Abstract
Objectives:
To conduct a needs assessment of Marshallese food processing workers and compare workplace experiences during the COVID-19 pandemic with Marshallese workers in other occupations.
Methods:
Marshallese adults residing in the continental United States (US) and Hawaii took part in an online survey. The sample was divided into two categories: food processing workers and workers in all other occupations. To examine differences between food processing workers and workers from all other occupations, we used Wilcoxon-Mann-Whitney U tests and Fisher’s Exact tests.
Results:
Of those employed at the time of the survey (n=113), 31 were employed in food processing plants, and 82 were employed in another occupation. Food processing workers and workers in other occupations differed significantly on level of education, length of residence in the US, English speaking ability, and health literacy. More food processing workers reported their employers installed barriers or provided shields (45%), provided temperature screenings (71%), and tested for COVID-19 (61%) compared with those in other occupations. A larger proportion of food processing workers reported having no sick leave compared with workers in other occupations, although they reported COVID-19 testing and being insured at similar rates.
Conclusion:
This is the first study to examine Marshallese food processing workers’ experiences during the COVID-19 pandemic. Our findings show that while some food processing employers implemented government-recommended guidelines to prevent the spread of COVID-19, preventative and protective measures were not comprehensively applied across the food processing industry, despite efforts by public health agencies and community partners.
Keywords: food processing, COVID-19, Marshallese, COFA migrant workers, occupational health, occupational safety
Introduction
The Food Processing Industry
Food processing industries play a significant role in the economic infrastructure of the United States (US) and perform a major function within the US grocery supply chain [1, 2]. Workers in these industries are responsible for putting food on the table for millions of Americans and are essential to the sustenance and stability of the US population and economy [1, 3].
Food processing, including meat and poultry processing, is often dangerous and difficult work; employees typically receive low pay, endure long shifts, and perform strenuous and repetitive labor in hazardous conditions [4-7]. Historically, the industry has employed marginalized workers, including women, minorities, and immigrants with limited English proficiency, whose status often renders them unable to reject work in undesirable conditions [3-5, 8]. Workers may experience language barriers that prevent them from understanding health and safety guidelines or pursuing their rights for safer work environments [3-5, 8]. Additionally, some food processing employees have limited access to healthcare and may not have paid sick leave [3, 4].
Workplace exposure and preventative measures have been key factors in the spread of COVID-19 [9]. US counties with food processing facilities have experienced increases in per capita infection rates ranging from 20% to 160% depending on the type of meat processing [10]. Between March 1 and May 31, 2020, Arkansas had the highest number (49) of meat processing workplaces affected by the virus in a study of 30 US states [11].
History of the Marshallese
The Marshallese are a Pacific Islander population from the Republic of the Marshall Islands (RMI) who are permitted to live and work in the US without a visa under the 1986 Compact of Free Association (COFA) agreement [12]. The COFA agreement was, in large part, a response to the US military’s nuclear weapons testing program in the RMI that exposed Marshallese to radioactive fallout. Between 2000 and 2010, the Marshallese population in the US more than tripled in size [13], with many COFA migrants moving to southern and Midwestern states, such as Arkansas, in search of employment opportunities. An estimated 14,000 Marshallese residents now reside in Arkansas [14], where one of the primary employers of Marshallese COFA migrants is the poultry processing industry [4, 15]. In a 2012 survey conducted with Marshallese residents in northwest Arkansas, 76% of the 120 respondents reported working for one of three major poultry processing industries in the area (Tyson Foods, George’s, and Butterball) [15].
Effects of COVID-19 on the Marshallese Population
Marshallese adults face health disparities that increase risk for severe COVID-19 infection and death, including high rates of overweight and obesity, type 2 diabetes mellitus (T2DM), and hypertension [16-18]. Marshallese have experienced persistent disparities in infections, hospitalizations, and deaths throughout the COVID-19 pandemic [19-21]. Between May and August 2020, the rate of new COVID-19 infection among Pacific Islanders increased, even as rates decreased among other racial and ethnic minority groups [19]. Decreases in disparities in new infections were noted from September to December 2020, although the decrease was largely due to the increase of infections among Whites [19].
One stark example of these COVID-19 disparities was found in northwest Arkansas, where the largest population of Marshallese in the continental US is located [13]. A July 2020 report from the Centers for Disease Control and Prevention (CDC) documented the effects of the COVID-19 pandemic on the Marshallese population. Despite representing only 2.4% of the total population, Marshallese individuals in Washington and Benton counties accounted for 19% of cases in the area [21]. Nationally, 144.1 per 100,000 of COVID-19 cases (< 0.01%) require hospitalization; however, among Marshallese COVID-19 cases in Arkansas in June 2020, 9% were hospitalized [21]. Additionally, 38% of all COVID-19 related deaths in the two-county region were Marshallese, and 28% of all COVID-19 infections among Marshallese in Washington and Benton counties were tied to employment in poultry processing plants [21].
COVID-19, Food Processing, and the Pillars of Ethical Action
Despite their importance in keeping food on the table for millions of Americans and their label as “essential critical infrastructure workers,” workers in the food processing industries were not afforded the same level of concern during the COVID-19 pandemic as workers in other essential occupations [1, 22, 23]. In the early stages of the COVID-19 pandemic, few food processing companies set up mandatory safety practices, such as physical distancing measures and shared-space sanitation procedures [1, 3, 22]. These practices are particularly important in the food processing industry, where employees share potentially contaminated work tools and surfaces, endure prolonged close contact in enclosed spaces, and share housing or transportation with other workers [22, 24]. Despite this combination of environmental factors that significantly heighten the risk of COVID-19 transmission, few companies implemented policies and procedures for COVID-19 testing, health screenings, sick leave, or health care [1].
Official guidance for suitable protections was only released after numerous large outbreaks had occurred in many areas of the country [1, 22], and even then, guidance was voluntary. In addition, uptake by food processing companies was often inconsistent and incomplete, leaving many food processing workers at high risk of infection but afraid to lose their source of income for missing work without sick leave [1, 22]. The CDC’s northwest Arkansas field report on the COVID-19 pandemic recommended a multisector and multiagency plan that would (1) “communicate and collaborate with employers, particularly with poultry processing facilities, on testing, case investigation, and contact tracing” and (2) “encourage employers to implement policies that support COVID-19 prevention and mitigation efforts,” reinforcing the need for additional protections for food processing workers [21].
To address the circumstances of workers in food processing plants, researchers at the University of Nebraska Medical Center (UNMC) developed three pillars of ethical action to address the needs of food processing workers during the COVID-19 pandemic and for use in future public health emergencies [1]. The three pillars of ethical action include: (1) worksite prevention and control; (2) community-based prevention and control; and (3) treatment [1]. These pillars of ethical action can be used to evaluate the needs of food processing workers and guide advocates in developing the best course of action to prioritize the health and safety of these essential workers [1].
The goal of this study was to 1) conduct an exploratory assessment of COVID-19 mitigation steps of Marshallese food processing workers and their employers, guided by the three key pillars of ethical action as defined by Ramos et al. [1] and 2) compare workplace experiences during the COVID-19 pandemic with Marshallese workers in other occupations.
Methods
This study was approved by University of Arkansas Institutional Review Board (#261131).
Data Collection
Data were collected as part of a survey of the impacts of COVID-19 on Marshallese communities. Data collection for the online survey began in July 2020 and concluded in November 2020. Survey data were collected through Research Electronic Data Capture (REDCap), a web-based software for data capture and management. The survey was written in English and Marshallese to accommodate all potential participants and took approximately 30 minutes to complete.
Marshallese adults (aged 18 and older) residing in the continental US and Hawaii were eligible to take part in the study. Recruitment for the online survey was conducted by sharing digital information flyers on the Facebook pages and websites of Marshallese-serving community-based organizations. The flyers were produced in both Marshallese and English and contained a link to the REDCap survey. Each participant completed an online consent prior to beginning the survey, and those who completed the survey received a $20 Walmart gift card.
Measures
To address the aims of the study, portions of the online survey utilized validated items from multiple sources. Demographic information was captured using items from the Behavioral Risk Factor Surveillance System [25], and questions related to the COVID-19 pandemic were captured using items from the PhenX Toolkit [26]. Differences were examined on domains related to the Pillars of Ethical Action: 1) Demographics, including socioeconomic and health-related topics [25]; 2) Worksite Prevention and Control, comprising preventative measures taken by employers (responses were dichotomous [No/Yes]) on the following prevention measures: required work from home, suggested work from home, required face masks, suggested face masks, provided face masks, required social distancing, installed barriers/shields, checked employee temperatures before entering, screened employees before entering, and tested for COVID-19) and the perceived risk of contracting COVID-19 at work (No/Yes) [26]; 3) Community-Based Prevention and Control, including personal preventative behaviors (responses were dichotomous [No/Yes]) on the following prevention behaviors: hand washing, avoiding touching eyes/nose/mouth, use of disinfectants to clean hands, staying home when sick, asking family/friends not to visit, covering coughs, wearing a face mask, physical distancing, staying home when not sick, and disinfecting surfaces) and information gathering (responses were dichotomous [No/Yes]) on the use and trust of various sources of information on COVID-19; see Table 3) [26]; and 4) Treatment, comprising availability of sick leave (No/Yes) [26], insurance status (uninsured/insured) [25], and access to COVID-19 testing (response options included: never tried to get tested; tried to get tested but couldn’t; yes, waiting on results; yes, tested negative; and yes, tested positive) [26].
Table 3.
Community-based prevention and control by employment industry
| Food processing workers (n=31) |
Workers in other occupations (n=82) |
p-value | |
|---|---|---|---|
| Personal preventative measures taken a,b | |||
| Wash hands for ≥20 seconds | 31 (100) | 76 (96.2) | 0.558 |
| Avoid touching eyes, nose, mouth | 30 (96.8) | 75 (94.9) | 0.877 |
| Use disinfectants to clean hands when soap not available | 31 (100) | 76 (96.2) | 0.524 |
| Stay home when sick or have cold | 30 (96.8) | 72 (91.1) | 0.153 |
| Asked family/friends not to visit | 28 (90.3) | 70 (89.7) | 0.829 |
| Cover mouth when coughing | 31 (100) | 75 (94.9) | 0.268 |
| Wear face mask | 31 (100) | 77 (97.5) | 0.527 |
| Practice physical distancing | 31 (100) | 75 (94.9) | 0.365 |
| Stay home when not sick | 24 (77.4) | 68 (86.1) | 0.022 |
| Disinfect surfaces | 31 (100) | 76 (96.2) | 0.368 |
| Sources used to stay informed about COVID-19 a,b | |||
| Facebook posts from UAMS | 23 (74.2) | 52 (63.4) | 0.279 |
| Educational flyers about COVID-19 in Marshallese | 11 (35.5) | 40 (48.9) | 0.205 |
| Flyers about resources, screening, and testing for COVID-19 in the area | 7 (22.6) | 47 (57.3) | 0.001 |
| COVID-19 Marshallese Screening Hotline | 8 (25.8) | 23 (28.1) | 0.812 |
| Dr. Sheldon Riklon's radio show or Facebook Live videos | 16 (51.6) | 48 (58.4) | 0.508 |
| Information from the Marshallese Consulate/RMI Ministry of Health | 13 (41.9) | 40 (48.8) | 0.515 |
| Other | 2 (6.5) | 14 (17.1) | 0.149 |
| None of the above | 1 (3.2) | 1 (2.4) | 0.816 |
| Most reliable sources of information about COVID-19 b | 0.937 | ||
| Your pastor | 1 (3.3) | 0 (0.0) | |
| UAMS | 3 (10.0) | 6 (7.9) | |
| Community Clinic | 4 (13.3) | 11 (14.5) | |
| Another healthcare organization | 0 (0.0) | 1 (1.3) | |
| Dr. Sheldon Riklon's radio show or Facebook Live videos | 3 (10.0) | 11 (14.5) | |
| News outlet | 1 (3.3) | 3 (4.0) | |
| US Department of Health (e.g., CDC, NIH, and other federal health agencies) | 12 (40.0) | 26 (34.2) | |
| Your state Department of Health | 5 (16.7) | 10 (13.2) | |
| Marshallese Consulate/RMI Ministry of Health | 1 (3.3) | 6 (8.0) | |
| Other | 0 (0.0) | 1 (1.3) | |
| None | 0 (0.0) | 1 (1.3) | |
Note: Data are n (%), unless otherwise indicated. CDC=Centers for Disease Control and Prevention; NIH=National Institutes of Health. RMI=Republic of the Marshall Islands; UAMS=University of Arkansas for Medical Sciences. p-values in bold indicate significance.
Respondents were asked to select all that apply.
Fisher’s Exact Test
Using responses to the question, “Prior to the COVID-19 outbreak, how would you best describe your primary job sector?” the sample was divided into two categories: food processing workers and workers in all other occupations.
Data Analysis
Quantitative data analysis of the survey results included descriptive and inferential statistical techniques. We used Wilcoxon-Mann-Whitney U tests and Fisher’s Exact tests to examine differences between workers in food processing plants and workers from all other occupations. For each analysis, each participant’s data are included if she or he responded to the relevant items, regardless of the amount of missing data for that participant on other items. No imputation was carried out for missing responses for any items. Responses of “Don’t Know” were excluded from analysis. Significance was set at alpha level 0.05. All analyses were carried out using SAS/STATv14.1 [27].
A total of 294 individuals were recruited to take part in the online survey and completed the eligibility screener. Of those, 41 were deemed ineligible per inclusion criteria. An additional five were determined to be duplicate records and were excluded. Per inclusion criteria, 248 individuals were deemed eligible to complete the survey. However, 128 individuals completed fewer than 10 survey items and were excluded from the final analytical sample.
Results
Characteristics of Participants
The final sample included 120 participants, with 113 reporting being employed at the time of the survey. Food processing was the most commonly reported employment sector. Of those currently employed at the time of the survey, 31 were employed in food processing plants and 82 were employed in another occupation (e.g., healthcare, education, service and sales). Food processing workers and workers in other occupations did not differ statistically in mean age, sex, marital status, household size, household income, place of birth, or health status (Table 1). However, less than one-third of food processing workers (29%) reported having attended or completed higher education compared to workers in other occupations (74.4%) (p<0.001). More workers in other occupations reported living in the US since birth or for more than 15 years (59.3%), speaking English very well (52.4%), and having a primary care provider (65.3%) compared with those employed in food processing plants (30.0%, 19.4%, 26.7%), (p=0.032, p=0.006, p<0.001, respectively). Fewer food processing workers felt ‘extremely’ or ‘quite a bit’ comfortable filling out medical forms (56.7%) compared to workers in other occupations (74.0%) (p=0.034).
Table 1.
Participant demographics by employment industry
| Food processing workers (n=31) |
Workers in other occupations (n=82) |
p-value | |
|---|---|---|---|
| Age, years a | 33.5 ± 7.8 | 36.1 ± 9.3 | 0.153 |
| Sex b | 0.228 | ||
| Female | 17 (54.8) | 55 (67.1) | |
| Male | 14 (45.2) | 27 (32.9) | |
| Education b | <0.001 | ||
| <HS | 4 (12.9) | 3 (3.7) | |
| HS diploma/GED | 18 (58.1) | 18 (22.0) | |
| >HS | 9 (29.0) | 61 (74.4) | |
| Marital status b | 0.121 | ||
| Single/widowed/separated/divorced | 4 (12.9) | 21 (25.9) | |
| Married/unmarried couple | 27 (87.1) | 60 (74.1) | |
| Household size a | 4.9 ± 2.6 | 4.8 ± 2.7 | 0.709 |
| Monthly household income b | 0.300 | ||
| $0-999 | 7 (26.9) | 9 (11.1) | |
| $1000-1999 | 7 (26.9) | 23 (29.9) | |
| $2000-2999 | 5 (19.2) | 22 (28.6) | |
| $3000+ | 7 (26.9) | 23 (29.9) | |
| Place of birth b | 0.147 | ||
| US/Other | 2 (6.5) | 15 (18.3) | |
| RMI | 29 (93.5) | 67 (81.7) | |
| Length of US residence b | 0.032 | ||
| <5 years | 5 (16.7) | 7 (8.6) | |
| 5 to <10 years | 8 (26.7) | 17 (21.0) | |
| 10 to <15 years | 8 (26.7) | 9 (11.1) | |
| 15+ years/Since birth | 9 (30.0) | 48 (59.3) | |
| How well do you speak English? b | 0.006 | ||
| Very well | 6 (19.4) | 43 (52.4) | |
| Well | 19 (61.3) | 31 (37.8) | |
| Not well/Not at all | 6 (19.4) | 8 (9.8) | |
| How comfortable are you filling out medical forms? b | 0.034 | ||
| Extremely/Quite a bit | 17 (56.7) | 57 (74.0) | |
| Somewhat/A little bit | 13 (43.3) | 16 (20.8) | |
| Not at all | 0 (0.0) | 4 (5.2) | |
| Self-rated health status b | 0.378 | ||
| Excellent | 7 (23.3) | 19 (23.5) | |
| Good | 19 (63.3) | 41 (50.6) | |
| Fair/Poor | 4 (13.3) | 21 (25.9) | |
| Has a personal doctor/primary care provider b | <0.001 | ||
| Yes | 8 (26.7) | 49 (65.3) | |
| No | 22 (73.3) | 26 (34.7) | |
Note: Data are n (%) or mean ± SD, unless otherwise indicated. Percentages have been rounded and therefore may not total to 100. HS=high school; RMI=Republic of the Marshall Islands. p-values in bold indicate statistical significance.
Wilcoxon-Mann-Whitney U Test
Fisher’s Exact Test
Worksite Prevention and Control
More of those employed in food processing plants reported their employers installed barriers or provided shields (45.2%) and tested for COVID-19 (61.3%) compared to those employed in other occupations (19.5%, 34.1%) (p=0.006, p=0.009). Less than half of workers in other occupations had their temperature taken by their employer (45.1%), compared to nearly three-fourths (71.0%) of food processing workers (p=0.014). Neither group was more likely to report that their job puts them at risk of getting COVID-19 (Table 2).
Table 2.
Worksite prevention and control by employment industry
| Food processing workers (n=31) |
Workers in other occupations (n=82) |
p-value | |
|---|---|---|---|
| Preventative measures implemented by employer a,b | |||
| Required work from home | 1 (3.2) | 26 (31.7) | 0.002 |
| Suggested work from home | 1 (3.2) | 22 (26.8) | 0.005 |
| Required face masks | 22 (71.0) | 49 (59.8) | 0.271 |
| Suggested face masks | 13 (41.9) | 28 (34.1) | 0.442 |
| Provided face masks | 20 (64.5) | 37 (45.1) | 0.066 |
| Required social distancing | 19 (61.3) | 39 (47.6) | 0.193 |
| Installed barriers/ shields to reduce contact | 14 (45.2) | 16 (19.5) | 0.006 |
| Take employee temps before entering | 22 (71.0) | 37 (45.1) | 0.014 |
| Screened employees before entering | 12 (38.7) | 27 (32.9) | 0.564 |
| Tested for COVID-19 | 19 (61.3) | 28 (34.1) | 0.009 |
| My job puts me at risk of getting COVID-19 b | 0.493 | ||
| Yes | 5 (16.1) | 18 (22.0) | |
| No | 26 (83.9) | 64 (78.0) | |
Note: Data are n (%), unless otherwise indicated. p-values in bold indicate statistical significance.
Respondents were asked to select all that apply.
Fisher’s Exact Test
Community-Based Prevention and Control
With the exception of staying home when not sick, food processing workers practiced personal preventative measures at the same rate as workers in other occupations, with both groups consistently reporting 90% or higher on all other preventative behaviors measured (hand washing, avoiding touching eyes/nose/mouth, use of disinfectants to clean hands, staying home when sick, asking family/friends not to visit, covering coughs, wearing a face mask, physical distancing, disinfecting surfaces; see Table 3). A higher proportion of those employed in other occupations reported staying home when they were not sick (86.1%) compared to those employed in food processing occupations (77.4%) (p=0.022).
Food processing workers and workers in other occupations used all but one information source at the same rate to stay informed about COVID-19. Workers in other occupations reported using educational flyers about resources, screening, and testing for COVID-19 (printed in Marshallese) at twice the rate of food processing workers (p=0.001). Food processing workers were equally likely as workers in other occupations to find each source of information reliable (p=0.937), with the highest proportion of both groups perceiving the US Department of Health (e.g. CDC, National Institutes of Health [NIH]) as a reliable source of information.
Treatment
A larger proportion of food processing workers (74.1%) reported having no sick leave available to them compared to workers in other occupations (48.7%; p=0.023). However, they reported being insured at similar rates (56.7% v. 66.2%; p=0.360) (Table 4).
Table 4.
Treatment by employment industry
| Food processing workers (n=31) |
Workers in other occupations (n=82) |
p-value | |
|---|---|---|---|
| Employer grants sick leave a | 0.023 | ||
| Yes | 7 (25.9) | 38 (51.4) | |
| No | 20 (74.1) | 36 (48.7) | |
| Insurance status a | 0.360 | ||
| Insured | 17 (56.7) | 49 (66.2) | |
| Uninsured | 13 (43.3) | 25 (33.8) | |
| COVID-19 testing a,b | |||
| Never tried to get tested | 2 (6.5) | 17 (20.7) | 0.092 |
| Tried to get tested but couldn’t | 0 (0) | 1 (1.2) | 1.00 |
| Yes, waiting on results | 2 (6.5) | 4 (4.9) | 0.665 |
| Yes, tested negative | 18 (58.1) | 32 (39.0) | 0.090 |
| Yes, tested positive | 11 (35.5) | 21 (25.6) | 0.351 |
| Changes due to COVID-19 a,b | |||
| Lost job permanently | 3 (9.7) | 8 (9.8) | 1.00 |
| Lost job temporarily | 3 (9.7) | 9 (11.0) | 1.00 |
| Reduced work hours | 4 (12.9) | 23 (28.0) | 0.137 |
| Income decreased | 14 (45.2) | 50 (61.7) | 0.137 |
| Lost insurance during COVID-19 | 6 (20.0) | 10 (13.0) | 0.376 |
Note: Data are n (%), unless otherwise indicated. Percentages have been rounded and therefore may not total to 100. p-values in bold indicate significance.
Fisher’s Exact Test
Respondents were asked to select all that apply.
The proportion of respondents reporting being tested for COVID-19 did not differ between the groups. Additionally, changes in employment, income, and insurance related to the COVID-19 pandemic did not differ by occupation group.
Discussion
The aims of this study were to conduct an exploratory assessment of COVID-19 mitigation steps of Marshallese food processing workers and their employers, guided by the three key pillars of ethical action as defined by Ramos et al. [1] and to compare workplace experiences during the COVID-19 pandemic with Marshallese workers in other occupations. Our findings highlight the efforts made to protect food processing workers later in the COVID-19 pandemic, as well as the differences and similarities in the experiences and personal preventative measures of Marshallese workers employed in food processing compared with those employed in other occupations.
Our results provide evidence of efforts to prevent the transmission of SARS-CoV-2 within food processing plants. A larger proportion of Marshallese food processing workers reported their employers installed barriers or provided shields, provided temperature screenings, and tested for COVID-19 compared to those employed in other occupations. Mass testing to contain the spread of COVID-19 among employees in pork processing plants has been reported, though testing strategies varied between plants and showed varying levels of efficacy [28]. Additionally, although not significantly different than workers in other occupations, a large percentage of food processing workers reported their employers required the use of face masks at work and required social distancing. This finding contrasts those of a similar study among food processing workers in which survey respondents reported their employers had implemented a universal mask policy but had not enforced physical distancing on the production lines [29].
Surprisingly, food processing workers did not report that their workplaces put them at risk of contracting COVID-19 at a higher rate than those in other occupations, despite outbreaks in meat and poultry processing plants across the country [11, 22, 24, 30], as well as within Marshallese communities in northwest Arkansas [21]. In contrast, in a similar study conducted in the Midwest, more than 70% of respondents reported they were at high risk of contracting COVID-19 [29]. These contrasting reports may indicate that some food processing plants implemented and enforced protective measures, while other food processors did not. Inconsistent implementation and enforcement of protective measures across the food processing industry demonstrates there is still a need for public health specialists to communicate and collaborate with employers to design policies that support prevention and mitigation efforts for COVID-19 and to create a comprehensive plan for future pandemics [31].
Food processing workers practiced personal preventative measures (e.g., washing hands, practicing physical distancing, and wearing facemasks) at the same rate as workers in other occupations, with the exception of more Marshallese employed in other occupations reporting staying home when they were not sick. Although the survey did not ask where individuals were spending their time during the COVID-19 pandemic, staying home as much as possible has been recommended to avoid infection. Marshallese communities are very collectivistic, and spending time with family members and friends is an important part of their culture, which may contribute to these findings [32, 33]. Moreover, isolation and staying home is difficult without the necessary resources. Community-based efforts in the Marshallese community in northwest Arkansas have shown the importance of providing support services to allow patients and their close contacts to self-quarantine upon exposure and positive test results, including the provision of essential items, such as food and medications, coordination with worksites, and coordination with community social and behavioral health services [32]. Future work will need to consider and address the needs of under-resourced communities when creating pandemic response plans.
Many of the Marshallese respondents reported relying on Facebook messaging from University of Arkansas for Medical Sciences (UAMS), Dr. Sheldon Riklon's radio and Facebook Live shows (Dr. Riklon is a Marshallese physician and a leader in the Marshallese community), and information from the Marshallese Consulate/RMI Ministry of Health to learn about COVID-19. However, Marshallese employed in other occupations reported using educational flyers about resources, screening, and testing for COVID-19 (printed in Marshallese) at twice the proportion of food processing workers. This finding is consistent with another study that found food processing workers had not received information related to proper mask use or asymptomatic/pre-symptomatic transmission, despite the posting of COVID-19-related signage throughout the facility [29]. Although the community-based prevention and control efforts (i.e., targeting multiple mediums for educational outreach) appear effective in both food processing workers and workers in other occupations, there is a need for additional printed materials on COVID-19 in the Marshallese language to be widely distributed [32].
A larger proportion of food processing workers reported having no sick leave available to them compared to workers in other occupations, although they reported COVID-19 testing and being insured at similar rates. Unfortunately this finding is not unique to our study. Food processing workers in another US study also reported a lack of paid sick leave [29]. This problem is also not unique to the US. Reports of migrant workers in meat processing facilities being excluded from illness benefits when forced to quarantine as a result of the pandemic have also been documented in Ireland [3]. Paid sick leave for employees is an important part of preventing the spread of infections like COVID-19 as it prevents presenteeism, through which infections can spread throughout a workforce [34]. Employers benefit from paid sick leave in numerous ways, including saving money by preventing employee turnover, increasing workforce productivity, and preventing sick employees from being injured on the job due to reduced physical or mental capacity [34].
Strengths & Limitations
The study is not without limitations. The sample is small, and although data were collected in multiple states, they may not be generalizable to the larger Marshallese population or to all workers in the food processing industry and should be interpreted with this limitation in mind. Further, the sample was uneven and had non-normal distributions; however, non-parametric statistical tests were used to address these limitations. Additionally, all of the measures were self-reported and, therefore, carry the risk of bias. This limitation is reduced through use of validated instruments, and prior work demonstrated limited effects even for sensitive questions (e.g., substance use) [35]. Additionally, recommended COVID-19 preventative measures and resulting behaviors have rapidly evolved, and given the cross-sectional nature of the study, we are not adequately able to account for changes in workplace policies or personal behaviors. Despite the limitations, this is the first article to examine the experiences of Marshallese working in food processing plants during the COVID-19 pandemic.
Conclusion
Food processing workers play a significant role in the economic infrastructure of the US and are essential to the sustenance and stability of the US population and economy [1-3]. However, our findings show that even as late as November 2020, preventative measures and protections were not consistently implemented across the food processing industry despite efforts across public health and community partners [32, 36]. This study is the first to report on the experience of Marshallese workers in food processing and will provide the basis for post-pandemic evaluations and future pandemic response planning.
Funding:
This study was supported by a Patient-Centered Outcomes Research Institute COVID-19 enhancement award (#AD-1603-34602). Additional support was provided by a Translational Research Institute award (#UL1TR003107) from the National Center for Advancing Translational Science of the National Institute of Health.
Footnotes
Disclosure: The authors have no conflicts of interest to disclose.
Data availability:
Participants in this study did not consent to their data being shared publicly, so supporting data are not available.
References
- 1.Ramos AK, Lowe AE, Herstein JJ, Schwedhelm S, Dineen KK, Lowe JJ. Invisible no more: The impact of COVID-19 on essential food production workers. Journal of agromedicine. 2020:1–5. [DOI] [PubMed] [Google Scholar]
- 2.Committee on a Framework for Assessing the Health Environmental and Social Effects of the Food System, Food and Nutrition Board, Board on Agriculture and Natural Resources, Institute of Medicine, National Research Council. A Framework for Assessing Effects of the Food System. Washington (DC): National Academies Press (US); 2015. [PubMed] [Google Scholar]
- 3.Reid A, Ronda-Perez E, Schenker MB. Migrant workers, essential workers, and COVID-19. American Journal of Industrial Medicine. 2021(64):73–7. doi: 10.1002/ajim.23209. [DOI] [PubMed] [Google Scholar]
- 4.Choi J, Constance D. Marshallese migrants and poultry processing. J Rural Soc Sci. 2019;34(1):1–29. [Google Scholar]
- 5.Grzywacz JG, Arcury TA, Marín A, Carrillo L, Coates ML, Burke B, et al. The organization of work: implications for injury and illness among immigrant Latino poultry-processing workers. Arch Environ Occup Health. 2007;62(1):19–26. Epub 2008/01/04. doi: 10.3200/aeoh.62.1.19-26. PubMed PMID: 18171643. [DOI] [PubMed] [Google Scholar]
- 6.Culp K, Brooks M, Rupe K, Zwerling C. Traumatic injury rates in meatpacking plant workers. J Agromedicine. 2008;13(1):7–16. doi: 10.1080/10599240801985373. [DOI] [PubMed] [Google Scholar]
- 7.Rosenbaum D, Mora D, Arcury T, Chen H, Quandt S. Employer differences in upper-body musculoskeletal disorders and pain among immigrant Latino poultry processing workers. J Agromedicine. 2014;19(4):384–94. doi: 10.1080/1059924X.2014.945710. [DOI] [PubMed] [Google Scholar]
- 8.Arcury T, Estrada J, Quandt S. Overcoming language and literacy barriers in safety and health training of agricultural workers. J Agromedicine. 2010;15(3):236–48. doi: 10.1080/1059924X.2010.486958. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Baker M, Peckham T, Seixas N. Estimating the burden of United States workers exposed to infection or disease: A key factor in containing risk of COVID-19 infection. PLoS One. 2020;15(4):e0232452. doi: 10.1371/journal.pone.0232452. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Saitone T, Schaefer K, Scheitrum D. COVID-19 morbidity and mortality in U.S. meatpacking counties. Food Policy. 2021;101:102072. doi: 10.1016/j.foodpol.2021.102072. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Waltenburg M, Rose C, Victoroff T, Butterfield M, Dillaha J, Heinzerling A, et al. Coronavirus disease among workers in food processing, food manufacturing, and agriculture workplaces. Emerg Infect Dis. 2021;27(1):243–9. doi: 10.3201/eid2701.203821. PubMed PMID: 33075274. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.108th United States Congress. Compact of Free Association Amendments Act of 2003 Washington, D.C.: U.S. Government Printing Office; 2003. Available from: http://www.gpo.gov/fdsys/pkg/PLAW-108publ188/html/PLAW-108publ188.htm. [Google Scholar]
- 13.Hixson L, Hepler B, Kim M. The Native Hawaiian and Other Pacific Islander population 2010. Washington, DC: United States Census Bureau, 2012. [Google Scholar]
- 14.United States Census Bureau. 2019 American Community Survey Demographic and Housing Estimates, Table DP05 Washington, DC: United States Census Bureau; 2020. Available from: https://data.census.gov/cedsci/table?q=arkansas&tid=ACSDP1Y2019.DP05. [Google Scholar]
- 15.Jimeno S, Rafael A. A Profile of the Marshallese Community in Arkansas, Volume 3. Little Rock, AR and Fayetteville, AR: Winthrop Rockefeller Foundation; University of Arkansas, 2013. [Google Scholar]
- 16.McElfish P, Rowland B, Long C, Hudson J, Piel M, Buron B, et al. Diabetes and hypertension in Marshallese adults: Results from faith-based health screenings. Journal of Racial and Ethnic Health Disparities. 2017;4(6):1042–50. Epub 2016/11/12. doi: 10.1007/s40615-016-0308-y. PubMed PMID: ; PubMed Central PMCID: Pmc5426989. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Williamson EJ, Walker AJ, Bhaskaran K, Bacon S, Bates C, Morton CE, et al. Factors associated with COVID-19-related death using OpenSAFELY. Nature. 2020;584(7821):430–6. doi: 10.1038/s41586-020-2521-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Vahey GM, McDonald E, Marshall K, Martin SW, Chun H, Herlihy R, et al. Risk factors for hospitalization among persons with COVID-19—Colorado. PloS one. 2021;16(9):e0256917. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Van Dyke ME, Mendoza MC, Li W, Parker EM, Belay B, Davis EM, et al. Racial and ethnic disparities in COVID-19 incidence by age, sex, and period among persons aged< 25 years—16 US jurisdictions, January 1–December 31, 2020. MMWR Morb Mortal Wkly Rep. 2021;70(11):382–388. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Kaholokula JK, Samoa RA, Miyamoto RES, Palafox N, Daniels SA. COVID-19 Special Column: COVID-19 Hits Native Hawaiian and Pacific Islander Communities the Hardest. Hawaii J Health Soc Welf. 2020;79(5):144–6. PubMed PMID: 32432218; PubMed Central PMCID: PMC7226312. [PMC free article] [PubMed] [Google Scholar]
- 21.Centers for Disease Control and Prevention. Summary Report CDC AR-3 Field Team COVID-19 among Hispanic and Marshallese communities in Benton and Washington Counties, Arkansas. Atlanta, GA: Centers for Disease Control and Prevention,, 2020. [Google Scholar]
- 22.Stephenson J. COVID-19 Outbreaks Among Food Production Workers May Intensify Pandemic’s Disproportionate Effects on People of Color. JAMA Health Forum. 2020;1(6):e200783–e. doi: 10.1001/jamahealthforum.2020.0783. [DOI] [PubMed] [Google Scholar]
- 23.Lancet. The plight of essential workers during the COVID-19 pandemic. Lancet (London, England). 2020;395(10237):1587. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Dyal J, Grant M, Broadwater K, Bjork A, Waltenburg M, Gibbins J, et al. COVID-19 Among Workers in Meat and Poultry Processing Facilities - 19 States, April 2020. MMWR Morb Mortal Wkly Rep. 2020;69(18):557–61. doi: 10.15585/mmwr.mm6918e3. PubMed PMID: 32379731. [DOI] [PubMed] [Google Scholar]
- 25.Centers for Disease Control and Prevention. 2019 BRFSS Questionnaire. Atlanta, GA: Centers for Disease Control and Prevention; 2019. Available from: https://www.cdc.gov/brfss/questionnaires/pdf-ques/2019-BRFSS-Questionnaire-508.pdf. [Google Scholar]
- 26.COVID-19 Protocols: PhenX Toolkit; 2020. Available from: https://www.phenxtoolkit.org/covid19/source. [Google Scholar]
- 27.SAS Institute Inc. SAS/STAT. 14.1 ed. Cary, NC; 2016. [Google Scholar]
- 28.VanderWaal K, Black L, Hodge J, Bedada A, Dee S. Modeling transmission dynamics and effectiveness of worker screening programs for SARS-CoV-2 in pork processing plants. medRxiv. 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Ramos AK, Lowe A, Herstein JJ, Trinidad N, Carvajal-Suarez M, Quintero S, et al. A Rapid-Response Survey of Essential Workers in Midwestern Meatpacking Plants: Perspectives on COVID-19 Responses in the Workplace. Journal of Environmental Health. 2021;84(1):16–25. [Google Scholar]
- 30.Waltenburg MA, Victoroff T, Rose CE, Butterfield M, Jervis RH, Fedak KM, et al. Update: COVID-19 among workers in meat and poultry processing facilities—United States, April–May 2020. Morbidity and Mortality Weekly Report. 2020;69(27):887. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Middleton J, Reintjes R, Lopes H. Meat plants—a new front line in the covid-19 pandemic. BMJ. 2020;370:m2716. doi: 10.1136/bmj.m2716. [DOI] [PubMed] [Google Scholar]
- 32.McElfish P, Purvis R, Willis D, Riklon S. COVID-19 Disparities among Marshallese Pacific Islanders. Preventing Chronic Disease. 2021;18(200407). doi: 10.5888/pcd18.20040. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.McElfish PA, Moore R, Laelan M, Ayers BL. Using CBPR to address health disparities with the Marshallese community in Arkansas. Ann Hum Biol. 2018;45(3):264–71. Epub 2018/06/08. doi: 10.1080/03014460.2018.1461927. PubMed PMID: 29877159. [DOI] [PubMed] [Google Scholar]
- 34.Asfaw A, Rosa R, Pana-Cryan R. Potential Economic Benefits of Paid Sick Leave in Reducing Absenteeism Related to the Spread of Influenza-Like Illness. J Occup Environ Med. 2017;59(9):822–9. Epub 2017/07/12. doi: 10.1097/jom.0000000000001076. PubMed PMID: 28692009; PubMed Central PMCID: PMC5649342. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Habecker P, Ivanich J. Unintended Interviewer Bias in a Community-based Participatory Research Randomized Control Trial among American Indian Youth. In: Interviewer Effects from a Total Survey Error Perspective. 2020. Boca Raton, FL: Chapman and Hall/CRC. [Google Scholar]
- 36.English E, Long CR, Langston K, Faitak B, Brown AL, Echegoyen A, et al. A community partnership for home delivery of food boxes to COVID-19 quarantined and isolated families. Journal of Hunger and Environmental Nutrition. 2021;16(1):19–28. doi: 10.1080/19320248.2020.1863284. [DOI] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Participants in this study did not consent to their data being shared publicly, so supporting data are not available.