Introduction
Like other essential workers, hospital employees, including but not limited to patient-facing healthcare workers, were especially susceptible to infection early in the coronavirus 2019 (COVID-19) pandemic. Reference Mutambudzi1 Ensuring employees’ safety during pandemics is crucial for sustaining healthcare systems. However, protecting employees from infection remains a challenge. Over the pandemic’s first 18 months, the World Health Organization estimated 80,000 to 180,000 healthcare workers worldwide died from COVID-19. 2
Several quantitative studies have evaluated sources of exposure for hospital employees with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, Reference Boomsma3–Reference Fell8 including a prior study by our team of 187 employees infected with SARS-CoV-2 at a U.S. tertiary care hospital during the pandemic’s first year. Reference Boomsma3 We assessed employees’ exposures to people with possible or known SARS-CoV-2 infection from patient, coworker, and community interactions. Although most employees could not identify a high-risk exposure, among those who could, high-risk exposures were most frequently reported in the community (consistent with other studies Reference Shaw, Suits, Steigerwald, Thomas and Formica4,Reference Zabarsky5,Reference Jacob9,Reference Alrabadi10 ) and secondarily from coworkers. Other studies suggest coworker exposures were more common than patient exposures and more likely to lead to transmission. Reference Farah11,Reference Shah12 Findings of these studies suggest that—while the focus early in the pandemic was on protecting hospital employees from infection during patient care—employees were more likely to get infected from community and coworker interactions.
Understanding employees’ perspectives on exposure risks during the COVID-19 pandemic may inform hospital strategies to reduce respiratory pathogen transmission during future pandemics. However, few studies have used qualitative research to explore hospital employees’ risks during the COVID-19 pandemic. Reference Utzet13–Reference Cherry15 Most qualitative studies focused on general experiences of healthcare workers and challenges faced very early in the pandemic, such as the uncertainty surrounding a novel virus. Reference Ashley14,Reference Cherry15 Few studies explored risks from coworker or community interactions, and, to our knowledge, studies did not interview employees who were not patient-facing or who had confirmed COVID-19.
In this study, we conducted qualitative interviews with patient-facing and non-patient-facing hospital employees to identify contextual factors contributing to SARS-CoV-2 infection risk from patient, coworker, and community interactions. We purposefully sampled from individuals who responded to our quantitative survey. Reference Boomsma3 We elicited reflections and recommendations focusing on the two weeks before employees were diagnosed with SARS-CoV-2. Our study reflects perspectives from the pandemic’s first year, before widespread vaccine availability. Our goal is to inform interventions to reduce risks for employees during future pandemic events.
Methods
Study setting
Tufts Medical Center (TMC) is a tertiary hospital with about 7000 employees in Boston, Massachusetts. Reference Boomsma3 Universal (non-N95) masking for employees, visitors, and patients was initiated on March 27, 2020. In the pandemic’s initial months, N95 respirator use was only recommended during contact with COVID-19 patients in intensive care units or for aerosol-generating procedures, following Centers for Disease Control and Prevention (CDC) guidelines during severe shortage. N95 respirator use by providers for all interactions with patients with suspected or confirmed COVID-19 was implemented on July 11, 2020. SARS-CoV-2 PCR testing for all patients at admission was initiated on May 19, 2020.
Research team characteristics and positionality
This study team comprised medical students (DK, DP, MCM, CB, JJ), a public health student (ZL), an undergraduate student (PS), public health faculty (RS, AMT), and attending physicians (SD, RS, AGW, EB), including one leading infection control (SD). Most team members were women (DP, MCM, CB, ZL, PS, AMT, SD, AGW, EB); the rest were men (DK, JJ, RS).
Initial survey and participant sampling for qualitative interviews
Study participants were TMC employees with a positive SARS-CoV-2 PCR test between March 1, 2020, and January 15, 2021. The survey’s methods were published previously. Reference Boomsma3 Of 573 TMC employees who tested positive for SARS-CoV-2, 187/573 (32%) participated in this survey, of whom 166/187 (89%) were willing to be contacted again and could be recruited for qualitative interviews.
Starting in July 2022, we recruited qualitative interview participants using purposeful sampling. To achieve a diverse sample, we categorized eligible participants by gender, race, ethnicity, and job type. Using criteria to achieve representation across characteristics, we contacted potential participants in groups of five by email and then a single phone call and voicemail if a person did not respond. Due to limited responses, recruitment was expanded to include all employees who had indicated willingness to be contact again in the survey. We stopped recruitment after completing 23 interviews, as initial analysis suggested thematic saturation.
Data collection
One-time interviews with each participant were conducted by authors DK and ZL between July and November 2022 using Zoom video-conferencing software. Each interview lasted 20-45 minutes and was recorded. As interviewers had not worked in the hospital, they were not previously known to participants. Zoom performed interview transcription. Interviewers corrected transcripts against the audio files and uploaded de-identified transcripts into Dedoose software. Transcripts were not returned to participants for verification.
Interview guides followed a framework eliciting information on perceived infection risks across the three settings of interest, and recommendations for hospital interventions to reduce risk. The period participants were asked to reflect upon was the two weeks before their SARS-CoV-2 diagnosis. As such, interview findings reflect experiences from the pandemic’s initial year, before vaccines or outpatient antiviral therapy were widely available and when hospitals were adapting to new information.
The interview guide started with open-ended questions asking participants how they believed they were infected (Supplementary Text). Subsequent sections asked open-ended and semi-structured questions about risks from patient, coworker, and community interactions, aligned with the framework of our prior survey and other studies. Reference Boomsma3–Reference Zabarsky5 The interview closed with questions about the hospital’s response and elicited recommendations.
Data analysis
We used a deductive approach to thematic analysis. A coding framework was developed based on quantitative survey findings and the interview guide structure (Supplementary Table). The framework was refined using post-interview impressions by interviewers DK and ZL. Code categories aligned with settings of transmission—ie, via community, patient, or coworker interactions—and a fourth category captured recommendations. Within each category, codes addressed setting-specific challenges—eg, patient masking codes for the patient setting, contact tracing codes for the coworker setting.
Each transcript was coded independently by two researchers (among DK, ZL, PS, DP, MM, and CB) using Dedoose software. 16 Inter-rater reliability was not assessed, as coding was reconciled for all transcripts through discussion. During initial coding, new codes were occasionally incorporated into the coding scheme. Transcripts were coded a second time to apply new codes. After completing coding, we identified recurring themes that were common or salient (ie, reported by a few participants, but that were thought to be important). Themes provided insights on factors contributing to SARS-CoV-2 exposure risk or informed recommendations for future hospital interventions. Factors represented a combination of exposures participants perceived as having contributed their own infection event and general perceived infection risks within each setting. Themes were refined through team discussions, and representative quotations were identified to illustrate each theme. Findings were not disseminated to participants.
Ethical approval
This study was approved by the Tufts Health Science Institutional Review Board. Participants signed an electronic consent, and verbal consent for audio recording was obtained.
Results
Participant characteristics
The median age of the 23 participants interviewed was 36 years (interquartile range 23 to 54 years). Most were women, of white race and non-Hispanic ethnicity, and had patient-facing jobs (Table 1).
Table 1. Demographic characteristics of the interview participants
Patient interactions—perceived infection risks
One recurring theme indicated low perceived infection risk with patient interactions during the pandemic’s first year, reflecting a sense of strong hospital support to ensure safety during patient care. For example, the following quotation reflects a theme of adequate personal protective equipment (PPE) availability during patient interactions that was expressed broadly by participants:
“There was great PPE [availability] at the hospital! I felt very safe. I felt covered.” (Registered nurse 1)
Participants also described a few challenges contributing to risk during patient interactions that were more relevant in the pandemic’s earliest weeks. Participants often acknowledged that these limitations resulted from evolving knowledge about COVID-19 or societal challenges (eg, supply chain deficits) experienced across U.S. hospitals.
For example, during the pandemic’s initial weeks, patient-facing employees did not use masks for patients without suspected or confirmed COVID-19, due to lack of a universal mandate for providers and patients, which contributed to perceived infection risk (Table 2, Q1). Participants reported some difficulties with mask availability (Table 2, Q2).
Table 2. Representative quotations on contextual factors contributing to perceived risk of SARS-CoV-2 infection during patient, coworker, and community interactions
Note. COVID, coronavirus disease; PPE, personal protective equipment; SARS-CoV-2, severe acute respiratory coronavirus 2.
a These challenges were reported only in the earliest weeks of the pandemic.
Participants described risks from lack of asymptomatic testing of patients at hospital admission and lack of confirmatory testing for patients with suspected COVID-19, due to test shortages in the pandemic’s initial weeks. This contributed to perceived risk by delaying placement of patients under enhanced infection precautions (Table 2, Q3 and Q4).
After universal masking of employees and patients became mandatory, suboptimal masking by patients contributed to perceived risk during patient interactions. This challenge was accentuated by patients’ medical conditions, which sometimes made masking prohibitive (Table 2, Q5 and Q6).
Coworker interactions—perceived infection risks
Multiple factors contributed to perceived infection risk during coworker interactions, including suboptimal masking with coworkers, inadequate breakroom space, and perceptions of contact tracing challenges. After the universal masking mandate, participants reported they were more likely to take their masks off around coworkers, due to “exhaustion” with mask-wearing or because they felt “safer” around coworkers (Table 2, Q7 and Q8).
Difficulties with coworker masking were compounded by limited rooms for physical distancing, especially while eating or drinking. Shared offices and breakrooms had limited space and suboptimal ventilation, which were recognized as structural challenges (Table 2, Q9 and Q10).
Participants reported perceived challenges with contact tracing, including sometimes not being contacted by the hospital to inform them that a contact (whether coworker or patient) had tested positive for SARS-CoV-2. Participants sometimes learned of a contact’s diagnosis by word of mouth, though they recognized the difficulties in implementing contact tracing, especially early in the pandemic (Table 2, Q11 and Q12).
In contrast, while universal asymptomatic testing of employees was not performed, consistent with policies in nearly all U.S. hospitals, participants felt the hospital made symptomatic testing free and easily accessible, which reduced perceived risk:
“They made the [SARS-COV-2] test available for employees who were considered exposed or had symptoms. It didn’t take me very long to get a test at all. It was very good, very good.”
(Respiratory therapist 1)
Community interactions—perceived infection risks
Perceived community risks related to other household members, small private gatherings, public transportation, and shopping. Household risks often related to household members (eg, family, roommates) who were also essential workers (eg, grocery store workers, hospital employees) and who therefore could not engage in physical distancing (Table 2, Q13 and Q14).
Participants reported exposures during small private gatherings, because of lower perceived risks with family or friends and social pressure to not engage in masking or physical distancing. Holidays were perceived as periods of increased transmission risk (Table 2, Q15 and Q16). A theme emerged of participants identifying public spaces, such as stores or public transportation, as potential sources of transmission risk, in the absence of another clear source of exposure (Table 2, Q17).
Recommendations for reducing infection risk among hospital employees
We organize recommendations into: (1) enhanced education and guidance by hospitals, and (2) structural changes requiring societal investment (Table 3). Participants emphasized the importance of education on the scientific rationale for masking (Table 3, Q18 and 19). Sustaining masking adherence as pandemic measures extended for several months was challenging. Participants encouraged cycles of education to reinforce mask use over time (Table 3, Q20 and Q21). Participants recognized the importance of transmission in non-patient-care settings (Table 3, Q22), and recommended enhanced education and support to promote behavior change around physical distancing (eg, caution around private gatherings) and masking (eg, education on community masking), as well as provision of masks to employees for use in community settings.
Table 3. Representative quotations on recommendations for reducing SARS-CoV-2 infection risk among hospital employees
Note. COVID, coronavirus disease; SARS-CoV-2, severe acute respiratory coronavirus 2.
Regarding structural challenges, recommendations focused on increasing space for physical distancing and improving ventilation in breakrooms, offices, and patient rooms. Participants recommended providing more spaces for employees to eat, especially when outdoor spaces are inaccessible due to weather (Table 3, Q23 and Q24). Participants noted that increasing the number of patient rooms might decrease transmission by reducing the need for patients to share rooms (Table 3, Q25).
Finally, participants felt that support for employees who got sick with COVID-19 was insufficient, particularly paid sick leave. Participants described using up paid time off days when they had to isolate due to COVID-19. Others noted that lack of paid sick leave and hospital staffing difficulties created “stigma” around calling in sick, which could increase transmission by creating social pressure to come to work when unwell (Table 3, Q26 and Q27).
Discussion
In this study, we identified factors contributing to SARS-CoV-2 infection risk among employees in a U.S. tertiary hospital, across patient, coworker, and community interactions. Consistent with our prior quantitative study Reference Boomsma3 and other studies Reference Shaw, Suits, Steigerwald, Thomas and Formica4–Reference Eyre7,Reference Jacob9,Reference Alrabadi10 —which found most high-risk SARS-CoV-2 exposures happened during community and coworker (rather than patient) interactions—participants reported low perceived risk with patient interactions. This perception was shaped by high levels of hospital education, support, and PPE provision for patient care. With the exception of suboptimal masking adherence by patients, other patient care risks were largely described during the pandemic’s early weeks, when the U.S. experienced nationwide testing and PPE shortages and when COVID-19 knowledge was rapidly evolving. These findings highlight the importance of national strategies to ensure testing and PPE availability to protect employees during the initial phase of future pandemics. Reference Feinmann17,Reference Montazeri and Sandbrink18
In contrast to patient care settings, participants described ongoing risks during coworker and community interactions throughout the pandemic’s first year. A cross-cutting theme was the perception that interactions with coworkers or during private gatherings were more “safe” than patient care interactions, despite the fact that hospital employees may be more likely to get infected in non-patient care settings. Reference Zeeb19–Reference Schepers22 Participants’ recommendations on education about masking directly aligned with these findings. Participants suggested that hospitals should educate employees about risks during coworker and community interactions, emphasize the importance of masking during non-patient interactions, provide high-quality masks for employees to use in the community, and sustain masking education over time. Notably, CDC guidelines on SARS-CoV-2 infection control in healthcare facilities have had limited guidance on provision of education or other strategies (eg, mask provision) hospitals should implement to protect employees in community settings, beyond those articulated in community guidelines for the general public. 23–25 Guidelines during future pandemics should have an expanded focus on mitigating risks to healthcare providers from coworker and community interactions.
Perceptions of infection risk during coworker interactions were shaped by structural challenges. In many U.S. cities, hospital infrastructure was built decades ago, resulting in employees having poorly ventilated breakrooms and offices with limited space for physical distancing, which may be key sites of transmission during pandemic and endemic phases. Employees would benefit from additional spaces for taking breaks to eat and from improved ventilation systems. However, hospitals are limited in their ability to rapidly adapt infrastructure to mitigate respiratory pathogen transmission in the midst of a pandemic. Reference Stichler26 Long-term societal investments are needed to retrofit existing hospitals and build future hospitals that are designed to reduced respiratory pathogen transmission. Reference Murphy, Mansfield and Barber27 Such hospital designs should aim to increase the number of single-occupancy patient rooms and provide more well-ventilated breakroom and office space.
Finally, the need for a better culture around taking time off from work was a recurring theme. Paid sick leave was a structural intervention recommended by participants that could improve employee health, the employee experience, and infection control, by making employees more comfortable to stay home when sick. Reference Pichler, Wen and Ziebarth28–Reference Kumar, Grefenstette, Galloway, Albert and Burke30 Participants experienced financial difficulties from taking off the required number of days for COVID-19 isolation. Workplace culture placed additional stigma on taking days off when patient volume was high. As many U.S. healthcare personnel do not have paid sick leave, Reference de Perio, Srivastav, Razzaghi, Laney and Black31 stronger regulations and societal investment may support hospitals in providing this benefit, which can financially benefit hospitals as well. Reference Asfaw, Rosa and Pana-Cryan32 Future studies should explore the role of presenteeism (ie, coming to work despite being sick) in contributing to transmission risk among hospital employees.
Strengths of this study include a focus on the COVID-19 pandemic’s first year (ie, before availability of vaccines and outpatient antiviral therapies), which enhances relevance of study findings for future pandemics. We also interviewed employees who had been diagnosed with SARS-CoV-2. Our interview guide had participants focus on the two weeks before their diagnosis, such that findings are more likely to reflect exposures that may have contributed to infection. Our findings add to the limited qualitative literature on experiences of the COVID-19 pandemic by hospital employees.
Study limitations include the fact that we collected interviews in 2022, such that findings may be limited by participant recall, potentially influencing details or the accuracy of reported risk perceptions and exposure sources. In addition, while our sample size is reasonable for a qualitative study, we may not have achieved thematic saturation for findings from employee subgroups (eg, based on their job role). Despite attempts to ensure a diverse sample reflecting the broader gender, racial, and ethnic composition of hospital employees, we were limited by those who responded from our recruitment sampling frame, which mostly comprised women (74%) and white non-Hispanic employees (64%). Reference Boomsma3 Notably, however, most TMC employees, as in other U.S. hospitals, are women. Finally, our sample did not include physicians who were not in training and had few non-patient-facing employees, which limits the external validity of our findings.
This qualitative study provides insights into perceived infection risks faced by hospital employees during the COVID-19 pandemic’s initial year, while eliciting recommendations that could inform hospital interventions during future pandemics. Perceived patient care risks were largely limited to the pandemic’s initial weeks, whereas coworker and community risks played a more substantial role throughout the first year. Hospital employees recommended provision of sustained education on mask use during future pandemics, with a focus on mitigating risk with coworkers and in community settings. Employees also emphasized the importance of structural challenges that require broader societal investment to improve future hospital design (eg, to increase space and ventilation), while enhancing regulations and supporting hospitals to expand paid sick leave. By listening to the voices of employees, hospitals may be better prepared to reduce their risk of infection and enhance employee well-being during future pandemics.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/ash.2025.39
Acknowledgments
We are grateful to the Tuft Medical Center Employee Health Department for facilitating access to information to support study implementation.
Financial support
Part of the time of the study team was supported by the Gorbach Family Fund associated with the Tufts University School of Medicine. Data collection was partially funded through the National Institute of Minority Health and Health Disparities (grant nos. 1K23MD015267-01 and AHRQ K08HS026008).
Competing interests
RS serves as an attending physician for a few weeks a year on the infectious disease service at Tufts Medical Center. AGW and EB were employed as physicians at Tufts Medical Center during the time period of this study. SD is the Chief Infection Control Officer at Tufts Medicine. All other authors report no conflicts of interest relevant to this article.
Open access
