To describe the work environment and COVID-19 mitigation measures for homeless shelter workers and assess occupational risk factors for COVID-19.
Between June 9-August 10, 2020, we conducted a self-administered survey among homeless shelter workers in Washington, Massachusetts, Utah, Maryland, and Georgia. We calculated frequencies for work environment, personal protective equipment use, and SARS-CoV-2 testing history. We used generalized linear models to produce unadjusted prevalence ratios (PR) to assess risk factors for SARS-CoV-2 infection.
Of the 106 respondents, 43.4% reported frequent close contact with clients; 75% were worried about work-related SARS-CoV-2 infections; 15% reported testing positive. Close contact with clients was associated with testing positive for SARS-CoV-2 (PR 3.97, 95%CI 1.06, 14.93).
Homeless shelter workers may be at risk of being exposed to individuals with COVID-19 during the course of their work. Frequent close contact with clients was associated with SARS-CoV-2 infection. Protecting these critical essential workers by implementing mitigation measures and prioritizing for COVID-19 vaccination is imperative during the pandemic.
Citation: Rao CY, Robinson T, Huster K, Laws RL, Keating R, Tobolowsky FA, et al. (2021) Occupational exposures and mitigation strategies among homeless shelter workers at risk of COVID-19. PLoS ONE 16(11): e0253108. https://doi.org/10.1371/journal.pone.0253108
Editor: David M. Ojcius, University of the Pacific, UNITED STATES
Received: July 8, 2021; Accepted: September 29, 2021; Published: November 1, 2021
This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.
Data Availability: All relevant data are within the manuscript and its Supporting information files.
Funding: The authors received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2), the virus that causes coronavirus disease 2019 (COVID-19), has spread rapidly among people experiencing homelessness in some homeless shelters throughout the United States [1,2]. In several reported outbreaks, homeless shelter workers have also been infected with the virus [3,4]. Homeless shelter workers provide critical infrastructure services  and often work in shared spaces with the potential for prolonged close contact with other staff and clients. A meta-analysis of COVID-19 in homeless shelters estimated a pooled prevalence among shelter workers of 14.8% during outbreaks and 1.55% during non-outbreak situations . Homeless shelter workers may be at increased risk for COVID-19 due to frequent exposure to disease and infectious agents in the workplace.
Homeless shelters provide an essential service and, like many congregate settings, have remained open during the COVID-19 pandemic. Homeless shelters have previously experienced outbreaks of tuberculosis, hepatitis A, and invasive bacterial disease [7–10]. However, very little information exists about the risks of exposure for homeless shelter workers. In a study of occupational exposures to infectious agents, an estimated 32.4% of community and social services sector workers are exposed > 1 time/month to infection or disease; 7.7% are estimated to be exposed > 1 time/week . Occupational health protections for shelter workers include following infection control processes, bringing their adult immunization status up to date, encouraging annual influenza and Hepatitis B vaccination, and screening/testing for tuberculosis [12,13].
Other crowded occupational environments, including long-term care facilities, correctional facilities, military facilities and cruise ships, have been associated with high risk of exposures and transmission of COVID-19 among workers [14–19]. Although various mitigation measures have been recommended to reduce homeless shelter worker COVID-19 risks , the implementation of these measures is unknown. To our knowledge, risk factors for COVID-19 among homeless shelter workers have not been previously described. We conducted a multi-site cross-sectional survey of homeless shelter workers to better understand SARS-CoV-2 occupational exposures, job practices, and COVID-19 mitigation measures.
Materials and methods
Homeless shelter selection and recruitment
Local public health and healthcare collaborators (e.g., local health departments, nongovernmental organizations, local government agencies) in Seattle, Washington; Boston, Massachusetts; Salt Lake City, Utah; Baltimore, Maryland; and across Georgia identified homeless shelters in their jurisdictions where at least one staff member had tested positive for SARS-CoV-2. We informed the shelter administrators of the survey objectives and requested participation of their staff. All staff who work at a shelter facility were eligible to participate. Public health partners or shelter administrators sent a recruitment email with a link to the online survey to all workers. The online survey was open between June 9 and August 10, 2020. Participation was voluntary and anonymous. At least two follow-up emails were sent to encourage participation.
The Centers for Disease Control and Prevention (CDC) and Public Health—Seattle & King County (PHSKC) developed the standardized online survey (S1 File) that included questions on demographics, work environment, possible SARS-CoV-2 exposures, and workplace COVID-19 mitigation strategies such as availability and use of personal protective equipment (PPE), hand hygiene facilities, and masks. The survey included questions about SARS-CoV-2 testing history, including number of testing events, test results, testing facility, symptoms around the time of testing, job practice while symptomatic, and medical services seeking behavior. We asked whether the test was a blood test, assuming that participants may not understand the terminology for antibody testing (i.e., blood test) versus molecular testing (i.e., nasopharyngeal swab). Study data were collected and managed using REDCap electronic data capture tools hosted at CDC.
We conducted the analyses using Stata/SE 16.0 and R version 4.0.2. Body Mass Index (BMI) was calculated by multiplying weight in pounds by a conversion factor of 703 and dividing by height in inches squared. Job titles and job descriptions were used to categorize job positions as primarily administrative (e.g., supervisors, office administration, information technology, accounting) or client engagement (e.g., case manager, food server, floor monitor, housing advocate, social worker). Individual shelters that were managed by the same organization were grouped into shelter networks for analyses. Because some workers were tested more than once, a worker was classified as positive for SARS-CoV-2 if he/she self-reported a positive non-blood test result for at least one testing event. Workers who reported “Don’t know” for a test result were categorized as a non-positive. We calculated frequencies, medians, and ranges to describe demographic and work characteristics of participants. To explore associations between SARS-CoV-2 positivity and participant characteristics, we used generalized linear models with a binary outcome (COVID-19 positivity according to at least one non-blood test) and a log link, clustered by shelter network (to allow for intragroup correlation), to produce unadjusted prevalence ratios (PR) and 95% confidence intervals (CI). Frequency of close contact, defined as less than 6 feet for more than 15 minutes at a time, was dichotomized into low (never, rarely, a few times a month) and high (a few times a week to a few times a day). We produced PRs comparing workers who reported testing positive at least once to workers who reported testing negative for every test by demographics (e.g., age, sex, ethnicity), work environment (e.g., hours worked, frequency of close contact, COVID-19 mitigation strategies implemented by facility), and attitudes (e.g., thought about quitting).
This activity underwent human subjects ethics review by CDC and was conducted consistent with applicable federal law and CDC policy (45 C.F.R. part 46, 21 C.F.R. part 56; 42 U.S.C. §241(d); 5 U.S.C. §552a; 44 U.S.C. §3501 et seq). Completing the survey was voluntary.
Among 17 shelter networks that represented 27 individual shelters, 106 homeless shelter workers (range per shelter network, 1–33) completed the online survey. The median age of participants was 42 years (range 21–67); 65 (61%) were female and 55 (52%) were non-Hispanic White (Table 1). Of the 106 participants, 23 (22%) were current smokers/vapers, 15 (14%) reported having a chronic lung disease, and median BMI was 28.2 (range 18.1–58.1). The median number of people living in workers’ households was 3 persons (range 1–9 persons). Participants reported working at the shelter for a median of 40 hours per week (range 1–60 hours) and for a median of 20 months (range 0–336 months); 41 workers (39%) worked at their shelter for ≤12 months with 20 (19%) for ≤6 months.
Sixty-three participants (59%) reported some sort of client engagement as part of their regular work duties, including case management, providing medical and mental health care, client intake, client outreach, client screening, serving food to clients, providing education and employment advice, monitoring clients while at the shelter, and janitorial activities (Table 1). Almost one third of participants reported close contact with clients a few times per day (34 participants; 32%); 65 participants (61%) reported that they never have direct physical contact (i.e., touching) with clients. Of the 43 participants categorized as administrative job duties, 22 (51.2%) reported close contact or direct physical contact with clients. Over half of participants reported close contact with coworkers a few times per day (61 participants; 58%). Many homeless shelter workers reported that cleaning was part of their normal duties (60 participants; 57%); of those, 34 (57%) reported receiving training on how to clean an area after a client with known COVID-19 leaves the shelter. Most workers (75%) reported being worried about becoming infected due to their jobs, and of those, 24% had thought about quitting. Although their families were supportive of their jobs, their families also worried about workers being infected due to their jobs (Table 1).
All 17 shelter networks had implemented at least one COVID-19 prevention measure. At the participant level, the most common mitigation measures were use of masks by staff or clients (97/106, 92%) and increased handwashing (88/106, 83%) (Table 1). Eighty-six percent of staff (n = 91) reported wearing a mask at least most of the time while at work. Among those who reported close contact with clients (n = 58), 47 (81%) reported wearing a mask most or all of the time. Among those who reported close contact with coworkers (n = 99), 62 (77%) reported wearing masks when in close contact with coworkers. Among those who reported having direct physical contact with clients (n = 39), 22 reported wearing gloves (56%). Among those who reported having close contact with a person with known COVID-19 (n = 38); 29 (76%) reported wearing a disposable mask most or all of the time, and 24 (63%) reported wearing gloves (Table 2). All believed that their close contact to a person with known COVID-19 occurred at work while 2 workers (5%) also said they had contact at home.
Of the 106 participants, 77 reported being tested for SARS-CoV-2 with 62% of participants (48/77) reported undergoing testing for SARS-CoV-2 more than once. For the 77 participants who reported being tested at least once, the median number of testing events per participant was 2 (range: 1–10). Of the 187 testing events reported by the 77 participants, 9 were blood tests (assumed to be serological testing), 170 were non-blood tests (assumed to be molecular testing), and 8 did not select a test type (Table 3). The average number of days for a participant to receive their test result was 3.1 days (range 0–10 days). Participants reported working while waiting for test results for 118 testing events (63%). Of the 187 tests, 114 testing events (61%) occurred at the workplace. Among 21 positive non-blood tests (4 workers tested positive multiple times), 16 (76%) were from participants who were symptomatic around the time of the test. There were 15 tests among 10 people who indicated “Don’t know” and/or did not answer for a test result, who were categorized as a non-positive. One worker who tested positive worked 1 day while symptomatic and while waiting for their result (Table 3). The worker initially had a sore throat and headache with additional symptoms after receiving the positive test result. Overall, 16 workers reported a positive test result with 46.7% reporting primarily administrative job duties and 53.3% reporting job duties with client engagement; 2 reported a positive blood test and 15 reported a positive non-blood test (one participant tested positive by both blood and non-blood test), for an overall prevalence rate of 15% (16/106). Of the 14 participants who answered the question about where they thought they were infected, 12 participants who tested positive believed they were infected at work.
When analyzing the 77 participants who reported at least one testing event, we identified demographic characteristics that were associated with testing positive for SARS-CoV-2, including having a BMI ≥30 (PR 1.86, 95% CI 1.31, 2.63; p = 0.001) or identifying as Non-Hispanic Black race (PR 2.00, 95% CI 1.23, 3.26; p = 0.01)(Table 4). Reporting frequent close contact with clients (PR 3.97, 95% CI 1.06,14.93; p = 0.04) or using gloves when interacting with a person with known COVID-19 (PR 3.90, 95% CI 1.36, 11.19; p = 0.01) was associated with increased risk of COVID-19 positivity, while facility mitigation measures of wearing masks or maintaining social distance was associated with decreased risk (PR 0.50, 95% CI 0.30, 0.84; p = 0.01; PR 0.52, 95% CI 0.32, 0.84; p = 0.01; respectively). Workers who thought about quitting their job because of concerns about COVID-19 had an increased risk of positivity (PR 1.84, 95% CI 1.02, 3.33; p = 0.04). We conducted a sensitivity analysis, excluding the 10 participants who said that they did not know their test results or who didn’t answer the question (which was categorized as negative in main analysis); BMI, identifying as Non-Hispanic Black race, close contact with clients, social distancing, and wearing a mask remained statistically significant.
This study sought to characterize homeless shelter worker job practices, occupational exposures to SARS-CoV-2, and COVID-19 mitigation measures in the workplace. In this sample of homeless shelter workers, participants reported close contact and direct physical contact with clients. Nearly 40% of workers reported having close contact with a person with known COVID-19 and all reported that they believed the contact occurred at work; 24% of those workers did not use masks all of the time during these interactions (Table 2). Workers who reported frequent contact with clients were more likely to test positive for SARS-CoV-2. Understanding how homeless shelter workers are exposed to COVID-19 at work is important to be able to implement mitigation strategies in this non-traditional workplace.
Recommendations for homeless service providers to help protect the staff and clients include hand hygiene and cleaning supplies, PPE (including masks), administrative controls (e.g., flexible work schedules), facility layout/ventilation considerations, and maintaining social distance . All homeless shelter networks in our study had implemented at least one mitigation measure to reduce risk of worker exposure, with masks for staff/clients and handwashing being most common (Table 1). Most workers reported using masks most or all of the time when at work. We found that workers who reported that their facility implemented the mitigation measure of maintaining social distance and using masks were less likely to be infected (Table 4). Although staff should avoid handling client belongings , we found that about half of the workers reported touching client belongings or shared items. Workers often did not receive training in infection control or cleaning procedures. For example, 43% of workers had not received training on cleaning surfaces for SARS-CoV-2 even though it was a part of their job duties. Homeless shelter workers’ training needs, both for their job tasks and for professional development, are often not prioritized . Shelter management should fully implement recommendations and provide additional training on both COVID-19 and cleaning/disinfection .
In this study, workers reported being worried about becoming infected at work and had thought about quitting. Nearly 40% of workers had been at their shelter less than one year. Homeless shelter workers tend to be a transient work population with a high burnout rate [21–25]. Also, homeless shelter workers may be in a vulnerable population themselves since workers are sometimes residents who are offered jobs at the shelter . Anecdotally, some homeless shelter workers in hourly wage unskilled jobs are, or have been, homeless before . Homeless shelters are already prone to understaffing due to high turnover  which is magnified by COVID-19 isolation requirements if multiple workers test positive. COVID-19 has likely exacerbated the stress associated with working in homeless shelters.
Homeless shelter workers have many different job roles, including case managers, janitors, administrative/managerial staff, cooks, security guards and floor monitors. In our sample, close contact with clients was not limited to workers whose job duties included known client engagement; more than 50% of workers with administrative job duties also reported close or direct physical contact with clients. Thirty-six percent of homeless shelter workers said that they had close contact with a known COVID-19 case at work. Of the 15 who tested positive by non-blood testing, 80% believed that they were infected at work. Homeless shelter workers, regardless of primary job duties, may be at increased risk for COVID-19, due to frequent close contact in congregate settings . Homeless shelter workers have been deemed essential critical infrastructure workers  thus would be recommended for COVID-19 vaccination in Phase 1B (i.e., frontline essential workers) .
Several demographic factors have been shown to increase risk of SARS CoV-2 infection in the general population, including older age, race, ethnicity, and obesity . In this survey, almost 50% of participants were people of color whereas Whites make up the majority of the labor force (77% in 2019) . Additionally, 36.8% of respondents were obese, which was slightly higher than obesity among U.S. employed adults (32.5% in 2016) . We found that respondents who identified as Non-Hispanic Black or were obese were more likely to test positive. People of color are more likely to be employed in occupations with close proximity to others . People with higher risk factors for SARS CoV-2 infection may be over-represented in the homeless shelter workforce when compared to the general U.S. labor force.
During the survey period, many homeless shelters in major cities were conducting serial testing for SARS-CoV-2 of clients and staff . In this study, 61% of participants were tested at the workplace and 69% did not experience symptoms around the time of testing, which supports the premise that many of these workers were tested as part of a universal screening process rather than because they were symptomatic. Universal screening could also explain why some workers were tested multiple times (range 1–10 tests). Over 60% continued to work while waiting for test results. It is not clear whether workers who tested positive would have been detected based solely upon symptoms. Of the 21 positive non-blood testing events, 76% were symptomatic around the time of testing, 6.3% worked while symptomatic and 63% sought medical care for their symptoms (Table 3). Serial testing at workplaces has been recommended as a control method to interrupt transmission of COVID-19 in possible outbreak situations .
This study is subject to several limitations. The questionnaire was online and self-administered where there may have been selection bias (e.g., access to internet, access to survey during working hours) and recall bias when reporting exposures, symptoms, and timing of testing. There were small number of respondents who reported positive tests which limited our ability to conduct more robust analyses to evaluate potential occupational risk factors associated with testing positive (Table 4). The respondents were a self-selected convenience sample where response rate was not able to be estimated. It is possible that workers who were more concerned with COVID-19 participated in the survey. Access to internet and ability to access survey during working hours may have been a factor as to who was able to participate. In addition, homeless shelter workers from multi-facility networks in large cities were invited to participate; thus, this sample may not be representative of all homeless shelter workers in the United States.
This is the first study that describes the work environment of homeless shelter workers in the context of COVID-19. We found that surveyed homeless shelter workers reported frequent close contact with clients; this was associated with having a positive test for SARS-CoV-2, while wearing masks and maintaining social distance at work were protective. Shelter management should continue to follow public health recommendations  by reinforcing mitigation measures in the workplace and training staff routinely on mitigation measures. Homeless shelter workers are essential workers with the potential for high-risk exposures, including close and direct physical contact with clients. Further research is needed on describing work environment, COVID-19 risks, and mitigation measures to reduce risk of infections among homeless shelter staff.
Meagan Kay, Matthew Hanson, Margaret D. Luckoff, Jody Rauch, Libby Page, Public Health-Seattle & King County; Angela McCauley, Baltimore City Mayor’s Office of Homeless Services; Julie L. Self, Centers for Disease Control and Prevention; Jessie Gaeta, Boston Health Care for the Homeless Program; Gerry Thomas, Boston Public Health Commission; Kate Tettamant, Georgia Department of Community Affairs; Tair Kiphibane, Salt Lake County Health Department.
Disclaimer: The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention/the Agency for Toxic Substances and Disease Registry.
- 1. Mosites E, Parker EM, Clarke KEN, Gaeta JM, Baggett TP, Imbert E, et al. Assessment of SARS-CoV-2 infection prevalence in homeless shelters—Four U.S. Cities, March 27-April 15, 2020. Morbidity and Mortality Weekly Report. 2020;69(17):521–2. pmid:32352957
- 2. Baggett TP, Keyes H, Sporn N, Gaeta JM. Prevalence of SARS-CoV-2 infection in residents of a large homeless shelter in Boston. Journal of the American Medical Association. 2020;323(21):2191–2. pmid:32338732
- 3. Yoon JC, Montgomery MP, Buff AM, Boyd AT, Jamison C, Hernandez A, et al. COVID-19 Prevalence among people experiencing homelessness and homelessness service staff during early community transmission in Atlanta, Georgia, April-May 2020. Clinical Infectious Diseases. 2020. pmid:32898272
- 4. Tobolowsky FA, Gonzales E, Self JL, Rao CY, Keating R, Marx GE, et al. COVID-19 outbreak among three affiliated homeless service sites—King County, Washington, 2020. Morbidity and Mortality Weekly Report. 2020;69(17):523–6. pmid:32352954
- 5. Cybersecurity and Infrastructure Security Agency. Advisory memorandum on ensuring essential critical infrastructure workers’ ability to work during the COVID-19 response Washington, DC: U.S. Department of Homeland Security,; December 16, 2020 [updated December 16, 2020. https://www.cisa.gov/publication/guidance-essential-critical-infrastructure-workforce.
- 6. Mohsenpour A, Bozorgmehr K, Rohleder S, Stratil J, Costa D. SARS-Cov-2 prevalence, transmission, health-related outcomes and control strategies in homeless shelters: Systematic review and meta-analysis. EClinicalMedicine. 2021:101032. pmid:34316550
- 7. Hofmeister MG, Xing J, Foster MA, Augustine RJ, Burkholder C, Collins J, et al. Hepatitis A person-to-person outbreaks: Epidemiology, morbidity burden, and factors associated with hospitalization—Multiple States, 2016–2019. Journal of Infectious Diseases. 2020.
- 8. Di Renzi S, Tomao P, Martini A, Capanna S, Rubino L, D’Amico W, et al. Screening for tuberculosis among homeless shelter staff. Am J Infect Control. 2012;40(5):459–61. pmid:21958841
- 9. Kerr EM, Vonnahme LA, Goswami ND. Impact of targeted local interventions on tuberculosis awareness and screening among persons experiencing homelessness during a large tuberculosis outbreak in Atlanta, Georgia, 2015–2016. Public Health Reports. 2020;135(1_suppl):90S–9S. pmid:32735200
- 10. Valenciano SJ, Onukwube J, Spiller MW, Thomas A, Como-Sabetti K, Schaffner W, et al. Invasive Group A Streptococcal infections among people who inject drugs and people experiencing homelessness in the United States, 2010–2017. Clinical Infectious Diseases. 2020. pmid:32803254
- 11. Baker MG, Peckham TK, Seixas NS. 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. pmid:32343747
- 12. Public Health-Seattle & King County. Tuberculosis Prevention and Control Guidelines for Homeless Service Agencies in Seattle-King County, Washington. In: Prevention Division—Tuberculosis Control Program, editor. 4th ed. Seattle, WA 2010.
- 13. Public Health-Seattle & King County. Recommended shelter health and safety best practice guidelines. Seattle, WA2005.
- 14. Ward JA, Parish K, DiLaura G, Dolovich S, Saloner B. COVID-19 Cases Among Employees of U.S. Federal and State Prisons. Am J Prev Med. 2021. pmid:33750599
- 15. Kakimoto K, Kamiya H, Yamagishi T, Matsui T, Suzuki M, Wakita T. Initial Investigation of Transmission of COVID-19 Among Crew Members During Quarantine of a Cruise Ship—Yokohama, Japan, February 2020. MMWR Morb Mortal Wkly Rep. 2020;69(11):312–3. pmid:32191689
- 16. Kasper MR, Geibe JR, Sears CL, Riegodedios AJ, Luse T, Von Thun AM, et al. An Outbreak of Covid-19 on an Aircraft Carrier. N Engl J Med. 2020;383(25):2417–26. pmid:33176077
- 17. Greene J, Gibson DM. Workers at long-term care facilities and their risk for severe COVID-19 illness. Prev Med. 2021;143:106328. pmid:33220398
- 18. Telford CT, Onwubiko U, Holland DP, Turner K, Prieto J, Smith S, et al. Preventing COVID-19 Outbreaks in Long-Term Care Facilities Through Preemptive Testing of Residents and Staff Members—Fulton County, Georgia, March-May 2020. MMWR Morb Mortal Wkly Rep. 2020;69(37):1296–9. pmid:32941413
- 19. Letizia AG, Ramos I, Obla A, Goforth C, Weir DL, Ge Y, et al. SARS-CoV-2 Transmission among Marine Recruits during Quarantine. N Engl J Med. 2020;383(25):2407–16. pmid:33176093
- 20. Centers for Disease Control and Prevention. Interim Guidance for Homeless Service Providers to Plan and Respond to Coronavirus Disease 2019 2020 [updated June 2021]. https://www.cdc.gov/coronavirus/2019-ncov/community/homeless-shelters/plan-prepare-respond.html.
- 21. Olivet J, McGraw S, Grandin M, Bassuk E. Staffing challenges and strategies for organizations serving individuals who have experienced chronic homelessness. Journal of Behavioral Health and Service Research. 2010;37(2):226–38. pmid:20052621
- 22. Waegemakers Schiff J, Lane AM. PTSD Symptoms, Vicarious Traumatization, and Burnout in Front Line Workers in the Homeless Sector. Community Mental Health Journal. 2019;55(3):454–62. pmid:30684127
- 23. Wirth T, Mette J, Nienhaus A, Schillmoller Z, Harth V, Mache S. "This Isn’t Just about Things, It’s about People and Their Future": A qualitative analysis of the working conditions and strains of social workers in refugee and homeless aid. International Journal of Environmental Research and Public Health. 2019;16(20).
- 24. Mullen J, Leginski W. Building the capacity of the homeless service workforce. Open Health Services and Policy Journal. 2010;3:101–10.
- 25. Hagen JL, Hutchison E. Who’s serving the homeless. Social Casework. 1988;69:491–7.
- 26. Boydell KM, Goering P, Morrell-Bellai TL. Narratives of identity: re-presentation of self in people who are homeless. Qual Health Res. 2000;10(1):26–38. pmid:10724750
- 27. Slotnick DE. She Works in a Homeless Shelter, and She Lives in One, Too. New York Times. 2021 April 16, 2021.
- 28. Centers for Disease Control and Prevention. Clinical Questions about COVID-19: Questions and Answers [https://www.cdc.gov/coronavirus/2019-ncov/hcp/faq.html.
- 29. Dooling K, McClung N, Chamberland M, Marin M, Wallace M, Bell BP, et al. The Advisory Committee on Immunization Practices’ Interim Recommendation for Allocating Initial Supplies of COVID-19 Vaccine—United States, 2020. Morbidity and Mortality Weekly Report. 2020;69(49):1857–9. pmid:33301429
- 30. de Lusignan S, Dorward J, Correa A, Jones N, Akinyemi O, Amirthalingam G, et al. Risk factors for SARS-CoV-2 among patients in the Oxford Royal College of General Practitioners Research and Surveillance Centre primary care network: a cross-sectional study. Lancet Infect Dis. 2020;20(9):1034–42. pmid:32422204
- 31. Labor force characteristics by race and ethnicity, 2019. Bureau of Labor Statistics. 2019 [cited August 2, 2021]. https://www.bls.gov/opub/reports/race-and-ethnicity/2019/home.htm.
- 32. Cawley J, Biener A, Meyerhoefer C, Ding Y, Zvenyach T, Smolarz BG, et al. Job Absenteeism Costs of Obesity in the United States: National and State-Level Estimates. J Occup Environ Med. 2021;63(7):565–73. pmid:33769330
- 33. Hawkins D. Differential occupational risk for COVID-19 and other infection exposure according to race and ethnicity. American Journal of Industrial Medicine. 2020;63(9):817–20. pmid:32539166
- 34. Centers for Disease Control and Prevention. Interim Considerations for Health Departments for SARS-CoV-2 Testing in Homeless Shelters and Encampments 2020 [updated March 2021. https://www.cdc.gov/coronavirus/2019-ncov/community/homeless-shelters/testing.html.