The United States food chain was among the sectors of the economy most affected by the COVID-19 pandemic. Millions of food services workers were laid off with little safety net to rely on,¹ while workers in other food chain sectors were deemed essential and continued working, often without adequate workplace safety measures,² resulting in negative mental health effects^3,4 and elevated COVID-19 infection and mortality rates compared to other workforces.^5–7 This drew attention to the health consequences of food chain workers’ employment and working conditions, and to their crucial role in ensuring the food supply for all.

Food chain workers constitute the single largest share of the US workforce, estimated at over 21 million workers, or 14% of the workforce, in 2016.⁸ The food chain has several main sectors (Appendix A, Supplementary Material): agricultural production (farm workers, fishers, and ranchers); manufacturing and processing (workers involved with converting raw goods into foods ready to sell or eat); wholesale and logistics (truck drivers, forklift operators, and warehouse managers); retail (stockers, cashiers, and in-house cooks involved in food sales in supermarkets, grocery stores, and convenience stores); and services (workers involved with food and drink preparation, serving, and facilities cleaning in commercial and institutional settings).^8,9 Each component of the food chain has a different profile in terms of worker demographics, working conditions, and employment conditions (including pay, benefits, job security, and union representation). Understanding these between-sector differences, and their impact on the health of workers in the different sectors, is key to identifying workforce-appropriate resources to address health inequities.

Conceptualizing work as a social determinant of health¹⁰ encourages a focus on the intersection of work and structural discrimination based on social position (including sex, race, ethnicity, and age) as co-drivers of social inequities in health. This approach also informs consideration of the potential contribution of both working conditions and employment conditions to those health inequities, and of the socio-political and economic forces that drive those conditions in each place. Thus, a social determinants of health framework brings attention to several characteristics and processes that might contribute to health inequities among US food chain workers. Overall, the US food chain is a highly racialized and gendered workforce, with women and Black, Latinx, and Asian people overrepresented but also often performing the lower paid, more manual, and more hidden forms of labor compared to people in the sector who are white.^11,12 Many food chain jobs involve risks, such as injury, workplace violence, or exposure to infectious disease, that are inherent in activities like working with sharp tools or heavy products, in proximity to animals or the public, or in fast-paced environments. In several sectors of the food chain, these risks are exacerbated by components of poor employment quality: lack of job security, inadequate pay, unpredictable schedules, few benefits, and limited workplace power^8,13 – all of which are associated with mental and physical health harms.^14,15 In the US, more than half of adults obtain health insurance through employment,¹⁶ and the level of other social safety net provisions such as unemployment insurance and retirement income are affected by employment as well. Notably, some groups of workers (including many food production and food services workers) have never had access to these provisions through their employment, while others have lost access over time with shifts in the economy and nature of work.¹⁷

Empirical studies suggest occupational stressors (e.g., workplace injuries, musculoskeletal symptoms), mental health, and physical health issues were common among food production,^18–21 retail,²² and service workers^23–25 even prior to the COVID-19 pandemic. However, because these are industry-wide studies, identifying priority areas for interventions is difficult. There are few peer-reviewed systematic assessments of food chain workers’ health, with the exception of an analysis of employer-reported occupational morbidity and mortality by Newman and colleagues (2015).²⁶ The researchers found that the occupational morbidity and mortality rates for food industries as a whole were higher than those of non-food industries combined. Within food industries, food warehousing and storage had the highest occupational morbidity rate, while food production had the highest occupational mortality rate.²⁶ Considering the food chain in its entirety, they argue, can help direct interventions towards this major part of the US economy.

To our knowledge, a social determinants of health perspective has not yet been used to systematically assess US food chain workers’ health. Most public health research involving food workers has focused on the implications of worker knowledge, skills, and beliefs for consumers’ food safety (e.g., Lipcsei et al., 2018; Feng and Bruhn, 2019; Yang et al., 2019; and Boyce and Schaffner, 2021)^27–30 – an important topic, but one that generally excludes the health needs and status of food chain workers themselves. Similarly, food system and food justice research, aimed at ensuring universal and environmentally sustainable access to nutritious, affordable, and culturally-appropriate food, often excludes the labor upholding the food system.³¹ While the relationship between specific occupational exposures and illness and injury is well established for certain segments of the food workforce, a social determinants of health framework¹⁰ pushes us to explore how other aspects of workers’ health and well-being might be linked with their work, and how these aspects might contribute to work-related health inequities across society.

This study seeks to fill these gaps in the literature, and to complement Newman and colleagues’ research, by assessing food chain workers’ healthcare access, health-related behaviors, and health outcomes, and by identifying needs for food chain subgroup-specific prevention and intervention efforts, including ones related to improving employment conditions. The COVID-19 pandemic brought acute and rapid changes to conditions, workforce sizes, and employer practices in some food industries that are still unfolding in many parts of the US. In order to understand pre-existing stressors associated with food chain work, we aim to describe the baseline health of workers along the food chain before the pandemic by using 2018–2019 data from the Behavioral Risk Factor Surveillance System (BRFSS).

Of the 222,771 respondents to the 2018–2019 BRFSS survey who indicated they were employed or self-employed, 185,114 met the industry and occupation criteria and were included in the analyses. Of these, 19,866 respondents were in the food chain industry and occupations of interest: 7,359 (production), 1,110 (manufacturing), 382 (wholesale), 2,551 (retail), 6,972 (commercial services), 1,042 (institutional services), and 450 (other food industries). The comparison group (all other workers) comprised 165,248 respondents.

Food chain workers differed from all other workers on several socio-demographic characteristics (Table 1). Food chain workers were more likely to be under the age of 35 (48%) than all other workers (30%). They were more likely to be Hispanic (32%) and less likely to identify with the other race/ethnicity categories available for analysis (50% non-Hispanic white, 10% non-Hispanic African American, and 7.5% non-Hispanic other). Food chain workers were 1.8 times more likely to rent their home and 1.8 times more likely to have a high school education or less than all other workers. Food chain workers were more likely to have household income of less than $50,000; 37% had household incomes of less than $25,000, compared to 14.3% of all other workers. Food chain workers were more likely to live in a nonmetropolitan urban (11%) or nonmetropolitan rural area (9%) than all other workers.

There were also significant socio-demographic differences comparing food chain workers by industry to all other workers. Food chain workers in commercial services, retail, and wholesale were more likely to be younger than 25 (33%, 24%, and 18%, respectively) than all other workers (9%), while institutional food services workers were more likely to be between 55 and 64 years of age (26%) than all other workers (19%). Food chain workers in production, manufacturing, and wholesale were more likely to be male (78%, 65%, and 85%, respectively), while institutional food services workers were more likely to be female (77%), than all other workers (54% male). Food chain workers in production, manufacturing, wholesale, and commercial services were more likely to identify as Hispanic (33%, 38%, 32%, and 37%, respectively), while those in institutional services were more likely to identify as Non-Hispanic African American (22%), compared to all other workers.

Compared to all other workers. a greater share of respondents had less than a high school education among production (35%), manufacturing (30%), retail (17%), and commercial services (22%), and a greater share of institutional services workers (50%) reported high school as their highest level of education. Food workers in production, retail, and commercial services were most likely to be low income; food workers in manufacturing and institutional services were generally middle income, and workers in food wholesale had the largest percentage of higher income workers.

Results from this examination of 2018–2019 BRFSS data indicate that, collectively, food chain workers had higher prevalences of barriers to healthcare access (lacking health insurance, lacking a personal doctor, and not having a flu shot or dental visit in the past year), smoking, no physical exercise, and poor self-reported health than all other workers. Patterns were similar for food chain workers in all industry sectors except wholesale, a group that was similar to all other workers. Additionally, commercial food services workers had higher prevalence of poor mental health, while institutional food services workers had higher prevalences of obesity, diabetes, and hypertension than all other workers. Most point estimates decreased somewhat after adjusting for income but retained statistical significance.

A focus on the entire food chain allows consideration of workers in different industries and occupations but who are unified by their role in the food system. A social determinants of health framework helps highlight the fact that unfortunately, though there are exceptions, many food chain workers are also unified by poor employment quality, as many food jobs are nonsupervisory positions with little job security, inadequate or unstable pay, unpredictable schedules, few benefits, and limited workplace power.^8,13 The food processing, wholesale, and retail industries have higher unionization rates (12.7%, 8.1%, and 13.9%, respectively, in 2015, based on the latest available comprehensive assessment conducted by the Food Chain Workers Alliance)⁸ than average for the food chain workforce as a whole (6%), allowing workers to negotiate for better employment and work conditions than those in non-unionized positions. However, unionization in food production (1.4%) and services (1.6%), especially commercial food services, remains dismally low,⁸ in context of low union membership nationwide.³³ The poor quality of employment in the food chain is attributed partly to historical exclusions of agricultural, service, and other workers from key labor protections (a practice with racist and anti-immigrant roots),¹¹ as well as increased consolidation of food chain components by large corporations,³⁴ whose use of temporary labor contracting to save costs, long common in food production where most workers are seasonal, has increased in food manufacturing, wholesale, and services. This strategy allows companies to avoid paying higher wages and providing health insurance and other benefits.^8,35

Guided by the understanding that work is a social determinant of health and health equity,^10,36 evidence of health issues from both employer-reported²⁶ and worker-reported data (the present study) among food chain workers strengthens the case for public health researchers and practitioners to turn our attention to the well-being of food chain workers, and not only to their role in preventing foodborne illness, in both our research and the types of interventions that should follow. In addition to workplace support for smoking cessation and physical exercise interventions, policies that ensure a living wage,³⁷ universal healthcare,^38,39 federal paid sick leave,⁴⁰ and protection of the right to organize,^41,42 regardless of full- or part-time employee status, would all likely help reduce the inequities in healthcare access, health behaviors, and health outcomes between food chain workers and the rest of the workforce. An assessment of the 100 largest food companies in the US found that few have pro-worker policies – for example related to a living wage, freedom of association, and training – and where these exist, they are often not implemented well or enforced.³⁴ Employers clearly have an important role in providing these protections. Policy approaches that decouple vital resources such as health insurance from employment and ensure the right of all workers to a living wage are also needed to ensure social protection for all. Strengthening workers’ influence in ensuring working conditions that enhance, rather than detract from, their health is also key.

Our results also highlight specific groups with the highest needs for workforce-specific resources such as prevention measures and interventions to improve working conditions and for effectiveness research to gauge the success of these changes. Though food wholesale workers, who are among the highest paid workers across the food chain,⁹ did not fare as poorly as other food chain workers, they did have higher prevalences of binge drinking and short sleep. There is limited prior health research on this specific workgroup; however, among workers in the transportation/warehousing industry generally, existing studies suggest elevated rates of short sleep,⁴³ but not excessive alcohol consumption.⁴⁴ Additional research is needed to corroborate and explore the reasons for both positive and negative findings among food wholesale workers.

Our finding of higher prevalence of poor mental health among commercial food services workers builds on prior research linking tipped service work with risk for developing mental illness.⁴⁵ Possible explanations include irregular and unstable scheduling, which results in income instability and has been linked with psychological distress among hourly workers in services.²² Workers in commercial food services can also experience stressors related to the emotional labor of interacting with customers and managing sexualized or hostile customer behavior, which is associated with adverse mental health.^25,46

Systematic data on institutional food jobs are lacking, as commercial and institutional food services workers are combined in most publicly available datasets, and there is limited prior research on the health of institutional food services workers. Institutional food services workers were more likely to be women between 55 and 64 years of age and to identify as African American than non-food chain workers. The regressions included age, gender, and race/ethnicity, but those variables may not fully account for differential exposure to stressors over the lifetime, including structural anti-Black racism, that increase barriers to healthcare and risk for a range of chronic diseases,⁴⁷ including the elevated levels of obesity, hypertension, and diabetes that we observed. Though institutional food services workers constitute a smaller share of the food workforce than commercial food services workers (e.g., in 2019 there were 402,480 institutional and cafeteria cooks compared to 1,401,890 restaurant cooks),⁴⁸ their higher prevalence of chronic health conditions and the strong role of the US government in setting standards and pay for these workers make them an important group to prioritize for intervention.

Among the study strengths, using both industry and occupation codes to isolate food chain workers allowed for more specificity than using industry alone, and disaggregating two types of food services workers revealed specific health needs among commercial and institutional food services workers that might not have emerged had the food services workers been grouped together. Strengths also pertain to the BRFSS’s large sample size, standardized weighting across states, and validated health measures. The COVID-19 pandemic highlighted the crucial role of food chain workers in society, as well as differences within the food chain that have implications for health inequities. BRFSS 2018–2019 data include the last pre-pandemic year and thus provide the best opportunity to obtain a baseline on the health and well-being of workers in different parts of the food chain. This baseline data creates the opportunity for future analyses of changes in food chain worker health and health inequities since the start of the pandemic that can take pre-existing stressors into account.

The study also has limitations. First, we excluded subsectors that are sometimes included in food chain conceptualizations, namely, those responsible for supplying manufactured inputs to the production sector (e.g., the agricultural chemical manufacturing industry), and for dealing with food loss, waste, and recovery. Second, the specific way industry and occupation are coded introduces some rigidity in the groups of food workers it was possible to isolate in the analysis (e.g., it is not possible to isolate food delivery workers classified as independent contractors). Misclassification bias is also possible because of errors in industry and occupation coding. Third, BRFSS is not nationally representative because 32 states used the optional industry and occupation module in 2018 and 2019, and the sample was designed to be representative across age, race/ethnicity, and gender, but not in terms of industry and occupation. Health indicators and outcomes are self-reported and thus subject to recall bias and social desirability bias; this may contribute to an underestimation of the true prevalence of harmful health behaviors and adverse health outcomes in this sample. Fourth, the income-adjusted models for the comparison of food chain sub-groups to all other workers are likely to be over adjusted, since income is part of the mechanism by which work influences health. However, the income-adjusted models are more useful for the first comparison, between all food chain workers and all other workers, given the variation in industries and occupations of the two groups.

The US food chain is a highly racialized and gendered workforce, with women and Black, Latinx, and Asian people both overrepresented and performing the lower paid forms of labor compared to whites in the sector.¹¹ There were clear signs of this occupational segregation in this analysis. Compared to the rest of the workforce, commercial food services workers were more likely to be younger, Hispanic, and to have less education and lower incomes; food manufacturing workers were more likely to be male, Hispanic, and to have less education but higher incomes; and institutional food services workers were more likely to be older, female, and African American, with higher levels of education and household income. This variation has implications for the ways work along the food chain might influence health risks, and for the types of interventions that might follow. This analysis points to the need for more granular work information in large-scale surveys such as the BRFSS to explore the relationship between work conditions, worker demographics and social stratification processes, and health. More broadly, conceptualizing work as a social determinant of health leads to the conclusion that addressing the health inequities observed in this study will require changing the socio-political and economic drivers of occupational segregation and low-quality employment.

A 2017 American Public Health Association policy statement notes that the millions of farm workers and food production workers in the US who help meet the public health goal of ensuring an accessible supply of nutritious food face poor employment and work conditions, contributing to health disparities.⁴⁹ The statement makes the case for expanding sustainable food systems goals of food safety, accessibility, environmental protection, and animal welfare to include just and equitable labor practices. The results of this study support extending this argument to the entire food chain and to health issues beyond occupational injuries, by integrating a concern for food chain workers’ health and well-being, and the employer practices and labor policies contributing to these outcomes, into the broader public health goals of ensuring sustainable and equitable food systems and reducing health inequities driven by social determinants including work.