Much of the discourse around COVID-19’s spread in Hawaiʻi has centered around the risk of tourists bringing the virus into the state. But less attention has been paid to the tourism industry — specifically, the risk of COVID-19 exposure that hospitality and food service workers face at their jobs.
Hawaiʻi has a disproportionately high number of non-healthcare workers at high risk of workplace exposure to COVID-19 — perhaps the largest share of any state — according to my analysis of publicly available data. Although many employers may be taking precautions, Hawaiʻi Department of Health reports have confirmed that some types of workplaces are still a significant source of transmission. This prevalence of systemic occupational risk undermines the notion that “personal responsibility” is enough to stop the spread of COVID-19, and underscores the direct connection between individual health and economic health.
Nearly 3 in 5 Hawaiʻi workers (outside of healthcare) have jobs with high degrees of workplace risk factors, according to state-level employment data applied to a framework published in March 2021 by the Council of State and Territorial Epidemiologists. Those risk factors — “routine in-person interaction with the public,” “working indoors,” and “working in close physical proximity to others” — are most common in the “essential” jobs that generally can’t be done from home. (In this framework, healthcare workers are presumed to already have high occupational risk for COVID-19 exposure.) The researchers write that understanding workplace risk factors is essential to “protect workers and, in turn, the communities in which they work and live.”
This aligns with the Hawaiʻi Department of Health (DOH) cluster reports, which have frequently homed in on transmission at workplaces — the same ones that the CSTE methodology considers high-risk. The reports frequently discuss clusters initiated between employees at high-risk occupational settings, including:
- Restaurants and bars (November 27, 2020; April 22, 2021; June 3, 2021; July 8, 2021; July 22, 2021; September 2, 2021; spread among restaurant patrons was also discussed on February 11, 2021)
- Hotel and accommodation settings (January 28, 2021; July 22, 2021)
- Construction and industrial settings (January 21, 2021; May 20, 2021; July 22, 2021)
- Agricultural settings (February 4, 2021)
Indeed, 26 of the 31 DOH cluster reports in 2021 include clusters at restaurants, and 24 reports include clusters in the “travel, lodging & tourism” category. (In comparison, 23 reports include clusters attributed to social gatherings. The most frequently mentioned cluster category is correctional facilities. Educational settings are also mentioned in 26 reports.)
California researchers have also found concrete evidence of the importance of workplace COVID-19 transmission. A groundbreaking study published in June 2021 found that COVID-19-related deaths in California were linked to occupational risk. “Our analysis of deaths among Californians between the ages of 18 and 65 shows that the pandemic’s effects on mortality have been greatest among essential workers, particularly those in the facilities, food/agriculture, manufacturing and transportation/logistics sectors,” wrote the researchers.
High-risk jobs — in service, hospitality, and other industries — share common factors
The risk of transmitting COVID-19 is highest in closed spaces with poor ventilation, crowded places, and close-contact settings, particularly if people aren’t wearing masks and aren’t vaccinated. These systemic, occupational risk factors are often out of the control of the workers who are most affected, such as cooks, housekeepers, food suppliers, and construction workers. It’s not hard to imagine why certain occupational settings present a higher intrinsic risk, even though some businesses may already be implementing mitigation measures. The DOH cluster reports provide some anecdotal insight:
Restaurants and bars: “Cooks and other kitchen staff often work in more cramped conditions and spend longer cumulative periods of time in close contact with their coworkers … Working for long periods over a hot cook line can also make it difficult to fully comply with mask mandates. If a few employees call off work, short-staffing can add further risk as remaining employees may be forced to cross-cover multiple roles leading to increased mixing among staff.” (April 21, 2021)
Hotel and accommodation: “Hotel workers might come into contact with the virus when in close contact … with other people at work, which can include both guests and coworkers. Because hotel industry employees have contact with travelers arriving from parts of the world that may be experiencing outbreaks of COVID-19, concern is heightened that those working in and adjacent to the hotel and accommodation industry could be a source of local outbreaks if they contract COVID-19 at work and then transmit it to household members and other community contacts.” (January 28, 2021)
Construction and industrial: “Their occupations can require sustained physical exertion and working in uncomfortable environments … Workers reported hot temperatures, poor ventilation, and removing their masks due to uncomfortable conditions inside the building. In addition, workers reported not being able to physically distance the recommended 6 feet while working and not practicing social distancing during lunch.” (May 20, 2021)
Agriculture: “Workers in these settings often have close and prolonged contact with one another in the fields, during breaks, when sharing transportation, or in shared housing. Depending on the setting and type of work, shared housing or transportation services for agricultural workers may be provided or arranged by the employer, a contracted service, or by the employees themselves.” (February 4, 2021)
The occupational risk framework from the Council of State and Territorial Epidemiologists characterizes about 13% of non-healthcare jobs as having the highest risk factors for workplace COVID-19 exposure. These are jobs in which employees work in very close proximity to others (either coworkers or the public), primarily indoors, and in public-facing roles. This category includes the jobs described above — like waiters, food service workers, cashiers, and hotel clerks — along with retail salespeople, teachers, and child care workers.
Hawaiʻi has a disproportionate number of workers employed in those highest-risk jobs, my analysis shows — over 30% of non-healthcare workers, which is again the highest of any state.
Another factor worth considering is a lack of paid sick leave. Research consistently shows that low-income and service-sector workers are less likely to have paid sick leave. A 2019 study found that just 45% of “leisure and hospitality” workers in Hawaiʻi had paid sick days. (There’s no national paid sick leave policy, and most states — including Hawaiʻi — don’t mandate paid sick leave, either.) “Given that many of these workers are also perilously close to a financial cliff, they face a strong incentive to work even when sick,” Daniel Schneider, a sociology professor at UC Berkeley, noted in 2020. Even though the pandemic hasn’t ended, the social safety net is already being scaled back — expanded unemployment benefits have ended, the state eviction moratorium has expired, and few isolation and quarantine units are available.
Workers have voiced similar concerns, like this fast food employee in California cited in a 2021 report: “Even though there have been many cases of COVID-19 among coworkers at this McDonald’s, I have never been notified that I was in close contact with anyone. This does not make sense to me because I work in the kitchen, and the kitchen is small, so we cannot maintain physical distance. And [others] also work in the kitchen with me, and they all had COVID-19.”
(Under the American Rescue Plan Act, certain employers that voluntarily provide workers with paid sick and family leave related to COVID-19 can receive federal tax credits to offset their costs. However, those employees are not entitled to paid leave — employers must opt-in — and these tax credits are also expiring on September 30.)
Finally, researchers argue that occupational exposure helps to explain racial and ethnic disparities in COVID-19 cases and deaths. One study found that “[p]eople of color were more likely to be employed in essential industries and in occupations with more exposure to infections and close proximity to others. Black workers in particular faced an elevated risk for all of these factors.” This was echoed by the California excess mortality study, which found that “[b]oth Black and Latino workers experienced substantial excess mortality during the pandemic.”
This is also true in Hawaiʻi: Native Hawaiians, Pacific Islanders, and Filipinos, which have been disproportionately affected by COVID-19, “are more likely than other racial/ethnic groups to work and live in environments that increase their risk of exposure to COVID-19,” according to a March 2021 report published by the DOH. “Collectively, these three groups make up nearly half of the essential workforce in Hawaiʻi, especially in the tourism, hospitality, retail, and food industries,” the report continues.
Reiterating those concerns, the Council of State and Territorial Epidemiologists paper suggests that policymakers take into consideration occupational risk factors to reduce health inequities:
“Given that people of color are disproportionately employed in service and manufacturing occupations, which unlike many professional jobs, cannot be conducted from home, this information is also critical to reducing health inequities. … These jobs are often precarious, low-paying, and lacking in social benefits. Fear of job loss, economic need, and lack of paid sick leave are disincentives for staying home when workers fall ill or are exposed to SARS-CoV-2. Lack of health insurance may prevent infected workers from seeking health care, delaying treatment, and leading to more severe disease. Language barriers, poor working conditions, and concerns about immigration status may also lead to disparities in morbidity and mortality. It is critical to consider ways to assure equitable and effective application of interventions across all populations.”
Vaccination, ventilation, and other mitigation strategies
Pandemic safety *requires* occupational safety! We need better rules to protect people in their workplaces!! https://t.co/Sl9rCA6IWl
— Dr Ellie Murray, ScD (@EpiEllie) August 13, 2021
Early, anecdotal evidence suggests that the burgeoning vaccination mandates for workers in, and patrons of, high-risk industries appear to be successful in giving people the nudge they need to get vaccinated, which is increasing vaccination rates. The COVID-19 vaccines, which are safe and effective, are the cornerstone of protecting individual workers in risky occupational settings, along with their families and communities. New federal rules requiring employers to provide paid time off to get vaccinated, and requiring employers with 100 or more employees to ensure their workers are vaccinated or tested recently, should also increase uptake. Wearing a mask — which has steadfastly remained a requirement under state policy — is also fundamental.
Over the past year, the Hawaiʻi DOH cluster reports have suggested additional mitigation strategies that employers can implement. Improving ventilation has been a frequent suggestion. For example, the April 21 report noted that “[b]ecause the physical layout of restaurant kitchens and the nature of prep and cooking work make physical distancing difficult, other mitigation strategies, such as engineering controls, adequate ventilation and proper mask usage, need to be applied in combination to the greatest extent possible.” The July 8 report also highlighted the risk of an “enclosed kitchen with poor ventilation and limited ability to physically distance,” as did the latest report on September 2, which cited “tight working spaces with poor ventilation, which probably contributed to the spread of COVID-19.”
DOH cluster reports also identify “employee work breaks” — when co-workers may eat and socialize without masks or physical distancing — as risky. Echoing CDC recommendations, the DOH writes that employers should “close shared employee spaces such as break rooms.” The CDC has recommendations on improving ventilation for workplaces in general and for restaurants specifically, and as I noted earlier this month, Hawaiʻi Governor David Ige has started emphasizing the risk of enclosed, poorly ventilated spaces. Some local restaurants have already started improving air quality.
Across the country, state-level data that analyzes COVID-19 cases by occupation or industry appears to be rare; more data could also provide more actionable insight. Washington released a report in December 2020, which found that the “health care and social assistance” industry had the highest percentage of cases, followed by retail, manufacturing, agriculture, construction, and accommodation and food services. This aligns with the California excess mortality study and the framework from the Council of State and Territorial Epidemiologists.
Technical notes: Data on workplace risk by state
In March 2021, the Council of State and Territorial Epidemiologists published a framework for “identifying occupations in which workers when present in the workplace are likely to be at increased risk of exposure to SARS-CoV-2.” Specifically, the CSTE paper focuses on non-healthcare workers, as state and federal authorities already consider healthcare workers to have a high risk of exposure. The authors identified three occupational factors that contribute to risk of exposure (“routine in-person interaction with the public,” “working indoors,” and “working in close physical proximity to others, either co-workers or the public”). They then scored about 700 occupations using those risk factors using data from O*NET, “a national database with information on occupational characteristics.”
The CSTE paper analyzed national employment data and suggested that “[s]tates electing to make use of [this framework] should consider using state level workforce estimates.” Such state-level workforce estimates from every state except Maine is available from the Projections Managing Partnership; Maine data is available through its state website. These workforce estimates use pre-pandemic data.
The maps in this blog post use that state-level employment data, categorized by the CSTE framework. Specifically, both maps use “Close Proximity Measure 2,” which combines O*NET scores with expert review. (Download the CSTE paper for more explanation about their methodology.) The first map uses all occupations in “high exposure combinations,” or “[o]ccupations designated as having the highest exposure level for at least two of the three exposure measures.” The second map uses occupations in “Combo 1,” which have the highest exposure levels in all three measures. The underlying data for both maps is available here.
The CSTE framework estimates risk for about 89% of all workers nationwide. The framework excludes 106 healthcare occupations (about 10% of workers) and another 47 occupations missing from O*NET (about 1% of workers).
The CSTE researchers note the limitations of their approach, such as a lack of data on “duration of exposure or population density in a work environment” and “what protective measures are in place.” This framework also does not analyze every single occupation. However, they note that this approach has nonetheless “proved to be a valuable method in occupational epidemiology in the absence of individual exposure data” and should be considered “baseline information on the potential risk of exposure to SARS-CoV-2 at work.”
