Abstract
Bristol Bay in Alaska is home to the world’s largest commercial salmon fishery. During an average fishing season, the population of the Bristol Bay region more than doubles as thousands of workers from out of state converge on the fishery. In the months leading up to the 2020 commercial fishery opening, as the COVID-19 pandemic exploded worldwide, great uncertainty existed about the health risks of opening the fishery. Bristol Bay residents had not yet experienced any cases of COVID-19, yet the livelihoods of most were closely tied to the commercial fishery opening. To better understand how COVID-19 risk perceptions affected decisions to participate in the fishery, we administered an online survey to community members and fishery participants. We collected standard socioeconomic data and posed questions to gauge risk perceptions related to COVID-19. We find that COVID-19 risk perceptions vary across race/ethnic groups by residency and income. People with below median income who are members of minority groups—notably, non-resident Hispanic workers and resident Alaska Native respondents—reported the highest risk perceptions related to COVID-19. This study highlights the important linkages among risk perceptions, socioeconomic characteristics, and employment decisions during an infectious disease outbreak.
1. Introduction
The COVID-19 epidemic altered the social fabric of the world. It is estimated that 18.2 million people worldwide died from the virus between January 1, 2020, and December 31, 2021, with over 1.1 million deaths in the United States alone [1]. In 2020, the global economy contracted by 3.3% [2], and personal consumption spending in the United States fell by 12.6% between February and April 2020 [3].
During this same period, in a relatively isolated corner of Alaska, community members, policymakers, and fishermen debated opening the world’s largest commercial sockeye salmon fishery [4]. Opening the fishery would mean welcoming thousands of out-of-state fishing industry workers (permit holders, fishing crew, and processors) into communities that had remained free from COVID-19 cases up to spring 2020, just prior to the start of the commercial fishery[5]. Keeping the fishery closed would likely lead to a significant negative income shock to thousands of households who depend on the commercial fishery, both in the Bristol Bay region and farther afield. Factoring into this decision were the limited healthcare options in remote communities; the Bristol Bay region has only one major hospital, with 16 beds and two ventilators. The impacts of the 1919 influenza epidemic that devastated the adult Alaska Native population of Bristol Bay villages [6,7] are still felt today, as the epidemic wiped out as many as 40% of the local population at the time, disrupting family structures and cultural practices. The people of Bristol Bay also remember more minor epidemics, such as tuberculosis, that were prevalent in western Alaska until the 1950s [6]. Alaska Native communities suffered from pandemics disproportionally relative to other areas of the United States in terms of per capita mortality, and this factored into local considerations [7].
Despite the uncertainty related to COVID-19, Native and non-Native community organizations - including the Bristol Bay Borough government - supported the State of Alaska’s decision to open the fishery. In late spring 2020, shortly after the decision to open the fishery, we administered an online survey to Bristol Bay fishery participants and residents to better understand how the perceived risk of COVID-19 influenced decision-making, such as whether to participate in the fishery that season. Our survey included questions about individual and household income and health characteristics, risk perceptions related to COVID-19, and the level of support for the commercial fishery’s opening under various mitigation measures. Using these data, we examined how risk perceptions and individual socioeconomic characteristics are associated with an individual’s decision to participate in the fishing industry during the first year of the COVID-19 pandemic. We found that risk perceptions are generally higher for Alaska Native residents and for out-of-state processor workers (most of whom are Hispanic), and we found that the highest concerns about severe illness are associated with membership in a racial/ethnic minority and among those with lower perceived job choice, lowest income, and non-resident status.
The Bristol Bay region of Alaska
The Bristol Bay wild commercial sockeye salmon fishery produces more value than any other wild salmon fishery worldwide and does so in just a few weeks during the summer [8]. During that time, wild anadromous sockeye salmon return from the ocean to spawn in the freshwater where they had hatched. The bulk of harvests and related processing occurs in about four intense weeks [9]. As in other fisheries, employees in commercial processing facilities and crew members on small fishing vessels work in close proximity and often in wet and challenging conditions. Consequently, practices to protect workers from COVID-19 that are common in other workplaces—such as social distancing and wearing masks and other personal protective equipment—are not practical in the Bristol Bay fishery. Therefore, in 2020, fishery managers implemented other measures to minimize the possibility of a mass outbreak and to reduce the likelihood of a spread to residents of the Bristol Bay region [10]. The measures included quarantine and mandatory tests before flying to the region, frequent testing upon arrival, quarantine flags, restricted shore access, and isolating processing workers and crew members to designated areas outside communities [4]. Although protective measures were effective in minimizing the likelihood of a large outbreak, individual fishery participants understood that they nevertheless faced a significant likelihood of being infected with COVID-19.
In 2020, about 36 commercial salmon processors and buyers were operating in Bristol Bay [11]. Of these, 26 were shore-based processing facilities [11]. Of those who participate in the Bristol Bay fishery, a large share are seasonal workers from other states and countries. In an average season, there are about 5,540 processing workers, only 1.6% of whom are local, and about 10% are Alaska residents [12]. In 2020, there were 4,908 processing workers, of whom 0.8% were local, with 6.6% from Alaska [13]. Overall, participation in the fishery decreased considerably in 2020 relative to previous years [14]. Because of international travel restrictions enacted in early 2020, a larger workforce share, compared to an average year, was from other U.S. states to replace the loss of international workers, who make up 25%-35% of the processing workforce in most years[15]Estimating the number of fishing captains and crew members is made challenging because many are self-employed, however in 2019 the it was estimated that 24% of all fishermen (captain and crew) were from Bristol Bay, and 54% were from Alaska[12]. Because of the lower use of international workers in these positions, the percentages for crew and captains were similar in 2020.[16,17]
Commercial fishing in Bristol Bay is capital-intensive. Fishers often have boat payments, which require considerable annual maintenance and equipment upgrades and permit payments that can often cost more than a boat (average permit cost: $225,000) [18]. During the pandemic, the Paycheck Protection Program (PPP) instituted under the Coronavirus Aid, Relief and Economic Security (CARES) Act required three years of business history. Some captains, especially newer entrants to the fishery, interviewed for this project were unable to provide the necessary paperwork to secure a PPP loan during the 2020 season and had to rely on earnings to make payments for boats and permits. Nevertheless, even with low projected prices to be paid, many fishers chose to participate in the season. In 2020, 1,521 gill nets were fished, slightly down from a previous 5-year average of 1,548 permits fished (1,605 in 2019). In addition, the average gross for each boat in 2020 was $113,625, down from a previous 5-year average gross of $138,092 and a 2019 average gross of $173,571. The year 2020 saw a major impact on the economic value of fishing because of a large reduction in demand for fish caused by restaurant and port closures. Bristol Bay was also affected by a reduced demand for sportfishing during 2020; throughout Alaska, the decline in sportfishing led to an estimated loss of $500 million to the Alaskan economy [19].
Socioeconomic factors and the coronavirus pandemic in 2020
With respect to COVID-19, higher-density living conditions, such as those found in close-quarters working environments in the Bristol Bay fishery, and poverty have been identified as risk factors for respiratory infection [20]. Minority populations, including African American and Hispanic, are more likely to live in poorer households [21] and in higher-density living conditions [22]. Recent coronavirus-related studies find that people who are Black or Hispanic have the highest risk factors for serious illness and the highest mortality rates for COVID-19 relative to other ethnic groups [23]. For instance, Karmakar, Lantz, & Tipirneni (2021) found that U.S. counties with higher percentages of racial/ethnic minority populations experienced a higher burden of COVID-19 incidence and mortality—specifically, counties with higher percentages of people identifying as African American, Hispanic or Latinx, American Indian or Alaskan Native (AIAN), and Asian. Similarly, Samuel et al. (2021) reported a higher risk of hospitalization and mortality in counties with a relatively higher proportion of African American and Hispanic residents with less than a high school degree [25]. Studies analyzing individual-level data had similar findings regarding race and ethnicity—that is, minority populations experienced a higher disease burden compared to White populations, and this difference persists after adjusting for area deprivation index [23]. Lo et al. (2021) also found that the risk of contracting COVID-19 was unequally distributed, disproportionately targeting racial and ethnic minority groups. Moreover, lower education and lower income as individual risk factors were also associated with a higher risk of COVID-19 infection [26].
Furthermore, in a systematic review, Khanijahani et al. (2021) found that socioeconomic factors were associated with worse outcomes for COVID-19, noting that “despite the significant incongruity among the studies, most of them showed that racial/ethnic minority groups had higher risks of COVID-19 infection and hospitalization, confirmed diagnosis, and death.” They further noted that “most of the studies cited factors such as low level of education, poverty, poor housing conditions, low household income, speaking in a language other than the national language in a country, and living in overcrowded households as risk factors of COVID-19 incidence/infection, death, and confirmed diagnosis” [27]
Essential workers (including fishermen and processors) tend to be disproportionately ethnic minorities, and they are more likely to work near others, putting them at greater risk of infection and serious illness [28]. During the pandemic, this was particularly true. Aswaf (2022), for instance, noted that the number of Hispanic workers in the farming, fishing, and forestry occupational category (from May 2020 through July 2021) was 134% greater than those in the total workforce and that their odds of teleworking were 85% less than workers in managerial occupations [29]. Similarly, Rogers et al. (2020) highlighted the disparities in the racial composition of essential workers, documenting significantly higher COVID-19 death rates among non-Hispanic Black and Hispanic essential workers relative to non-Hispanic White essential workers. Seafarers in the global fishing industry were particularly disadvantaged during 2020 and 2021 because of greater restrictions on shore services and limitations on crew changes; in fact, a large number of seafarer crews were reported stranded at sea [30]. Related health information was known to the general public early in the pandemic. National surveys conducted in March 2020 revealed that people who are Black or Hispanic had a stronger perception of COVID-19 as a threat to their health than did people who are White [31].
Data
Our online Qualtrics surveys were designed in collaboration with local leaders, community members, and commercial fishery participants. The survey was advertised through an established network of residents who could help distribute the survey. In addition, all post boxes in Dillingham received an advertisement with a QR code for the survey. Our local researchers also put up fliers at fishing and processing facilities in Dillingham. This study was deemed exempt from formal Institutional Review Board review by the Office of Research Protections of the Pennsylvania State Universities and the Universities of Alaska Anchorage and Fairbanks.
We collected demographic data, including age, gender, income, educational attainment, race, ethnicity, group quarters, and place of residence (including Bristol Bay communities), along with identifying the respondent as a resident or non-resident of Bristol Bay and whether they participated in the fishery (e.g., crew, processor, captain, or permit holder). The survey was conducted from May through July 2020. Survey respondents received a gift card for participation in the survey. The average payment was $22.06, and payments varied in light of responses to a risk elicitation question. We checked IP addresses of respondents, eliminated duplicates, and ran other checks for invalid survey responses, such as question response consistency and duplicate payment emails. After eliminating robot and duplicate responses, we had 927 valid responses to the online survey.
Of the survey’s 927 respondents, 420 were from the Bristol Bay region. Eighty-seven respondents were from other parts of Alaska, and 420 respondents were from states outside Alaska. No survey respondents listed their home country as being outside the United States in summer 2020. Place of residence of respondents is shown in Figure 1.
Figure 1: Place of residence of respondents.
Inset (a) shows distribution of residence within Alaska, (b) shows distribution within Bristol Bay, and (c) shows residence by state of non-Alaskan respondents.
In addition to place of residence, we asked questions related to race/ethnicity, gender, educational attainment, and other personal characteristics. Of survey respondents, 26.1% identified as non-Hispanic White,10.1% identified as non-Hispanic Black, 15.7% identified as Alaska Native, 48.2% identified as Hispanic, and 73.6% were male. By place of residence, 29% of Bristol Bay residents identified as Hispanic, with 16% of residents from greater Alaska identified as Hispanic, and 74% of out-of-state workers identified as Hispanic. The survey over-represented Hispanics—in the Bristol Bay Borough and Dillingham Census area population, Hispanic people comprised 3.1% of all Census respondents in 2020 [32]. Regarding educational attainment, 680 respondents had at least a high school diploma, and 296 respondents had an associate’s degree or higher.
To investigate concerns related to COVID-19, we also asked respondents to report perceived likelihood and concern about exposure, severe illness, or death. To measure employment opportunities, we asked questions about the potential for other work opportunities and the importance of income from the Bristol Bay fishery to household income:
COVID-19 exposure: On a scale of 0–10, with 0 being unlikely and 10 being highly likely, how likely do you think it is that you or someone else working for the processor or fishing crew will be exposed to COVID-19 during the 2021 commercial fishing season? (asked of captains, crew, and processors)
COVID-19 infection: On a scale of 0–10, if someone working for the processor or fishing crew contracts COVID-19, how likely do you think it is that you will contract the virus? (asked of captains, crew, and processors)
COVID-19 serious illness: On a scale of 0–10, if you contract COVID-19, how likely do you think it is that you will experience serious health complications? (asked of captains, crew, processors, and all other residents)
Employment alternatives: If you didn’t work for a Bristol Bay processor (crew), could you have found other employment this summer (Likert-scale response, where 1 = definitely not, and 5 = definitely yes)? (asked of crew and processors)
Employment earnings: Compared to other summer employment opportunities, the amount of money you can earn working for a processor in 2021 is … (Likert scale, where 1 = definitely not as great as other options, and 5 = definitely greater than other options)
Employment and health: On a scale of 0–10, where 0 means I don’t agree at all, how much do you agree with the following statement? “Coming to work is more important to me than my health.” (asked of crew, processors, and all other residents)
Table 1 presents descriptive statistics for respondents’ residence status, income, presence of comorbidity, education level, work status in the fishing industry, gender, and family status. Each characteristic is calculated by race/ethnicity. As indicated in Table 1, Hispanic respondents had the lowest percentage of Alaska residence. More than half of the respondents who are White or Black have income greater than the median, with more than half of Alaska Native or Hispanic respondents having below median income.
Table 1:
Demographics and socioeconomic statuses of survey respondents by race/ethnicity.
| Variable | Non-Hispanic White |
Non-Hispanic Black |
Non – Hispanic AIANa |
Hispanic | Total |
|---|---|---|---|---|---|
| Alaska resident | 170 (70.8%) | 59 (63.4 %) | 144 (100 %) | 131 (29.6%) | 504 (54.8%) |
| Income > $67,500 | 175 (73.0%) | 67 (72.0%) | 56 (38.9%) | 127 (28.7%) | 425 (46.2%) |
| Comorbidity | 26 (10.8%) | 4 (4.3%) | 16 (11.1%) | 39 (8.8%) | 85 (9.2%) |
| Education > High School | 214 (89.2%) | 63 (67.7%) | 103 (71.5%) | 295 (66.6%) | 675 (73.4%) |
| Work in fishing | 201 (83.8%) | 89 (95.7%) | 119 (82.6%) | 439 (99.1%) | 848 (92.2%) |
| Male | 181 (75.4%) | 60 (64.5%) | 76 (52.8%) | 350 (79.0%) | 667 (72.5%) |
| Has children | 182 (75.8%) | 67 (72.0%) | 114 (79.2%) | 419 (94.6%) | 782 (85%) |
| Total | 240 (26.1%) | 93 (10.1%) | 144 (15.7%) | 443 (48.2%) | 920 (100%) |
American Indian Alaska Native
Table 2 presents summary statistics about the response to the six outcome variables, by race/ethnicity. Here we see that respondents who are Alaska Native and Hispanic had higher mean responses for concern about severe disease than non-Hispanic White or non-Hispanic Black respondents. Alaska Native and Hispanics also lower average scores for perceived opportunity for other work opportunities outside of fishing than non-Hispanic White or Black respondents did.
Table 2:
Descriptive statistics of outcome variables for regression models by race/ethnicity.
| Non-Hispanic White |
Non-Hispanic Black |
Non- Hispanic AIANa |
Hispanics | |
|---|---|---|---|---|
| Perceived risk of exposure (0-10) | 4.6 (1.25) | 4.7 (1.46) | 5.70 (1.63) | 6.01 (1.54) |
| [205] | [89] | [122] | [439] | |
| Perceived risk of disease (0-10) | 5.54 (1.61) | 6.00 (1.42) | 6.52 (1.62) | 4.60 (2.51) |
| [205] | [89] | [122] | [439] | |
| Perceived risk of severe disease (0-10) | 4.16 (1.98) | 3.62 (1.90) | 5.16 (2.69) | 6.10 (2.04) |
| [232] | [93] | [143] | [442] | |
| Opportunity for other work (1-5) | 3.61 (0.77) | 3.21 (0.88) | 2.63 (0.97) | 2.75 (1.16) |
| [185] | [89] | [114] | [412] | |
| Work is good earning opportunity (1-5) | 4.2 (0.78) | 4.52 (0.83) | 4.52 (0.60) | 4.33 (0.61) |
| [185] | [89] | [114] | [412] | |
| Values work over health (0-10) | 5.95 (2.37) | 5.52 (2.07) | 4.27 (2.13) | 5.04 (1.96) |
| [222] | [92] | [143] | [411] |
Notes:
American Indian Alaska Native.
Standard deviations are in (SD). Sample sizes are in [N]. The responses about perception of risk of exposure, risk of disease, risk of severe disease and agreeing with the statement that you value work over health are in a scale of 0–10, with 0 being unlikely, and 10 being highly likely (asked of captains, crew, and processors). The responses about perception of risk of severe disease are in a scale of 0–10, with 0 being unlikely, and 10 being highly likely (asked of captains, crew, and processors). The response about opportunity for other work are on a Likert-scale response where 1 = definitely not, and 5 = definitely yes (asked of crew and processors). The responses about employment alternatives if unable to work in the fishing industry are a Likert-scale response, where 1 = definitely not, and 5 = definitely yes (asked of crew and processors).
Analysis and Results
We used simple ordinary least squares regression analysis with robust standard errors to explore links between risk perceptions and decisions to participate in the fishery. Dependent variables used in the analysis include the three COVID-19 questions related to exposure, infection, and serious illness, along with the three questions related to employment. Regression results are presented in Tables 3 and 4. We begin by discussing relationships between socioeconomic characteristics and COVID-19 risk perceptions (Table 3) and employment opportunities (Table 4).
Table 3:
Estimated Responses to questions about perceived risk of exposure to COVID-19, risk of infection if exposed, and risk of severe illness if infected among resident community members and resident and non-resident fishing industry workers.
| Outcome: | Risk of exposure | Infection, if exposed | Severe symptoms, if infected |
|---|---|---|---|
| Range: | 1-10 | 1-10 | 1-10 |
| Sample Frame | Captains, Fishers, Processors | Captains, Fishers, Processors | Captains, Fishers, Processors, Residents |
| N | 848 | 848 | 910 |
| Constant | 4.81 (4.39, 5.22) | 5.45 (5.00, 5.90) | 6.08 (5.38, 6.79) |
| > Median incomea | −1.04 (−1.24, −0.84) | −0.48 (−0.70, −0.26) | −1.93 (−2.20, −1.66) |
| Race (non-Hispanic White = ref) | |||
| Non-Hispanic Black | 0.13 (−0.2, 0.45) | 0.49 (0.14, 0.85) | −0.25 (−0.68, 0.18) |
| AIAN b | 0.91 (0.59, 1.22) | 1.04 (0.70, 1.39) | 0.54 (0.14, 0.94) |
| Hispanic | 0.50 (0.26, 0.75) | 0.36 (0.10, 0.63) | 0.72 (0.39, 1.04) |
| Do you reside in Alaska | −0.63 (−0.84, −0.42) | −0.78 (−1.01, −0.55) | −0.78 (−1.07, −0.50) |
| Do you have one or more comorbidity? | −0.41 (−0.77, −0.05) | −0.39 (−0.78, 0.00) | 0.96 (0.52, 1.40) |
| Education Greater than High School | 0.08 (−0.12, 0.28) | 0.18 (−0.04, 0.4) | 0.08 (−0.19, 0.36) |
| Industry Category | |||
| Resident | (NIU) | (NIU) | Reference |
| Processor | Reference | Reference | −1.87 (−2.43, −1.31) |
| Crew | 0.2 (−0.05, 0.46) | 0.22 (−0.06, 0.5) | −2.20 (−2.78, −1.62) |
| Captain Owner | 0.13 (−0.21, 0.47) | −0.40 (−0.77, −0.03) | −1.74 (−2.39, −1.09) |
| Any Children? | 1.28 (1.01, 1.56) | 1.06 (0.76, 1.35) | 1.94 (1.60, 2.29) |
| Male (Female reference) | −0.24 (−0.51, 0.04) | −0.11 (−0.41, 0.18) | 0.17 (−0.18, 0.52) |
Notes:
The median income of $67,500 was used.
American Indian Alaska Native
Results are on a 0–10 scale with a 95% confidence interval
Table 4:
Responses to questions about opportunities for other employment (outside of fishing industry), earning potential of fishing work compared to other possible work, and feelings about whether work is more important than health, among resident community members and resident and non-resident fishing industry workers.
| Outcome: | Other employment. (1-5 scale) |
Earning opportunity (1-5 Scale) |
Work more important to health (1-10 scale) |
|---|---|---|---|
| Range | 1-5 | 1-5 | 0-10 |
| Sample Frame | Captains, Fishers, Processors (1-5) | Captains, Fishers, Processors (1-5) | Captains, Fishers, Processors, Residents (1-10) |
| N | 799 | 799 | 868 |
| Constant | 2.88 (2.59, 3.18) | 4.49 (4.27, 4.70) | 1.54 (0.83, 2.25) |
| > Median incomea | 0.73 (0.59, 0.88) | 0.06 (−0.05, 0.17) | 0.99 (0.70, 1.27) |
| Race (non-Hispanic White = ref) | |||
| Non-Hispanic Black | −0.37 (−0.6, −0.14) | 0.30 (0.13, 0.47) | −0.78 (−1.23, −0.33) |
| AIAN b | −0.86 (−1.09, −0.63) | 0.44 (0.27, 0.61) | −1.67 (−2.09, −1.26) |
| Hispanic | −0.33 (−0.51, −0.15) | 0.17 (0.04, 0.3) | −0.73 (−1.08, −0.38) |
| Do you reside in Alaska | 0.55 (0.39, 0.7) | −0.19 (−0.31, −0.08) | 1.03 (0.72, 1.35) |
| Do you have one or more comorbidity? | 0.48 (0.23, 0.74) | −0.1 (−0.28, 0.09) | 0.52 (0.08, 0.97) |
| Education Greater than High School | −0.10 (−0.24, 0.04) | 0.13 (0.02, 0.23) | −0.45 (−0.73, −0.17) |
| Industry Category | |||
| Resident | (NIU) | (NIU) | Reference |
| Processor | Reference | Reference | 3.66 (3.11, 4.22) |
| Crew | −0.07 (−0.25, 0.11) | 0.03 (−0.10, 0.17) | 3.55 (2.97, 4.13) |
| Captain Owner | −0.24 (−0.54, 0.06) | 0.08 (−0.14, 0.30) | 2.41 (1.69, 3.12) |
| Any Children? | −0.32 (−0.51, −0.12) | −0.34 (−0.48, −0.2) | 0.04 (−0.32, 0.40) |
| Male (Female reference) | 0.26 (0.07, 0.46) | −0.08 (−0.22, 0.07) | 0.42 (0.06, 0.78) |
Notes:
The median income of $67,500 was used.
American Indian Alaska Native.
In general, we found that perceived risk of exposure to COVID-19, infection risk, and risk of serious disease (Table 3) is associated with several factors, including income, race/ethnicity, place of residence, and living with children. Alaska Native and Hispanic respondents consistently expressed greater concerns about exposure risk than non-Hispanic White respondents. Furthermore, respondents with children and relatively lower incomes also indicated relatively greater concern about likelihood of infection, exposure, and serious illness. Respondents with a known comorbidity also indicated a higher concern about risk of serious illness. Gender was not associated with perceived risk of exposure, infection, or severe illness. Across all three categories—perceived risk of exposure, infection, and serious illness—respondents indicating an Alaska residency had a lower perception of risk than those from outside the state. Captains showed a slightly lower concern about being infected if exposed compared to processors. Those working in the fishing industry, whether processor, crew, or captain, indicated a lower perceived risk of severe illness, compared to Bristol Bay residents not directly working in the fishing industry.
In Table 4, we highlight socioeconomic characteristics associated with employment opportunities. African Americans, Alaska Native, and Hispanic respondents indicated that their alternative employment opportunities (columns 1 and 2) were significantly less than non-Hispanic White respondents did. Similarly, African Americans, Alaska Native, and Hispanic respondents considered personal health to be more important than work, as compared to non-Hispanic Whites. Males expressed higher confidence in finding other work opportunities, and that work was more important than health. All industry categories considered work to be more important than health, compared to non-industry residents. Full results for questions about work are found in Table 4.
In table 5, we explored in more detail factors correlated with relatively higher risk perceptions of serious illness, particularly with respect to race and income. We regressed responses to the question about serious illness due to COVID-19 on interaction terms for race/ethnicity and income, stratified by residence and other employment opportunities and report the marginal values for estimates.
Table 5.
Perceived serious health complications if contracted COVID-19a by race/ethnicity, residence, income, and employment opportunities.
| Non-resident of Alaska | Resident of Alaska | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Income | Race/Ethnicity | Est | 95% low |
95% high |
Income | Race/Ethnicity | Est | 95% low |
95% high |
| Below | Non-Hispanic White | 3.43 | 2.47 | 4.39 | Below | Non-Hispanic White | 5.29 | 4.70 | 5.89 |
| median | Non-Hispanic Black | 4.67 | 3.43 | 5.90 | median | Non-Hispanic Black | 4.00 | 2.89 | 5.11 |
| incomeb | Non-Hispanic AIANc | NIUd | income | Non-Hispani AIAN | 6.00 | 5.49 | 6.51 | ||
| Hispanic | 7.23 | 7.07 | 7.39 | Hispanic | 5.06 | 4.61 | 5.52 | ||
| Above | Non-Hispanic White | 3.45 | 2.97 | 3.92 | Above | Non-Hispanic White | 4.18 | 3.84 | 4.52 |
| median | Non-Hispanic Black | 3.00 | 2.42 | 3.58 | median | Non-Hispanic Black | 3.39 | 2.94 | 3.84 |
| income | Non-Hispanic AIAN | NIUd | income | Non-Hispanic AIAN | 3.84 | 3.17 | 4.51 | ||
| Hispanic | 4.51 | 3.92 | 5.10 | Hispanic | 4.20 | 3.86 | 4.55 | ||
| If not working in Bristol Bay, could you have found other employment? Definitely Not, Probably Not (n = 317) |
If not working in Bristol Bay, could you have found other employment? Maybe, Yes, Probably Yes (n = 593) |
||||||||
| Income | Race/ethnicity | Est | 95% low |
95% high |
Income | Race/ethnicity | Est | 95% low |
95% high |
| Below | Non-Hispanic White | 5.00 | 3.12 | 6.88 | Below | Non-Hispanic White | 4.64 | 4.07 | 5.22 |
| median | Non-Hispanic Black | 5.92 | 4.54 | 7.30 | median | Non-Hispanic Black | 3.21 | 2.41 | 4.02 |
| incomeb | Non-Hispanic AIANc | 7.14 | 6.71 | 7.57 | income | Non-Hispanic AIAN | 4.43 | 3.60 | 5.27 |
| Hispanic | 7.45 | 7.32 | 7.58 | Hispanic | 5.45 | 5.06 | 5.83 | ||
| Above | Non-Hispanic White | 4.36 | 3.43 | 5.30 | Above | Non-Hispanic White | 3.94 | 3.65 | 4.24 |
| median | Non-Hispanic Black | 3.33 | 2.32 | 4.35 | median | Non-Hispanic Black | 3.30 | 2.90 | 3.69 |
| income | Non-Hispanic AIAN | 2.33 | 1.45 | 3.22 | income | Non-Hispanic AIAN | 4.13 | 3.38 | 4.88 |
| Hispanic | 4.00 | 2.87 | 5.13 | Hispanic | 4.37 | 4.03 | 4.71 | ||
Notes: a The survey question is: “On a scale of 0–10, if you contract COVID-19, how likely do you think it is that you will experience serious health complications?” The questions were asked to captains, crew, processors, and residents.
The median income of $67,500 was used.
American Indian Alaska Native.
No non-Hispanic AIAN respondents resided outside of the state of Alaska.
We found that non-resident Hispanic respondents (mean 7.23; 95% CI: 7.03, 7.43) and resident Alaska Native respondents (mean 6.00; 95% CI: 5.58, 6.42) making below median income had a greater perception of perceived risk for serious COVID-19 illness compared to all other race/ethnic groups (Table 5a). Also, we found that among residents, those below median income had a stronger perception of risk for each category of race/ethnicity compared to residents with above median income.
Similarly, as shown in Table 5b, marginal values of risk perceptions are highest for those with few alternative employment opportunities and relatively low incomes. Of the low-income/few employment options group, Alaska Native (mean 7.13; 95% CI: 6.79, 7.48) and Hispanic respondents (mean 7.45; 95% CI: 7.28, 7.61) had the highest reported values of perceived risk of serious illness. Results from Table 5a and 5b are shown in Figure 2. All analyses were conducted in Stata/MP 17.0, using robust standard errors.
Figure 2: Perceived risks of severe illness by race/ethnicity.
Marginal responses for non-Hispanic White, non-Hispanic Black, Alaska Native, and Hispanic to the question “On a scale of 0–10, if you contract COVID-19, how likely do you think it is that you will experience serious health complications?” (a) by residence (Alaska resident versus non-resident) and income (higher or lower than median income of $67,500) (b) by perceived employment opportunities (definitely not or probably not versus maybe, probably yes, yes) and income.
Conclusions and Discussion
This study centers on the Bristol Bay region of Alaska, home to the world’s largest sockeye salmon commercial fishery during the early months of the COVID-19 pandemic. During this time frame, the Bristol Bay borough and Dillingham census area reported only 22 cases of COVID-19 [33], and there were no reports of outbreaks within the community. However, in addition to risks of the impacts of the disease, community members and leaders were concerned about potential underreporting. During that early period of the pandemic, there were no over-the-counter rapid antigen tests, which received an emergency use authorization in December 2020 [34], so underreporting may have been an issue. However, whether by chance or out of the concerns and actions of the community, there were no reports of major outbreaks in the fishing community during the survey period. Our survey of 927 people found that concerns about COVID-19 and the constraints on job choice are associated with residence, income, and race/ethnicity. Overall, non-resident workers who make less than the median income and are members of ethnic minorities (non-Hispanic Black, non-Hispanic AIAN, and Hispanic) had greater concerns about exposure to COVID-19, contracting COVID-19, and suffering severe effects from it. These concerns are consistent with existing research that finds ethnic minority populations are at greater risk of serious illness due to COVID-19. Further, these same racial groups perceived fewer outside employment opportunities or jobs with wages equivalent to jobs in the fishing industry.
Combining these factors, we found that, of all race/ethnic groups, those reporting the highest concern for serious illness were Alaska Native and Hispanic respondents, who also reported relatively low incomes and few outside employment opportunities—that is, lower income, higher-risk people participated in the fishery largely because it was the best income-earning opportunity available. Although fishermen have been known to be less risk-averse than the average person in previous studies in Vietnam [35] and the U.S. state of Maine [36], it seems reasonable to conclude that lower-income respondents who perceived higher-than-average risk of serious illness would have taken other employment opportunities had they been available.
Previous research has found that perception of infection risk and risk of severe outcomes for COVID-19 are higher in people who are Black [37] and people who are Hispanic [38]. Overall, these perceptions align with actual risk; specifically, risk of infection and of ICU admission were higher in people who are Black or Hispanic compared to people who are White [23,39]. Our evidence here suggests that Alaska Native and Hispanic respondents, who generally make less than median income, perceived COVID-19 illness as a serious threat but still chose to work in an environment where the likelihood of exposure to COVID-19 was quite high. Their decisions were driven in large part by income and employment considerations. While COVID-19 cases were generally low during this study, and females were under-sampled, future work using other data sources might further examine the intersection between gender, occupation, and COVID-19 case counts. Additionally, given the importance of this income to seasonal international workers who participate in the fishing industry, future work would benefit by exploring their views on risk and benefits of disease and employment.
The risk perceptions and income needs of fishermen and processor workers in our survey may explain why compliance and support for the 2020 COVID-19 mitigation procedures were so great. Local municipalities, the Bristol Bay Borough, and the State of Alaska implemented the guidelines described in the introduction and, by all accounts, fishermen complied with the mitigation procedures at multiple levels [40]. Indeed, in our survey, satisfaction with the Bristol Bay Borough’s mitigation procedures was high, with a mean score of 6.4 on a scale of 0–10 (SD = 1.4), and high marks were given to processors (mean = 6.9; SD = 1.2) and to the State of Alaska (mean = 6.6, SD = 1.3).
This study adds further detail to the research into COVID-19 risk perceptions and associated employment choices among fishermen working in a relatively high-risk industry. This work highlights the necessity for a better understanding of the linkages between risk perceptions during an infectious disease outbreak and employment and income considerations in order to design effective public health policy. These considerations are particularly important when employment opportunities and disease risk vary across factors of race, ethnicity, and income.
Acknowledgments
This project is supported in part by the National Science Foundation (Awards #2032790, #2032787, #2033493, #1927827, #2121909, #2207436, and #2220219), the USDA National Institute of Food and Agriculture (#PEN04796, #PEN04965, Accession #7003407, Accession #7006628), and the Eunice Kennedy Shriver National Institute of Child Health and Human Development (Award # P2C HD041025).
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Declarations of interest: none
References
- [1].Wang H, Paulson KR, Pease SA, Watson S, Comfort H, Zheng P, Aravkin AY, Bisignano C, Barber RM, Alam T, Fuller JE, May EA, Jones DP, Frisch ME, Abbafati C, Adolph C, Allorant A, Amlag JO, Bang-Jensen B, Bertolacci GJ, Bloom SS, Carter A, Castro E, Chakrabarti S, Chattopadhyay J, Cogen RM, Collins JK, Cooperrider K, Dai X, Dangel WJ, Daoud F, Dapper C, Deen A, Duncan BB, Erickson M, Ewald SB, Fedosseeva T, Ferrari AJ, Frostad JJ, Fullman N, Gallagher J, Gamkrelidze A, Guo G, He J, Helak M, Henry NJ, Hulland EN, Huntley BM, Kereselidze M, Lazzar-Atwood A, LeGrand KE, Lindstrom A, Linebarger E, Lotufo PA, Lozano R, Magistro B, Malta DC, Månsson J, Mantilla Herrera AM, Marinho F, Mirkuzie AH, Misganaw AT, Monasta L, Naik P, Nomura S, O’Brien EG, O’Halloran JK, Olana LT, Ostroff SM, Penberthy L, Reiner RC Jr, Reinke G, Ribeiro ALP, Santomauro DF, Schmidt MI, Shaw DH, Sheena BS, Sholokhov A, Skhvitaridze N, Sorensen RJD, Spurlock EE, Syailendrawati R, Topor-Madry R, Troeger CE, Walcott R, Walker A, Wiysonge CS, Worku NA, Zigler B, Pigott DM, Naghavi M, Mokdad AH, Lim SS, Hay SI, Gakidou E, Murray CJL, Estimating excess mortality due to the COVID-19 pandemic: a systematic analysis of COVID-19-related mortality, 2020–21, Lancet. 399 (2022) 1513–1536. https://doi.org/ 10.1016/S0140-6736(21)02796-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].World Bank., Global Economic Prospects, Washington, DC, 2022. https://doi.org/doi: 10.1596/978-1-4648-1843-1. [DOI] [Google Scholar]
- [3].U.S. Bureau of Economic Analysis, Personal Consumption Expenditures [PCE], (2022). https://fred.stlouisfed.org/series/PCE (accessed July 15, 2022).
- [4].Wiebold K, Seafood processing and Covid-19, Alaska Econ. Trends Nov 2020 (2022) 4–7. https://laborstats.alaska.gov/trends/nov20art1.pdf (accessed July 15, 2022). [Google Scholar]
- [5].Thompson T, Bristol Bay announces 3 additional cases of COVID-19 since Friday, Alaska Public Media. (2020). https://alaskapublic.org/ (accessed September 7, 2022). [Google Scholar]
- [6].deValpine MG, Influenza in Bristol Bay, 1919: “The saddest repudiation of a benevolent intention,” SAGE Open. 5 (2015). 10.1177/2158244015577418. [DOI] [Google Scholar]
- [7].Smiley S, Remembering 1919, one hundred years later, KLDG Public Radio Bristol Bay. (2019). https://www.kdlg.org/education/2019-07-29/remembering-1919-one-hundred-years-later (accessed September 1, 2022). [Google Scholar]
- [8].Knapp G, Guettabi M, Goldsmith OS, The economic importance of the Bristol Bay salmon industry, Institute of Social and Economic Research, Anchorage, 2013. [Google Scholar]
- [9].AK Department of Fish and Game, Inseason Commercial Salmon Harvest Timing Charts, (2022). https://www.adfg.alaska.gov/index.cfm?adfg=commercialbyfisherysalmon.harvest_charts (accessed January 9, 2022).
- [10].Herz N, Alaska’s fishing communities have plans to contain COVID-19. Now they’ll be put to the test., Alaska Public Media; - Anchorage. (2020). https://alaskapublic.org/2020/06/02/alaskas-fishing-communities-have-plans-to-contain-covid-19-now-theyll-be-put-to-the-test/ (accessed July 1, 2022). [Google Scholar]
- [11].Tiernan T, Elison T, Sands T, Head J, Vega S, Neufeld G., 2020 Bristol Bay annual management report. Fishery Management Report No. 21-16., Anchorage., 2021. [Google Scholar]
- [12].McKinley Research, The Economic Benefits of Bristol Bay Salmon, (2021) 29. https://www.mcdowellgroup.net/wp-content/uploads/2021/03/economic-benefits-of-bristol-bay-salmon.pdf (accessed July 5, 2022).
- [13].State of Alaska Department of Labor, Local Seafood Processing Workforce, State Alaska Dep. Labor (2022). https://live.laborstats.alaska.gov/seafood-processing-local-workers?field_seafood_data_ne_region_target_id=82 (accessed February 8, 2022). [Google Scholar]
- [14].Schultz G, Kreiger R, COVID-19 and worker counts, Alaska Econ. Trends Mag (2022) 10–11. https://laborstats.alaska.gov/trends/mar22art3.pdf (accessed September 1, 2022). [Google Scholar]
- [15].Smith C, Alaska seafood processing jobs draw workers from around the world through temporary work visa program, (2023). https://www.kdlg.org/fisheries/2023-07-06/alaska-seafood-processing-jobs-draw-workers-from-around-the-world-through-temporary-work-visa-program (accessed October 9, 2023). [Google Scholar]
- [16].Krieger R, Whitney S, Nonresidents working in Alaska 2020, 2022. [Google Scholar]
- [17].Krieger R, Whitney S, Nonresidents working in Alaska 2019, 2021. [Google Scholar]
- [18].Coleman J, Carothers C, Donkersloot R, Ringer D, Cullenberg P, Bateman A, Alaska’s Next Generation of Potential Fishermen: a Survey of Youth Attitudes Towards Fishing and Community in Bristol Bay and the Kodiak Archipelago, Marit. Stud 18 (2019) 47–63. https://doi.org/ 10.1007/s40152-018-0109-5. [DOI] [Google Scholar]
- [19].Viechnicki J, COVID-19 takes half-billion dollar bite from Alaska commercial fisheries revenue, Alaska Public Media. (2021). https://www.alaskapublic.org/2021/01/19/alaska-sees-nearly-half-billion-dollar-loss-in-commercial-fisheries-revenue/ (accessed September 1, 2022). [Google Scholar]
- [20].Patel JA, Nielsen FBH, Badiani AA, Assi S, Unadkat VA, Patel B, Ravindrane R, Wardle H, Poverty, inequality and COVID-19: the forgotten vulnerable, Public Health. 183 (2020) 110–111. 10.1016/j.puhe.2020.05.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [21].US Census Bureau, Table 2. Poverty Status of People by Family Relationship, Race, and Hispanic Origin: 1959 to 2020, (2022). https://www.census.gov/data/tables/time-series/demo/income-poverty/historical-poverty-people.html (accessed September 1, 2022).
- [22].US Census Bureau, Historic Income Tables: Households; Table F-8. Size of Family--All Families by Median and Mean Income: 1947 to 2020, (2022). https://www.census.gov/data/tables/time-series/demo/income-poverty/historical-income-households.html (accessed September 5, 2022).
- [23].Magesh S, John D, Li WT, Li Y, Mattingly-app A, Jain S, Chang EY, Ongkeko WM, Disparities in COVID-19 Outcomes by Race, Ethnicity, and Socioeconomic Status: A Systematic Review and Meta-analysis, JAMA Netw. Open 4 (2021) e2134147–e2134147. 10.1001/jamanetworkopen.2021.34147. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [24].Karmakar M, Lantz PM, Tipirneni R, Association of Social and Demographic Factors With COVID-19 Incidence and Death Rates in the US, JAMA Netw. Open 4 (2021) e2036462–e2036462. 10.1001/jamanetworkopen.2020.36462. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [25].Samuel LJ, Gaskin DJ, Trujillo AJ, Szanton SL, Samuel A, Slade E, Race, ethnicity, poverty and the social determinants of the coronavirus divide: U.S. county-level disparities and risk factors, BMC Public Health. 21 (2021) 1250. 10.1186/s12889-021-11205-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [26].Zhu Y, Duan M-J, Dijk HH, Freriks RD, Dekker LH, Mierau JO, Association between socioeconomic status and self-reported, tested and diagnosed COVID-19 status during the first wave in the Northern Netherlands: a general population-based cohort from 49 474 adults., BMJ Open. 11 (2021) e048020. 10.1136/bmjopen-2020-048020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Khanijahani A, Iezadi S, Gholipour K, Azami-Aghdash S, Naghibi D, A systematic review of racial/ethnic and socioeconomic disparities in COVID-19, Int. J. Equity Health 20 (2021) 248. 10.1186/s12939-021-01582-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [28].Rogers TN, Rogers CR, VanSant-Webb E, Gu LY, Yan B, Qeadan F, Racial Disparities in COVID-19 Mortality Among Essential Workers in the United States., World Med. Heal. Policy 12 (2020) 311–327. 10.1002/wmh3.358. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [29].Asfaw A, Racial and Ethnic Disparities in Teleworking Due to the COVID-19 Pandemic in the United States: A Mediation Analysis., Int. J. Environ. Res. Public Health 19 (2022). 10.3390/ijerph19084680. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [30].Vandergeest P, Marschke M, MacDonnell M, Seafarers in fishing: A year into the COVID-19 pandemic, Mar. Policy 134 (2021) 104796. https://doi.org/ 10.1016/j.marpol.2021.104796. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [31].Niño M, Harris C, Drawve G, Fitzpatrick KM, Race and ethnicity, gender, and age on perceived threats and fear of COVID-19: Evidence from two national data sources, SSM - Popul. Heal 13 (2021) 100717. https://doi.org/ 10.1016/j.ssmph.2020.100717. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [32].Department of Labor and Workforce Development: Research and Analysis, 2020 Census Data for Redistricting, (2022). https://live.laborstats.alaska.gov/cen/2020/downloads (accessed June 3, 2022).
- [33].The New York Times;, Coronavirus (Covid-19) Data in the United States., Coronavirus Data United States. (2021). https://github.com/nytimes/covid-19-data (accessed January 12, 2023).
- [34].Dawson L, Kates J, Rapid Home Tests for COVID-19: Issues with Availability and Access in the U.S, KFF. (2021). https://www.kff.org/report-section/rapid-home-tests-for-covid-19-issues-with-availability-and-access-in-the-u-s-issue-brief/ (accessed November 1, 2023). [Google Scholar]
- [35].Nguyen Q, Leung P, Do Fishermen Have Different Attitudes Toward Risk? An Application of Prospect Theory to the Study of Vietnamese Fishermen, J. Agric. Resour. Econ 34 (2009) 518–538. http://www.jstor.org/stable/41548431. [Google Scholar]
- [36].Davis ME, Perceptions of occupational risk by US commercial fishermen, Mar. Policy 36 (2012) 28–33. https://doi.org/ 10.1016/j.marpol.2011.03.005. [DOI] [Google Scholar]
- [37].Kumar V, Encinosa W, Racial Disparities in the Perceived Risk of COVID-19 and in Getting Needed Medical Care., J. Racial Ethn. Heal. Disparities (2021) 1–10. 10.1007/s40615-021-01191-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [38].Jamieson T, Caldwell D, Gomez-Aguinaga B, Doña-Reveco C, Race, Ethnicity, Nativity and Perceptions of Health Risk during the COVID-19 Pandemic in the US., Int. J. Environ. Res. Public Health 18 (2021). 10.3390/ijerph182111113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [39].Gu T, Mack JA, Salvatore M, Sankar SP, Valley TS, Singh K, Nallamothu BK, Kheterpal S, Lisabeth L, Fritsche LG, Mukherjee B, COVID-19 outcomes, risk factors and associations by race: a comprehensive analysis using electronic health records data in Michigan Medicine., MedRxiv Prepr. Serv. Heal. Sci (2020). 10.1101/2020.06.16.20133140. [DOI] [Google Scholar]
- [40].Adams A, COVID-19 threatened Alaska’s fishermen. Here’s how they persevered., Natl. Geogr. Mag (2021). https://www.nationalgeographic.com/ (accessed September 9, 2022). [Google Scholar]