Vaccine-Associated Shifts in SARS-CoV-2 Infectivity Among the Native Hawaiian and Other Pacific Islander Population in Hawaii

Abstract

Native Hawaiians and other Pacific Islanders (NHPIs) across the country have experienced significant disparities because of the COVID-19 pandemic. The Pacific Alliance Against COVID-19 used a community-based participatory approach involving academic and community partners to expand sustainable COVID-19 testing capacity and mitigate the severe consequences among NHPI communities in Hawaii. We describe the approach of this one-year study, some of the results, and how the data are being used to inform next steps for the communities. Clinical Trials.gov identifier: NCT04766333. (Am J Public Health. 2022;112(S9):S896–S899. https://doi.org/10.2105/AJPH.2022.306973)

Communities nationwide are experiencing a resurgence in COVID-19. The stagnating COVID-19 vaccination rates, emergence of new variants, reversal of prevention mandates, and overall pandemic fatigue contribute to this resurgence, especially in vaccine-hesitant populations.¹ For Native Hawaiians and other Pacific Islanders (NHPIs) and Filipinos across the country, the pandemic has had a profound impact.^2–4 In Hawaii from March 2020 to October 2021, NHPIs and Filipinos made up the majority of COVID-19 cases in Hawaii.⁵ Currently, hospitalization and death rates remain significantly higher among NHPIs and Filipinos, suggesting that these communities avoid testing and vaccination.

INTERVENTION AND IMPLEMENTATION

To address these disparities, we assembled a multidisciplinary team of academic and community investigators, along with long-standing community partners across Hawaii, to form a collaborative called the Pacific Alliance Against COVID-19 (PAAC) to participate in the National Institutes of Health Rapid Acceleration of Diagnostics-Underserved Populations (RADx-UP) Initiative.⁶ Partners included the Accountable Healthcare Alliance of Rural Oahu (AHARO), a consortium of five federally qualified community health centers (FQHCs), and public K-12 schools (kindergarten through grade 12) that serve communities on three islands with large proportions of NHPIs. Our objectives were to (1) increase COVID-19 testing in communities across Hawaii and (2) collect and analyze data to inform the development of public health policies.

Once funded, the budget was divided equally between academic and community partners. Each FQHC hired and trained community members to implement the study and build on community engagement and COVID-19 mitigation already being implemented. Participants who were enrolled in our ongoing COVID-19 testing program consented to participate in the RADx-UP survey, which queries demographics, vaccination status, trust in sources of COVID-19 information, and attitudes toward COVID-19 vaccination. Participants then completed rapid COVID-19 antigen tests (BinaxNow, Abbott, ME) administered by trained PAAC staff.

PLACE, TIME, AND PERSONS

The study was implemented from March 2021 through March 2022 in communities served by the AHARO FQHCs. In 2021, these FQHCs served 71 698 patients, the majority of whom were economically disadvantaged NHPIs and Asians, primarily Filipino.

PURPOSE

The primary goal of this study was to strengthen Hawaii community–academic partnerships, building on preexisting networks and relationships, to create a sustainable research program that enables data-driven strategies to reduce COVID-19 health disparities.

EVALUATION AND ADVERSE EFFECTS

Over the study period, PAAC trained more than 45 community members across the five sites to implement the study and provide bidirectional feedback to both researchers and FQHC administrators. In total, PAAC implemented 310 testing events, facilitating 16 064 COVID-19 antigen and reverse transcription polymerase chain reaction (RT-PCR) tests among 5662 individuals. We also disseminated an elective survey among these participants, 23% of whom returned completed responses. The rate of COVID-19 positivity tracked closely with that of the state, peaking at 4% during the Delta variant period and 20% during the Omicron variant period. Our PAAC program engaged a high percentage of NHPIs. Although NHs and PIs make up 21% and 4% of the population, respectively, 54% and 13% of tests were among individuals of NH or PI ancestry, respectively.

To examine changes in attitudes and behaviors related to COVID-19 over time, we further characterized our cohort of 1304 individuals from whom we tested and received complete survey information. Notably, we observed a significantly higher proportion of vaccinated individuals during the Omicron compared with the Delta period (74% vs 30%, respectively). However, even during the Omicron period, NHPIs had the lowest percentage of vaccinated individuals (69%) compared with all other ethnic groups (74%–89%).

Despite increased vaccinations, there was an increase in COVID-19 positivity during the Omicron period compared with the Delta period (29% vs 7.8%, respectively). To better understand this, we further stratified individuals based on timing of vaccination, comparing those more recently vaccinated (≤ 6 months from study entry) with those less recently vaccinated (> 6 months from study entry). We observed that 31% of individuals less recently vaccinated tested positive for SARS-CoV-2 (the causative agent of COVID-19) during the Omicron period, compared with only 17% of those more recently vaccinated (Table 1).

TABLE 1—

COVID-19 Testing and Positivity, by Racial/Ethnic Status and Vaccination Status: Hawaii, March 2021–March 2022

COVID-19 Antigen Tests	Overall, No. (%)	Predominant Variant Period, No. (%)		P
		Delta	Omicron
Total	2893	1595	1298
Racial/ethnic group
White	292 (10)	176 (11)	116 (9)	.3
NHPI	1927 (67)	1049 (66)	878 (67)
Asian	503 (17)	272 (17)	231 (18)
Other	171 (6)	98 (6.1)	73 (5.6)
COVID-19 positivity by racial/ethnic group				< .001
Overall	256 (8.8)	33 (2.1)	223 (17)
White	20 (7)	2 (1.1)	18 (16)
NHPI	180 (9.3)	26 (2.5)	154 (18)
Asian	42 (8.4)	4 (1.5)	38 (16)
Other	14 (8.2)	1 (1.0)	13 (18)
COVID-19 vaccination status				< .001
Less recently vaccinated	430 (15)	232 (15)	198 (15)
More recently vaccinated	850 (29)	294 (18)	556 (43)
Never vaccinated	1613 (56)	1069 (67)	544 (42)
COVID-19 positivity by vaccination status				< .001
Less recently vaccinated	61 (14)	0 (0)	61 (31)
More recently vaccinated	95 (11)	3 (1.0)	92 (17)
Never vaccinated	100 (6.2)	30 (2.8)	70 (13)

Note. NHPI = Native Hawaiian and other Pacific Islander; PAAC =  Pacific Alliance Against COVID-19. The PAAC testing program facilitated 16 064 tests. This table reflects COVID-19 tests, positivity rate, and vaccination status by racial/ethnic group for 18% of the tests (n = 2893) from participants who completed all elective survey questions. The percentages shown for racial/ethnic group and vaccination status are computed among such groups for the overall, Delta, and Omicron periods, respectively. The percentages shown for COVID-19 positivity are the positivity rate for the specific racial/ethnic or vaccination status group for the overall, Delta, and Omicron periods, respectively. The P values shown are from tests of differences in population proportions between the Delta and Omicron periods using the χ² test.

Given the lower vaccine coverage within the NHPI population, we examined the extent to which trust might be associated with vaccine uptake. Data on perceptions of trust of a variety of COVID-19 information sources were collected using the standardized RADx-UP Likert Scale ranging from “not at all” to “a little,” “somewhat,” or “a great deal.” We compared trust in “official” sources of information—which included the US government, health care providers, the US Coronavirus Task Force, and news on radio, TV, online, and newspaper—with trust in “unofficial” sources, which included coworkers or other acquaintances, social media contacts, close friends and family, and faith leaders. We observed that trust in each “official” source independently increased the probability of vaccination significantly higher than trust in each “unofficial” source (Figure 1). The association between trust in faith leaders and vaccine uptake was not significant.

FIGURE 1—

Marginal Effects (Probability Changes) in Vaccination Uptake for Different Sources of COVID-19 Information: Hawaii, March 2021–March 2022

Note. CI = confidence interval. Although trust in almost all types of information sources evaluated is significantly positively associated with vaccine uptake, trust in official sources of information—such as the US Coronavirus Task Force, doctors, and health care providers—exhibits larger marginal effects than that of unofficial sources such as friends and family members and contacts in social media.

*P < .01.

For many, shifting levels of trust during the study were associated with the probability of receiving vaccination (Figure 1). For instance, individuals who increased their trust in COVID-19 information from their health care provider from “not at all” to “a great deal” increased their probability of vaccination by 63% (P < .001; 95% confidence interval [CI] = 52%, 75%). We note that trust in each source of official information had a larger marginal effect on vaccine uptake than each source of unofficial information.

Additionally, composite indexes of trust were computed by averaging the values of the four official and four unofficial sources of COVID-19 information. Individuals whose trust shifted from “not at all” to “a great deal” in all four official information sources increased their probability of vaccination by 76% (P < .001; 95% CI = 64%, 89%; Figure 1). In comparison, individuals who increased their trust from “not at all” to “a great deal” in all four unofficial sources of information increased their probability of vaccination by 30% (P < .001; 95% CI = 18%, 42%). These results suggest that official sources of information have a larger effect on vaccine uptake than unofficial sources of information.

SUSTAINABILITY

Academic and FQHC partners have found the partnership to be productive and beneficial to their communities. As such, the partners have agreed to continue their collaborative work. To date, they have received two additional federal grants to directly inform ongoing COVID-19 mitigation programs at the FQHC communities.

PUBLIC HEALTH SIGNIFICANCE

Academic–community partnerships can provide actionable data effectively deployed to inform community strategies that mitigate the impact of the COVID-19 pandemic. The significant association between trust in official sources of COVID-19 information that we observed underscores the crucial influence that the health care sector has on individual-level decision-making among vaccine-hesitant populations, including NHPIs. Fostering trust in official sources of information may be essential to promoting vaccine uptake. This study provides an example of the application of the granular RADx-UP common data elements, collected from an understudied population, to guide community-relevant and culturally relevant interventions that reduce COVID-19 disparities.

We recognize that our major findings are largely driven by understudied NHPIs, with more limited representation of Hawaii’s other major race/ethnic groups. Another limitation is that the strong association between trust in official sources of COVID-19 information and vaccine uptake observed (Figure 1) was derived from cross-sectional data analysis. Thus, the degree to which trust in these sources of information might serve as mediators or modifiers of vaccine uptake requires further examination, including change over time. Such longitudinal assessments remain especially important considering changes in COVID-19 public health policy and lifting of many pandemic restrictions.