Psychosocial and Behavioral Outcomes and Transmission Prevention Behaviors: Working During the Coronavirus Disease 2019 Pandemic.

OBJECTIVE: To investigate the impact of coronavirus disease 2019 (COVID-19) on psychosocial and behavioral responses of the non-health care workforce and to evaluate transmission prevention behavior implementation in the workplace. PARTICIPANTS AND METHODS: We deployed the baseline questionnaire of a prospective online survey from November 20, 2020, through February 8, 2021 to US-based employees. The survey included questions on psychosocial and behavioral responses in addition to transmission prevention behaviors (e.g., mask wearing). Select questions asked employees to report perceptions and behaviors before and during the COVID-19 pandemic. Data were analyzed descriptively and stratified by work from home (WFH) percentage.
RESULTS: In total, 3607 employees from 8 companies completed the survey. Most participants (70.0%) averaged 90% or more of their time WFH during the pandemic. Employees reported increases in stress (54.0%), anxiety (57.4%), fatigue (51.6%), feeling unsafe (50.4%), lack of companionship (60.5%), and feeling isolated from others (69.3%) from before to during the pandemic. Productivity was perceived to decrease for 42.9% of employees and non-work-related screen time and alcohol consumption to increase for 50.7% and 25.1% of employees, respectively, from before to during the pandemic. Adverse changes were worse among those with lower WFH percentages. Most employees reported wearing a mask (98.2%), washing hands regularly (95.7%), and physically distancing (93.6%) in the workplace.
CONCLUSION: These results suggest worsened psychosocial and behavioral outcomes from before to during the COVID-19 pandemic and higher transmission prevention behavior implementation among non-health care employees. These observations provide novel insight into how the COVID-19 pandemic has impacted non-health care employees.

Coronavirus disease 2019 (COVID-19), caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has resulted in high morbidity and mortality rates since being declared a global pandemic in March 2020.1, 2, 3 Because most SARS-CoV-2 infections are likely transmitted by asymptomatic individuals before symptom onset, infection control measures have been and continue to be vital even as effective vaccines are administered. Across many workforce sectors, companies have implemented mitigation strategies, including mandates that employees work from home (WFH) if possible, while also requiring engagement in SARS-CoV-2 transmission prevention behaviors (eg, physical distancing, mask wearing).5, 6, 7 Although these mitigation strategies can help reduce COVID-19 infection rates, unintended negative consequences on employees can arise (eg, stress/anxiety, adverse behavior changes).

The COVID-19 pandemic has led to increased prevalence8, 9, 10, 11, 12, 13, 14 of self-reported stress, anxiety, depression, episodes of acute panic, obsessive behaviors, and posttraumatic stress disorder, among others. Widespread quarantine and fear of self/loved ones getting COVID-19 are among the largest contributors to observed poorer psychosocial health., Mental health problems are also higher among those diagnosed as having COVID-19 compared with those not directly affected. In the workplace, past literature noted that being a health care worker or caregiver with SARS-CoV in 2003 was associated with a higher likelihood of negative psychosocial outcomes, exacerbating the impact of the COVID-19 pandemic on this employee cohort. How past infectious diseases and the COVID-19 pandemic have impacted the psychosocial health of non–health care employees has scantly been studied.

SARS-CoV-2 spreads primarily via small airborne infectious particles (i.e., aerosols) that infected individuals can generate when breathing, coughing, sneezing, or talking.16, 17, 18, 19, 20, 21, 22 A less common transmission route is via larger respiratory droplets that deposit onto nearby surfaces.22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33 A recent meta-analysis and several standalone investigations35, 36, 37, 38, 39, 40 have found that mask wearing, physical distancing, and hand/surface disinfection mitigate the transmission of various coronaviruses. Thus, SARS-CoV-2 transmission prevention behaviors in the workplace are critical to reduce COVID-19 infection rates, with engagement in these behaviors also associated with lower levels of stress, anxiety, and depression.

Governmental recommendations are in place for SARS-CoV-2 transmission prevention in the workplace. Yet, few data exist across non–healthcare employee workforce sectors regarding (1) psychosocial outcomes (e.g., stress, anxiety, safety) and perceptions related to working during the COVID-19 pandemic, (2) behavioral outcomes (e.g., smoking, alcohol consumption, physical activity), and (3) implementation of SARS-CoV-2 transmission prevention behaviors in the workplace. Survey collection of such data would inform workplace strategies to reduce negative psychosocial outcomes among employees as more individuals return regularly to the workplace.

The purpose of this study was to examine COVID-19–related psychosocial outcomes, prevention practices, and health behaviors among employees across numerous workforce sectors.

Methods

This study was approved by the University of Minnesota and Mayo Clinic Institutional Review Boards. All the study participants gave informed consent.

Study Design and Survey Development

We designed a prospective online survey titled the Characterizing Awareness of SARS-CoV-2 PrevenTion and Understanding Responses and Experiences (CAPTURE) Survey, which was deployed at baseline, 3 months, 6 months, and 1 year. Because the follow-up surveys are ongoing, only baseline results are presented herein. The CAPTURE Survey consisted of 48 questions regarding (1) the socioeconomic impact of COVID-19, (2) personal SARS-CoV-2 transmission prevention behaviors in the workplace, (3) perceptions of the importance/efficacy of SARS-CoV-2 transmission prevention behaviors, (4) workplace COVID-19 culture and practices, (5) psychosocial experiences and perceptions before and during the COVID-19 pandemic, and (6) health behaviors before and during the COVID-19 pandemic. See the Measures subsection and Supplemental Appendix 1, available online at http://mcpiqojournal.org, for a description of survey components/questions. We pilot tested the CAPTURE Survey twice with experts at the Well Living Lab, Mayo Clinic, and the University of Minnesota before deployment. The CAPTURE Survey completion time was approximately 12 minutes.

The CAPTURE Survey was deployed using the Mayo Clinic’s Qualtrics Platform. Survey responses were deidentified and given a unique study ID that consisted of a series of numbers that had no meaning to employees’ personal identifying characteristics. We used this ID to track each individual employee in the research database.

Inclusion and Exclusion Criteria

Any public or private US company in business at the time of baseline survey distribution was eligible for study participation. We assessed these company-level criteria before contacting each company regarding their employees’ potential participation in the CAPTURE Survey. Employee-level inclusion criteria were as follows: (1) 18 years or older, (2) English speaking, (3) currently employed by the company, and (4) working 50% or more of their workweek indoors given that SARS-CoV-2 is most likely to spread indoors. Employees were excluded if they were (1) younger than 18 years, (2) no longer working for the company contacted, (3) working primarily outdoors, or (4) not working within the United States. We excluded employees working outside the United States given the differing cultures, practices, and severity of the COVID-19 pandemic between the United States and other countries.

Measures

Consent/Screening, Demographics, Job Descriptions, and General Health

We gathered each employee’s age, employment status, employer (i.e., company), and US location. Average percentage of workweek time spent working indoors was assessed on a scale from 0% to 100% in 1% increments. Demographic information, job descriptions, and general health-related information were adapted from the Stand and Move at Work (SMW) group randomized trial and the Coronary Artery Risk Development in Young Adults (CARDIA) study.,

Socioeconomic Impact of the COVID-19 Pandemic

Questions assessed the average number of hours per week that the employee worked in total and face-to-face in the workplace. We assessed the average percentage of workweek hours spent WFH on a scale from 0% to 100% in 1% increments, with this question asked in parallel to acquire before and during COVID-19 WFH percentages. Questions adapted from the SMW trial and the Environmental Influences on Child Health Outcomes (ECHO) study examined health insurance coverage and doctor visits before and during the COVID-19 pandemic.,

Psychosocial Outcomes and Perceptions

We used 5-point Likert scales to assess employees’ frequency of feeling stressed, anxious, fatigued, and unsafe before and during the COVID-19 pandemic; employees’ perceived productivity before and during COVID-19; whether employees felt threatened or afraid of COVID-19; employees’ fear of catching the disease when around other people; and employee engagement in health behaviors before and during COVID-19, including physical activity, non–work-related screen time, sleep, and alcohol consumption. These questions were adapted from the SMW trial.

SARS-CoV-2 Prevention Behaviors in the Workplace

We included questions that asked about the frequency of (1) SARS-CoV-2 transmission prevention behaviors (mask wearing, glove wearing, handwashing, physical distancing, surface disinfecting) that an employee and their coworkers had taken when in the workplace and (2) company provision of SARS-CoV-2 transmission prevention supplies. These questions were asked on a 5-point Likert scale. We also assessed types of SARS-CoV-2 transmission prevention behavior training provided by employees’ companies and whether the employee perceived SARS-CoV-2 transmission prevention behaviors to be effective at preventing the spread of COVID-19, the latter on a 4-point Likert scale.

COVID-19 Symptoms and Diagnosis

We inquired whether the employee had experienced any COVID-19 symptoms and/or been diagnosed as having COVID-19 using questions adapted from (1) the ECHO study, Epidemic-Pandemic Impacts Inventory; (2) the CARDIA study, Three-Item Loneliness Scale; and (3) the Perceived Coronavirus Threat Questionnaire from the Social Psychology Survey of COVID-19 created by researchers at the University of Montana.42, 43, 44, 45, 46, 47

Recruitment and Survey Deployment

We contacted a convenience sample of 234 US companies regarding CAPTURE Survey participation through word of mouth, emails, LinkedIn messages, and website form submissions. Figure 1 reviews the number of companies contacted, interested, and participating. We spoke with human resource personnel, managers, and/or supervisors within companies returning correspondence. A short informational presentation was provided to these key contacts at interested companies that outlined the survey aims, the company and employee participation requirements, and the potential benefits of the company’s participation. Companies that agreed to have their employees participate were given 2 survey deployment choices: (1) the company sends out a company-wide email with the survey link and informational materials inviting employee participation or (2) the company collects the email addresses of interested employees and shares those email addresses with us for survey deployment. Four companies chose to generate their own company-wide deployment email, and we sent out company-provided emails to four companies. Although employees from participating companies were strongly encouraged to participate, they were told that participation was completely voluntary and had no effect on their company employment or relationship with us. The survey was sent to employees of participating companies starting November 20, 2020, and closing February 8, 2021. A reminder email was sent approximately 5 to 7 business days after initial deployment, with the survey open to employees of a given company for up to 3 weeks after the initial deployment date.

Figure 1

Flowchart of the companies expressing interest and participation status.

Statistical Analyses

CAPTURE Survey responses were downloaded from Mayo Clinic’s Qualtrics Platform. Data were cleaned using Microsoft Excel (Microsoft Corp) and uploaded to Python 3.9 for further cleaning and analyses using Jupyter Notebook for Windows 1.3.1093. Duplicates were examined by the date of first survey completion and CAPTURE Survey progress completion for each duplicate entry. Of the duplicates, we kept the CAPTURE Survey entry with the highest progress completion.

In addition to determining prevalence and mean responses for the entire cohort, we also stratified by WFH categories as 25% or less, 26% to 50%, 51% to 75%, and greater than 75%. Furthermore, we compared employee responses to questions asking about psychosocial experiences, perceptions, and health behaviors before and during the COVID-19 pandemic. Likert responses were coded numerically, and the mean (95% confidence interval) change between responses before and during the pandemic was calculated for the whole sample and by WFH category.

Results

A total of 3607 employees from 8 separate companies across the United States completed the survey. Seven companies were considered industry and included industries such as transportation, manufacturing, and commercial facilities management. One company was an academic institution. The overall mean response rate from employees within the 8 companies was 14.9%, with intracompany response rates ranging from 6.8% to 63.8%. When analyses were stratified by company type (industry vs academic), results were not materially different.

Demographic Results

Demographic results are given in Table 1. Most employees were from the Midwest (83.5%). The mean ± SD participant age was 44.7±12.1 years, with most classified as professionals (49.5%), followed by executive, administrator, or senior manager (13.5%) and clerical and administrative support (11.8%). Perceived health was “good,” “very good,” or “excellent” for most participants (91.7%). The percentage of employees reporting that their doctor had told them they had a medical condition ranged from 2.9% for heart disease and stroke (combined) to 24.0% for mood disorder. Most employees (62.1%) reported reduced in-person contact with family who live outside the home, friends, colleagues at work, and events in the community because of the COVID-19 pandemic.

Table 1

Demographic Results of the Baseline CAPTURE Surveya

Characteristic	Participants (No. [%])
Sex
Female	2292 (67.4)
Male	1063 (31.3)
Other/nonbinary	33 (1.0)
Age (y)
<35	896 (26.9)
35–65	2309 (71.8)
>65	125 (3.8)
Hispanic or Latino/Latina/Latinx	124 (3.5)
Which of the following best describes you?
Asian, Black, or African American	217 (6.4)
Other	216 (6.4)
White	2924 (86.0)
Current marital status
Married or partnered	2469 (72.6)
Highest level of schooling completed
Less than Bachelor’s degree	364 (10.7)
Bachelor's degree	1260 (37.1)
Master's, Professional, or Doctoral degree	1762 (51.8)

“Prefer not to answer” responses were not included in the table and were less than 2% of the responses.

Socioeconomic Impact of the COVID-19 Pandemic

Most employees (85.7%) stated that they spent more than 90% of their time indoors when working, 90.5% stated that they spent an average of 33 to 60 hours per week working, and 58.8% stated that they currently had no face-to-face, in-person interactions with coworkers or the public while completing their job-related duties. Few employees (0.2%) reported being furloughed or furloughed previously (or temporarily laid off). Approximately 68% of employees believed that COVID-19 had a negative effect on their work. Few (0.7%) lost health insurance or other coverage for medical care. As shown in Supplemental Figure 1, available online at http://mcpiqojournal.org, with WFH percentage grouped into the four 25% categories, the distribution shifts dramatically from before to during the COVID-19 pandemic.

Psychosocial Outcomes and Perceptions

During the pandemic, many employees reported feeling stressed (53.1%), anxious (43.8%), and fatigued (41.1%) “quite a bit” or “all the time.” The prevalence rates of employees agreeing that they felt threatened about COVID-19 or afraid of COVID-19 were 39.4% and 60.0%, respectively, and 60.3% reported being stressed around other people because they worried that they would catch COVID-19. In total, 32.4% reported being “moderately worried” or “very worried” about contracting COVID-19 while at work, and 34.1% reported being “moderately worried” or “very worried” about being an asymptomatic carrier.

Table 2 shows the changes in psychosocial and behavioral responses from before to during the COVID-19 pandemic, with mean change and 95% confidence intervals for each parameter. More than half of the employees reported an increase in stress, anxiety, fatigue, feeling unsafe, lack of companionship, and feeling isolated from others. For the behavioral outcomes, a high percentage of employees reported a decrease in productivity and physical activity, and increases were observed for non–work-related screen time, sleep, and alcohol consumption.

Table 2

Psychosocial and Behavioral Changes From Before to During the Coronavirus Disease 2019 Pandemic

	Change (No. [%])			Mean Change	95% Confidence Interval
	Decreased	None	Increased
Psychosocial responses
Stressa	336 (10.0)	1217 (36.1)	1821 (54.0)	0.56	0.53 to 0.60
Anxietya	245 (7.3)	1192 (35.4)	1934 (57.4)	0.70	0.67 to 0.73
Fatiguea	389 (11.5)	1243 (36.8)	1742 (51.6)	0.58	0.54 to 0.61
Feeling unsafea	214 (6.4)	1455 (43.2)	1697 (50.4)	0.74	0.70 to 0.78
A lack of companionshipc	155 (4.6)	1179 (34.9)	2040 (60.5)	0.76	0.73 to 0.79
A feeling of being left outc	247 (7.3)	1716 (50.9)	1408 (41.8)	0.43	0.40 to 0.46
A feeling of being isolated from othersc	119 (3.5)	918 (27.2)	2338 (69.3)	0.94	0.91 to 0.97
Behavioral responses
Productivitya	1454 (42.9)	1467 (43.3)	467 (13.8)	–0.51	–0.55 to –0.47
Physical activitya	1398 (41.1)	1256 (36.9)	750 (22.0)	–0.29	–0.33 to –0.25
Non–work-related screen timeb	92 (2.7)	1583 (46.6)	1725 (50.7)	0.56	0.54 to 0.59
Sleepa	545 (16.0)	1778 (52.2)	1085 (31.8)	0.21	0.18 to 0.24
Alcohol consumptiond	338 (10.0)	2193 (64.9)	848 (25.1)	0.22	0.19 to 0.25

Questions were asked on a 5-point Likert scale.

Questions were asked on a 4-point Likert scale.

Questions were asked on a 3-point Likert scale.

Questions were asked on a 9-point Likert scale.

Figure 2 shows the mean change and 95% confidence intervals for the psychosocial and behavioral responses stratified by WFH categories of greater than 75% (n=2682) vs 25% or less (n=509) of worktime. As shown, mean increases were higher in the 25% or less WFH group for stress, anxiety, fatigue, and feeling unsafe. Decreases in productivity were larger for WFH 25% or less, and an increase in sleep was observed only for WFH greater than 75%. We also ran additional analyses comparing those 100% WFH (n=2017) with those 0% WFH (n=214) with the greater than 75% WFH (n=2682) and 25% or less WFH (n=509) groups. This sensitivity analysis did not change the results materially, and therefore, to preserve statistical power, we provided only the results for the greater than 75% WFH and 25% or less WFH groups.

Figure 2

Psychosocial and behavioral responses (mean, 95% confidence intervals) from before to during the coronavirus disease 2019 pandemic stratified by work from home (WFH) status (≤25% WFH [n=509] vs >75% WFH [n=2682]).

SARS-CoV-2 Prevention Behaviors in the Workplace

Employees reported “often” or “always” engaging in personal SARS-CoV-2 transmission prevention behaviors in the workplace (Table 3). Glove wearing was an exception, with most reporting “never” or “rarely.” Table 3 also shows personal protective equipment, instructions, and practices provided and promoted by the employers. Most employees reported that their employer provided web training (33.4%), reading materials (24.1%), or both (25.5%).

Table 3

Personal COVID-19 Prevention Behaviors, Workplace Culture, Practices, and PPE Provided and Promoted During the COVID-19 Pandemicab

Variable	Participants (No. [%])
	Never/rarely	Sometimes	Often/always
Personal COVID-19 prevention behaviors
Worn a mask of any type	18 (0.9)	18 (0.9)	1982 (98.20)
Worn gloves	1225 (64.1)	353 (18.5)	332 (17.4)
Washed my hands regularly	21 (1.0)	66 (3.3)	1932 (95.7)
Physically distanced from coworkers or public	27 (1.4)	100 (5.0)	1870 (93.6)
Disinfected surfaces	334 (17.5)	371 (19.4)	1209 (63.2)
Monitored symptoms before work	293 (14.9)	203 (10.3)	1474 (74.8)
PPE and instructions provided by company
N95 masks	1750 (50.4)	152 (4.4)	279 (8.0)
Surgical masks	1230 (35.4)	198 (5.7)	855 (24.6)
Cloth masks	741 (21.2)	593 (16.9)	1287 (36.8)
Gloves	1077 (30.9)	240 (6.9)	959 (27.5)
Hand sanitizer	270 (7.7)	174 (5.0)	2265 (64.8)
Handwashing instructions	130 (3.7)	163 (4.7)	2512 (71.8)
Physical distancing instructions	77 (2.2)	160 (4.6)	2652 (75.8)
Cleaning/disinfecting products for surfaces	265 (7.6)	246 (7.0)	2068 (59.1)
Tools to monitor symptoms before work	1286 (36.8)	136 (3.9)	936 (26.8)
PPE and practices promoted by company
N95 masks	1425 (41.1)	353 (10.2)	761 (21.9)
Surgical masks	831 (23.9)	281 (8.1)	1589 (45.7)
Cloth masks	136 (3.9)	182 (5.2)	2805 (80.2)
Gloves	1276 (36.7)	497 (14.3)	932 (26.8)
Hand sanitizer	95 (2.7)	208 (6.0)	2863 (81.9)
Handwashing	38 (1.1)	90 (2.6)	3105 (88.9)
Physical distancing	27 (0.8)	80 (2.3)	3137 (89.7)
Surface cleaning/disinfecting	184 (5.3)	337 (9.6)	2529 (72.4)
Monitoring of symptoms before work	241 (6.9)	243 (7.0)	2549 (73.0)

COVID-19, coronavirus disease 2019; PPE, personal protective equipment.

Row totals do not add to 100% because of a lack of applicability to working at home.

When asked about perceptions of the importance of specific prevention behaviors and practices, most employees reported wearing a mask, handwashing, and physical distancing to be “‘very important’” (93.3%, 89.1%, and 93.5%, respectively). Disinfecting surfaces was reported as “‘very important’” by some (45.5%), whereas most employees reported that wearing gloves was “not important” (36.2%).

COVID-19 Symptoms and Diagnosis

The COVID-19 symptoms reported as lasting several hours more than usual were headaches (34.3%); unusual fatigue (27.1%); malaise or general feeling of illness, discomfort, or uneasiness (23.2%); and muscle aches (22.3%). In total, 13.1% of employees suspected that they had a COVID-19 infection or had COVID-19 symptoms. Most reported having a negative nasal swab or saliva test (52.9%) and a negative blood/antibody test (9.9%). Few had a positive nasal swab or saliva test (4.0%) and a positive blood/antibody test (1.0%).

Discussion

We observed large increases in stress, anxiety, fatigue, and feeling unsafe among non–healthcare employees due to the COVID-19 pandemic. We also noted an increase in lack of companionship, feeling left out, or isolation when completing work-related duties, with decreased productivity and physical activity reported. Most employees were afraid of COVID-19 and stressed about acquiring the disease in general. Interestingly, however, not as many employees were worried about contracting COVID-19 at work or being an asymptomatic carrier of COVID-19 while at work, regardless of WFH status. A high percentage of employees believed that wearing a mask, handwashing, and physical distancing were important in preventing the spread of COVID-19. Whereas productivity and physical activity decreased for many employees, non–work-related screen time increased. Sleep also increased in a small percentage of employees and, in particular, those engaged in greater than 75% WFH each week. Alcohol consumption increased in one-quarter of respondents, possibly related to stress.

Research has identified adverse psychosocial changes during the COVID-19 pandemic, with higher levels of anxiety, depression, posttraumatic stress disorder, and stress observed in the general population in multiple countries. In the United States, a survey conducted from March through June 2020 found that distress increased as the pandemic first emerged in the United States, including increases in anxiety and depression. Literature has suggested psychosocial support to be critical to mitigating these adverse changes, as individuals reporting having no psychosocial support were more vulnerable to mental health problems during the COVID-19 pandemic. These observations align with the present findings from employees across the United States and suggest that employers might need to make concerted efforts to assist employees in dealing with mental health issues as more individuals return to the physical workplace. Notably, the present observations suggest that the focus of employers might need to be most concentrated on those who have worked the least amount of time at home during the COVID-19 pandemic.

The present behavioral outcome observations were also largely consistent with previous research on the COVID-19 pandemic. Increased non–work-related technology use and screen time have been seen in previous studies. We saw a slight increase in hours of sleep per night among employees greater than 75% WFH each week but not among employees 25% or less WFH each week. Despite the increased sleep duration observed among some, other research during the COVID-19 pandemic has suggested that individuals’ sleep quality has been poorer. Reductions in physical activity were also observed in existing research. More investigation of these phenomena, as well as changes in alcohol consumption, might be warranted. We believe that the ongoing follow-up CAPTURE Surveys will provide more insight.

A secondary aim of CAPTURE was to assess personal SARS-CoV-2 transmission prevention behaviors taken in the workplace to mitigate the spread of COVID-19. Although wearing a mask, washing hands regularly, physically distancing from coworkers, and monitoring symptoms were reported in high percentages, wearing gloves and disinfecting surfaces were not as common. This can be explained by the changing viewpoints of SARS-CoV-2 spread and associated mitigation strategy guidance set forth by the Centers for Disease Control and Prevention and the World Health Organization suggesting that surface-based transmission is less common. Importantly, employees reported their companies to be highly supportive of SARS-CoV-2 transmission prevention behavior engagement. Indeed, many companies provided COVID-19 prevention training for their employees which might have also aided in the high percentage of prevention behaviors reported.

Study limitations should be noted. First, participants were a convenience sample. More than 200 companies were contacted, but the companies that did partake in the survey had previously worked with, or were connected to, the research institutions. This convenience sample skewed the educational status of participants, with more than half of the participants having a graduate degree. Second, although we asked about job status, we did not ask whether employees identified as essential or nonessential workers due to the multidimensional nature of the companies. Future questionnaires should include perceptions of essential worker status or have a defined strategy to determine whether employees are considered essential or nonessential workers. Finally, this was a cross-sectional analysis of a prospective survey. The questions asking about outcomes and perceptions before COVID-19 were retrospective and subject to recall bias. However, follow-up studies will describe prospective observations of the ongoing CAPTURE Survey.

Nevertheless, the study had its strengths, including (1) a large sample size, (2) a wide range of WFH percentages, and (3) a sample of non–health-care employees. These strengths provided unique insight into psychosocial and behavioral responses of employees working outside of the health care workforce during COVID-19. Increases in stress, anxiety, fatigue, and feeling unsafe due to work-related duties, as well as adverse changes in health behaviors (e.g., increased non–work-related screen time, decreased physical activity), suggest that companies might consider how to support their employees as employees begin to return to the office in greater numbers.