Introduction

According to the World Health Organization (WHO), nearly 36.3 million people are currently confirmed to have Novel Coronavirus (2019-nCoV, or COVID-19) at the time of writing in October 2020. COVID-19, which is deceptively similar to pneumonia and influenza, has impacted healthcare, shifted the political climate, and created a financial crisis. New areas of the world report cases daily, some level of quarantine or social distancing is recommended (if not required) where the virus is present, and on March 11, 2020 the WHO officially identified COVID-19 as a pandemic. Social distancing efforts seek to decrease the number of people in contact with one another with the goal of minimizing the spread of infection. As a more drastic measure, quarantine involves complete isolation, with recommendations directing people testing positive or showing symptoms of COVID-19 to quarantine for 14 days. The present investigation seeks to identify effective strategies for managing distress during this stressful period of worldwide uncertainty, particularly among those facing the psychological challenges of quarantine conditions in Wuhan, China and other affected Chinese cities in February 2020.

Method

Participants and procedure

This study (#NJUPSY202002002) was approved by the Ethics Committee of Nanjing University, and consent was obtained via an online document (participants clicked "continue" if they consented to participate in the survey). Participants in China (N = 5115; see Table 1 for demographic information) were recruited online between February 12–19, 2020. We recruited as many participants as possible during the relevant period of time. We did not conduct a power analysis; however, our sample size is more than sufficient for the relevant analyses at nearly any effect size.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Demographic information.

https://doi.org/10.1371/journal.pone.0242043.t001

The online survey was hosted on IQEQ (Intelligence Quotient and Emotional Quotient), a dedicated online research platform with a mobile-friendly interface for smartphones and tablets, which was developed by one of the co-authors (R.Z.) at Nanjin University. Participant recruitment was targeted at college students via social media (WeChat platform). No course credit or monetary reward was offered for participants. The survey took participants an average of 9 minutes to complete.

Results

We tested our hypotheses with multiple regression analyses predicting health and well-being from quarantine length, flow, mindfulness, and the interactions between quarantine length and flow/mindfulness. All variables were grand-mean centered. We controlled for gender, age, education, region, sibling status, family income, dispositional optimism, intolerance of uncertainty, and satisfaction with life in all analyses. Given the large sample size (N = 5115), we define statistical significance for our purposes as effects at p < .01. Table 2 presents bivariate correlations between our variables.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Bivariate correlations.

https://doi.org/10.1371/journal.pone.0242043.t002

Table 3 presents the full regression results for each measure of health and well-being. Note that quarantine length was a significant predictor of all but one measure (positive emotion). In all but one case (healthy behavior), a longer quarantine was associated with poorer well-being.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Keys results of multiple regression models.

https://doi.org/10.1371/journal.pone.0242043.t003

Flow

Looking first at flow’s overall association with well-being, the results were mixed (see Table 3 for key results; full results available on the Open Science Framework along with study materials). Participants who reported greater flow also reported more positive emotion, less severe depressive symptoms, less loneliness, more healthy behaviors, and fewer unhealthy behaviors. However, interactions with quarantine length were more consistent. With the exception of positive emotion and healthy behaviors, flow significantly moderated the relationship between quarantine length and all other measures of well-being. Table 4 presents the associations between quarantine length and well-being at +/-1 SD for flow. As hypothesized, these associations are quite robust for those who reported relatively little flow, whereas they weakened considerable (nearly disappearing in some cases) among those who reported relatively high levels of flow.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Associations between quarantine length and well-being, modeled at +/-1 SD for flow.

https://doi.org/10.1371/journal.pone.0242043.t004

Discussion

This study provides a unique opportunity to understand the experiences of people at ground zero for a worldwide pandemic, nearly in real-time. As we write this paper in October 2020, COVID-19 has been nearly eradicated in China; however, the number of new cases in the U.S. continues to rise rapidly, and millions of people worldwide are currently in or will soon enter (or re-enter) some form of quarantine or isolation. As such, our findings provide timely guidance to those experiencing distress associated with quarantine conditions.

Specifically, the findings reveal three key insights about this unique and stressful experience, which may generalize to similar periods of open-ended uncertainty. First, people who had endured a longer quarantine reported poorer well-being by many definitions: more worry, more negative emotions, more depressive and anxious symptoms, more loneliness, and more unhealthy behaviors. Second, experiencing flow and mindfulness was associated with better well-being during the stressful circumstances in which people found themselves in China in February 2020, regardless of quarantine status or length. Comparing effect sizes in Table 1, flow seemed particularly associated with reduced loneliness, engagement in healthy behaviors, and avoidance of unhealthy behaviors, whereas mindfulness was more associated with fewer negative emotions and anxious symptoms—although mindfulness was also associated with greater loneliness and more unhealthy behavior. Flow and mindfulness were approximately equally associated with elevated positive emotions and reduced depressive symptoms.

Third, and perhaps most novel, flow—but not mindfulness—moderated the link between quarantine length and well-being deficits. In fact, people in a lengthy quarantine who reported higher-than-average flow experiences in the past week (i.e., +1 SD) were no worse off than people who were not yet quarantined on many measures of well-being, statistically speaking (see Table 2). Length of quarantine remained significantly associated with worry and loneliness among those high in flow (albeit a far weaker association compared to those low in flow), which suggests the possibility that worry and loneliness may be particularly stubborn forms of distress during quarantine. In contrast, mindfulness did not moderate any link between quarantine length or status and well-being.

Why did flow so consistently moderate what appear to be deleterious effects of quarantine length—and equally important, why did mindfulness not? In short, flow is a form of exquisite distraction, perfect for situations in which the primary goal is to make time pass more quickly [12, 46]. As a quarantine wears on, people may find that the tedium of isolation allows their worries to run wild, with little else to keep their mind occupied. If instead people can find activities that absorb their attention, the days feel shorter and the weeks, therefore, more tolerable. Of course, it is likely that not all forms of distraction are so consistently beneficial to well-being; for example, if a problem needs to be addressed or emotions processed, distraction may be detrimental to well-being in the long-run. However, flow is a particularly beneficial form of distraction, perhaps because it entails active engagement rather than avoidance or suppression.

Mindfulness, on the other hand, draws attention toward one’s present circumstances and internal states. As in daily life, our findings suggest that mindfulness is good for reducing distress in a given moment or on a given day during a widespread health crisis. However, mindfulness is nearly antithetical to distraction (in fact, a reverse-coded item on the scale used in this study is “I have been easily distracted”), and flow-inducing distractions may be the very best medicine to bolster well-being during a lengthy quarantine.

We would note that although we asked participants to reflect on their experiences of mindfulness and flow during the previous week, the measure likely tapped into both state and trait aspects of these variables. That is, a person who reported a high degree of mindfulness during the past week is likely also someone who is dispositionally mindful. We attempted to mitigate the influence of dispositional characteristics by controlling for conceptually-related individual differences, namely dispositional optimism, intolerance of uncertainty, and satisfaction with life. Nonetheless, our findings may also point to well-being advantages among those naturally disposed toward mindfulness and flow.

Although our study had a number of strengths, most notably the large sample size and the timely, highly significant real-world context, several weaknesses are worth noting. The study was cross-sectional, thus rendering our data incapable of testing causal hypotheses. We argue that our interpretation of the direction of effects is the most parsimonious, given our efforts to control for likely third variables and the well-established causal links between flow/mindfulness and well-being in other contexts. However, it is possible that some relationships are reciprocal, such that experiencing distress disrupts people’s ability to find flow and remain mindful, or that we failed to assess relevant third variables. Similarly, it is possible that people who were more distressed opted into quarantine (i.e., reverse causality) rather than quarantine conditions causing deterioration of well-being. We propose that the most parsimonious explanation is a deleterious effect of isolation and restricted movement, particularly given other research that found such effects in other pandemics [e.g., 2, 4] and in the current pandemic [in France; 5] and given the relatively strict policies regarding quarantine during that period in China, quite different from the United States and other countries with more lenient policies. That said, quarantine status was almost certainly a function of where our participants lived within China, thus introducing the possibility of third-variable explanations in those associations as well.

In addition, we would note that a large majority of our sample was not (yet) in quarantine at the time they completed the survey, although presumably all of our participants were keenly aware of the growing health crisis in their country and experienced significant disruptions to their lives. Ideally, we would have followed participants longitudinally as they passed in and out of quarantine restrictions, or at least had more even coverage across various lengths of quarantine. Furthermore, our sample was skewed toward relatively educated participants (nearly all had a Bachelor’s degree), given the nature of the data collection strategy.

Another unavoidable feature of our sample was the relatively short average length of quarantine (weeks rather than months). Our study cannot address the usefulness of flow and mindfulness across longer periods of distancing and isolation like those that are now the norm in the United States and other countries. On the other hand, our study is well-suited to address a particular type of isolation that remains common, namely the more extreme quarantine recommendations following potential exposure to someone with COVID-19 or a positive test for the virus. Our findings may be most helpful to people in these relatively short, time-limited, and highly restrictive conditions.

Finally, we were limited in the length of our survey by practical constraints, and thus we do not have information about the activities that most effectively or consistently promoted flow states. Other studies have found that although people seem capable of reporting on their flow states, they are less successful at identifying activities that are particularly flow-inducing [12]. Although flow activities are highly idiosyncratic, activities that provide the right level of challenge (not too hard, not too easy) and offer feedback on one’s progress are most reliable at creating flow states [6]. A key direction for future research is to more clearly identify the types of activities that work best for the most people, as well as examining the importance of person-activity fit, during periods of quarantine-like conditions.

Despite these limitations, this study offers a rare glimpse inside a community at the initial epicenter of a once-in-a-century pandemic. At the time the data were collected, Wuhan was unique in its strict restrictions and widespread lockdown. In the relatively brief time since then, the crisis has grown and spread exponentially, such that people across the world are experiencing conditions like those of our participants. Our findings provide a glimmer of hope for people struggling with the challenges of social distancing or even complete quarantine: The time may pass more quickly if you find your flow.