Associations Between Daily Activities, Stress, and Sleep Among Adults During the COVID-19 Pandemic

Abstract

The COVID-19 pandemic introduced lifestyle changes that appeared to impact adults’ engagement in health-promoting activities. Researchers observed increased stress, sleep problems, and sedentary activity, and decreased physical activity during the pandemic. This secondary analysis examined the frequency that adults engaged in different daily activities and links between activity types, stress, and sleep during the pandemic. Fifty-three adults completed a single online survey in January 2021 where they reported demographics, COVID-19 pandemic experiences, daily activities, stress, and sleep. Sum scores were calculated to indicate total frequency of different activity types. Regressions were used to examine whether the frequency of different activity types predicted sleep duration and stress. There was a trend-level relationship between the frequency of activities done outdoors and stress scores, such that doing outdoor activities more often was linked with less stress (P = .098). Higher income also tended to be linked with more stress (P = .053), as did younger age (P = .028). When restricting the outdoor activity variable to physical outdoor activities only, the relationship with stress became significant (P = .013). These results suggest that engaging in physical activity outdoors may reduce stress during the COVID-19 pandemic.

“The most reported COVID-19 impacts were working from home more, working with children at home, and knowing someone who tested positive for COVID-19.”

Background

During the COVID-19 pandemic, most people’s lifestyles changed considerably. Globally, both health and health-promoting behaviors were negatively impacted as people attempted to navigate these unprecedented times. This included increased stress levels, increased sleep problems, greater participation in sedentary activities, and less participation in physical activity.^1–7 Risk factors for experiencing increased stress during the pandemic included younger age, female gender, and caregiver status. ⁸ Women in particular were more likely to experience increased stress and decreased physical activity during and after COVID-19 associated quarantine, as compared to men. ⁹ Previously, sedentary activities have been linked with increases in stress and sleep problems, while physical activity has been found to reduce them. ¹⁰ Examining associations between different types of activities and health and well-being can help researchers and practitioners understand how to encourage people to engage in healthy lifestyles during the COVID-19 pandemic years and more broadly.

Understanding physical activity changes during the COVID-19 pandemic is critical to knowing how to promote health. Maintaining physical activity routines throughout the pandemic was associated with decreased anxiety and depression, as those who reported more time participating in moderate to vigorous physical activity during lockdowns had lower chances of presenting depressive symptoms and anxiety. ¹¹ People also engaged more frequently in sedentary activities during the pandemic, potentially exacerbating the negative health consequences of isolation. Social media use, for instance, was associated with higher anxiety, even when used for social support and information seeking. ⁶ Vogel et al ¹² found that active adults reported less stress than inactive adults during the COVID-19 pandemic enforced quarantine, and that active adults who managed their stress with physical activity (either completed indoors or outdoors), or by reading, had the lowest levels of stress. On the contrary, managing stress by watching television/movies, sleeping, and eating was associated with increased stress, and inactive adults were more likely to engage in these activities. Researchers have posited that the changes in stress, physical activity, sedentary behavior, and associated health outcomes catalyzed by the COVID-19 pandemic may have lasting implications, highlighting the importance of understanding predictors of positive health outcomes in this context.^9,13

Relatedly, there is a body of research on potential benefits of time spent outdoors in nature and physical activity completed outdoors. Pre-pandemic research highlighted the positive outcomes of nature interventions (regardless of the level of physical activity), such as decreases in stress, better mental health, and improvement in sleep quality and quantity. A systemic review concluded that green space exposure, whether studied via a self-reported questionnaire or through an intervention, such as an outdoor walking program or gardening, was associated with improvement in both sleep quality and quantity. ¹⁴ Additional unanticipated benefits of nature-based interventions have included the ability to form social networks or spend time with others, increased connectedness to nature, aid in healing or recovery, and restoration of mental and emotional health.^15,16

Participating in physical outdoor activities has even stronger health advantages than physical activity completed indoors, or inactive time spent in nature. Pre-pandemic research found active engagement with nature, such as participating in an outdoor hike, as compared to watching nature scenes while walking on a treadmill, had greater positive effects on calmness, reduced fatigue and anxiety, and improvement in mood and decreased stress.^17,18 Greater enjoyment was also observed with participation in outdoor green vs virtual green exercise and is a motivating factor to engage with nature, along with spending time with loved ones.^19,20 Although most literature has shown positive impacts of outdoor physical activity, mixed findings have also been reported. A meta-analysis of 25 studies provided some evidence of positive benefits, specifically on self-reported emotions, of a walk or run in a natural environment as compared to a synthetic environment. There is limited research supporting consistent reductions in cortisol concentrations across study groups, suggesting the need for further investigation into the link between physical outdoor activity and some outcomes, like stress. ²¹

Purpose and Aims

While most of the extant research in this area occurred prior to the COVID-19 pandemic, recent studies revealed potential benefits of promoting regular physical activity in green spaces during the pandemic: After a structured park-based physical activity intervention, researchers noted significant improvements in vitality among both men and women, as well as anxiety, depressed mood, self-control, and positive well-being in women. ²² Another study found that previously inactive adults who spent more time engaged in outdoor physical activity had lower anxiety than those who spent less time doing so, suggesting that participation in outdoor physical activity may be most impactful to those who do not normally receive its associated benefits. ²³ This analysis seeks to build upon the extant research to further understand links between different types of activities and health and well-being during the COVID-19 pandemic.

The primary aim of this study was to examine if the frequency adults engage in outdoor activities was linked with their stress levels and total daily sleep duration during the COVID-19 pandemic. We hypothesized that the more adults engaged in outdoor activities, the lower their stress levels would be and the greater their total daily sleep duration would be. A secondary aim was to determine if the frequency adults engage in different activity types (sedentary, social, and physical) was linked with their stress levels and total daily sleep duration. We predicted that the more adults engaged in physical activities, the lower their stress levels, and the greater their total daily sleep duration would be. We also examined whether those physical activities that were done outdoors were particularly protective, with even stronger associations with stress and sleep than overall physical activity done indoors and outdoors. There were no specific hypotheses for links between sedentary or social activities, stress, and sleep. To our knowledge, this analysis is the first study that seeks to simultaneously explore the potential protective health effects of each of these different activity types among adults during the COVID-19 pandemic; it is also novel in that it focuses on activities during wintertime in a cold-weather climate.

Methods

Participants

A total of 53 adults from the Western New York region were recruited in January 2021 via advertisements on social media and lab listserv emails. Eligibility criteria included adults 18 years or older who were English speaking, had online access, reported living in the Western New York region, were interested in receiving suggestions for ways to get outside, stay active, and stay busy during COVID-19, had no health problems precluding participation, and were not currently involved in regular (i.e., at least weekly) hikes or nature walks. Inclusion criteria were developed for a larger study of which this secondary analysis is a part. A total of 346 individuals completed an eligibility survey, 185 of whom were eligible. Recruitment stopped after enrollment of 53 participants based on the a priori sample size goals of the overarching study. ²⁴

Procedures

Participants completed an online survey in January 2021, as part of a larger intervention study designed to examine if providing access to a regional winter hiking challenge had positive health outcomes during the COVID-19 pandemic. ²⁴ The present analysis focuses on the overarching study’s baseline survey only, prior to the implementation of any interventions. The survey link was provided to participants via email. Study procedures were reviewed and approved by the University at Buffalo Institutional Review Board, with the study granted an exempt status.

Measures

The online survey consisted of 122 total questions about demographics, COVID-19 experiences, daily activities, stress, and sleep. Those relevant to the present analysis are described in more detail below.

Demographics

Participants completed demographic survey questions about themselves, including their sex, age, race/ethnicity, education, marital status, and annual household income. They also reported on their family, including the total number of adults and children under 18 years old in the household, and the ages of any children.

COVID-19 Experiences

Additional questions from the National Institutes of Health Common Data Element bank allowed participants to report on how the COVID-19 pandemic affected them in the past month. ²⁵ This included whether anyone close to them tested positive for COVID-19 and ways the COVID-19 pandemic impacted their lives, including employment, childcare, and finances. They also had an opportunity to answer an open-ended question about additional ways the COVID-19 pandemic impacted them in the past month.

Activities

An adapted Pleasant Activities List assessed how often participants engaged in a list of 91 activities in the past 30 days on a scale of 0 (never) to 4 (more than once a week) and where each activity took place (outdoors, indoors). ²⁶ Activities were categorized into the following 4 types: sedentary enriching (n = 28; ex. reading), sedentary non-enriching (n = 25; ex. watching TV), physical (n = 20; ex. going for a run or jog), and social (n = 18; ex. going to a sports event). Categorizations were adapted from a previous study. ²⁷

Stress

The 10-item validated Perceived Stress Scale (PSS) assessed stress during the prior month. Participants rated how often they felt a certain way (ex: upset because of something that happened unexpectedly) on a Likert scale of zero (never) to 4 (very often). ²⁸ A total score was calculated from the 10 items, with higher scores indicating greater stress. The possible range of scores was from 0–40, with scores between 14–26 indicating moderate stress.

Sleep

An item from the Pittsburgh Sleep Quality Index was used to assess participants’ typical hours of nighttime sleep over the past month. ²⁹

Statistical Analysis

Sum scores were created from Pleasant Activities List items to represent the total frequency of each activity type of interest in the past month: outdoor, sedentary enriching, sedentary non-enriching, physical, and social. Physical activities that participants reported completing outdoors were also examined, based on prior research suggesting that these may be particularly beneficial. Based on National Sleep Foundation ¹⁴ recommendations that adults should receive between 7–9 hours of sleep per night, sleep was dichotomized between participants who reported <7 hours per night and participants who reported >7 hours of sleep per night. Descriptive statistics were conducted to examine means and standard deviations (continuous variables) and frequencies (categorical variables) for all variables of interest. Analysis of Covariance models (ANCOVAs) were used to examine whether frequency of each activity type of interest predicted stress scores, and logistic regressions were used to examine whether activity types predicted sleep duration (>7 hours or not). Assumptions were checked to ensure residuals were normally distributed. Given observed relationships between variables of interest and participant demographics, income, age, and household size were considered as covariates, with backwards deletion of those that did not predict outcomes at P < .10.

Results

Descriptive Statistics

Demographics

Demographic characteristics are shown in Table 1. Most of the sample was white women, and the mean participant age was 47.8 years (SD = 10.6). Nearly all (>95%) of participants had a college degree, with 56.6% having earned a graduate degree. Most participants (77.4%) were married. The average total number of individuals in the household was 3.4 + 1.4 and 54.7% of participants had children in the household under 18 years old.

Table 1.

Demographic Characteristics of Participants (n = 53).

	Frequency (%) or Mean ± SD
Sex
Female	92.5
Male	7.6
Age (years)	47.8 ± 10.6
Race/ethnicity
White	94.3
Black	3.8
Asian and white	1.9
Education
Associate’s degree	5.6
Bachelor’s degree (BA/BS)	37.7
Graduate degree	56.6
Marital status
Married	77.4
Single	9.4
Living with a partner	7.6
Divorced	5.7
Annual household income
<$10,000	3.7
$50,000–$74,999	18.9
$75,000–$99,999	22.6
$100,000–$149,999	35.9
>$150,000	9.4
Prefer not to answer	9.4
Total individuals in household	3.4 ± 1.4
Children <18 years in household
None	45.2
One	18.9
Two	22.6
Three	13.2

Note: In cases where percentages do not total 100, this is due to rounding.

COVID-19 Experiences

Reported experiences with COVID-19 are shown in Table 2. The most reported COVID-19 impacts were working from home more (54.7% of participants), working at home while a child or children were home with them (51.7% of participants who had children living in the household with them), and having a friend or acquaintance test positive for COVID-19 (43.4%). Additionally, 28.3% of participants reported working atypical hours (more or less hours than usual) and 18.9% had their pay reduced and/or could not afford food and/or gas. When responding to an open-ended question about other impacts of COVID-19, most participants listed at least one additional impact as shown in Table 2.

Table 2.

COVID-19 Related Experiences of Participants (January 2021; n = 53).

	Frequency (%)
Has anybody close to you tested positive for COVID-19 in the past month?
Yes, I have	3.8
Yes, at least 1 family member has	28.3
Yes, at least 1 friend/acquaintance has	43.4
Experienced any of the following as a result of COVID-19 in the past month?
Worked remotely or from home more than usual	54.7
Worked more hours than usual	17.0
Worked reduced hours	11.3
Was not able to work due to COVID-19 illness	5.7
Became unemployed	3.8
Difficulty arranging childcare	5.7 a
Worked with children at home	28.3 b
Income or pay reduced	13.2
Not enough money for gas	1.9
Not enough money for food	3.8

Note: Impacts that no participants endorsed were: increased costs for childcare, not paid at all, had serious financial problems, not enough money for rent, not enough money for medications, no regular place to sleep or stay. Responses to an open-ended question about other impacts of COVID-19, included having to shut down businesses, normal activities being canceled, anxiety and worry, limited opportunities for self and/or children to socialize, distance from family members, limited travel, frequent testing, less spending, being quarantined, children learning remotely, college-aged children at home, no time to themselves, stress, isolation, and limited shopping/dining out.

^a10.3% of those with children.

^b51.7% of those with children.

Variables of Interest

Means and standard deviations for key variables of interest are shown in Table 3. Participants engaged in sedentary non-enriching activities the most often (45.6 ± 8.5) and reported engagement in physical activities (14.9 ± 5.8) and outdoor activities (10.7 ± 6.8) the least frequently. On average, participants’ average stress scores were 16.3 ± 6.4 (moderately stressed). Forty-three percent of participants reported getting <7 hours of sleep, and 56.6% reported getting the recommended >7 hours.

Table 3.

Descriptive Statistics for Variables of Interest (n = 53).

Variable	Mean ± SD
Frequency of outdoor activities	10.7 ± 6.8
Frequency of sedentary enriching activities (n = 28; ex: reading)	32.5 ± 8.6
Frequency of sedentary non-enriching activities (n = 25; ex: watching TV)	45.6 ± 8.5
Frequency of physical activities (n = 20; ex: going for a run or jog)	14.9 ± 5.8
Frequency of social activities (n = 18; ex: going to a sports event)	21.5 ± 5.7
Perceived stress scale total score	16.3 ± 6.7

Note: The means and standard deviations recorded are reflective of what participants responded for the past month (30 days). Participants ranked how often they completed the activity types listed above on a scale of 0 (this has not happened in the past 30 days (0 times))–4 (this has happened very often in the past 30 days (more than 1 time per week)), such that the frequency ranges for each activity are equal to 0–4 times n. n is listed next to each activity type on the table, reflective of how many activities from the adapted Pleasant Activities List (n = 91) fell into each category. The exception here is the frequency of outdoor activities, which was self-reported as participants were asked to report where they completed an activity (indoors or outdoors) if they reported engagement in it, and thus has a maximum range of 0–4 times 91, but would vary between each participant. Sum scores for each activity type frequency were calculated by adding how often participants reported participating in each activity from each category, such that higher frequencies indicate more engagement in each activity type. For, the Perceived Stress Scale total score, participants rated how often they felt a certain way on a Likert scale of zero (never) to 4 (very often). A total score was calculated from the 10 items, with higher scores indicating greater stress. Unlike the other variables in this table, sleep was treated as categorical as compared to continuous, based on National Sleep Foundation recommendations for hours of sleep per night. ¹⁴

Associations Between Activity Types and Health Outcomes

Activity Types and Stress

Results from models examining links between activity types and stress are shown in Table 4. There was a trend-level relationship between the frequency participants engaged in outdoor activities in the last 30 days and their perceived stress levels, such that those doing outdoor activities more often reported less stress (P < .10). When restricting the outdoor activity variable to those activities done outdoors that were also physical, the inverse relationship with stress became significant (F = 6.61, P = .013). No other activity types (i.e., social, sedentary enriching, sedentary non-enriching) were linked with stress. Higher income and younger age were linked with more stress across models, with the specific relationships in the context of each model shown in Table 4.

Table 4.

General Linear Model Results for Each Activity Type and Stress.

	Estimate	Standard Error	F Value	P Value
Outdoor activities
Activity type frequency	–.208	.122	2.85	.098
Income	.942	.475	3.94	.053
Age	–.180	.078	5.16	.028
Sedentary enriching activities
Activity type frequency	.071	.102	.490	.489
Income	.979	.488	4.02	.051
Age	–.180	.083	4.67	.036
Sedentary non-enriching activities
Activity type frequency	–.044	.103	.180	.669
Income	.936	.488	3.68	.061
Age	–.155	.082	3.52	.067
Social activities
Activity type frequency	–.119	.149	.600	.441
Income	.992	.489	4.11	.048
Age	–.161	.080	4.07	.049
Physical activities
Activity type frequency	–.077	.148	.270	.606
Income	.991	.495	4.01	.051
Age	–.162	.080	4.11	.048

Note: Only covariates that were retained throughout the ANCOVA after backwards deletion are included.

Activity Types and Sleep

There was no relationship between any of the assessed activity types and dichotomized nightly sleep duration (Table 5).

Table 5.

Logistic Regression Results for Each Activity Type and Sleep.

Activity Type Frequency	Point Estimate	Standard Error	95% CI		P Value
			LL	UL
Outdoor	1.008	.041	.929	1.093	.849
Physical	1.008	.048	.917	1.108	.862
Sedentary enriching	1.017	.033	.954	1.084	.611
Sedentary non-enriching	1.019	.033	.955	1.088	.562
Social	1.054	.051	.954	1.164	.300

Note: Only covariates that were retained throughout the logistic regression after backwards deletion are included (none in this case). Abbreviations in the chart stand for the following: CI; confidence interval, LL; lower limit, UL; upper limit.

Discussion

The present analysis investigated whether the frequency of engagement in different types of activities (outdoors, sedentary enriching, sedentary non-enriching, physical, and social) predicted stress levels and nightly sleep duration among a convenience sample of adults about 1 year into the COVID-19 pandemic (January 2021). On average participants were moderately stressed, and about half were not meeting daily sleep recommendations of 7 hours per night. The most reported COVID-19 impacts were working from home more, working with children at home, and knowing someone who tested positive for COVID-19. More frequent outdoor activity was linked with less stress, and this inverse association became statistically significant when restricted to physical activities that were done outdoors. Younger age and higher income were also associated with more stress. None of the examined activity types predicted nightly sleep duration.

Our findings fit with prior findings that stress, sleep problems, and sedentary activity, as well as decreased physical activity, have characterized the COVID-19 era.^1,3 For example, in the present study, about half of participants reported receiving less than the recommended 7 hours of sleep per night, and sedentary activities were the most frequent activity, while physical and outdoor activities were the least frequent. Lower frequencies of physical and outdoor activities are notable, given prior demonstrated benefits of physical activity and outdoor time. The extant, largely pre-pandemic, literature has shown the greatest positive impacts on stress when doing physical activity outdoors, as compared to being physically active indoors and/or simply viewing images of nature.^17,18 The present findings fit with this pattern, as outdoor physical activity was the only significant predictor of stress across all activity types examined. This finding, coupled with pre-pandemic intervention research demonstrating positive effects of nature-based interventions, increases confidence in the encouragement of outdoor physical activity during the COVID-19 era. In contrast to prior findings, ¹¹ the present results did not support the idea that sedentary enriching activities like reading might have similar beneficial impacts on stress.

Along with the significant associations between outdoor physical activity and stress, two covariates demonstrated relevance in the models predicting stress: age and income. The observed link between younger age and greater stress aligned with prior findings that youth was associated with increased stress during the pandemic.^1,8 For the other covariate, which was linked with stress at a trend level, results were more surprising, as higher income is not typically associated with greater stress. ³⁰ Perhaps families with a higher income experienced greater changes from their usual routines or in resources typically available to them as a result of the pandemic.

Overall, the study sample was experiencing moderate levels of stress at the time of the study. Average perceived stress scores for our study (16.3 ± 6.7) were similar to results from surveys of adults from 41 countries during the pandemic, where the average stress score was also moderate at 17.4 ± 6.4. ¹ Considering that the vast majority of participants in the present study identified as female, results suggest that previously-observed links between outdoor physical activity and stress extend to this population, which is relevant given evidence of increased stress and unhealthy lifestyle behaviors among women during the COVID-19 pandemic. ⁹ Promoting outdoor physical activity may help mitigate impacts of the COVID-19 pandemic on stress and its negative health effects.

In addition to stress, we examined links between activity types and sleep duration, noting no link between frequency of any activity type and hours of sleep per night in this study. Given the literature supporting links between active engagement with nature and sleep quality and quantity, we expected associations between physical activity and/or outdoor activity and sleep duration. ¹⁴ While previous literature has reported that engagement in social activity can also improve sleep, for other activity types, research in this area is limited and our current findings underscore the need to further investigate this area. One reason we may not have observed any associations between activity types and sleep in this study is that there was not much variability in self-reported sleep duration in our sample, which could constrain the ability to detect a relationship.

It is encouraging that the present findings linking outdoor physical activity with less stress are consistent with prior research, given some differences between these studies. To our knowledge, the present analysis is the first of its kind, elucidating associations between different activity types and stress and sleep during COVID-19 using the Pleasant Activities List to collect data on and then aggregate across many specific activities. Our study does have several limitations, including the relatively small convenience sample. In addition, recruitment was primarily completed online using social media, which may have restricted the diversity of the sample and thus the generalizability of results. Likewise, because this was a secondary analysis of a larger intervention study, part of the eligibility criteria included adults who were interested in receiving suggestions for ways to get outside, stay active, and stay busy during COVID-19, which also may affect generalizability of results. This analysis was also cross-sectional, using baseline data before the implementation of any interventions, and thus conclusions cannot be definitively made about directionality or causality. Self-report survey measures were used to measure outcomes of interest, and thus responses may have been impacted by social desirability or other biases vs more objective measurements. Finally, although a precedent was followed in categorizing activities, another limitation of this present study was that not all activities fit perfectly within one of the activity categories. For example, text messaging family, friends, or acquaintances was considered to be a social activity in the present analysis, but an argument could be made for it to fall into a sedentary activity category. Future research could build upon the current study and address these limitations, such as by conducting longitudinal examinations of links between outdoor physical activity and various health outcomes in larger, more diverse samples.

This study was conducted during the winter, a time of year when it is difficult to get outside in cold-weather climates, which may have affected outdoor physical activity, stress levels, and the present results. Yet, the timing of the study, completed during the first winter of the COVID-19 pandemic (i.e., January 2021), is a time that has not been frequently studied and thus provides further insight on seasonal health behaviors that occurred during the pandemic. Furthermore, our study may have implications for interventions that could be developed during times that people tend to stay indoors and be less active, including the winter season. More research is needed to solidify conclusions about engagement in activities such as sedentary and social activities with stress and sleep; these variables were not linked in the present analysis but are understudied areas of research in literature both prior to and during the COVID-19 pandemic.

Implications for Practice and/or Policy and Research

In summary, about one year into the COVID-19 pandemic, adults in this sample were moderately stressed, and about half did not meet sleep recommendations. Results suggest that engaging in outdoor activity may reduce stress, and this may be particularly true for physical activity done outdoors, consistent with intervention research conducted prior to the pandemic. Interventions that promote outdoor physical activity during times these activities may be in decline should be considered for future studies. Results of our study are consistent with previous observations of heightened stress among younger women during the pandemic, which is relevant because it provides insight as to groups who would most likely benefit from such interventions going forward. Hiking and nature walks offer opportunities to engage in outdoor physical activity and could be a useful intervention during times people may otherwise stay indoors, such as pandemics or the winter season in cold-weather climates. Future research could examine if results from the present analysis are consistent in longitudinal research with larger, more diverse samples and leverage these findings in future intervention research promoting access to safe, accessible outdoor physical activity as a health promotion strategy.

ORCID iD

Callista Zayatz https://orcid.org/0000-0002-6037-6611