Precision public health offers the promise of improving health equity by delivering the “right intervention at the right time, every time to the right population.”1 But the COVID-19 pandemic underscored how far the United States is from meeting that promise, especially for marginalized urban and rural American Indian/Alaska Native (AI/AN) populations. The reasons for this are many, including failure to collect relevant data, barriers to data dissemination, and less than optimal use of data to effectively inform public policy. In Table 1, we present a report card on COVID-19 data availability compiled by the Urban Indian Health Institute in 2021. This exposes some of the reporting limitations as experienced by the AI/AN populations.2 However, this report card does not begin to capture the long-term consequences for AI/AN persons deriving from the loss of so many elders. Gone are key leaders, including the keepers of unwritten language, and enduring are disruptions of tribes, nations, bands, pueblos, communities, native villages, and families who face complicated waves of grieving. For AI/AN populations, who are especially sensitive to threats of genocide, COVID-19 reawakens awareness of structural racism as a weapon for their destruction.
TABLE 1—
Report Card on the Centers for Disease Control and Prevention (CDC) and State COVID-19 Surveillance Data for American Indians/Alaska Natives: United States, January 2020–January 2021
| State Information | State-Reported COVID-19 Information | CDC COVID-19 National Surveillance Data | Overall Grade | |||
| State | AI/AN Populationa | Is AI/AN Population Included on State Dashboard? | Grade (% of Cases With Complete Racial Information Reported on Dashboard) | Grade (% of Confirmed Cases From the State Reported to CDC) | Grade (% of Confirmed Cases With Complete Racial Information Reported on CDC Database) | |
| Alabama | 69 283 | No | F (49) | D– (62) | D– (62) | F (43) |
| Alaska | 148 222 | Yes | D– (62) | . . .b | F (59) | C (74) |
| Arizona | 458 422 | Yes | C– (72) | A (95) | D (63) | B (83) |
| Arkansas | 61 824 | Yes | B– (82) | A (95) | B (85) | A– (91) |
| California | 1 089 251 | Yes | D (65) | A+ (99) | F (37) | C (75) |
| Colorado | 159 162 | Yes | C (73) | C (76) | F (38) | C– (72) |
| Connecticut | 43 195 | Yes | F (53) | B (84) | F (40) | D+ (69) |
| Delaware | 13 516 | No | B+ (89) | A (96) | F (0) | F (46) |
| Florida | 219 895 | No | C (76) | F (50) | D (65) | F (48) |
| Georgia | 122 051 | Yes | C (76) | C+ (77) | F (43) | C (74) |
| Hawaii | 37 751 | No | D (65) | B (83) | F (50) | F (49) |
| Idaho | 51 467 | Yes | F (59) | A– (92) | F (59) | C+ (77) |
| Illinois | 141 473 | Yes | D+ (67) | A– (92) | F (52) | C+ (78) |
| Indiana | 66 617 | No | C (75) | D (66) | D– (61) | F (50) |
| Iowa | 33 753 | Yes | D (65) | B+ (89) | D+ (69) | B– (81) |
| Kansas | 69 645 | Yes | C+ (77) | B– (82) | C+ (78) | B (84) |
| Kentucky | 38 568 | No | C (74) | F (40) | B– (80) | F (48) |
| Louisiana | 65 461 | Yes | B (85) | F (6) | F (59) | D (63) |
| Maine | 20 865 | Yes | B (84) | A (94) | B (85) | A– (91) |
| Maryland | 81 228 | No | B– (81) | F (41) | F (35) | F (39) |
| Massachusetts | 75 027 | Yes | D+ (67) | A+ (98) | F (59) | B– (81) |
| Michigan | 158 391 | Yes | C– (70) | F (59) | B– (80) | C+ (77) |
| Minnesota | 124 345 | Yes | B+ (88) | A+ (100) | B (83) | A (93) |
| Mississippi | 31 669 | Yes | B (83) | F (29) | B (84) | C (74) |
| Missouri | 87 760 | Yes | D– (60) | F (13) | C– (70) | D– (61) |
| Montana | 90 472 | Yes | C– (70) | A (96) | C– (72) | B (84) |
| Nebraska | 43 760 | Yes | F (58) | F (27) | C+ (78) | D (66) |
| Nevada | 85 953 | Yes | B+ (87) | B+ (89) | F (53) | B– (82) |
| New Hampshire | 12 534 | No | F (57) | F (20) | B– (81) | F (39) |
| New Jersey | 102 441 | No | D (63) | A+ (98) | F (48) | F (52) |
| New Mexico | 257 858 | Yes | C (76) | F (35) | D (64) | D+ (69) |
| New York | 318 858 | No | F (0) | D+ (69) | F (39) | F (27) |
| North Carolina | 245 724 | Yes | C– (70) | A+ (99) | C– (70) | B (85) |
| North Dakota | 51 664 | Yes | D (63) | F (50) | F (6) | F (55) |
| Ohio | 107 899 | Yes | C– (70) | . . .b | C (73) | B– (81) |
| Oklahoma | 553 509 | Yes | C (74) | B– (80) | C+ (77) | B (83) |
| Oregon | 146 851 | Yes | F (56) | A (94) | F (57) | C+ (77) |
| Pennsylvania | 117 073 | No | F (59) | A+ (99) | D (63) | F (55) |
| Rhode Island | 20 190 | Yes | C (75) | F (27) | F (27) | F (57) |
| South Carolina | 58 171 | No | D (66) | A+ (99) | D+ (67) | F (58) |
| South Dakota | 92 686 | Yes | B+ (89) | B (84) | B (84) | B+ (89) |
| Tennessee | 76 883 | Yes | C (73) | A+ (99) | C– (71) | B (86) |
| Texas | 485 363 | No | F (3) | F (3) | C– (72) | F (20) |
| Utah | 73 697 | Yes | B (83) | . . .b | B– (82) | B+ (88) |
| Vermont | 8 088 | Yes | B+ (88) | A– (92) | A– (90) | A (93) |
| Virginia | 109 216 | Yes | C (74) | B– (81) | D (64) | B– (80) |
| Washington | 264 596 | Yes | F (54) | A+ (100) | F (43) | C (74) |
| West Virginia | 15 137 | No | D (64) | F (8) | A– (92) | F (41) |
| Wisconsin | 106 202 | Yes | B (86) | B– (82) | B+ (87) | B+ (89) |
| Wyoming | 22 024 | Yes | F (57) | F (2) | B (83) | D– (61) |
| United States Overallc | 6 935 690 | C– | D+ (68) | D+ (69) | D (63) | D– (68) |
Note. AI/AN = American Indian/Alaska Native, defined as American Indian/Alaska Native only + American Indian/Alaska Native in combination.
Source. Urban Indian Health Institute.2
aPopulation numbers are believed to be an undercount and should not be interpreted to represent tribal enrollment numbers.
bState reported a greater number of cases to the CDC than reported on dashboard; therefore, we were unable to identify the percentage of confirmed cases sent to the CDC.
cUnited States overall grades averaged across states: 72% of states included AI/AN populations on their scoreboards; across states, 68% of state-reported cases included complete racial information reporting and 69% of confirmed state cases were reported to the CDC; 63% of confirmed cases in the CDC databases had complete racial information. Averaging these percentage-based grades is a score of 68, receiving a grade of D–.
Problems With Data Collection and Compilation
At the beginning of the COVID-19 pandemic, there could be no doubt that the AI/AN community would be a highly vulnerable population for whom accurate data could drive effective policy interventions. Two factors were already known. One is that AI/AN populations are often made invisible in data collection efforts—for example, AI/AN persons are sometimes classified as other or White race.3 This is especially problematic for the 71% of AI/AN persons who live in urban areas and access services within mainstream health systems.4 The second is that, in the service of federal, state, and local laws related to data privacy and protection, results for small populations are often suppressed or aggregated with others. This reduces capacity to provide policy-setting predictions.2,3 But a third factor was also missing data on race completely—early efforts at COVID-19 tracking failed to either measure or report the race of infected individuals.5
A fourth challenge for public health planning lies in the creation of an appropriate social vulnerability index (SVI). In a recent study of COVID-19 rates among AI/AN persons,6 the authors combined the Centers for Disease Control and Prevention’s (CDC) SVI with a set of risk conditions unique to tribal conditions within New Mexico. Using census and other data sources, the study added new vulnerability measures (e.g., absence of telephone service or Internet, incomplete plumbing, presence of abandoned uranium mines) to demonstrate that expanded SVI measures are highly correlated with COVID-19 infection rates at the zip code level. These authors also showed that higher levels of racial segregation and density of racial/ethnic minority populations are predictive of higher COVID-19 infection rates. Finally, using data from 23 states, Hatcher et al.7 found that underlying conditions for the AI/AN population may explain why early pandemic infection rates among the AI/AN community were 3.5 times that of Whites. They speculate that both underlying health conditions and reliance on shared transportation contributed to the early spread of the virus within AI/AN populations.
Data Dissemination Barriers
Even when appropriate data are collected, public health laws and policies can work to limit data sharing with the 12 tribal epidemiology centers (TECs), even though data are shared with states. TECs are the public health organizations of tribal and urban AI/AN communities, serving similar roles as local public health departments. A recent US Government Accountability Office (GAO) report found that more than half of TECs experienced data access problems, as some CDC and Indian Health Service officials were unaware that the Department of Health and Human Services (DHHS) is required by federal law to provide data in its possession to TECs.8 Even when these requests were addressed, some took a year to fulfill. This is particularly problematic for smaller tribal communities that do not have the resources to track and update local community data in a timely manner; the CDC and Indian Health Service are usually the only reliable and timely data source for these underresourced tribal nations.
Although the federal government’s public health laws are designed to protect the security, privacy, and confidentiality of health data, less well-known is that the federal government also is bound by its trust doctrine to assist tribal groups, such as TECs, in matters of well-being. The trust doctrine reflects federal responsibility to Indian nations, requiring that it support tribal self-government and economic prosperity. Tribal nations, in turn, have the responsibility to provide health care services and ensure the survival and welfare of Indian tribes and people.3,7,9 Consequently, the GAO report recommended that the DHHS work to resolve policy lapses and that both the CDC and Indian Health Service develop clear guidance for data sharing with TECs.
Although this is a necessary set of actions, COVID-19 data equity might be best achieved by also engaging the principles of the Global Indigenous Data Alliance, the voices of indigenous data warriors. CARE (Collective benefit, Authority to control, Responsibility, and Ethics) principles for indigenous data governance were developed as a framework for data management and sharing. These are complementary with FAIR (Findable, Accessible, Interoperable, Reusable) principles.10 TEC’s requests for COVID-19 data are consistent with CARE principles and reflect that data equity is best served when it advances indigenous innovation and governance efforts that emanate from sovereignty and self-determination. Recognition of CARE principles is also needed for state and local data entities. Indeed, adhering to the Global Indigenous Data Alliance’s FAIR and CARE principles is fundamental to data equity. How to apply these principles is well laid out in the Urban Indian Health Institute report.2
Data Usability
Data equity requires that data be usable and meet clinically meaningful use standards for both public health entities and populations. CARE principles also underscore that data collection and its use must not bring unnecessary harm to those providing the data. In particular, identifying high infection rates for specific tribal nations and other small communities could generate potential risks for targeted racism and violence. Data collection and surveillance methods must plan for this possibility and act to safeguard vulnerable populations.11 It is also essential that AI/AN and other small and marginalized populations be included in this planning to ensure effective use and to protect against unanticipated harm.12 County and local health and public health entities can optimize careful public messaging by not expecting data to speak for themselves, but rather helping people to have awareness of what the data mean. For example, when data are aggregated into personally unrecognizable categories (e.g., “other”) or grouped where those within the group are heterogenous for risks and resources,13 as is true of AI/AN groups, it is difficult to effectively use public health data for risk reduction. Data equity requires “sense making” by public health agencies in which numbers are interpreted in the context of the lives, risks, and stories of those whom the data are meant to help.
Finally, data equity also obligates us to plan for exceptions to the “average” case and to recognize the unique needs of small populations. At the start of the COVID-19 pandemic, when our public health knowledge base and armamentarium were sparse, enacting stay-at-home orders and public health advisories such as masking, hand washing, and physical distancing were reasonable strategies for the whole. But the viability of these strategies for communities that lack indoor plumbing, are isolated from public health messaging, experience water scarcities, and whose households lack isolating spaces were underappreciated challenges. We could have done better. For example, tribal reservations could have been seen as warranting deliveries of personally protective equipment and water resources.14 An early study showed that initially high rates of COVID-19 infection in rural tribal groups were related to the prevalence of indoor plumbing on tribal reservations and English language use.15 These results illustrate that housing and general infrastructure information might be exceptionally predictive of infection rates for some AI/AN populations. Additionally, the prevalence of public health messaging and notices in a few languages may not have served certain populations well and may have left some with little to no information—at least at the early stages of a pandemic. However, AI/AN communities recognized the deficits of this messaging and quickly rallied scarce resources to successfully launch COVID-19 prevention campaigns that used regional context, language, and imagery, resulting in better-informed AI/AN communities nationwide.
We would be remiss to end this article without highlighting the incredible public health success, born of the principles of self-determination, that occurred when AI/AN tribal and urban groups asked the federal government to step aside and provide them with the COVID-19 vaccine.16 These groups engaged in consultation and education-first activities with AI/AN persons, which resulted in vaccination rates in some tribal groups as high as 80% to 90%. This is an achievement that many states cannot claim. FAIR and CARE principles are an effective public health tool for COVID-19 prevention, mitigation, and recovery that may be integral to protecting AI/AN and other racial/ethnic, rural, and low-income groups in the next pandemic. We cannot withstand another million deaths.
ACKNOWLEDGMENTS
Partial funding for this work was provided by the National Institute of Minority Health Disparities, National Institutes of Health (NIH; grant MD 006923) and the National Institute of Mental Health, NIH (grant MH 115344).
CONFLICTS OF INTEREST
None of the authors have any conflicts of interest to report.
Footnotes
REFERENCES
- 1.Khoury M.2018. https://blogs.cdc.gov/genomics/2018/05/15/precision-public-health-2
- 2.Urban Indian Health Institute. 2021. https://www.uihi.org/projects/data-genocide-of-american-indians-and-alaska-natives-in-covid-19-data
- 3.Small-Rodriguez D, Akee R. Identifying disparities in health outcomes and mortality for American Indian and Alaska Native populations using tribally disaggregated vital statistics and health survey data. Am J Public Health. 2021;111(suppl 2):S126–S132. doi: 10.2105/AJPH.2021.306427. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Urban Indian Health Institute, Seattle Indian Health Board. 2021. https://www.uihi.org/resources/community-health-profile-national-aggregate-of-urban-indian-organization-service-areas
- 5.Krieger N, Testa C, Hanage W. US racial and ethnic data for COVID-19 cases: still missing in action. Lancet. 2020;396(10261):E81. doi: 10.1016/S0140-6736(20)32220-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Yellow Horse A, Yang T, Huyser K. Structural inequalities established the architecture for COVID-19 pandemic among Native Americans in Arizona: a geographically weighted regression perspective. J Racial Ethn Health Disparities. 2022;9(1):165–175. doi: 10.1007/s40615-020-00940-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hatcher S, Agnew-Brune C, Anderson M, et al. COVID-19 among American Indian and Alaska Native persons—23 states, January 31–July 3, 2020. MMWR Morb Mortal Wkly Rep. 2020;69(34):1166–1169. doi: 10.15585/mmwr.mm6934e1d. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.US Government Accountability Office. 2022. https://www.gao.gov/products/gao-22-104698
- 9.Hoss A. Federal Indian law as a structural determinant of health. J Law Med Ethics. 2020;47(S4):34–42. doi: 10.1177/1073110519898041. [DOI] [PubMed] [Google Scholar]
- 10.GO FAIR. 2016. https://www.go-fair.org/fair-principles
- 11.Ford C, Amani B, Harawa N, et al. Adequacy of existing surveillance systems to monitor racism, social stigma and COVID inequities: a detailed assessment and recommendations. Int J Environ Res Public Health. 2021;18(24):13099. doi: 10.3390/ijerph182413099. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Carroll S, Akee R, Chung P, et al. Indigenous peoples’ data during COVID-19: from external to internal. Front Sociol. 2021;6:617895. doi: 10.3389/fsoc.2021.617895. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Anderson T, Awan V, McDonnell D, Mousseau H.2021. https://harderco.com/foundations-of-data-equity-learnings-from-a-workshop
- 14.US Government Accountability Office. Drinking water and wastewater infrastructure: opportunities exist to enhance federal agency needs assessment on tribal projects. 2018. https://www.gao.gov/products/gao-18-309
- 15.Rodriguez-Lonebear D, Barceló N, Akee R, Carroll S. American Indian reservations and COVID-19: correlates of early infection rates in the pandemic. J Public Health Manag Pract. 2020;26(4):371–377. doi: 10.1097/PHH.0000000000001206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Haroz E, Kemp C, O’Keefe V, et al. Nurturing innovation at the roots: the success of COVID-19 vaccination in American Indian and Alaska Native communities. Am J Public Health. 2022;112(3):383–387. doi: 10.2105/AJPH.2021.306635. [DOI] [PMC free article] [PubMed] [Google Scholar]