1.67) | 1.09 (0.56, 2.11) | 1.08 (0.55, 2.13) | 0.97 (0.49, 1.91) |
| Associate, occupational, technical, or vocational degree | 0.75 (0.35, 1.64) | 0.83 (0.39, 1.74) | 0.83 (0.38, 1.82) | 0.95 (0.45, 2.00) | 1.02 (0.49, 2.16) | 0.83 (0.40, 1.87) |
| College degree or higher | 1.10 (0.60, 2.01) | 1.27 (0.73, 2.22) | 1.26 (0.69, 2.30) | 1.37 (0.74, 2.56) | 1.39 (0.74, 2.61) | 1.15 (0.58, 2.28) |
| Marital status | ||||||
| Married or member of an unmarried couple | Reference | Reference | Reference | Reference | Reference | Reference |
| Divorced, separated, widowed | 0.85 (0.45, 1.62) | 1.09 (0.58, 2.05) | 1.10 (0.59, 2.06) | 1.49 (0.86, 2.58) | 1.69 (0.99, 2.87) | 1.66 (0.97, 2.82) |
| Never married | 1.55 (0.97, 2.47) | 1.20 (0.77, 1.87) | 1.25 (0.80, 1.96) | 1.14 (0.67, 1.92) | 1.09 (0.64, 1.86) | 1.11 (0.66, 1.88) |
| Global self-rated health | ||||||
| Excellent | Reference | Reference | Reference | Reference | Reference | Reference |
| Very good | 0.78 (0.34, 1.80) | 1.10 (0.48, 2.48) | 1.12 (0.50, 2.52) | 1.71 (0.69, 4.22) | 1.84 (0.71, 4.73) | 1.73 (0.67, 4.43) |
| Good | 1.35 (0.62, 2.92) | 1.91 (0.91, 4.00) | 1.87 (0.88, 3.98) | 1.51 (0.63, 3.65) | 1.69 (0.67, 4.26) | 1.49 (0.59, 3.74) |
| Fair/Poor | 0.95 (0.40, 2.26) | 1.63 (0.68, 3.92) | 1.58 (0.64, 3.89) | 0.98 (0.34, 2.80) | 1.19 (0.40, 3.53) | 1.02 (0.36, 2.94) |
| Pandemic-related exposures | ||||||
| COVID-19 infection | ||||||
| No | Reference | Reference | Reference | Reference | Reference | Reference |
| Yes | 1.36 (0.81, 2.26) | 1.17 (0.74, 1.86) | 1.21 (0.75, 1.94) | 0.96 (0.50, 1.83) | 0.90 (0.47, 1.74) | 0.84 (0.43, 1.65) |
| COVID-19 perceptions | ||||||
| Not at risk / Not concerned | Reference | Reference | Reference | Reference | Reference | Reference |
| Not at risk / Concerned | 1.05 (0.46, 2.40) | 1.18 (0.57, 2.44) | 1.09 (0.53, 2.28) | 0.90 (0.35, 2.28) | 1.14 (0.44, 2.95) | 1.02 (0.40, 2.60) |
| At risk / Not concerned | 0.88 (0.47, 1.64) | 0.96 (0.55, 1.68) | 0.93 (0.52, 1.64) | 1.23 (0.62, 2.43) | 1.37 (0.70, 2.68) | 1.33 (0.68, 2.60) |
| At risk / Concerned | 1.03 (0.57, 1.87) | 1.09 (0.63, 1.86) | 0.99 (0.56, 1.77) | 1.12 (0.58, 2.18) | 1.22 (0.63, 2.34) | 1.02 (0.52, 2.00) |
| Effect of pandemic on everyday life | ||||||
| Not at all / A little / Somewhat disruptive | Reference | Reference | Reference | Reference | Reference | Reference |
| Very / Extremely disruptive | 1.38 (0.89, 2.15) | 1.23 (0.82, 1.84) | 1.21 (0.81, 1.82) | 1.31 (0.84, 2.05) | 1.28 (0.82, 2.01) | 1.27 (0.81, 2.00) |
| Count of 6 concerns about cultural impacts of pandemic | 1.05 (0.92, 1.19) | 1.02 (0.91, 1.15) | 1.01 (0.89, 1.14) | 1.11 (0.99, 1.24) | 1.08 (0.96, 1.22) | 1.08 (0.96, 1.23) |
Unweighted sample size, results weighted for sampling and non-response.
Model 1 includes the indicated variable as the only independent variable; Model 2 includes indicated variable and all sociodemographic and health variables; Model 3 includes all sociodemographic and health variables and all pandemic-related exposures.
Prevalence ratio for a 5-year increase in age.
p < 0.05; PR = prevalence ratio; CI = confidence interval.
3.3. Effect modification by cultural identity
There was also little evidence for differences in effects among participants with low vs high cultural identity (Table 4 ); however, confidence intervals were wide and should be interpreted with caution.
Table 4.
The association of changes in emotional health and substance use since beginning of COVID-19 pandemic with pandemic-related exposures stratified by cultural identity, assessed among urban American Indians and Alaska Native people Jan-May 2021 (N=6401).
| Emotional health |
Increased substance use |
Decreased substance use |
||||
|---|---|---|---|---|---|---|
| Cultural identity2 |
Cultural identity2 |
Cultural identity2 |
||||
| Low |
High |
Low |
High |
Low |
High |
|
| PR (95% CI) | PR (95% CI) | PR (95% CI) | PR (95% CI) | PR (95% CI) | PR (95% CI) | |
| Pandemic-related exposures | ||||||
| COVID-19 infection | ||||||
| No | Reference | Reference | Reference | Reference | Reference | Reference |
| Yes | 1.01 (0.74, 1.36) | 1.33 (0.93, 1.89) | 1.09 (0.63, 1.90) | 2.34 (0.76, 7.23) | 0.78 (0.33, 1.88) | 1.26 (0.42, 3.77) |
| P-value for effect modification | 0.21 | 0.43 | 0.60 | |||
| COVID-19 perceptions | ||||||
| Not at risk / Not concerned | 0.95 (0.65, 1.40) | 0.31 (0.11, 0.93) | 1.05 (0.60, 1.83) | 0.54 (0.11, 2.60) | 1.28 (0.60, 2.73) | 0.22 (0.04, 1.18) |
| Not at risk / Concerned | 0.65 (0.40, 1.08) | 0.96 (0.57, 1.61) | 1.15 (0.55, 2.41) | 0.79 (0.07, 8.47) | 0.94 (0.28, 3.19) | 1.06 (0.39, 2.88) |
| At risk / Not concerned | 0.83 (0.63, 1.08) | 1.12 (0.72, 1.75) | 0.80 (0.48, 1.33) | 2.04 (0.53, 7.85) | 1.65 (0.90, 3.06) | 0.11 (0.01, 0.99) |
| At risk / Concerned | Reference | Reference | Reference | Reference | Reference | Reference |
| P-value for effect modification | 0.22 | 0.58 | 0.34 | |||
| Effect of pandemic on everyday life | ||||||
| Not at all / A little /Somewhat disruptive | Reference | Reference | Reference | Reference | Reference | Reference |
| Very / Extremely disruptive | 1.96 (1.49, 2.57) | 2.25 (1.48, 3.40) | 1.31 (0.84, 2.05) | 3.05 (0.93, 10.0) | 1.36 (0.82, 2.23) | 0.37 (0.12, 1.09) |
| P-value for effect modification | 0.70 | 0.69 | 0.74 | |||
| Count of 6 concerns about cultural impacts of pandemic | 1.12 (1.03, 1.21) | 1.06 (0.94, 1.20) | 1.03 (0.92, 1.15) | 0.84 (0.65, 1.09) | 1.10 (0.96, 1.26) | 1.20 (0.86, 1.67) |
| P-value for effect modification | 0.92 | 0.82 | 0.99 | |||
Unweighted sample size, results weighted for sampling and non-response.
Cultural identity defined by how much participant reported following the American Indian or Alaska Native way of life: Low = Not at all, A little, Somewhat; High = A lot.
4. Discussion
In this cross-sectional examination of patients seen at five health organizations serving primarily AI/AN people in urban settings, we describe changes in emotional health and substance use among urban AI/AN people since the onset of the COVID-19 pandemic. Nearly half of the sample reported worsened emotional health and 20% had increased substance use. Those with disruptive pandemic experiences and concerns about cultural impacts reported higher prevalence of worsened emotional health. This relationship persisted after adjustment for sociodemographics and other pandemic experiences. To our knowledge, this is the first study identifying AI/AN cultural concerns as associated with worsened emotional health in the pandemic.
Poor pandemic-related emotional health in urban AI/AN people is consistent with the literature in other racial/ethnic groups. A 2020 study of over 60,000 patients in the general population diagnosed with COVID-19 found 5.8% received a first recorded psychiatric diagnosis, relative to 2.5–3.4% among matched controls with non-COVID-19 acute illness [11]. As previously noted, people with prior diagnoses had a 65% increased risk of COVID-19 infection [11]; inequities were further compounded in AI/AN people with known elevated baseline prevalence of poor emotional health [14,25]. The 2005–2014 National Survey on Drug Use and Health described nearly 25% prevalence of past year mental illness among AI/AN adults [14] and an older national survey estimated 70% of AI/AN men and 63% of women had a lifetime mental illness diagnosis, 8–10% higher than white people [25].
Given the impact of COVID-19 on emotional health, steps need to be taken to address underlying factors influencing inequities, barriers to care, and build upon community strengths. Probability of infection can be modified by vaccination, a strategy with great success. AI/AN people are one of the groups with the highest vaccination rates in the US [4]; as of May 2022, 72.6% received at least one vaccine dose and 60.6% were fully vaccinated [26]. Indian Health Service facilities have been remarkably effective in distribution of vaccines, with >3/4ths of over 2.5 million doses successfully administered to patients [27]. AI/AN tribal leadership and community efforts have been critical to these vaccination successes [28], highlighting the value of self-determination and community informed approaches. Over time, these community efforts will help to reduce the negative COVID-19 outcomes.
While we did not directly observe protective effects of culture on emotional health, we evaluated pandemic-related feared cultural harms. Because concern regarding pandemic-related cultural impacts were associated with worsened emotional health, novel approaches to maintaining cultural connections and identity in times of social distancing are crucial. Inequities in social determinants of health, structural inequities, and historical trauma are present among AI/AN communities. Such inequities are associated with poor emotional health in general [29]. Evidence suggests AI/AN peoples' historical trauma and lower social support are associated with COVID-19 related stress [8]. Furthermore, pandemic social distancing and isolation are likely to exacerbate the effect of limited social support and thus worsen emotional health. Barriers in access to care can be mitigated by the implementation of telemedicine programs, which have previously been documented as feasible and effective in AI/AN people for psychiatric care [30,31].
Perception of COVID-19 risk may be related to the information the person has received. It is important to educate individuals about COVID-19 risks and undertake strategies such as vaccination and social distancing to reduce risk. Although the effect of risk perception was no longer significant after adjustment for other pandemic exposure variables, pandemic misinformation is known to be detrimental to public health [32]. Trusted community leaders are well positioned to disseminate evidence based health guidelines to combat misinformation [33].
Impacts of COVID-19 were common among participants, with over half reporting very or extremely disruptive impacts on everyday life. While reduction of community mortality is a larger goal for pandemic mitigation efforts, other tribal and social programs may be important sources of support. The significant associations between feared negative cultural impacts and emotional health should be interpreted judiciously, as the confidence interval approaches 1 (PR 1.11; 95% CI 1.03, 1.19) and further study is needed to confirm clinical significance, especially for those with lower levels of feared cultural harms. Regardless, this may further highlight the value of cultural connections in AI/AN communities and warrant further study.
Despite evidence of effects on emotional health, the findings suggest little evidence that the studied pandemic- related exposures were related to changes in substance use. However, it should also be noted that wide confidence intervals lack precision to evaluate the consistency of results with previous studies. The literature suggests the pandemic has influenced substance use, with a June 2020 survey report indicating 13% of respondents starting or increasing substance use to cope with related stresses [34], which further increased to 15% at follow-up in September [35] and is similar to the 20% increase in the study. The lack of an observed association of the exposures with changes in substance use (COVID-19 infection status, perception of COVID-19 risk, life impacts, and feared effects on AI/AN culture) may be occurring for several reasons. Changes in substance use may be driven by factors besides those examined in this paper, such as baseline use characteristics and the type of substance. The data identified changes in substance use but did not quantify which substances and what amount. For this reason, we could not detect the magnitude of change among distinct subgroups of substances that may be better analyzed separately. Substance use is multifactorial in influence and the data did not capture variables such as access to substances, influence from family and peers, social isolation, or ongoing vs. newly initiated substance use treatment.
The CONCERTS study has several limitations. The cross-sectional design prevents establishing definite temporality between the measures, although the outcomes were worded to report relative change pre−/ post-pandemic onset. Because the measures are primarily survey based, reporting bias may be a factor as stigma can limit reporting of emotional health symptoms, even in some anonymous reporting approaches [36]. Subjective self-reports of emotional health and substance use may be less accurate compared to clinical measurement-based approaches. Additionally, social-desirability bias could lead to perceived “correct” answers when reporting emotional health and substance use. Individuals with more severely impacted emotional health may have difficulties accessing care or be less likely to participate in the study. While the overall sample size for the study was large (n = 642 unweighted), some cells for categorical variables were small, particularly limiting the power for effect modification analyses. Some prevalence ratios were relatively imprecise (for example, the association of global self-rated health with the outcomes) and should be interpreted with caution. Although the broad range of participants is a strength of CONCERTS, heterogeneity among tribes and between urban and rural residents limit generalizability of the findings to AI/AN people in other settings. Detailed investigation of specific psychiatric illnesses and symptoms is warranted, as these likely have detrimental effects on health behaviors and COVID-19 outcomes.
In conclusion, the COVID-19 pandemic has impacted urban AI/AN people, with reported worsening pandemic emotional health and substance use. Emotional health inequities exacerbated by pandemic disruptions and interrupted urban AI/AN cultural practices warrant further study to clarify these relationships and identify actionable interventions.
Previous presentation
University of Colorado Psychiatry Department Annual Poster Session; no prior print publication.
Funding and Acknowledgments
Research reported in this Rapid Acceleration of Diagnostics – Underserved Populations (RADx-UP) publication was supported by the National Institutes of Health under Award Number 3U54MD011240-06S1. Dr. Haskins is a member of the Pathways Resident Research Track, Department of Psychiatry, University of Colorado School of Medicine, which is supported by the National Institute of Mental Health R25MH125758.
CONCERTS Community acknowledgement: Community Organizations for Natives: COVID-19 Epidemiology, Research, Testing, and Services (CONCERTS) would not have been possible without the extraordinary efforts of the clinic staff, administrators, providers, and patients from each of the partnering Tribal Health Organizations. We are extremely grateful for their hard work and commitment to CONCERTS.
Banner authorship: CONCERTS Collaborative: Dedra Buchwald, Richard MacLehose, Spero M. Manson, Odile Madesclaire, Katie Nelson, Austin Henderson, Carolyn Noonan, Talia Quandelacey, Southcentral Foundation, Denver Indian Health and Family Services, Indian Health Board of Minneapolis, First Nations Community Healthsource, Hunter Health Clinic, and the Urban Indian Center of Salt Lake.
Declaration of Competing Interest
None. Authors Haskins, Noonan, MacLehose, Buchwald, and Manson have no competing interests to report.
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.jpsychores.2023.111424.
Appendix A. Supplementary data
Supplementary material - Association of factors with increased or decreased substance use since beginning of COVID-19 pandemic in February 2020, assessed among urban American Indians and Alaska Native people Jan-May 2021 (N=3631) [Substance non-users excluded from analysis]
References
- 1.Centers for Disease Control Interpretive Summary for September 9. 2022. https://www.cdc.gov/coronavirus/2019-ncov/covid-data/covidview/index.html#print CDC.gov2022 [Available from]:
- 2.Centers for Disease Control Disparities in COVID-19-Associated Hospitalizations: Racial and Ethnic Health Disparities Centers for Disease Control and Prevention. 2021. https://www.cdc.gov/coronavirus/2019-ncov/community/health-equity/racial-ethnic-disparities/disparities-hospitalization.html Available from:
- 3.Arrazola J., Masiello M.M., Joshi S., Dominguez A.E., Poel A., Wilkie C.M., et al. COVID-19 mortality among American Indian and Alaska native persons—14 states, January–June 2020. Morbidit. Mortalit. Week. Rep. 2020;69(49):1853. doi: 10.15585/mmwr.mm6949a3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Silberner J. Covid-19: how native Americans led the way in the US vaccination effort. BMJ. 2021;374 doi: 10.1136/bmj.n2168. [DOI] [PubMed] [Google Scholar]
- 5.Substance Abuse and Mental Health Services Administration . SAMHSA.gov; 2021. Key Substance Use and Mental Health Indicators in the United States: Results from the 2020 National Survey on Drug Use and Health. [Google Scholar]
- 6.Centers for Disease Control, Fatal Injury and Violence Data, Injury Prevention and Control Description: Web-based Injury Statistics Query and Reporting System.CDC.gov, 2023 Available from: https://www.cdc.gov/injury/wisqars/fatal.html.
- 7.Hedegaard H., Miniño A.M., Spencer M.R., Warner M. NCHS Data Brief; 2021. Drug Overdose Deaths in the United States, 1999–2020. [PubMed] [Google Scholar]
- 8.John-Henderson N.A., Ginty A.T. Historical trauma and social support as predictors of psychological stress responses in American Indian adults during the COVID-19 pandemic. J. Psychosom. Res. 2020;139 doi: 10.1016/j.jpsychores.2020.110263. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Kading M.L., Hautala D.S., Palombi L.C., Aronson B.D., Smith R.C., Walls M.L. Flourishing: American Indian positive mental health. Soc. Mental Health. 2015;5(3):203–217. doi: 10.1177/2156869315570480. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Whitesell N.R., Beals J., Crow C.B., Mitchell C.M., Novins D.K. Epidemiology and etiology of substance use among American Indians and Alaska natives: risk, protection, and implications for prevention. American J. Drug Alcohol Abuse. 2012;38(5):376–382. doi: 10.3109/00952990.2012.694527. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Taquet M., Luciano S., Geddes J.R., Harrison P.J. Bidirectional associations between COVID-19 and psychiatric disorder: retrospective cohort studies of 62 354 COVID-19 cases in the USA. Lancet Psychiatry. 2021;8(2):130–140. doi: 10.1016/S2215-0366(20)30462-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Rogers A.H., Shepherd J.M., Garey L., Zvolensky M.J. Psychological factors associated with substance use initiation during the COVID-19 pandemic. Psychiatry Res. 2020;293 doi: 10.1016/j.psychres.2020.113407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Wainwright J.J., Mikre M., Whitley P., Dawson E., Huskey A., Lukowiak A., et al. Analysis of drug test results before and after the US declaration of a national emergency concerning the COVID-19 outbreak. JAMA. 2020;324(16):1674–1677. doi: 10.1001/jama.2020.17694. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Park-Lee E., Lipari R., Bose J., Hughes A., Greenway K., Glasheen C., et al. 2018. Substance Use and Mental Health Issues among US-Born American Indians or Alaska Natives Residing on and off Tribal Lands. [Google Scholar]
- 15.Beals J., Manson S.M., Whitesell N.R., Mitchell C.M., Novins D.K., Simpson S., et al. Prevalence of major depressive episode in two American Indian reservation populations: unexpected findings with a structured interview. Am. J. Psychiatr. 2005;162(9):1713–1722. doi: 10.1176/appi.ajp.162.9.1713. [DOI] [PubMed] [Google Scholar]
- 16.Beals J., Novins D.K., Whitesell N.R., Spicer P., Mitchell C.M., Manson S.M., et al. Prevalence of mental disorders and utilization of mental health services in two American Indian reservation populations: mental health disparities in a national context. Am. J. Psychiatr. 2005;162(9):1723–1732. doi: 10.1176/appi.ajp.162.9.1723. [DOI] [PubMed] [Google Scholar]
- 17.Spicer P., Beals J., Croy C.D., Mitchell C.M., Novins D.K., Moore L., et al. The prevalence of DSM-III-R alcohol dependence in two American Indian populations. Alcohol. Clin. Exp. Res. 2003;27(11):1785–1797. doi: 10.1097/01.ALC.0000095864.45755.53. [DOI] [PubMed] [Google Scholar]
- 18.Beals J., Spicer P., Mitchell C.M., Novins D.K., Manson S.M., Team A-S Racial disparities in alcohol use: comparison of 2 American Indian reservation populations with national data. Am. J. Public Health. 2003;93(10):1683–1685. doi: 10.2105/ajph.93.10.1683. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Jones D.S. The persistence of American Indian health disparities. Am. J. Public Health. 2006;96(12):2122–2134. doi: 10.2105/AJPH.2004.054262. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Warne D., Lajimodiere D. American Indian health disparities: psychosocial influences. Soc. Personal. Psychol. Compass. 2015;9(10):567–579. [Google Scholar]
- 21.Joe J.R., Maresca T., Trafzer C.E., Trujillo M.H., Slater E., Waukazoo M., et al. Urban Indian Health Commission; 2007. Invisible Tribes: Urban Indians and Their Health in a Changing World. [Google Scholar]
- 22.Oetting E.R., Beauvais F. Orthogonal cultural identification theory: the cultural identification of minority adolescents. Int. J. Addict. 1991;25(sup5):655–685. doi: 10.3109/10826089109077265. [DOI] [PubMed] [Google Scholar]
- 23.Cummings P. The relative merits of risk ratios and odds ratios. J. Pediat. Adoles. Med. 2009;163(5):438–445. doi: 10.1001/archpediatrics.2009.31. [DOI] [PubMed] [Google Scholar]
- 24.StataCorp. StataCorp LLC; College Station, TX: 2021. Stata Statistical Software: Release 17. [Google Scholar]
- 25.Brave Heart M.Y.H., Lewis-Fernández R., Beals J., Hasin D.S., Sugaya L., Wang S., et al. Psychiatric disorders and mental health treatment in American Indians and Alaska natives: results of the National Epidemiologic Survey on alcohol and related conditions. Soc. Psychiatry Psychiatr. Epidemiol. 2016;51(7):1033–1046. doi: 10.1007/s00127-016-1225-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Centers for Disease Control . CDC.gov; 2022. Trends in Demographic Characteristics of People Receiving COVID-19 Vaccinations in the United States.https://covid.cdc.gov/covid-data-tracker/#vaccination-demographics-trends Available from: [Google Scholar]
- 27.Indian Health Services Coronavirus (COVID-19) IHS.gov. 2021. https://www.ihs.gov/coronavirus/ December 30, 2021:Available from:
- 28.Foxworth R., Redvers N., Moreno M.A., Lopez-Carmen V.A., Sanchez G.R., Shultz J.M. Covid-19 vaccination in American Indians and Alaska natives—lessons from effective community responses. N. Engl. J. Med. 2021;385(26):2403–2406. doi: 10.1056/NEJMp2113296. [DOI] [PubMed] [Google Scholar]
- 29.Allen J., Balfour R., Bell R., Marmot M. Social determinants of mental health. Int. Rev. Psychiat. 2014;26(4):392–407. doi: 10.3109/09540261.2014.928270. [DOI] [PubMed] [Google Scholar]
- 30.Legha R.K., Moore L., Ling R., Novins D., Shore J. Telepsychiatry in an Alaska native residential substance abuse treatment program. J. Telemed. e-Health. 2020;26(7):905–911. doi: 10.1089/tmj.2019.0131. [DOI] [PubMed] [Google Scholar]
- 31.Shore J.H., Brooks E., Savin D., Orton H., Grigsby J., Manson S.M. Acceptability of telepsychiatry in American Indians. J. Telemed. e-Health. 2008;14(5):461–466. doi: 10.1089/tmj.2007.0077. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Roozenbeek J., Schneider C.R., Dryhurst S., Kerr J., Freeman A.L., Recchia G., et al. Susceptibility to misinformation about COVID-19 around the world. J. Royal Soc. Open Sci. 2020;7(10) doi: 10.1098/rsos.201199. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Bavel J.J.V., Baicker K., Boggio P.S., Capraro V., Cichocka A., Cikara M., et al. Using social and behavioural science to support COVID-19 pandemic response. Nat. Hum. Behav. 2020;4(5):460–471. doi: 10.1038/s41562-020-0884-z. [DOI] [PubMed] [Google Scholar]
- 34.Czeisler M.É., Lane R.I., Petrosky E., Wiley J.F., Christensen A., Njai R., et al. Mental health, substance use, and suicidal ideation during the COVID-19 pandemic—United States, June 24–30, 2020. Morbidit. Mortalit. Week. Rep. 2020;69(32):1049. doi: 10.15585/mmwr.mm6932a1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Czeisler M.É., Lane R.I., Wiley J.F., Czeisler C.A., Howard M.E., Rajaratnam S.M. Follow-up survey of US adult reports of mental health, substance use, and suicidal ideation during the COVID-19 pandemic, September 2020. JAMA Netw. Open. 2021;4(2) doi: 10.1001/jamanetworkopen.2020.37665. e2037665-e. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Fear N.T., Seddon R., Jones N., Greenberg N., Wessely S. Does anonymity increase the reporting of mental health symptoms? BMC Public Health. 2012;12(1):1–7. doi: 10.1186/1471-2458-12-797. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary material - Association of factors with increased or decreased substance use since beginning of COVID-19 pandemic in February 2020, assessed among urban American Indians and Alaska Native people Jan-May 2021 (N=3631) [Substance non-users excluded from analysis]