Prevalence of and Risk Factors Associated With Mental Health Symptoms Among the General Population in China During the Coronavirus Disease 2019 Pandemic

1In December 2019, the coronavirus disease 2019 (COVID-19) outbreak occurred and aroused global attention. It has infected more than 6.1 million people and caused 376 000 deaths worldwide across 216 countries, areas, or territories according to June 3, 2020, data from the World Health Organization. In contrast to previous epidemics and pandemics, such as severe acute respiratory syndrome and Middle East respiratory syndrome, COVID-19 is more contagious and spreads faster.

2The World Health Organization officially announced that COVID-19 was a pandemic and called on the whole world to work together to confront this infection. Compulsory measures, such as containment, quarantine, community control, and business and school closures, have been implemented in China and other countries to impede further escalation of the pandemic.,,, Facing this large-scale infectious public health event and these enormous disruptions to daily life, people are under unprecedented pressure and are experiencing severe psychological distress.

3Few studies have investigated COVID-19–related mental health symptoms in the general population. One pilot study in China reported that nearly one-half of the respondents considered the mental health impact of COVID-19 as moderate or severe, and one-third exhibited anxiety symptoms. To set priorities for public health policies and implement effective health care interventions, the prevalence of mental health symptoms and factors that are associated with them during the COVID-19 outbreak need to be determined urgently. Thus, we conducted a large-sample, cross-sectional online survey study to investigate the prevalence of symptoms of depression, anxiety, insomnia, and acute stress and potential risk factors in the general population in China.

4This cross-sectional online study was conducted from February 28, 2020, to March 11, 2020. 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 (AAPOR) reporting guideline.

5A self-designed online survey was released via the health page in the Chinese website Joybuy, a large ecommerce and information service platform that provides online health products and services in China. The survey link was posted on the website. We used a convenience sampling method. The registered members clicked the link on the platform and responded to the survey voluntarily until the convenience sample covered all 34 province-level regions in China. This was an anonymous survey, and confidentiality of data was ensured.

6The respondents were all registered members of Joybuy. A total of 71 227 individuals clicked on the survey link, and 57 006 individuals commenced the survey, among whom 74 individuals refused to provide informed consent and 56 932 participants provided informed consent and submitted the questionnaires. One hundred fifty-six questionnaires that did not provide valid age information were excluded from the analysis. Ninety-seven respondents who were younger than 18 years were also excluded because obtaining online informed consent from their parents was not realistic under the present conditions.

7The survey lasted approximately 15 minutes and had 4 parts. The first part gathered demographic information of the participants, including gender, age, living area (urban vs rural), level of education, marital status, monthly family income, geographic region, history of chronic diseases, history of psychiatric disorders, family history of psychiatric disorders, and occupation. The second part asked epidemic-related questions. The third part evaluated isolation conditions and social attitudes toward the COVID-19 pandemic. The information about these 3 parts is listed in eTable 1 in the Supplement.

8The fourth part of the questionnaire consisted of 4 standardized scales, including the Chinese versions of Patient Health Questionnaire–9, Generalized Anxiety Disorder–7, Insomnia Severity Index, and Acute Stress Disorder Scale, which measured symptoms of depression, anxiety, insomnia, and acute stress, respectively. The total scores of these scales were interpreted as follows: Patient Health Questionnaire–9, normal (0-4), mild (5-9), moderate (10-14), and severe (15-27) depression; Generalized Anxiety Disorder–7, normal (0-4), mild (5-9), moderate (10-14), and severe (15-21) anxiety; Insomnia Severity Index, normal (0-7), subthreshold (8-14), moderate (15-21), and severe (22-28) insomnia; and Acute Stress Disorder Scale, acute stress symptoms (a dissociative cluster score ≥9 and cumulative re-experiencing, avoidance, and arousal cluster scores ≥28). In the present study, cutoff scores of 5 for the Patient Health Questionnaire–9, 5 for the Generalized Anxiety Disorder–7 scale, and 8 for the Insomnia Severity Index were adopted to detect symptoms of depression, anxiety, and insomnia.

9In this study, frontline workers are considered to be individuals who directly participated in the control of COVID-19, covering a wide range of occupations (eg, medicine, research, public health, media, security work, police, community work, emergency material delivery services, charity, construction, management, and psychological interventions). Frontline workers were presumed to have direct contact with patients with COVID-19. They were also presumed to possess personal protective equipment when they were in contact with patients with COVID-19. People with occupational exposure risks are individuals who did not necessarily participate in COVID-19 control but were required to contact people at work and could subjectively conceive of threats of being infected, including sanitary workers, taxi drivers, and supermarket workers, among others. They were presumed to be poorly equipped with self-protective measures.

10Descriptive statistics were used to present demographic data. The prevalence of symptoms of depression, anxiety, insomnia, and acute stress was calculated using the aforementioned cutoff scores and reported as the percentages of cases in different populations. The 95% CIs were produced by the exact binomial methods. χ2 tests were used to compare the prevalence of mild-to-severe mental health symptoms in different populations. To explore factors potentially associated with depression, anxiety, insomnia, and acute stress, unadjusted logistic regression and multiple logistic regression analyses were performed, and odds ratios (ORs) and 95% CIs are presented. Two-sided Wald tests were conducted to determine whether the ORs in the regression models were statistically significant. All of the variables that were statistically significant in the unadjusted regression analysis and those that might convey important information were entered into the multivariable model. Detailed information of variables in the logistic regression is listed in eTable 2 in the Supplement. The level of significance was set to P <.05. All of the statistical analyses were performed using SPSS statistical software version 22 (IBM Corp). Data analysis was performed from March to May 2020.

11Data from a total of 56 679 eligible participants were included in the final analysis, for a participation rate of 79.9% (56 932 of 71 227 participants). The participants included in the analysis represented all 34 province-level regions in China. Of the total sample, 27 149 participants (47.9%) were male, and the mean (SD) age was 35.97 (8.22) years; 39 468 participants (69.6%) were aged 18 to 39 years. Of the total number of respondents, 47 139 (83.2%) had a university degree or higher, and 43 763 (77.2%) were married. Most of the participants (52 839 participants [93.2%]) lived in urban areas, and 2352 (4.1%) were from Hubei province (ie, the province most severely affected by COVID-19 in China). This survey included data from 100 individuals (0.2%) with confirmed or suspected cases of COVID-19, 219 (0.4%) people who were in close contact with patients with COVID-19, and 9725 (17.2%) frontline workers. Of the total number of respondents, 608 (1.1%) had at least 1 family member or friend who was infected with COVID-19, 17 587 (31.0%) had at least 1 family member or friend who was a frontline worker, 16 454 (29.0%) had quarantine experience, 19 754 (34.9%) were not yet back to work, and 2904 (5.1%) had occupational exposure risks. Additional demographic and epidemic-related characteristics are presented in Table 1.

12The prevalence of symptoms for the 4 mental health conditions among the total sample was 27.9% (95% CI, 27.5%-28.2%) for depression (15 802 participants total, including 9688 participants [17.1%] with mild depression and 6114 participants [10.8%] with moderate-to-severe depression), 31.6% (95% CI, 31.2%-32.0%) for anxiety (17 897 participants total, including 12 026 participants [21.2%] with mild anxiety and 5871 participants [10.4%] with moderate-to-severe anxiety), 29.2% (95% CI, 28.8%-29.6%) for insomnia (16 564 participants total, including 13 308 participants [23.5%] with subthreshold insomnia and 3256 participants [5.7%] with moderate-to-severe insomnia), and 24.4% (95% CI, 24.0%-24.7%) for acute stress (13 817 participants). The prevalence of symptoms of the 4 mental health was high among patients with COVID-19 (depression, 75.0%; anxiety, 71.0%; insomnia, 68.0%; acute stress, 71.0%), frontline workers (depression, 30.4%; anxiety, 34.0%; insomnia, 32.4%; acute stress, 27.3%), family members or friends of patients with COVID-19 (depression, 46.7%; anxiety, 49.3%; insomnia, 48.4%; acute stress, 42.4%), family members or friends of frontline workers (depression, 29.3%; anxiety, 33.1%; insomnia, 31.6%; acute stress, 25.3%), residents of Hubei province (depression, 40.8%; anxiety, 44.9%; insomnia, 38.3%; acute stress, 33.7%), participants who were in close contact with patients with COVID-19 (depression, 53.9%; anxiety, 52.1%; insomnia, 56.2%; acute stress, 42.9%), participants who had been to Hubei province during the past 2 months (depression, 37.5%; anxiety, 41.5%; insomnia, 36.5%; acute stress, 30.2%), participants who experienced both centralized quarantine (depression, 38.0%; anxiety, 42.6%; insomnia, 43.3%; acute stress, 35.8%) and home quarantine (depression, 32.7%; anxiety, 36.4%; insomnia, 32.9%; acute stress, 28.3%), participants who had not yet returned to work (depression, 30.4%; anxiety, 33.6%; insomnia, 30.5%; acute stress, 25.5%), and participants who had occupational exposure risks (depression, 40.2%; anxiety, 45.0%; insomnia, 37.1%; acute stress, 37.8%). Additional details on the prevalence of the mental health symptoms in different populations are presented in Table 2.

13The results of the unadjusted analysis of demographic and epidemic-related variables are presented in eTable 3 in the Supplement. In the multivariable analysis, being younger than 40 years, having lower income levels, and having a personal history of psychiatric disorders were still found to be associated with the symptoms of depression, anxiety, insomnia, and acute stress. Male participants and unmarried people displayed a remarkably higher risk for depression, insomnia, and acute stress symptoms.

14Individuals with confirmed or suspected COVID-19 had at least twice the risk for the 4 mental health symptoms compared with those not infected with COVID-19 (adjusted ORs, 3.27 [95% CI, 1.84-5.80] for depression, 2.48 [95% CI, 1.43-4.31] for anxiety, 3.06 [95% CI, 1.73-5.43] for insomnia, and 3.50 [95% CI, 2.02-6.07] for symptoms of acute stress). Family members or friends of patients with COVID-19 were also susceptible to symptoms of depression (adjusted OR, 1.53; 95% CI, 1.26-1.85), anxiety (adjusted OR, 1.53; 95% CI, 1.27-1.84), insomnia (adjusted OR, 1.62; 95% CI, 1.35-1.96), and acute stress symptoms (adjusted OR, 1.77; 95% CI, 1.46-2.15). In addition, associations were identified between potential occupational exposure risks to patients with COVID-19 and the 4 mental health outcomes (adjusted ORs, 1.96 [95% CI, 1.77-2.17] for depression, 1.93 [95% CI, 1.75-2.13] for anxiety, 1.60 [95% CI, 1.45-1.77] for insomnia, and 1.98 [95% CI, 1.79-2.20] for acute stress symptoms). Respondents with centralized quarantine or home quarantine experience had higher risks than those without quarantine experience of having depression (adjusted ORs, 1.33 [95% CI, 1.10-1.61] vs 1.30 [95% CI, 1.25-1.36]), anxiety (adjusted ORs, 1.46 [95% CI, 1.22-1.75] vs 1.28 [95% CI, 1.23-1.34]), insomnia (adjusted ORs, 1.63 [95% CI, 1.36-1.95] vs 1.24 [95% CI, 1.19-1.30]), and acute stress symptoms (adjusted ORs, 1.46 [95% CI, 1.21-1.77] vs 1.29 [95% CI, 1.24-1.35]). Hubei residents also demonstrated a higher risk than participants from other provinces, with adjusted ORs of 1.42 (95% CI, 1.19-1.68) for depression, 1.54 (95% CI, 1.30-1.82) for anxiety, 1.20 (95% CI, 1.01-1.42) for insomnia, and 1.49 (95% CI, 1.25-1.79) for acute stress symptoms. Nonetheless, being at work was associated with a lower risk of having symptoms of depression (adjusted OR, 0.85; 95% CI, 0.79-0.91), anxiety (adjusted OR, 0.92; 95% CI, 0.86-0.99), and insomnia (adjusted OR, 0.87; 95% CI, 0.81-0.94). Frontline workers had an elevated risk for insomnia (adjusted OR, 1.06; 95% CI, 1.00-1.12) and acute stress symptoms (adjusted OR, 1.08; 95% CI, 1.01-1.14), and being a close contact of a patient with COVID-19 remained statistically significant in the multivariable model of insomnia (1.55; 95% CI, 1.11-2.15). The detailed results of multivariable analysis are shown in Table 3.

15The present study investigated the prevalence of and factors associated with mental health symptoms among the general population in China during the COVID-19 pandemic based on a nationwide, large-sample survey. Approximately one-quarter (24.4%) to one-third (31.6%) of respondents exhibited symptoms of depression, anxiety, insomnia, and acute stress. We also identified several psychologically vulnerable populations, such as individuals with confirmed or suspected COVID-19 and their relatives, respondents with occupational exposure risks, those who experienced quarantine, those who lived in Hubei province, and those who had not yet returned to work. These findings provide a comprehensive profile of psychological status in the general population in China during the COVID-19 outbreak and may contribute to developing population-specific mental health management and intervention strategies.

16The prevalence of mental health symptoms in the present study is consistent with the results of an epidemiological study that was conducted in early February 2020 among the general population in China, which indicated that nearly 35% of the respondents manifested psychological distress during the COVID-19 outbreak. Another preliminary online survey that was performed in late January 2020 found that nearly one-third of respondents experienced moderate-to-severe mental health outcomes. The prevalence of these mental health symptoms during this period of the pandemic was higher than before the outbreak, during which rates of moderate-to-severe depression, moderate-to-severe anxiety, and insomnia in China were approximately 6.0%, 5.3%, and 15.0%, respectively.,,, The prevalence of psychiatric morbidities during the severe acute respiratory syndrome epidemic in Singapore was reported to be 22.9%, and one-third of people exhibited psychological distress during the equine influenza outbreak in Australia. The prevalence of symptoms of depression, anxiety, insomnia, and acute stress in Hubei province was notable and was in accordance with previous studies of such psychiatric problems during other disasters., These findings indicate that severe emotional distress occurs among the general population during public health events and underscore the importance of preventing and treating mental health problems during the COVID-19 outbreak.

17The present study identified several population groups who were likely to develop mental health symptoms. Compared with uninfected people, those with confirmed or suspected COVID-19 were 2 to 3 times more likely to report mental health symptoms. The prevalence of symptoms of the 4 mental health conditions was high among patients with COVID-19 (depression, 75.0%; anxiety, 71.0%; insomnia, 68.0%; acute stress, 71.0%). This is partially consistent with findings among patients with severe acute respiratory syndrome. The higher risk of mental health symptoms among patients experiencing an epidemic may be attributable to their distressing circumstances, physical pain, and adverse effects of medications that are used to treat infections. Moreover, family members and friends of patients with COVID-19 were susceptible to mental health disturbances. We found that nearly 50% of family members or friends of patients with COVID-19 experienced mild-to-severe mental health symptoms. Loved ones experiencing negative outcomes during disasters has been shown to be a strong catalyst of mental health disorders., In addition, relatives of patients with COVID-19 may worry about becoming infected themselves, may be quarantined, and may feel stigmatized, all of which may exacerbate psychological distress.,

18The present study also found that people with occupational exposure risks reported greater symptoms of depression, anxiety, insomnia, and acute stress. People who work in high-risk environments often report more fatigue, health worries, and fear. COVID-19 may be symptomless during the incubation period, and its clinical manifestations can be easily confused with those of normal influenza., Therefore, people may understandably feel a threat of becoming infected by being exposed to general patients, thereby affecting their psychological well-being. However, in the present study, frontline work was not significantly associated with depression or anxiety. A previous study suggested that people at moderate infection risk (eg, individuals who might come in contact with patients with suspected cases) had more adverse mental health outcomes than those at high risk (eg, individuals who worked in infectious wards). The high infection risk group may be more aware of the risk, have better coping skills, have less uncertainty, and have more access to personal protective equipment and social support.

19Another prominent finding was the substantial impact of quarantine experience on mental health, which is consistent with prior studies. Quarantine can contribute to poor mental health in both children and adults., People may experience fear of infection, frustration, and boredom during quarantine. Insufficient basic supplies and disruptions of information flow can increase both fear and anxiety. In the present study, both centralized quarantine and home quarantine enhanced the odds of adverse mental health outcomes. Moreover, centralized quarantine can have a more pernicious outcome because of fear of infection, being in an enclosed space, and being in an unfamiliar and crowded environment. The environment plays a vital role in maintaining healthy emotions and sleep., An unfamiliar and crowded environment may be a catalyst for the unique association between centralized quarantine and poor mental health status.

20A high probability of symptoms of depression and anxiety was found among people who had not yet returned to work. Among people who had returned to work, those who worked at home had a higher risk of mental health symptoms compared with those who did not work at home. Being occupied with work activities can serve as a distraction from epidemic-related information. Interpersonal interactions that occur with conventional styles of work, in contrast to working at home, can ameliorate depression and lower the risk of mental disorders., Our results imply that accelerating people’s return to normal work may have a positive influence on mental health.

21Some demographic factors may also influence mental health during the COVID-19 pandemic. Unmarried status and lower income were identified as risk factors for poor mental health outcomes, which is consistent with previous studies.,, However, opposite trends were observed for gender and age. The higher susceptibility of male participants to mental health symptoms during the pandemic may be attributable to their more frequent risky behaviors during epidemics (eg, going to crowded places or not wearing masks) and higher infective rate., The higher risk of mental health symptoms among younger people was also identified in this study, which is consistent with a previous study during the severe acute respiratory syndrome outbreak. This may be attributable to the fact that young adults frequently engage in social media and may be more exposed to misinformation online, which can trigger psychological distress., In addition, the proportion of self-reported history of mental disorders in this study was low, which might be attributed to stigma toward mental illnesses and the low recognition rate of mental health symptoms in China. We cannot exclude the possibility that the representativeness of the sample may influence the proportion of self-reported mental disorders.

22The strengths of this study include its extensive geographic coverage across China, large sample size, and the special study period. Participants with different demographic and COVID-19 epidemic–related characteristics were recruited from all 34 province-level regions in China. In addition, to the best of our knowledge, this is the first study that has systematically investigated the prevalence of and factors associated with mental health symptoms (ie, symptoms of depression, anxiety, insomnia, and acute stress) by standardized rating scales among the general population during the COVID-19 pandemic in China. Our findings may provide more helpful information for policy making, recognition of high-risk populations, and framework design for population-specific psychological crisis management.

23Our study has several limitations. First, this was an online survey, and we used a convenience sampling method. Aside from posting our questionnaires on the main Joybuy website, we also sent proportional questionnaire links to registered members to increase the response rate. Although this study had extensive geographic coverage across China and a large sample size, it was conducted among internet users who were young and highly educated; thus, the representativeness of the sample might be limited. Second, the status of mental health symptoms was based on the respondents’ self-reports rather than clinical diagnoses. Third, this was a cross-sectional study. Therefore, associations between mental health symptoms and risk factors cannot necessarily be considered causal relationships. Fourth, the results may only reflect mental health status during the epidemic. Follow-up studies are needed to determine the possible long-term mental health outcomes associated with the COVID-19 pandemic.

24The prevalence of symptoms of depression, anxiety, insomnia, and acute stress was high in this sample of study participants from China during the COVID-19 pandemic, especially among patients with confirmed or suspected COVID-19 and their family members and friends, people with occupational exposure risks, and residents of Hubei province. Pervasive intervention measures, including quarantine and delays in returning to work, were closely associated with negative mental health outcomes. These findings suggest that the COVID-19 pandemic may have severe mental health repercussions. Population-specific mental health interventions are urgently needed to meet demand during this outbreak. Future studies are needed to explore the association of the COVID-19 pandemic with mental health in other countries and its long-term outcomes.