The mediating effect of meditation and physical activity behaviors on the associations of COVID-19 related worry, attention to news, and stress with mental health in mobile app users in the United States: Cross-sectional survey

Background: The Coronavirus Disease 2019 (COVID-19) pandemic has been declared an international public health emergency with potentially long-lasting effects on mental health. There is a need to identify effective strategies to buffer the negative mental health impact of COVID-19. Objective: To examine the regional differences in mental health and perceptions of COVID-19 in light of the state-level prevalence of COVID-19 cases, estimate the associations between perceptions of COVID-19 and health behavior engagement (i.e., physical activity, mindfulness meditation) and mental health, and explore the mediating effect of health behavior engagement on the relationship between perceptions of COVID-19 and mental health. Methods: A cross-sectional survey was distributed to a sample of US adult paying subscribers to the Calm app (data collection April 22 to June 3, 2020). The survey assessed perceptions of COVID-19, health behavior engagement, and mental health (i.e., perceived stress, post-traumatic stress disorder, and anxiety and depression). Statistical analyses were performed using R software. Differences in perceptions of COVID-19 and mental health by location were assessed using t-tests and chi-square tests. Logistic and OLS models regressed mental health and health behavior on COVID-19 perceptions, and Causal Mediation Analysis was used to estimate the significance of mediation effects. Results: The median age (N=8392) was 47 (SD=13.8) years. Mid-Atlantic region (New Jersey, New York, Pennsylvania) participants reported higher levels of stress, more severe depression symptoms, were more worried about COVID-19, paid more attention to COVID-19 news, and had more stress-related to social distancing recommendations than participants living in other regions. The relationship between worry about COVID-19 and perceived stress was significantly mediated by changes in physical activity (P<.001) and strength of meditation habit (P<.001). The relationship between worry about COVID-19 and PTSD symptoms was significantly mediated by changes in physical activity (P<.001) and strength of meditation habit (P<.001). Conclusions: Our findings describe the mental health impact of COVID-19 and outline how continued participation in health behaviors such as physical activity and mindfulness meditation may buffer against worsening mental health due to the COVID-19 pandemic. These data have important implications for public health agencies and health organizations to promote the maintenance of health habits to reduce the residual mental health burden of the COVID-19 pandemic. (JMIR Preprints 04/03/2021:28479) DOI: https://doi.org/10.2196/preprints.28479 Preprint Settings 1) Would you like to publish your submitted manuscript as preprint? https://preprints.jmir.org/preprint/28479 [unpublished, peer-reviewed preprint] JMIR Preprints Green et al Please make my preprint PDF available to anyone at any time (recommended). Please make my preprint PDF available only to logged-in users; I understand that my title and abstract will remain visible to all users. Only make the preprint title and abstract visible. No, I do not wish to publish my submitted manuscript as a preprint. 2) If accepted for publication in a JMIR journal, would you like the PDF to be visible to the public? Yes, please make my accepted manuscript PDF available to anyone at any time (Recommended). Yes, but please make my accepted manuscript PDF available only to logged-in users; I understand that the title and abstract will remain visible to all users (see Important note, above). I also understand that if I later pay to participate in <a href="https://jmir.zendesk.com/hc/en-us/articles/360008899632-What-is-the-PubMed-Now-ahead-of-print-option-when-I-pay-the-APF-" target="_blank">JMIR’s PubMed Now! service</a> service, my accepted manuscript PDF will automatically be made openly available. Yes, but only make the title and abstract visible (see Important note, above). I understand that if I later pay to participate in <a href="https://jmir.zendesk.com/hc/en-us/articles/360008899632-What-is-the-PubMed-Now-ahead-of-print-option-when-I-pay-the-APF-" target="_blank">JMIR’s PubMed Now! service</a> service, my accepted manuscript PDF will automatically be made openly available. https://preprints.jmir.org/preprint/28479 [unpublished, peer-reviewed preprint] JMIR Preprints Green et al


Table of Contents
The mediating effect of meditation and physical activity behaviors on the associations of COVID-19 related worry, attention to news, and stress with mental health in mobile app users in the United States: Cross-sectional survey

Original Manuscript
international public health emergency with potentially long-lasting effects on mental health. There is a need to identify effective health behaviors to mitigate the negative mental health impact of COVID-

19.
Objective: The objectives were to: (1) examine the regional differences in mental health and COVID-19 related worry, attention to news, and stress, in light of the state-level prevalence of COVID-19 cases, (2) estimate the associations between COVID-19 related worry, attention to news, and stress and health behavior engagement (i.e., physical activity, mindfulness meditation) and mental health, and (3) explore the mediating effect of health behavior engagement on the associations between COVID-19 related worry, attention to news, and stress with mental health.

Methods:
A cross-sectional survey was distributed to a sample of US adult paying subscribers to the Calm app (data collection April 22 to June 3, 2020). The survey assessed COVID-19 related worry, attention to news, and stress, health behavior engagement, and mental health (i.e., perceived stress, post-traumatic stress disorder, and anxiety and depression). Statistical analyses were performed using R software. Differences in COVID-19 related worry, attention to news, and stress and mental health by location were assessed using t-tests and chi-square tests. Logistic and OLS models regressed mental health and health behavior on COVID-19 related worry, attention to news, and stress, and Causal Mediation Analysis was used to estimate the significance of mediation effects.

Results:
The median age (N=8392) was 47 (SD=13.8) years. Mid-Atlantic region (New Jersey, New York, Pennsylvania) participants reported higher levels of stress, more severe depression symptoms, were more worried about COVID-19, paid more attention to COVID-19 news, and had more stress-related to social distancing recommendations than participants living in other regions.
The association between worry about COVID-19 and perceived stress was significantly mediated by

Introduction
In January 2020, the World Health Organization declared the novel Coronavirus Disease 2019 (COVID-19) an international public health emergency, [1] and the negative mental health effects of the ongoing COVID-19 pandemic are expected to be a significant, long-lasting global health problem. [2,3] In an April 2020 review of the existing literature on COVID-19 and mental health, moderate to severe levels of depressive symptoms and anxiety were reported in 16-28% of the general population and medical staff in response to the COVID-19 pandemic. [4] Additionally, fear and worry about COVID-19 is common [5] with many citing worries related to personal infection or the infection of family members, [6,7] an overrun health care system, financial losses without recovery anytime soon, [8,9] and long-lasting isolation and movement restrictions. [10] There is previous data to suggest regional differences across the US in COVID- 19  Services, many respondents were concerned about how they would access mental health support, as many previously available in-person services have been discontinued as a result of COVID-19.
[11] Stakeholders were also concerned with the capacity to handle the increased demand for mental health services and the lack of emphasis on mental health as compared to treatment of COVID-19 and its physical health impacts. [11] Importantly, self-management strategies including digital approaches to improve mental health have become an area of interest for policymakers, as many individuals fail (or are unable) to participate due to the pandemic (e.g., facility closures, reduced client load, social distancing). [12] Self-management strategies may also empower individuals by taking a more active role in their healthcare (i.e., recognizing and managing their own health problems). [13] Additionally, self-management strategies are cost-effective and can be utilized as a preventative tool (rather than prescriptive or treatment focused) that may mitigate more debilitating mental health issues from developing. [12] Common and evidence-based self-management strategies for improving mental health are physical activity and mindfulness meditation, [14,15,16] both of which can still be maintained during social distancing policies and stay-at-home orders, and can be adapted and delivered digitally. However, it is unknown the extent to which individuals have maintained their participation in these self-management strategies during the COVID-19 pandemic.
Physical activity has been widely adopted as a beneficial way to self-manage mental health and may attenuate the mental health decline resulting from COVID-19. [17,18] Physical activity has been shown to be as effective as antidepressants in decreasing stress, improving mood, and enhancing self-esteem. [17,18] Despite the known benefits of physical activity, currently more than 60% of US adults do not engage in the recommended amount of physical activity (i.e., 150 minutes of moderate to vigorous physical activity per week) and 25% are not active at all (i.e., sedentary).
[19] Social distancing, quarantine/social isolation, and closing public spaces due to COVID-19 are likely to increase these rates of physical inactivity, [20,21] and the degree to which adults change the type or duration of their physical activity during COVID-19 is also unknown. Thus, there is a need to determine how self-management strategies such as physical activity have changed as a result of the pandemic, and specifically if reductions in physical activity are associated with worsening mental health status.
Mindfulness meditation, another self-management strategy, has also been evidenced to improve mental health and maintenance (or habituation) of this behavior, particularly during COVID-19, may reduce worsening of mental health overtime. Evidence suggests that mindfulness meditation reduces stress, improves mood (e.g., symptoms of anxiety and depression), and enhances well-being. [14,22] Though an optimal amount of mindfulness meditation has not been established, evidence suggests more benefits accrue with greater frequency of practice. [23] The habituation of mindfulness meditation practice (i.e., behavioral automaticity) [24,25] has yet to been examined in the literature [26] but may have important implications for whether individuals maintain their practice during the COVID-19 pandemic. Those with stronger meditation habits may be more likely to continue meditation practice even when daily life is disrupted, as habits are known to persist when motivation declines and other distractions are present. [25] Mindfulness meditation delivered via mHealth may be another useful way to help people maintain their practice, especially when many public facilities are closed or limit participation, [21,22,23] and popular meditation apps such as Calm or Headspace have shown promise to reduce stress and improve health. [21,24,25,26,27,28] Interestingly, Calm's sales in April 2020 were 62% higher than February 2020 and 32% higher than March 2020 (unpublished sales data provided by internal Calm team), indicating that more people are accessing this type of self-management strategy in response to COVID-19. Though mHealth meditation apps have promise to help people self-manage mental health, there is a lack of understanding about how continued participation in meditation is impacted during COVID-19 and if meditation habits are associated with improved mental health.
Given the benefits of physical activity and mindfulness meditation on mental health, the purpose of this cross-sectional survey (data collected between April 22 nd to June 3 rd 2020) was to 1) examine the differences in COVID-19 related worry, attention to news, and stress from social distancing and mental health (i.e., stress, PTSD, depression, and anxiety) by region of the US, 2) explore the associations between COVID-19 related worry, attention to news, and stress and health behavior engagement (i.e., strength of meditation habit, and changes in mindfulness meditation and physical activity) and mental health, and 3) estimate the mediating effect of health behavior engagement on the associations between COVID-19 related worry, attention to news, and stress with mental health. We hypothesized that 1) there would be differences in COVID-19 related worry, attention to news, and stress and mental health in higher prevalence COVID-19 states, 2) greater COVID-19 related worry, attention to news, and stress would be associated with lower health behavior engagement and greater levels of mental health, 3) health behavior engagement would mediate the associations between COVID-19 related worry, attention to news, and stress with mental health.

Ethics Approval
The Institutional Review Board at Arizona State University (STUDY00011867) approved the study. All participants provided electronic consent before participating in the study. The datasets generated and analyzed during the study are available at https://osf.io/dsgrv/.

Study Design
This is a descriptive and national longitudinal study, using a non-random convenience sample of adult paying subscribers to the mindfulness meditation mobile app, Calm. Participants of this study initially completed a cross-sectional, baseline survey called the "COVID-19 Health and Well-being Survey" and agreed to complete four follow-up surveys over the subsequent 12 months.
The data presented in this paper come from the baseline survey administered from April 22 nd to June 3 rd , 2020.

Participant Recruitment and Selection
Emails inviting Calm subscribers to participate in the study were sent on April 22 nd , April 29 th , and May 6 th , 2020. Subscribers were eligible if they had opened an email from Calm at least once in the last 90 days and used Calm at least once in the last 90 days, were 18 years or older, able to read and understand English, and resided in the US.

Procedures
Interested individuals were directed to a Qualtrics eligibility screener (~1 minute to complete). Once eligibility was determined, participants completed an electronic informed consent form and the baseline survey. There were no incentives for participation in the first wave (baseline) of the study, but incentives were offered for continued participation in subsequent waves of this study (i.e., random drawing for 20 $50 Amazon gift cards at months 2-4 and 50 $50 Amazon gift cards at month 12).

COVID-19 Health and Well-being Survey
The baseline survey was also administered using Qualtrics and included both investigatordeveloped and validated questionnaires. The investigator-developed portion of the survey included a total of 15 questions related to COVID-19 worry, attention to news, and stress from social distancing and 20 questions related to health behavior engagement (i.e., strength of meditation habit, and changes in mindfulness meditation and physical activity; see Table 1 for questions used in current analyses

COVID-19 related worry, attention to news, and stress from social distancing
Participants were asked how worried they were about: personally getting COVID-19, a family member getting COVID-19, and the spread of COVID-19 in their area (See Table 1). Worry about COVID-19 was operationalized as the sum of responses to these three questions. Scores ranged from 3-15 where higher scores indicate greater levels of worry about COVID-19. Participants were also asked how they would rate their attentiveness to information about ongoing changes and news regarding COVID-19 (See Table 1). Scores at or below the median were used to identify individuals with low attention to COVID-19, while scores above the median were categorized as high attention to COVID-19. Finally, participants were asked if recommendations for socially distancing have caused them stress (See Table 1). Scores at or below the median were used to identify participants with low stress, scores above the median indicated high stress.

Mindfulness Meditation Practice
Participants were asked about their meditation practice using Calm prior to COVID-19 as well as their current use of Calm. If participants indicated that they meditated using Calm prior to COVID-19, they were asked to what extent has COVID-19 changed their meditation practice (See Table 1). Participants who indicated they no longer participate in meditation were categorized as "stopped meditation," while all other participants were considered as "continuing meditation".

Physical Activity Behavior
Participants were asked to select how many days/week (scale of 0-7) of physical activity they participated in prior to COVID-19 as well as their current participation. Participants were also asked to what extent has COVID-19 changed the frequency and/or duration of physical activity (See Table 1). Changes in physical activity were calculated as the difference between the participants' reported number of days of physical activity currently and prior to COVID-19.

Self-report Habit Index (SRHI)
The SRHI includes 12 items reflecting on three proposed characteristics of habit (i.e., automaticity, frequency, and relevance to self-identity). Response options range from 1= strongly disagree to 5=strongly agree. The SRHI is a reliable and valid measure that has demonstrated Cronbach's α=.89-.92. Items are summed to produce a total score with higher values indicating stronger habits. [40] For the current study, only the four-item automaticity subscale of the SRHI was used in the analyses and produced a Cronbach's α=.88.

Perceived Stress Scale (PSS)
The PSS includes 10 items that measure the degree of self-appraised stress in one's life within the past month. [29,30] Response items are a 5-point Likert scale from 0 = "never" to 4 = "very often." Items are summed to produce a total score from 0-40 with higher scores indicating higher levels of perceived stress. The PSS is a reliable and valid measure that has demonstrated good internal consistency (Cronbach's α=.74-.91) [43]. For the current study, the Cronbach's α=.89.

Impact of Events Scale-6 (IES-6)
The IES-6 is a six-item abbreviated version of the IES-R (22-items) that assesses PTSD symptoms over the past seven days. [44] Response items are a 5-point Likert scale from 0 = "not at all" to 4 = "extremely." The score is calculated as a mean of the six items. Scores range from 0-5 with a binary cut off score of 1.75 indicating clinically important PTSD symptoms. The IES-6 is a valid and reliable measure with excellent internal consistency α=0.91. For the current study, the Cronbach's α=.86.

Hospital Anxiety and Depression Scale (HADS)
The HADS is a 14-item scale measuring levels of anxiety and depression.

Statistical Analysis
All statistical analyses were performed using R software (version 4.0.0). [

Sample Characteristics
The sample (N=8392) was primarily White, non-Hispanic, and female (see Table 2). The majority of participants had Bachelor-or graduate-level degree, were employed, and had an annual household income exceeding $100,000. Approximately one third of participants reported having a least one medical condition associated with increased risk of severe illness from COVID-19, but over 80% perceived themselves to be in good overall health. The self-reported changes in preventative health behaviors are shown in Table 3, where most participants are found to have increased or maintained their physical activity and meditation habits during the initial period of the COVID-19 pandemic. There was a significant correlation between strength of meditation habit and changes in meditation during COVID-19, such that those with the strongest habits were the most likely to increase or maintain their meditation practices (r=.37, P<.001).

Associations between health behavior engagement and COVID-19 related worry, attention to news, and stress
Participants who were more worried about COVID-19, paid more attention to COVID-19 news and updates, and experienced more stress due to COVID-19 social distancing recommendations had greater decreases in physical activity and lower strength of meditation habit (see Table 6). Attention to news and updates about COVID-19 and stress due to social distancing recommendations was also associated with stopping meditation.
Men, White and non-Hispanic respondents, respondents with higher levels of education and higher household incomes were also less likely to decrease their engagement in physical activity.
Strength of meditation habit was also generally stronger among respondents who were older, White, non-Hispanic, women, more educated, and had higher annual household incomes. Younger participants and men were more likely to report that since COVID-19, they had stopped meditating.
Living in a state with high COVID-19 prevalence was associated with decreases in physical activity and lower strength of meditation habit, but not with stopping their meditation practice. Table 6. Association between health behavior engagement and COVID-19 related worry, attention to news, and stress Note. This table presents the coefficients from OLS regression models for the continuous outcomes, changes in physical activity and meditation habit strength, on COVID related worry, attention to news, and stress, and logistic regression models for the stopped meditation outcome on COVID related worry, attention to news, and stress. Standard errors are in parentheses and are estimated using heteroscedasticity-robust procedures. The covariates listed in the first panel (age, racial minority status, female, Hispanic, high-school education only, undergraduate education, income <$80,000, income $81,000-$100,000, unemployed, underlying medical condition, and living in state with high COVID-19 prevalence) were included in all models. State-levelCOVID-19 prevalence is based on the number of COVID-19 cases per 100,000 at the time of the first survey distribution (April 22, 2020); high-prevalence states were California, Colorado, Illinois, Massachusetts, New Jersey, New York, Washington.

Perceived stress and COVID-19 related worry, attention to news, and stress via engagement in health behaviors
The mediating effect of health behavior changes on the associations between COVID-19 related worry, attention to news, and stress from social distancing with stress is demonstrated by the regression analyses presented in Table 7. The first row and first column of Table 7 Table 7). The second panel of Table 7 presents a similar mediation analysis for the association between attention to COVID-19 news and updates and perceived stress. The bootstrapped standard error calculations found that changes in physical activity (P=.02) and stopping meditation (P=.002) partially mediated the association between attention to COVID-19 news and updates and perceived stress, but strength of meditation habit was not a significant mediator (P=.11). Finally, the association between perceived stress and stress due to COVID-19 social distancing recommendations was significantly mediated by changes in physical activity (P=.01), stopping meditation (P=.02), and strength of meditation habit (P<.001).  Note. This table presents the coefficients from OLS regression models of PSS stress score on COVID-19 worry, COVID-19 attention, and Stress from social distancing as well as the indicated health behavior changes. Age, male, Hispanic, income <$80,000, income $81,000-$100,000, unemployed, and underlying medical condition were included as covariates in all models, which also estimate heteroscedasticity-robust standard errors.

PTSD symptoms and COVID-19 related worry, attention to news, and stress via engagement in health behaviors
The mediating effect of health behavior changes on the associations between COVID-19 related worry, attention to news, and stress with PTSD symptoms is outlined by the regression analyses presented in Table 8. The first row and first column of Table 8 Table 8). The second panel of Table 8    Note. This table presents the coefficients from OLS regression models of PTSD score on COVID-19 related worry, attention to news, and stress, COVID-Age, male, Hispanic, income <$80,000, income $81,000-$100,000, unemployed, and underlying medical condition were included as covariates in all models.

Discussion
Our findings describe the mental health impact of COVID-19 and outline how continued participation in health behaviors such as physical activity and mindfulness meditation reduce worsening mental health due to the COVID-19 pandemic. The aim of this baseline survey was to first examine the regional differences in mental health and COVID-19 related worry, attention to news, and stress in light of the state-level prevalence of COVID-19 infections at the time of this survey. We additionally sought to estimate the associations between COVID-19 related worry, attention to news, and stress, health behavior engagement, and mental health, as well as explore the mediating effect of health behavior engagement on the associations between COVID-19 related worry, attention to news, and stress with mental health.

Mental health and COVID-19 related worry, attention to news, and stress by region
Our findings indicate that participants living in the Mid-Atlantic region (i.e., New Jersey, New shown that greater concern about COVID-19 was associated with greater anxiety and depression levels [54] and decreased physical activity and increased sedentary time during COVID-19 was associated with poor mental health (i.e., higher stress, anxiety, depression symptoms). [35,41] Additionally, both physical activity and meditation have been recommended as a healthy way to cope with stress during COVID-19 underscoring the importance of maintaining these behaviors during a time of heightened stress. [56] Our data demonstrate that participation in physical activity and meditation are important mechanisms in reducing the increased mental health problems associated with COVID-19 related worry. Other studies have shown that both mindfulness-based and physical activity interventions may be protective against the development of trauma-related psychopathology (e.g., PTSD) by enhancing cognitive function, reducing arousal, normalizing hypothalamic pituitary axis (HPA) function, and/or reducing inflammatory markers. [57,58] There is a need for more research, especially longitudinal data, to examine the long-term changes in health behaviors due to COVID-19, to better disentangle the causal relationships between these psychological and behavioral outcomes, and identify strategies for helping people maintain health promoting behaviors.
In summary, our findings support existing evidence of the beneficial health effects of physical activity and meditation on mental health outcomes. [16,[59][60][61] Because the COVID-19 pandemic has profoundly impacted daily routines (e.g., social distancing, quarantine, businesses closures) and may negatively impact performance of health behaviors, it is important to continue promoting self-management of health behaviors such as physical activity and meditation that can reduce worsening mental health during the COVID-19 pandemic, particularly in regions with a heightened sense of worry about COVID-19. Encouraging participation in physical activity and meditation should be an important public health objective during the current COVID-19 pandemic, especially because reduced physical and poor mental health has been shown to make people more susceptible to COVID-19 infection and disease transmission. [45,46] Public health agencies may consider providing strategies to help people maintain or adapt their current health behaviors. For example, digital or mHealth interventions for both physical activity and meditation have shown promise for their feasibility, scalability, and physical and mental health benefits. [63][64][65][66] Digital and mHealth interventions are also convenient and often budget-friendly ways to encourage participation in both physical activity and meditation, and more research is needed to better understand their efficacy and applicability during the COVID-19 pandemic.

Limitations
Despite this study being one of the first to describe the associations between COVID-19 related worry, attention to news, and stress, mental health, and self-management health behaviors, there are important limitations to be noted. First, our sample was primarily female, non-Hispanic, White, high-income, highly educated, and were paid subscribers of Calm, which limits the generalizability of these data. Second, this survey was cross-sectional and therefore causal relationships cannot be determined from these analyses. The broader study plans to implement four more surveys over the next 12 months to provide a more comprehensive longitudinal assessment of the impact of COVID-19 on mental health and health behaviors.

Conclusions
Our findings underscore the importance of maintaining self-management health behaviors such as physical activity and meditation for sustaining one's mental health during the COVID-19 pandemic. These results suggest that public health agencies and health organizations should promote the maintenance of health habits with strategies such as digital and mHealth approaches that can be more easily adapted during stay-at-home orders and/or social distancing mandates.
Future research is needed to identify the causal relationships between these psychological and behavioral outcomes and to evaluate strategies for helping people adapt their current meditation and physical activity practices for the restrictions on daily life imposed by COVID-19 public health policies.