High-Risk Outreach for COVID-19 Mortality Reduction in an Indigenous Community

Abstract

Indigenous populations have been disproportionally affected by COVID-19, particularly those in rural and remote locations. Their unique environments and risk factors demand an equally unique public health response. Our rural Native American community experienced one of the highest prevalence outbreaks in the world, and we developed an aggressive management strategy that appears to have had a considerable effect on mortality reduction. The results have implications far beyond pandemic response, and have reframed how our community addresses several complicated health challenges. (Am J Public Health. 2021;111(11):1939–1941. https://doi.org/10.2105/AJPH.2021.306472)

More than 22% of our rural Native American community was diagnosed with COVID-19 between April 1, 2020 and February 15, 2021, an incidence nearly three times higher than that reported for the rest of the United States.¹ Medical care is provided by a single 22-bed hospital covering an area approximately the size of Delaware. The unique environment and scale of the outbreak necessitated an efficient and creative response to the pandemic.

INTERVENTION

We developed an early outreach and field medicine program to augment contact tracing and act as a bridge between community surveillance and hospital care.

PLACE AND TIME

Our program is based out of Whiteriver, Arizona, and serves approximately 18 000 people living on and around the White Mountain Apache Reservation. Outreach began in March 2020, and remains ongoing.

PERSON

Community members infected with or exposed to SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2, the causative agent of COVID-19).

PURPOSE

The high transmissibility and evasive clinical presentation of SARS-CoV-2 often leads to patients being infected for days before they are traced and tested. Many people then enter medical care late in the disease process, when medical therapeutics are far less effective. Our goal was to quickly identify, monitor, and support those at highest risk of death.

IMPLEMENTATION

COVID-19 was first diagnosed in our community on April 1, 2020. Within weeks, our Indian Health Service facility was identifying more than 250 new cases per 100 000 persons daily. Quickly outpaced, we focused our efforts on finding the highest-risk contacts of each person and testing them immediately.

Our community’s low phone ownership necessitated that most tracing activities be performed in the field. Testing was performed in contacts’ homes at the time of tracing, which often meant that newly discovered cases were still in the asymptomatic period. As tracers returned to deliver test results or drop off isolation supplies, they noticed a concerning pattern among patients in the earliest stages of COVID-19. It was clear that the window of time between asymptomatic infection and critical illness was small for many, and particularly so for elders.

Short-interval follow-up was needed, but the usual methods of health care delivery would have overwhelmed local resources. Our facility was already operating near capacity, seeing about 1000 primary care encounters, 500 emergency department visits, and 50 admissions or transfers each week. We therefore partnered with tribal government to integrate public health outreach and direct medical care.

We built a High-Risk COVID Outreach Program to conduct repeated home visits for every person older than 60 years (and most older than 40 years) with known or suspected to have COVID-19. Hospital staff were paired with community health workers, and each team typically saw about 10 to 20 patients a day. Depending on the case volume, we would dispatch up to four teams each morning. The field physician would provide remote medical supervision for each team while also directly visiting patients with more complicated health needs. Visits for all patients continued until symptoms improved or the patient met criteria for hospital admission (most often ambulatory hypoxia).

Hospitalized patients would also receive a visit from the field team within 24 hours of discharge. This was critically important for patients who had been hospitalized at lower-elevation facilities. Blood oxygen levels often dropped considerably when patients returned home to our mountainous region. Patients could then be rehospitalized locally or set up with home oxygen and close follow-up depending on their risk of further decompensation. The latter option not only allowed patients to remain in a more comfortable environment, but also freed up scarce hospital beds for sicker patients.

EVALUATION

Integrating public health outreach and direct medical care appears to have had a considerable impact on our communitywide case fatality rate (CFR). It was 1.3% as of February 15, 2021, lower than Arizona’s general population CFR of 1.9%, and just over a third of that for all other Native Americans statewide (3.2%).²

Our community had 3904 cases of COVID-19 by February 15, 2021. Of these, 1348 (34.5%) with a test date of January 15, 2021 or earlier were identified through contact tracing and followed in our program. The median age was 55 years (interquartile range = 41–64). This group is among those at highest risk of death from COVID-19,³ yet the CFR for all 1348 patients was 1.3%. This is lower than some reported general-population estimates.⁴

We wondered if we had simply identified more mild cases relative to other communities, but this does not seem to be the case. The COVID-19 hospitalization rate in our community was 24.7% (n = 966), markedly higher than that reported for the rest of the country (6.2%).¹ The CFR for patients in our program aged older than 60 years who were eventually hospitalized was 5.2% (n = 10). A major study reported a CFR of 25.7% in a general hospitalized population with similar age distribution.⁵ Within our own community, the CFR for hospitalized individuals aged older than 60 years who self-presented for medical care outside of the high-risk program was 14.3% (n = 18). The relative risk of death for those in the program was 0.36 (95% confidence interval = 0.17, 0.76; P < .01) despite having the same geographic distribution and similar average age (70 vs 72 years for program participants and those self-presenting, respectively), average body mass index (31.1 vs 29.6 kg/m²), diabetes rate (58.1% vs 58.8%), and hypertension rate (70.1% vs 76.5%).

ADVERSE EFFECTS

No patients reported adverse effects. Staff members encountered impassable roads, wild animals, and the occasional patient who declined home visits (< 1%). By traveling during daylight hours in teams of at least two, the effects of these hazards were minimal.

SUSTAINABILITY

The partnership between our hospital, tribal leaders, and community members created a strong sense of unity and purpose that has strengthened other multidisciplinary partnerships addressing mental health and noncommunicable diseases unrelated to COVID-19. In fact, it led to the formation of a new Preventive Medicine Department that aims to apply the lessons learned from this program to other health conditions that have been challenging to address through usual care delivery models.

Staff members were initially recruited from departments that became less active during the pandemic. The full-time equivalents of one physician and one physician assistant were assigned to the program (of a total medical staff of about 30), along with one nurse and three medical assistants. Partial salary reimbursement came from billing for home visits and telehealth services, with additional investment from our hospital’s operating budget to offset the difference. The benefits from improved survival rates, reduced admissions, and enhanced community partnerships are difficult to quantify in purely financial terms, but we feel they are worth the modest cost of our program.

PUBLIC HEALTH SIGNIFICANCE

COVID-19 remains disproportionately hazardous to rural Indigenous communities. Our experience has shown that brief, frequent evaluations of people at risk for severe COVID-19 can extend the capacity of limited health care resources and have considerable impact on communitywide mortality. As a stand-alone intervention, this program had a better mortality risk reduction than commonly used medical treatments.^5–7 But early outreach alone does not save lives, it simply allows therapeutics and health professionals to work more effectively when there is a scarcity of both.

The dichotomy between public health and medicine has always been artificial. Integrating the two saves lives, and our experience has been a striking reminder of this. As we rebuild our postpandemic health systems, the populations we serve will benefit from expanded coordination between public health agencies and traditional health care facilities. The shared goals have never been more apparent or more necessary.