Food Insecurity in Urban American Indian and Alaska Native Populations During the COVID-19 Pandemic

Background Food insecurity is an important social determinant of health that was exacerbated by the COVID-19 pandemic. Both food insecurity and COVID-19 infection disproportionately affect racial and ethnic minority groups, particularly American Indian and Alaska Native communities; however, there is little evidence as to whether food insecurity is associated with COVID-19 infection or COVID-19 preventive behaviors such as vaccination uptake. The purpose of this study was to evaluate associations between food insecurity, COVID-19 infection, and vaccination status among urban American Indian and Alaska Native adults seen at 5 clinics serving urban Native people. Methods In partnership with health organizations in Alaska, Colorado, Kansas, Minnesota, and New Mexico, the study team conducted a cross-sectional survey in 2021 to assess food security status and attitudes, barriers, and facilitators for COVID-19 testing and vaccination. Logistic regression was used to examine the association of food security status with sociodemographic factors and COVID-19 infection and vaccination status. Marginal standardization was applied to present results as prevalence differences. Results Among 730 American Indian and Alaska Native adults, the prevalence of food insecurity measured during the pandemic was 38%. For participants who reported persistent food security status before and during the pandemic (n=588), the prevalence of food insecurity was 25%. Prevalence of COVID-19 infection and vaccination did not vary by food security status after adjustment for confounders. Conclusions High rates of food insecurity among American Indian and Alaska Native communities likely increased during the COVID-19 pandemic. However, despite the high prevalence of food insecurity, community-led efforts to reduce COVID-19 infection and increase vaccination uptake across Indian Health Service and Tribal healthcare facilities may have mitigated the negative impacts of the pandemic for families experiencing food insecurity. These successful approaches serve as an important reference for future public health efforts that require innovative strategies to improve overall health in AIAN communities.


INTRODUCTION
2][3] Food insecurity, in particular, is a complex and important social determinant of health that affects AIAN communities at signi cantly higher rates than other racial groups. 4The forced relocation of AIAN communities from traditional lands disrupted their way of life and access to traditional healthy foods, contributing to a high prevalence of food insecurity and its subsequent negative health outcomes.Food insecurity is de ned as having unreliable or limited access to affordable, nutritious food and is associated with higher risk of physical and mental health conditions, including type 2 diabetes, hypertension, chronic stress, anxiety, and depression. 5,6AIAN people experience disparities in these health conditions, which also confer a high risk of severe complications, hospitalization, and death following COVID-19 infection. 7,8Despite these challenges, Tribal entities were quick to design and implement COVID-19 vaccination campaigns, which contributed to high rates of vaccination in AIAN communities early in the pandemic. 9ior to the COVID-19 pandemic, AIAN communities across the U.S. experienced a higher prevalence of food insecurity (27%) than African American (21%), Asian American and Paci c Islander (6%), Hispanic (17%), and non-Hispanic White populations (9%). 10However, during the COVID-19 pandemic, food insecurity initially sharply increased throughout the U.S. due to job or income loss, school closures, social isolation, and food supply chain disruptions. 11Food insecurity during the pandemic was also associated with a high likelihood of COVID-19 infection in older U.S. adults and in counties with large African American or large Alaska Native populations. 12,13One study in an Iranian population found that after controlling for confounders, food insecurity increased the risk of COVID-19 threefold. 14Data from the 2021 National Health Interview Survey revealed that food insecurity was independently associated with COVID-19 after controlling for sociodemographic and health-related factors. 15Food insecurity may also negatively impact vaccination uptake. 16Little research has been conducted on the prevalence of food insecurity among AIAN people during the pandemic, despite their elevated prevalence of food insecurity.Likewise, individual-level associations among food insecurity, COVID-19 infection, and COVID-19 vaccination have not been examined in any AIAN population.
We conducted a cross-sectional survey in collaboration with 5 geographically dispersed health organizations that serve urban AIAN populations to examine the association between sociodemographic factors and food insecurity.We also evaluated the association between food insecurity and COVID-19 infection, and food insecurity and vaccination uptake mid-pandemic.We hypothesized that food insecurity would be a risk factor for COVID-19 infection and associated with lower rates of vaccination, after adjusting for potential confounding by sociodemographic, cultural, and health-related factors.As an exploratory analysis, we examined the association between persistent food security and COVID-19 infection and vaccination.

Study Setting
Community Organizations for Natives: COVID-19 Epidemiology, Research, Testing, and Services (CONCERTS) was designed to identify and remove barriers to COVID-19 testing and vaccination among urban AIAN adults.In partnership with 5 geographically dispersed health organizations serving primarily AIAN people in urban settings, the CONCERTS study team created and implemented a cross-sectional survey to assess attitudes, barriers, facilitators, and risk factors for COVID-19 testing and vaccination.The survey was developed using the National Institutes of Health RADx-UP Common Data Elements and PhenX Toolkit with feedback from each site to ensure the appropriateness and relevance of survey questions. 17The urban areas served by the participating sites include Albuquerque, NM, Anchorage, AK, Denver, CO, Minneapolis-St.Paul, MN, and Wichita, KS.Surveys were administered between January 2021 and May 2021.
Eligibility for the study included being a patient at any of the 5 primary care clinics within the prior year, ≥ 18 years of age, identi ed as American Indian or Alaska Native, and not diagnosed with dementia or other serious cognitive issues (ICD-10 codes F01-04, G30, or G31) in the Electronic Health Record system (EHR).Using the EHR, we strati ed sampling by age (18-54 vs. 55 and over) and sent eligible clinic patients an email invitation to participate in the survey.Patients without an email address were mailed invitations.Potential participants received up to 4 reminders over a 14-day period.Our goal was to enroll up to 150 participants per clinic.If this was not achieved after the rst round of recruitment, we selected a new random sample of eligible clinic patients.The survey took approximately 40 minutes to complete, and participants received a $100 gift card as compensation.This study was approved by the Washington State University IRB (#18590), the Alaska Area Institutional Review Board (#2020-11-044), Tribal research review committees, and the Alaska Native Tribal Health Consortium.Tribal approval was also obtained for dissemination.

Survey Measures
Food Insecurity.Food security status was assessed using the U.S. Department of Agriculture's 6-item short form of the Food Security Survey Module. 18This survey has been shown to identify households experiencing food insecurity with reasonably high speci city and sensitivity and minimal bias compared with the 18-item USDA measure, and is less burdensome for respondents.Survey questions are found in Table 3.The number of a rmative responses was summed and scores of 0-1 were coded as food secure whereas scores of 2-6 were coded as food insecure, in alignment with the published scoring procedures.Table 3 USDA 6-item short form food security survey questions.

Questions
Answer Options In the past 12 months, the food that my household bought just didn't last, and there wasn't enough money to get more Often true; Sometimes true; Never true In the past 12 months, my household couldn't afford to eat balanced (healthy) meals.
Often true; Sometimes true; Never true In the past 12 months, did you or anyone in your household ever cut down the size of your meals or skip meals because there wasn't enough money for food?

Yes; No
In the past 12 months, how often did this happen?Almost every month; Some months but not every month; Only 1 or 2 months In the past 12 months, did you ever eat less than you felt you should because there wasn't enough money to buy food?

Yes; No
In the past 12 months, were you ever hungry but didn't eat because you couldn't afford enough food?

Yes; No
We created an additional variable hereafter called "persistent" food security status by combining income change and food security status during the pandemic.Income change was determined using the question: "Since the beginning of the COVID-19 pandemic in February 2020, has your income…" with response categories of stayed the same, increased, or decreased.Persistent food security status was considered insecure if a participant was classi ed as food insecure during the pandemic (using the Food Security Survey Module described above) and reported either the same or increased income since the onset of the pandemic.Persistent food security status was considered secure if a participant was classi ed as food secure during the pandemic and reported either the same or decreased income since the onset of the pandemic.Participants who were classi ed as food insecure during the pandemic and reported decreased income since the onset of the pandemic (n = 127) and participants who were classi ed as food secure during the pandemic and reported the same or increased income since the onset of the pandemic (n = 65) had persistent food security status set to missing since their status may have changed during the pandemic.Participants with persistent food insecurity were assumed to have been food insecure both before and during the pandemic while participants with persistent food security were assumed to have been food secure both before and during the pandemic.

Statistical Analysis
We rst examined the association between sociodemographic factors and food security status using logistic regression.Separate models were t for each sociodemographic factor as the only independent variable in the model and food security status as the dependent variable.Although age was measured as a continuous variable, it was included in the model as a categorical variable (18-35, 36-54, 55-64, 65 + years) for ease of interpretation.Next, we estimated the association of food security status (independent variable) with selfreported COVID-19 infection and vaccination status (dependent variables) using separate logistic regression models for each outcome.For each association, we t an unadjusted model and a model adjusted for health organization (as a categorical variable), sex at birth (as a categorical variable), age (as a continuous variable), education (as a categorical variable), and marital status (as a categorical variable).Marginal standardization was used to calculate the predicted prevalence of each outcome by food security status and the prevalence difference according to food security status. 19,20These results are interpreted as the average difference in the prevalence of the outcome that would be expected comparing a group of individuals experiencing food insecurity to individuals not experiencing food insecurity, adjusted for the distribution of confounders as in the study sample.
Exploratory analyses were restricted to participants for whom we could estimate persistent food security status.We used the logistic models to evaluate associations between persistent food insecurity (vs.persistent food security) and COVID-19 infection and vaccination.All analyses incorporated inverse probability weights to account for age-based sample selection and nonresponse according to age and sex.Weights were scaled so that each of the ve participating sites contributed equally to the nal analyses.All analyses were conducted in STATA and included participants with complete-case data. 21

RESULTS
Of 4,603 eligible clinic patients contacted, 788 (17%) were enrolled and completed the survey.Of these, 730 had complete data to assess food insecurity and were included in the analyses.The prevalence of food insecurity across all health organizations was 38%.Table 1 presents the distribution of sociodemographic factors and the prevalence of food insecurity by sociodemographic factors.The prevalence of food insecurity was lower among participants aged 55 years and older compared to those aged 18 to 35 years.Food insecurity was 19% less prevalent among individuals with a college degree compared to those with less than a high school degree (95% CI: -32%, -6%).The prevalence of food insecurity was also lower among those with higher compared to lower income.Conversely, the prevalence of food insecurity was 11% higher among females compared to males (95% CI: 2%, 20%) and 18% higher among individuals who were divorced, separated, or widowed compared to those who were married or a member of an unmarried couple (95% CI: 7%, 30%).Similarly, individuals who were unemployed and disabled had 20% (95% CI: 5%, 35%) and 26% (95% CI: 9%, 44%) higher prevalence of food insecurity, respectively, than their counterparts who were employed.The association of food insecurity with COVID-19 infection and vaccination status, unadjusted and adjusted for sociodemographic characteristics is shown in Table 2.The prevalence of COVID-19 infection was 4% higher among individuals classi ed as food insecure compared to those classi ed as food secure (95% CI: -3%, 12%), and adjustment for confounders had no impact on the magnitude of the difference (PD: 4%, 95% CI: -3%, 11%).Prevalence of COVID-19 vaccination was 11% lower among individuals classi ed as food insecure compared to those classi ed as food secure (95% CI: -21%, -2%); however, adjustment for confounders reduced the magnitude of the difference (PD= -2%; 95% CI: -10%, 6%).-2% (-13%, 9%) 1 Unweighted sample size, results weighted for sampling and nonresponse; 2 Adjusted for health organization clinic, sex, age, education, and marital status; 3 Restricted to the subset of respondents categorized as having persistent status (food insecure, food secure) before and during the pandemic.PD = prevalence difference; CI = con dence interval We included 558 participants who were classi ed as having persistent food security or insecurity status in an exploratory analysis to examine associations between food insecurity and COVID-19 testing and vaccination speci cally for participants whose food security status likely did not change as a result of the pandemic.Among those with persistent food security status (i.e., those likely to have had the same food security status prior to and during the pandemic), the prevalence of food insecurity captured at the time of the study was 25%.Sociodemographic factors among those who had persistent food security status, strati ed by food security status at the time of the survey, are presented in Table 4.The exploratory analysis showed similar associations between persistent food insecurity and COVID-19 infection and vaccination as the full cohort analysis (Table 2).Total household income in past 12 months, % <$15,000 14% 39% 1 Unweighted sample size, results weighted for sampling and nonresponse some degree of food insecurity over a 4-month period in late 2020. 23,24Similarly, few studies have examined associations between food insecurity and COVID-19 infection or vaccine uptake.However, our ndings that food insecurity was not associated with COVID-19 infection or vaccination, differ from prior research among other populations.Prior studies in older adults and Latinx families suggest higher prevalence of COVID-19 infection in households experiencing food insecurity. 12,30However, in one of the studies, the relationship was no longer signi cant after adjusting for house size and number of residents. 30ile research in other communities across the U.S. indicates that people with food insecurity were less likely to receive COVID-19 vaccines, this was not evident in this sample of urban AIAN adults after adjusting for confounders. 31,32Many Tribal organizations responded swiftly to the COVID-19 pandemic and were early adopters of public health vaccination campaigns, with timely distribution of vaccines to Indian Health Service and Tribal healthcare facilities. 9Federal data from April 2021 showed that more AIAN people (32%) were fully vaccinated against COVID-19 than Whites (19%), Asian Americans (16%), and African Americans (12%). 33espite a high prevalence of food insecurity in this sample, the community-led vaccination campaigns across Indian Health Service and Tribal healthcare facilities may have mitigated the negative impacts of food insecurity on COVID-19 preventive behaviors such as vaccination. 9However, the unadjusted results suggest that public health strategies that direct resources to food assistance programs may be useful in improving access to COVID-19 vaccinations for families experiencing food insecurity (for example, offering vaccinations at food pantries or Supplemental Nutrition Assistance Program sites).
The implementation and scaling up of secondary interventions that address food insecurity among urban AIAN communities is a growing priority.Two interventions have reduced food insecurity among AIAN children. 34,35ne intervention with the Navajo Nation included a fruit and vegetable prescription and empowerment program, while the other intervention aimed to foster Indigenous culture and pride, including culturally-based gardening practices.The Indigenous Food Circle described an indigenous-led network to address the immediate needs of food insecurity exacerbated during the pandemic 36 that highlighted the importance of advocacy, Indigenous-led initiatives, and culturally appropriate frameworks to address the systemic issues of food insecurity in AIAN communities.A study conducted among Alaska Native women receiving the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) assistance showed that intake of traditional foods was associated with improved diet quality. 37While an association between food security and traditional food intake was not observed, incorporating efforts to increase access to traditional foods and subsistence practices could strengthen programs and policies aimed at reducing health disparities experienced by AIAN communities.
These studies suggest multiple strategies may reduce food insecurity among Native communities.As the food environment and federal nancial interventions (e.g., stimulus checks and child tax credit) are discontinued, understanding the impacts on food insecurity status is critical to reducing food insecurity among AIAN people.Lastly, validation of standardized measurements of food insecurity among rural and urban AIAN communities is critical to create appropriate strategies to combat food insecurity and its negative impacts on health. 4mitations of the present study include a cross-sectional design, which prevents our ability to establish causation, temporality, and proximity between food insecurity and COVID-19 infection or vaccination.Surveys were only sent to individuals who visited the clinic in the year prior to the pandemic, so our sample may have included more people who seek out consistent healthcare.Further, our response rate was low, raising the possibility of unknown selection bias.9][40] These practices are not re ected in the U.S. Department of Agriculture food insecurity screening questions, which focus on the ability to purchase food, rather than provide food.As experiencing food insecurity likely changed for many households during the pandemic, we developed a variable combining income change and food security as a proxy measure to assess change.While exploratory, this proxy measure may not have accurately assessed change, and additional research exploring food insecurity as a uctuating (and less stable) variable is necessary.Lastly, we collected data from January to May 2021, about one year after the onset of the pandemic.Access to vaccines varied across the partner sites, and we were not able to distinguish whether participants were not vaccinated due to access or availability.However, our study included a large sample of AIAN adults across the age range from geographically diverse areas, making it the largest and most comprehensive study to date of food insecurity among urban AIAN communities.

CONCLUSIONS
This study sheds light on the pervasive issue of food insecurity among AIAN communities during the COVID-19 pandemic.Despite the high prevalence of food insecurity among the study participants, our ndings show that food insecurity was not signi cantly associated with an increased risk of COVID-19 infection or decreased vaccination uptake.While the community-led efforts to alleviate the impacts of the pandemic appear to have mitigated the impact of food insecurity on COVID-19 preventive behaviors, the unadjusted results suggest that additional public health strategies may also be bene cial for preventing COVID-19 infection among those experiencing food insecurity.Addressing the systemic issues of food insecurity in AIAN communities will require a multifaced approach that incorporates culturally appropriate initiatives and measures and efforts to increase access to traditional foods.

Abbreviations
1) "Has a doctor, nurse or other health care professional ever told you that you were infected with or have COVID-19?"and 2) "Did any test indicate that you had COVID-19?"A participant was considered to have had a COVID-19 infection if s/he responded yes to either question.
COVID-19 Vaccination.COVID-19 vaccination status was assessed by asking "Have you received a COVID-19 vaccine?" Response categories included yes or no.Sociodemographic Factors.Sociodemographic characteristics included in the analysis were self-reported age, sex assigned at birth, current marital status, level of completed education, current employment status, and household income in the past 12 months.

Table 2
Association of food insecurity with COVID-19 infection and vaccination status among urban American Indian and Alaska Native survey participants, January-May 2021.

Table 4
Sociodemographic factors according to food security status among participants classi ed as having persistent food security status before and during the pandemic.