Changes in Teleworking and Physical Activity Behaviors in the United States Before and After Emergence of COVID-19

Abstract

Objective:

The aim of the study is to determine the prevalence of perceived decreases in three physical activity (PA) behaviors and meeting the PA guideline, by changes in telework.

Methods:

US workers (N =2393) reported teleworking and PA behaviors before and after COVID-19 emergence. Those reporting more and less telework were compared with those reporting stable telework on prevalence of (1) decreasing behaviors and (2) meeting the aerobic guideline (≥150 min/wk of moderate-intensity PA).

Results:

Compared with workers with stable telework, those with increased telework were more likely to report decreases in any PA (by 61%), active transportation (65%), and park use (52%). Workers who deceased telework were also more likely to report decreases in these behaviors. Groups were equally likely to meet the guideline.

Conclusions:

Changes in teleworking status—either more or less—may be associated with decreased participation in PA behaviors.

Physical activity (PA) is associated with greater mental health, higher quality of life, and reduced incidence of many diseases. These benefits of PA occur across the lifespan,^1,2 including among working-age adults.³ Guidelines for aerobic activity from the US Department of Health and Human Services¹ and the World Health Organization² encourage adults to move more and sit less, with a goal of accruing at least 150 min/wk of moderate-intensity aerobic activity, 75 min/wk of vigorous-intensity aerobic activity, or an equivalent combination.

In addition to the physical and mental health benefits,^1,2 regular PA may reduce job-related absenteeism and improve productivity—particularly among office workers with largely sedentary work routines.^4–6 Because physical inactivity was associated with greater frequency and duration of COVID-19 symptoms^7,8 and hospitalization,^8–10 the effect on absenteeism may have been pronounced during the pandemic. This coincided with other pandemic-related changes across society, including stay-at-home orders, school closures, and an expansion of telework across many sectors of the economy.¹¹ (We use the term telework to capture the concept also described as telecommuting, remote work, and work from home.) Several studies around the world have analyzed the impact of these pandemic-related workplace disruptions on PA levels and behaviors, including active transportation.^12–15 These small studies have yielded mixed results. Understanding the relationship between teleworking and PA is particularly important because telework, unlike social distancing and other public health mitigation measures,¹⁶ may not return to prepandemic levels.^17,18

The primary objective of this study was to examine the prevalence of perceived decreases in three behaviors (any PA, active transportation, and use of parks and trails) by changes in teleworking status during the first year of the COVID-19 pandemic among workers in the United States. The second objective was to examine the prevalence of workers’ meeting the aerobic PA guideline by changes in teleworking status.

METHODS

Study Sample

This cross-sectional study used data from Porter Novelli’s Styles database—a series of web-based surveys sent to members of the Ipsos KnowledgePanel. Panel members are randomly recruited via a probability-based sampling strategy according to home address, resulting in a representative sample of the noninstitutionalized US population with approximately 60,000 panelists.¹⁹ SummerStyles, the survey used in this analysis, was conducted during May 21-July 6, 2022, and included 5990 randomly invited participants; 4156 responded (in-panel response rate = 69.3%).²⁰ Of these, 2462 reported currently working, and 2391 (97.1%) workers had complete data for analysis (Fig. 1).

FIGURE 1.

Flow diagram of the participants.

Survey Questions

The survey introduced questions on perceived changes in PA and telework behaviors with this stem: “COVID-19 was declared a pandemic March 11th, 2020. A pandemic is a global outbreak of disease. In the past year, compared to the year before the COVID-19 pandemic, have you changed the amount you do the following activities?” It then listed the following three physical activities: (1) “Walking or biking to travel to and from places” (ie, active transportation); (2) “Using parks, trails, or greenways for physical activity” (which we abbreviate herein as parks and trails); (3) “Being physically active or exercising for any reason” (ie, any PA). It also listed one work activity: “Teleworking or working remotely.” For each prompt, participants could select “I do more of this activity,” “I do the same amount of this activity,” “I do less of this activity,” or “I have never done this activity.”

The survey panel includes information on several sociodemographic characteristics. Respondents were classified by age (18–44, 45–64, and ≥65 years), sex (male or female), race/ethnicity (non-Hispanic White, non-Hispanic Black, Hispanic, and non-Hispanic other), education level (high school or less, some college, and college graduate), marital status (married; divorced, separated, or widowed; and never married), household income (<$50,000, $50,000-$99,999, and ≥$100,000), and employment status (full-time or part-time).

Leisure-time aerobic PA during a “usual week” was assessed based on questions from the National Health Interview Survey. ²¹ Weekly duration of moderate-to-vigorous PA (MVPA) was calculated as the sum of vigorous minutes multiplied by two and moderate minutes.^1,2 Participants’ PA level was classified as inactive, insufficient (1–149 min/wk of MVPA), sufficiently active (150–300 min/wk), and highly active (>300 min/wk). Analyses used the Physical Activity Guidelines for Americans, Second Edition, which defines recommended aerobic PA as reporting ≥ 150 min/wk of MVPA. ¹

Statistical Analysis

The sample was described using provided analytic weights. Respondents were weighted according to 2021 Current Population Survey proportions for age group, sex, race/ethnicity, education, household income and size, parental status of adolescent children, Census region, and metro status (metro or nonmetro). Weighted prevalence and 95% confidence intervals (CIs) of changes in each of the three PA behaviors were estimated using log-binominal regression. To determine whether teleworking status adversely affected these behaviors, weighted prevalence of reporting a decrease in each behavior was also calculated, stratified by changes in teleworking status: (1) stable telework (ie, those reporting never teleworked or the same amount of telework), (2) more telework, and (3) less telework. The second and third groups were compared with the first group with respect to decreasing each PA behavior. This was accomplished with log-binomial regression to calculate unadjusted and adjusted prevalence ratios (aPR)—the latter accounting for age (as a continuous variable), race/ethnicity, education, marital status, income, and employment status (full-time or part-time). Those reporting more and less telework were also compared with those reporting stable telework with respect to meeting the aerobic guideline at the time of survey administration. We also assessed perceived decreases in PA and meeting the aerobic guideline stratified by sex; we compared sex-specific aPRs using a test for interaction. We also repeated analyses with “never teleworked” as the referent group. Significance was established at a two-sided a value of 0.05. All analyses were conducted in SAS v9.4 (SAS Institute, Cary, NC), and the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline for cross-sectional studies was used to prepare this report.²²

Approvals

All participants in SummerStyles provided written consent to participate in the Ipsos KnowledgePanel. This study used existing survey data and was exempt from institutional review board approval.

RESULTS

Participant Characteristics

The weighted sample was predominantly aged 18 to 44 years (55.6%), male (54.7%), non-Hispanic White (61.3%), and married (55.8%). A plurality reported never teleworking (37.5%), followed by more telework (32.0%) and the same amount of telework (22.0%); only 8.5% reported less telework. Most participants were either sufficiently active (23.2%) or highly active (31.0%) (Table 1).

TABLE 1.

Baseline Characteristics of the Employed US Adults—SummerStyles 2022 (N = 2,391)

	n (Weighted %)
Age, yr
18–44	1,052 (55.6)
45–64	1,091 (37.1)
≥65	248 (7.3)
Sex
Male	1,313 (54.7)
Female	1,078 (45.3)
Race/ethnicity
NH White	1,670 (61.3)
NH Black	211 (10.3)
Hispanic	294 (18.8)
NH other	216 (9.7)
Education
HS or less	536 (28.4)
Some college	629 (26.2)
College grad	1,226 (45.4)
Marital status
Married	1,578 (55.8)
DSW	297 (11.3)
Never married	516 (32.9)
Income
<$50K	396 (20.1)
$50K-99.9K	687 (30.7)
≥$100K	1,308 (49.2)
Employment
Full-time	1,893 (77.2)
Part-time	498 (22.8)
Teleworking statusa
Never teleworked	856 (37.5)
Same telework	545 (22.0)
More telework	789 (32.0)
Less telework	201 (8.5)
PA, min/wkb
Inactive (0)	371 (17.1)
Insufficient (1–149)	669 (28.7)
Sufficient (150–300)	582 (23.2)
Highly active (>300)	769 (31.0)

^aComparing the year before the survey to the year before the COVID-19 pandemic.

^bModerate-intensity equivalent leisure-time PA.

DSW, divorced, separated, and widowed; HS, high school; NH, non-Hispanic; PA, physical activity.

Perceived Changes in PA Behaviors

Approximately half or more perceived maintaining a consistent level of participation in any PA (59.9%), active transportation (49.9%), and use of parks and trails (51.1%). The prevalence of increasing (16.4%) and decreasing (16.6%) any PA was similar. Whereas 8.7% reported increased active transportation, 15.3% reported decreased active transportation. Regarding parks and trails, 13.3% reported increased use, while 16.6% reported decreased use (Table 2).

TABLE 2.

Perceived Changes in Three PA Behaviors Among the Employed US Adults—SummerStyles 2022 (N = 2,391)

	Any PA, n (Weighted %)	Active Transportation, n (Weighted %)	Use of Parks and Trails, n (Weighted %)
Increased behavior	394 (16.4)	206 (8.7)	306 (13.3)
Remained same	1,485 (59.9)	1,228 (49.9)	1,289 (51.1)
Decreased behavior	364 (16.6)	330 (15.3)	359 (16.6)
Never participated	148 (7.0)	627 (26.1)	437 (19.0)

PA, physical activity.

Perceived Changes in PA Behaviors by Telework Status

For each of the three behaviors, those who reported more telework and less telework, compared with those who reported stable levels (ie, same or never), were more likely to perceive a decrease in any PA, active transportation, and use of parks and trails. Compared with 13.0% of those with stable telework, 19.2% of those with more telework reported decreased PA (aPR = 1.61; 95% CI: 1.30–1.98), and 32.3% of those with less telework reported decreased PA (aPR = 2.43; 95% CI: 1.91–3.10). Similar differences were observed for perceived decreases in active transportation and in use of parks and trails (Table 3).

TABLE 3.

Perceived Decreases in PA Behaviors by Reported Changes in Telework Among the Employed US Adults—SummerStyles 2022 (N = 2391)

	Prevalence, % (95% CI)	Unadjusted PR (95% CI)	Adjusted PRa (95% CI)
Decreased PA
Stable teleworkb	13.0 (10.7–15.3)	1.00	1.00
More telework	19.2 (15.9–22.5)	1.48 (1.21–1.80)	1.61 (1.30–1.98)
Less telework	32.3 (24.1–40.5)	2.49 (1.96–3.16)	2.43 (1.91–3.10)
Decreased active transportation
Stable teleworkb	11.9 (9.7–14.1)	1.00	1.00
More telework	16.8 (13.6–20.1)	1.41 (1.14–1.74)	1.65 (1.32–2.07)
Less telework	33.4 (25.1–41.8)	2.80 (2.21–3.56)	2.66 (2.10–3.38)
Decreased use of parks and trails
Stable teleworkb	12.8 (10.6–15.0)	1.00	1.00
More telework	16.7 (13.5–19.9)	1.30 (1.06–1.61)	1.52 (1.22–1.89)
Less telework	42.5 (33.9–51.2)	3.32 (2.69–4.10)	3.05 (2.49–3.74)

^aFor age, sex, race/ethnicity, education, marital status, income, and full- or part-time employment.

^bIncludes participants who reported never teleworking and the same amount of teleworking.

CI, confidence interval; PA, physical activity; PR, prevalence ratio.

Meeting the Aerobic PA Guideline by Telework Status

At the time of survey completion, 51.4% of participants who had stable telework reported a level of leisure-time PA that meets the aerobic guideline, compared with 60.0% of those who had reported more and 51.7% of those who reported less. Prevalence differences were not statistically significant in the adjusted model (Table 4).

TABLE 4.

Changes in Teleworking and Meeting the Aerobic PA Guideline^a Among Employed US Adults—SummerStyles 2022 (N = 2,391)

	Prevalence, % (95% CI)	Unadjusted PR (95% CI)	Adjusted PRb (95% CI)
Stable teleworkc	51.4 (48.2–54.7)	1.00	1.00
More telework	60.0 (55.8–64.1)	1.17 (1.08–1.26)	1.03 (0.95–1.11)
Less telework	51.7 (43.1–60.3)	1.01 (0.87–1.16)	1.02 (0.88–1.17)

^aReported ≥150 min/wk of moderate-intensity equivalent leisure-time PA.

^bFor age, sex, race/ethnicity, education, marital status, income, and full- or part-time employment.

^cIncludes participants who reported never teleworking and the same amount of teleworking.

CI, confidence interval; PA, physical activity; PR, prevalence ratio.

Sensitivity Analyses

The association between less telework and decreased PA was stronger among males than females (P for interaction = 0.04), but no sex differences were observed for the association between telework changes and meeting the aerobic PA guideline (Supplementary Table 1, http://links.lww.com/JOM/B347). Patterns largely held when never telework, instead of stable telework, was used as the referent group (Supplementary Table 2, http://links.lww.com/JOM/B347).

DISCUSSION

Among surveyed US workers, 32.0% reported more telework during the first year of the COVID-19 pandemic, while 8.5% reported less telework. Compared with those with stable telework, and adjusted for sociodemographic factors, participants who reported more or less telework perceived lower participation in any PA, active transportation, and use of parks and trails. Nonetheless, at the time of survey completion, these groups were equally likely to meet the aerobic PA guideline in leisure time.

The fact that both more and less telework, compared with stable telework, was associated with declines in perceived PA behaviors, suggests that work arrangement changes—rather than telework itself—may influence participation in multiple forms of PA. An accelerometer-based study conducted in summer 2020 found that Swedish office workers had different 24-hour movement compositions on days working from home compared with days working at the office.¹⁵ Work schedule changes may be particularly detrimental when experienced as uncontrollable. Flexible work arrangements, in which employees choose where and when to work, were associated with improved perceptions of physical health in a recent meta-analysis.²³ In addition, in a longitudinal study of workers in Japan, those who perceived less control over their work time were more likely to develop insomnia, fatigue, and depressive symptoms.²⁴ Studies of nurses have found that rotating shift work, compared with fixed-schedule work, is associated with higher body mass index²⁵ and lower levels of PA.²⁶ Despite heterogeneity in occupations and outcomes, these studies collectively suggest that imposed and inconsistent work routines may have detrimental health effects in some individuals. It is unknown whether predictable and consistent telework would mitigate these health effects.

Work routine inconsistency is not the only plausible explanation of these findings. Other factors unrelated to work routine may have affected PA participation among persons who had to increase or decrease telework during the pandemic. For example, those required to telework more may have previously commuted to work by walking or bicycling, lost access to PA opportunities at or around their usual workplace, or assumed new childcare requirements at home. Meanwhile, those required to telework less may have experienced time shift from leisure-time PA to work obligations in frontline occupations, or whose motivation for PA was curbed by pandemic-related occupational stressors. The strong associations between less telework and decreased PA behaviors may also reflect pandemic-related changes in upstream determinants of PA, including interpersonal and environmental supports.²⁷ For example, frontline workers may have decreased their PA if they experienced less social support or had reduced access to public transit and parks. Sex-based differences in such effects may also be possible, evidenced in this study in which more males exhibited a stronger association between less telework and a perceived decrease in PA.

Across the US labor force, 33% of employers increased telework during the pandemic, with sector-specific gains ranging from 3% (accommodation and food services) to 60% (educational services).¹¹ Unsurprisingly, studies during the COVID-19 pandemic have focused on the effect of more (vs less) telework on PA. Their results have varied. A Brazilian study found that individuals who were required to telework, compared with those who remained at their usual worksites, were more likely to be physically active.¹² Conversely, working from home was associated with greater sedentary time among Turkish employees.²⁸ Studies of Japanese workers found that teleworkers experienced large declines in active commuting—declines which were not offset by commensurate gains in noncommute active transportation.¹⁴ They also performed less PA during work time, relative to those who remained in the office, with the share of worktime spent teleworking inversely correlated to PA participation.¹³ Adult participants in the Cancer Prevention Study-3 decreased moderate- and light-intensity PA in the summer of 2020, with no statistical differences across work and home arrangements. In other words, pandemic-related changes in teleworking status and childcare responsibilities did not affect PA participation in this longitudinal US cohort.²⁹

Attempting to summarize the literature on changes in PA early in the pandemic, Wunsch and colleagues³⁰ suggested that “work-from-home orders” contributed to decreased PA in many countries; however, the impact of telework could not be cleanly disentangled from other pandemic repercussions. In the present cohort, those who experienced more (and less) telework, while perceiving decreased PA in the pandemic’s first year, were equally likely to meet the aerobic PA guideline when assessed in summer 2022. Because our dataset does not include prepandemic PA participation, we cannot compare changes in the proportion meeting the guideline.

Many economists concur that teleworking will persist in the coming years.^31,32 As industries adapt to the postpandemic work landscape, employers may wish to consider the consequences of inconsistent telework schedules on PA. This may be especially important to employers given the impact of PA on occupational outcomes. In the first systematic review of modifiable risk factors for absenteeism, Wagner and colleagues⁵ concluded, based on strong evidence, that PA is inversely related to work-related disability and absenteeism. A more recent systematic review found that workplace interventions to increase PA not only reduce absenteeism but may also reduce nonproductive work time.⁶ At the macroeconomic level, the implications of reduced PA are considerable. One analysis estimated that population-wide achievement of the aerobic guideline would increase the global gross domestic product by 0.15% to 0.24% annually by 2050, resulting in a cumulative economic dividend of 6.0 to 8.6 trillion (in 2019 US dollars).³

Short and simple fitness programs in the workplace are associated with positive productivity and fiscal outcomes,³³ but this evidence base was established before the pandemic. A recent scoping review concluded that new evidence is needed to understand how these prepandemic workplace interventions can be translocated to the remote or hybrid work environment.³⁴ Several approaches have been offered for increasing PA and reducing sedentary behavior in these new work arrangements, including active workstations (eg, sit-stand desks), employee incentive programs (eg, gym memberships), activity trackers, screen time control devices, and computer-based prompts for movement.^29,34 Active People, Healthy Nation,^SM an initiative by the Centers for Disease Control and Prevention, promotes evidence-based strategies to increase PA across the population. Some of these strategies, such as community design interventions and social support,³⁵ could be used to support the PA of teleworkers. This study suggests that public policy interventions to increase PA among workers should consider the impact of schedule changes. Greater perceived control over one’s work schedule may have salutary effects on PA participation by reducing barriers, such as psychological distress, somatic symptoms, and social constraints.^23,24

The results of this study should be interpreted considering a few limitations. First, the findings are based on self-report, which are susceptible to recall and social desirability biases. Recalling changes in telework and PA behaviors could be challenging because participants were asked to compare two periods, both of which ended more than a year before survey completion. Second, other pandemic effects, unrelated to changes in telework—such as greater caregiving responsibilities and poorer sleep quality^36–38—may have contributed to these findings. Third, despite random recruitment of panel members and application of survey weights, the study sample was not designed to be representative of all sectors of the US labor force. Fourth, for the secondary objective only, the survey timing may have introduced a complication. The prevalence of meeting the aerobic PA guideline was assessed at the time of survey completion, which was noncontiguous with the periods used to assess changes in teleworking. Because PA levels may have changed during the interlude, these results should be interpreted cautiously.

CONCLUSIONS

In a nationwide survey conducted in 2022, nearly one third of workers in the United States reported more telework during the first year of the COVID-19 pandemic, compared with one in 12 who reported less telework. Both groups of workers—those who either increased or decreased their amount of telework—were more likely to report concomitant declines in PA, active transportation, and use of parks and trails. Given the importance of PA for workers’ physical health, mental wellbeing, and economic productivity, employers may wish to consider the impact of telework schedule changes on opportunities to be physically active. This might include tailoring effective worksite wellness programs for teleworkers. Public policy efforts to promote PA—such as community design and access to places for PA—might consider unique challenges faced by workers with changing schedules.

Supplementary Material

Supplemental digital contents are available for this article. Direct URL citation appears in the printed text and is provided in the HTML and PDF versions of this article on the journal’s Web site (www.joem.org).

LEARNING OUTCOMES.

• Understand how teleworking changes before and after COVID-19 emergence may have affected physical activity, active transportation, and park use in the United States

• Ensure physical activity opportunities for all workers, with special attention to those who experience telework schedule changes

• Determine whether known workplace interventions that promote physical activity are effective in remote or hybrid work environments