Introduction

In early December, 2019, Coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-Cov-2) occurred and has now rapidly spread around the world. As of July 17, 2020, the COVID-19 outbreak has resulted in 13 million confirmed cases including with 58,5727 deaths globally [1]. A range of public health interventions including traffic restriction, social distancing, home confinement and centralized quarantine, and improvement of medical supplies have enormously contributed to the quick containment of the epidemic in China and set an encouraging example for other countries being affected [2]. However, quarantine and isolation measures urgently adopted to control the pandemic might potentially have negative psychological and social effects especially on those most vulnerable, such as frontline medical workers, children, and older adults [3,4,5]. Most of the anticipated direct consequences of quarantine and associated social and physical distancing, including financial insecurity, boredom, frustration, feeling a burden, loneliness, and fear, are risk factors for mental health issues including anxiety, depression, suicide, and self-harm [3]. In circumstances such as these, the caution and actions about protecting the mental health and boosting psychological wellbeing of population placed under quarantine are warranted [6].

Several cross-sectional studies have reported a high prevalence of symptoms of psychological distress during the outbreak, estimating that nearly half of health care workers exposed to COVID-19 experienced symptoms of depression and anxiety [7], and 20% of students restricted to home in Wuhan reported having anxiety and depressive symptoms [8]. Previous evidence also suggested that quarantine measures used in the outbreak of severe acute respiratory syndrome (SARS) and H1N1 influenza pandemic are associated with increased risk of psychological outcomes [9,10,11] and may have long-term consequences [3, 12]. However, to our knowledge, no study has yet investigated the association of quarantine measures with the mental health status among general population during the pandemic. Research is needed to ascertain the psychological effect of quarantine and identify factors associated with mental health outcomes among population quarantined to further inform interventions of mitigating mental health risk especially for vulnerable groups under pandemic conditions.

In this study, rates of depression, anxiety, insomnia, and acute stress symptoms were reported and compared for population quarantined with those not quarantined to evaluate the associations of quarantine measures and mental health outcomes during the outbreak of COVID-19, and factors associated with psychological symptoms among population quarantined were identified, which could serve as evidence base for policy makers to carefully weight against the potential psychological risks when develop protocols and implement quarantine, and to support psychological wellbeing especially in vulnerable groups.

Methods

Participants

The study was approved by the ethics committee of Peking University Sixth Hospital (Institute of Mental Health). Written informed consent was received online before the respondents began the questionnaire. This study follows the American Association for Public Opinion Research reporting guideline.

This is a cross-sectional, nationwide study conducted via an online survey from February 28 to March 11, 2020. During this period, following massive city lockdown with traffic restriction and quarantine implemented since early February, the COVID-19 outbreak in China was temporally controlled. A self-report questionnaire was designed to investigate mental health status of general population during the outbreak and delivered through an online crowdsourcing platform (http://www.jd.com/), as detailed elsewhere [13]. Joybuy platform is a large e-commerce and information service corporation with 0.44 billion active users by 2020 in China. Briefly, 71,227 people clicked on the survey page and 56,932 participants submitted the questionnaire voluntarily in 12 days, with a participation rate of 79.9%. After the quality control, 56,679 participants from 34 provinces in China were included, with an effective rate of 99.6%.

Outcomes and covariates

Data on demographic characteristics (e.g., age, sex, educational attainment, income level, occupation, marital status, geographic location, and living area), medical comorbidities (e.g., chronic disease and mental disorder), and information related to COVID-19 (e.g., infection status of participants and their relatives, condition of contact with infected or suspected patients, attitude and respond toward the epidemic, whether participant in frontline work related to COVID-19 including medical care, scientific research, disease control and management, and supply support, status of work or school resumption, risk of expose to patients due to occupational reasons, fear of infection, experience of public health interventions including quarantine, traffic restriction and community confinement were collected via questionnaire).

The primary psychological outcomes included symptoms of depression, anxiety, insomnia, and acute stress measured by the 9-item Patient Health Questionnaire (PHQ-9) [14], the 7-item Generalized Anxiety Disorder (GAD-7) scale [15], the 7-item Insomnia Severity Index (ISI) [16], and the Acute Stress Disorder scale (ASDS) [17], respectively. All measures were validated for use in Chinese [14,15,16]. Severity categories of mental health status were divided according to the total scores of measures: PHQ-9, normal (0–4), mild (5–9), moderate (10-14), and severe (≥15) depression; GAD-7, normal (0–4), mild (5–9), moderate (10-14), and severe (≥15) anxiety; ISI, normal (0-7), subthreshold (8-14), moderate (15-21), and severe (≥22) insomnia; and a score for dissociative cluster of ASDS ≥ 9 and a score for re-experiencing, avoidance and arousal cluster of ASDS ≥ 28 indicate the symptom of acute stress. The cutoff points for detecting symptoms of major depression, anxiety, and insomnia were 10, 10, and 15, respectively. Scores of participants greater than the cutoff threshold indicate potential psychological issues.

Statistical analysis

All analyses were conducted using SAS software, version 9.4 (SAS Institute, Inc), and two-sided P < 0.05 indicated significance. The scores of measure tools not normally distributed are presented as medians with interquartile ranges (IQRs) and categorical variables are presented as numbers and percentages. The Mann–Whitney U test was used to compare continuous variables not normally distributed between two groups.

Multivariable logistic regression analysis was performed to calculate the odds ratios (ORs) and 95% CI of risk of experiencing mental health issues, after adjusting potential confounders, including age, sex, educational attainment, income level, occupation, marital status, geographic location, living area, comorbidity of chronic diseases, history of mental disorders, infection status of COVID-19, experience of traffic restriction and community containment, and participation of frontline work related to the outbreak, status of work or school resumption, risk of expose to patients due to occupational reasons, and fear of infection.

Results

Demographic characteristics

In this cross-sectional, nationwide study, a total of 56,679 participants from 34 provinces in China completed the survey. Of the participants included (mean [SD] age, 36.0 [8.2] years), 27,149 (47.9%) were male and 16,454 (29.0%) ever experienced home confinement or centralized quarantine during COVID-19 outbreak. The basic characteristics of participants by quarantine condition are shown in Table 1. Most participants were aged 18–40 years (42,966 [75.8%]), were married (45,033 [79.4%]), had an educational level of college/undergraduate or less (50,311 [88.8%]), and lived in urban areas (52,839 [93.2%]). A total of 9725 participants (17.2%) participated in frontline work related to COVID-19 and participants (36745 [65.1%]) have returned to work or were constantly working. A considerable proportion of participants reported experiencing symptoms of depression (15,802 [27.9%]), anxiety (17,897 [31.6%]), insomnia (16,564 [29.2%]), and acute stress (13,817 [24.4%]). Sixteen thousand, five hundred eighty-three participants (29.2%) reported having one of the moderate to severe mental health issues including symptoms of depression, anxiety, insomnia, and acute stress.

Table 1 Demographic characteristics of responders by quarantine condition.

The psychological impact of quarantine

Participants who experienced quarantine measures reported higher prevalence rates of moderate-to-severe symptoms of depression (2243 [13.6%] vs 3871 [9.6%]; P < 0.001), anxiety (2188 [13.3%] vs 3683 [9.2%]; P < .001), insomnia (1192 [7.2%] vs 2064 [5.1%]; P < .001), and acute stress (4702 [28.6%] vs 9115 [22.7%]; P < .001) than those who were not quarantined (Table 2). Regarding total mental health issues, 34.1% of participants quarantined reported one of the psychological symptoms, compared with 27.3% in the population without quarantine (P < 0.001).

Table 2 Severity categories of depression, anxiety, insomnia. and acute stress symptoms by quarantine condition.

The median (IQR) scores on the PHQ-9, the GAD-7, the ISI, and the ASDS for all participants were 0.0 (0.0–6.0) for depression, 1.0 (0.0-7.0) for anxiety, 4.0 (1.0-8.0) for insomnia, and 23.0 (19.0-38.0) for acute stress. Participants who had been quarantined reported higher scores in scales measuring symptoms of depression (median [IQR] PHQ-9 score: 1.0 [0.0-8.0] vs 0.0 [0.0–5.0]; P < .001), anxiety (median [IQR] GAD-7 score: 2.0 [0.0-7.0] vs 1.0 [0.0–6.0]; P < .001), insomnia (median [IQR] ISI score: 5.0 [1.0-9.0] vs 4.0 [1.0-8.0]; P < .001) and acute stress (median [IQR] ASDS score: 25.0 [19.0-38.0] vs 23.0 [19.0-37.0]; P < .001) than those who were not quarantined (Table 3).

Table 3 Scores of depression, anxiety, insomnia and acute stress symptoms by quarantine condition.

Compared those without quarantine and adjusted for potential confounders, home confinement and centralized quarantine were associated with increased risk of total psychological outcomes (prevalence, 34.1% vs 27.3%; OR, 1.34; 95% CI, 1.28–1.39; P < .001), and of having moderate-to-severe symptoms of depression (prevalence, 13.6% vs 9.6%; OR, 1.42; 95% CI, 1.34–1.50; P < .001), anxiety (prevalence, 13.3% vs 9.2%; OR, 1.48; 95% CI, 1.40–1.57; P < .001), insomnia (prevalence, 7.2% vs 5.1%; OR, 1.44; 95% CI, 1.34–1.55; P < .001), and acute stress (prevalence, 28.6% vs 22.7%; OR, 1.34; 95% CI, 1.28–1.40; P < .001) (Table 4).

Table 4 The impact of quarantine on mental health issues.

Associated factors of mental health status for population quarantined

Multivariable logistic regression analyses showed that, after adjusting for potential confounders, male (OR, 1.27; 95% CI, 1.18–1.36; P < 0.001), those having a household income less than 5000 RMB per month (OR, 1.12; 95% CI, 1.03–1.21; P = 0.005), those with history of mental disorders (OR, 2.03; 95% CI, 1.46–2.82; P < 0.001) or having chronic physical diseases (OR, 1.26; 95% CI, 1.10–1.43; P = 0.001), those who were infected or suspected of COVID-19 (OR, 3.74; 95% CI, 1.80–7.76; P < 0.001) and those having suspected or infected relatives and friends (OR, 1.81; 95% CI, 1.39–2.36; P < 0.001), those who experienced fear of infection (e.g., worried vs not worried: OR, 2.05; 95% CI, 1.85–2.27; P = 0.005), those who are in Wuhan (OR, 1.50; 95% CI, 1.23–1.84; P = 0.002), frontline workers (OR, 1.16; 95% CI, 1.06–1.28; P = 0.002), and those who were exposed to patients with general diseases except COVID-19 (OR, 1.28; 95% CI, 1.04-1.58; P = .043), or to suspected or diagnosed COVID-19 patients because of occupational condition (OR, 1.44; 95% CI, 1.24–1.69; P < 0.001) had significantly higher risk of total psychological outcomes including moderate-to-severe symptoms of depression, anxiety, insomnia, and acute stress. Compared with those who lack of clarity about information related to the COVID-19, participants who have a good understanding of the information on the outbreak reported lower risk of psychological symptoms (basically understand, OR, 0.61; 95% CI, 0.48–0.78; P < 0.001; very understand, OR 0.46; 95% CI, 0.36–0.59; P < 0.001). Work resumption or working persistently (OR, 0.88; 95% CI, 0.82-0.95; P = 0.001), complying with quarantine protocol (OR, 0.47; 95% CI, 0.37–0.61; P < 0.001), occupational exposure to general population (OR, 0.78; 95% CI, 0.71–0.85; P < 0.001), and experience of community containment (OR, 0.77; 95% CI, 0.64–0.91; P = 0.003) were associated with lower risk of psychological outcomes (Table 5).

Table 5 Risk factors for mental health issues in quarantine population.

Discussion

In this nationwide survey study, 34.1% of participants with an experience of quarantine during COVID-19 outbreak reported having at least one of the psychological symptoms including anxiety, depression, insomnia, and acute stress, which is higher than those who were not quarantined (27.3%), indicating that the pandemic and quarantine measures related to COVID-19 are having adverse effects on mental health. Quarantine measures were associated with increased risk of experiencing mental health burden, especially for vulnerable groups including people with pre-existing mental or physical illnesses, frontline workers, those in Wuhan, those who are infected or at risk of infection, those who are less financially well-off, and those who experienced fear of infection. Complying with quarantine, being able to take part in usual work, and having adequate understanding of information related to the outbreak were associated with less mental health issues. Health officials and policy makers should take supportive measures, such as providing sufficient and transparent information on the condition of outbreak in question, and advising possible activities (such as resuming usual work or study through internet and telephone when applicable) for people who are quarantined to reduce boredom and improve connection with others, to achieve an optimal balance between the possible costs of mental health and containment of the outbreak when implementing quarantine and lockdown strategy. Interventions to reduce mental health consequences of quarantine and empower wellbeing especially in vulnerable groups under pandemic conditions need to be urgently identified and informed.

The psychological impact of quarantine measures

This study is, to our knowledge, the first to systematically explore the likely impacts of quarantine measures during the COVID-19 pandemic on mental health in general population, and the associated factors that may contribute to, or mitigate these effects. Recently published epidemiological studies during the outbreak investigating the effect of COVID-19 on mental health were restricted to the rate of anxiety, depression, insomnia, psychological distress, and other mental health issues in general population [18] or vulnerable groups such as medical care workers [7] or children [8], failing to consider the potential consequences of COVID-19 quarantine measures on mental health, which are one of the important research priorities for informing management of COVID-19 [4]. This study found that 29.2% of general population reported mental health issues, which is comparable with another investigation showing that 35% of the general population in China experienced psychological distress [18]. Against the backdrop of expected rise in psychological symptoms during these extraordinary circumstances, our findings suggest that quarantine measures used to manage the COVID-19 pandemic have negative effects on mental health and psychological wellbeing as previous quarantine and lockdown measures do during the outbreak of SARS and H1N1 influenza [19, 20]. Most of the adverse effects on mental health might come from the potential fallout of quarantine [3, 21,22,23,24], such as increased social isolation and loneliness, lack of belongingness, feeling a burden, financial insecurity, restriction of liberty, and fear of infection, which are associated with increased risk of mental health issues across the lifespan [4]. Given the evolving situation with coronavirus and related public health interventions, specific attention is required regarding the mental health status during and after quarantine for those who undergo it. In addition, effective interventions should be put in place as part of science-based quarantine strategy to mitigate mental health consequences and sustain containment of COVID-19.

Vulnerable groups of population quarantined

Our findings suggest that quarantine measures might disproportionately affect those most vulnerable and exacerbate health inequalities within populations. People with pre-existing mental health issues or chronic diseases, suspected or infected patients, frontline works especially those who are exposed to patients with general diseases and COVID-19, those who are less financially well-off, and those in the most severely affected areas in China should be particularly considered and given extra support during and after quarantine. Having a history of psychiatric illness is associated with psychological distress after experiencing traumatic events [25, 26], which might be exacerbated by lack of access to mental health support and services during COVID-19 mass quarantine. People with chronic diseases might be affected psychologically by isolation, loneliness, and lack of routine health care. For frontline workers vulnerable to high risk of infection, they might be affected by fear of infection and transmitting the virus to others, stigmatizing attitudes from others, and work stress [27,28,29]. Suspected or diagnosed patients of COVID-19 quarantined in dedicated facility or hospital may be affected by worry about physical symptoms that may related to the infection [30], fear of the consequences of infection when exposed to a potentially fatal infectious disease [24], and concern about infecting family members and friends [25, 31]. Furthermore, adverse effect of treatment for COVID-19, such as insomnia caused by corticosteroids, and symptoms of the infection such as fever and hypoxia, might increase vulnerability during quarantine. People in Wuhan, the area most severely affected by COVID-19 in China, might be most and directly affected by experiencing major public health emergency and the earliest public transport lockdown [7], alongside inadequate supplies such as food and medical resource, fearing the worst due to lack of clarity about the new virus in the early phase of the outbreak, stigma, and guilty. Those who are less financially well-off might be affected by the socioeconomic effect of quarantine including increased unemployment and financial insecurity [32, 33].

What can be done to mitigate the mental health risk?

Our study found that complying with quarantine, being able to take part in usual work, and having adequate understanding of information related to the outbreak were associated with less mental health issues. In addition, we found that experiencing community confinement (only residents of the community are allowed to access) was associated with less psychological symptoms. Further study is needed to investigate whether other public health interventions such as social distancing and community confinement might be more favorable. Previous evidence suggests that disruptions of usual daily activities and social networks could cause a feeling of isolation, boredom and frustration, that is often distressing to those under quarantine [21, 22]. As such, participating in daily routine when possible and keeping in touch with others via the internet and mobile phone to activate the social network could reduce the sense of isolation and distress. There is also the evidence that mental health support services using online applications and phone lines specifically for those under quarantine would help reassure people and make them feel connected to others [34]. Lack of transparency and accessible information about the situation of the outbreak, and difficulty with complying with quarantine protocols were associated with post-traumatic stress symptoms during the SARS epidemic [24]. Ensuring that people under quarantine have a good understanding of the severity of the outbreak and receive adequate information related to the disease and the reasons for quarantine from health and government authorities, should be encouraged. Regarding compliance with quarantine measures, having a sense of altruism, as well as feeling that quarantine is helping to keep others safe and epidemic control are likely to increase adherence and make stressful experiences easier to bear [10]. Informing the public about the benefits of quarantine for protecting or restoring public health and controlling epidemic, while making every effort to ensure that the burden quarantine is bearable for people, should be priorities to alleviate mental health issues and prevent long-term consequences.

Implications and recommendations

In light of our findings and previous evidence, we make recommendations that may help mitigate the impact of COVID-19 quarantine on mental health to inform optimal quarantine strategies and further interventions to promote wellbeing. First, at the early stage where quarantine measures are deemed necessary, reminding public about the necessity and benefit of quarantine measures, providing clear guidelines of quarantine, and ensuring adequate support of supply especially for people on low income and financial insecurity, could be helpful to increase compliance and reduce harmful effects of quarantine. Preventive measures such as online psychoeducational sessions and easy access to digital mental health care could be helpful. Second, effective and timely mental health and social support, including counseling services provided by hotline and online applications, population-level policies and guidelines, and remotely delivered psychological interventions, should be informed and provided to people under quarantine, especially those might be disproportionately affected such as people with a history of mental illness, suspected or diagnosed patients, and frontline workers. Providing clear communication with regular and transparent updates about the COVID-19 outbreak, advising people activating their social network to improve connection with others and maintaining usual daily routine when applicable, and ensuring basic supplies could be helpful to alleviate the feelings of isolation and boredom. Third, it is likely that quarantine measures might have long-term effects on mental health after quarantine. Regular screening for psychological symptoms especially for vulnerable groups after quarantine are needed to prevent long-term consequences and protect mental health wellbeing. Further evaluations are also needed to develop and inform population-level approaches for the prevention and treatment of mental health symptoms targeting at factors that are causally related to poor psychological outcomes and modifiable by interventions.

Strengths and limitations

Strengths of this study is using a large, nationwide population-based survey to investigate the psychological effect of quarantine measures and the associated factors that may contribute to, or mitigate these effects during the COVID-19 pandemic. We also adjusted for several sociodemographic variables, comorbidities, and variables related to the COVID-19 outbreak and additional public health interventions such as traffic restriction and community confinement, to reduce residual confounding. Several limitations of this study are worth noting. Current study with a cross-sectional design could not evaluate whether COVID-19 quarantine measures have long-term consequences on mental health. Further longitudinal studies are needed to clarify whether these outcomes will be long-lasting after the COVID-19 outbreak. In addition, psychological outcomes were measured in an online survey and defined by symptom scales rather than clinical diagnosis. However, the diagnosis of mental disorders by psychiatrists may not be feasible in a large general population under pandemic conditions. The use of clinical interviews should be encouraged in future study to improve the understanding about the psychological effect of COVID-19 and related interventions. Finally, although the response rate and completeness rate of the survey were 80.0% and 99.7%, our results might be subject to potential bias if the nonparticipants were too distressed to participate or were those with poor access to internet resources.

Conclusions

Quarantine measures during COVID-19 pandemic are associated with increased risk of experiencing mental health burden, especially for vulnerable groups including people with pre-existing mental or physical illnesses, frontline workers, those in the most severely affected area, and those who are less financially well-off. Reminding public about the necessity and benefit of quarantine measures, providing clear communication with regular and transparent updates about the COVID-19 outbreak, and advising people activating their social network to improve connection with others and maintaining usual daily routine when applicable, might be helpful to alleviate psychological distress. Achieving the appropriate balance between infection control and mitigation of the potential adverse psychological effects when implementing quarantine measures are crucial and immediate priorities for policy makers in health response to the COVID-19 outbreak. Further study is needed to establish interventions to reduce psychological effect of quarantine and empower wellbeing especially in vulnerable groups under pandemic conditions, and investigate the potential long-term consequences of COVID-19 quarantine and lockdown on mental health.