The subjects were in a quarantined hotel in Shenzhen from August 2020 to November 2021, all of whom were inbound from Hong Kong. After finishing the first nucleic acid testing at the customs, they would stay 14 days in the quarantine hotel. Inclusion criteria: 1) Occupancy date from August 2020 to November 2021; 2) Good cognitive ability, good reading comprehension ability, and able to complete the answers normally; 3) Complete the questionnaire survey on the day of check-in; 4) Informed consent to the questionnaire survey and voluntary participation. Exclusion criteria: 1) infants and illiterate persons; 2) Unable to understand the questionnaire’s content or incapable of completing the questionnaire due to cognitive impairment.

The questionnaire was divided into two parts. The first part collected sociodemographic data, including gender, age, education, marital status, whether quarantine alone, history of physical illness, history of mental illness, history of psychiatric medication, moods at check-in, and attitude towards quarantine policies. Another part of the study focused on standard scales to measure depression, anxiety, and insomnia symptoms, such as PHQ-9, GAD-7, and ISI. There were nine questions on the PHQ-9 scale, and each question was scored 0–3 points, a total of 28 points. 0 to 4 points were for normal. In this paper, subjects were divided into 0 to 4 points for the normal group and >4 points for the depression group. There were seven questions on the GAD-7 scale, and each question was scored 0–3 points out of 21 points. 0 to 4 points were for normal people. This paper divided subjects into 0 to 4 points for the normal group and >4 points for the anxiety group. There were seven questions on the ISI scale, each graded on a scale of 0–4 points out of 28. 0–7 points for normal and 8–14 points for subclinical insomnia. In this paper, subjects were divided into 0–14 points for the normal group and >14 points for the insomnia group. An invalid questionnaire would be marked if the response time was less than 1 min or if there were more than 20% incomplete responses.

The statistical analysis was performed using SPSS 20.0. Continuous variables were expressed as mean ± standard deviation. Continuous variables were expressed by mean ± standard deviation. Classified variable data were expressed by numerical value or percentage. To compare the count data between groups, we used the Chi-square test. Fisher’s exact test was adopted when the expected frequency of grouped variables was less than 5. The correlation of data was analyzed using Spearman correlation analysis. Multivariate analysis was performed by binary logistic regression. Odds ratios (OR) and 95% confidence intervals (95% CI) of independent variables were calculated. p < 0.05 was considered statistically significant.

A total of 1,360 valid questionnaires were collected for data analysis. 639 were female (47.0%), and 721 were male (53.0%). The average age was 41.82 ± 11.4 years old. 77.7% were quarantined alone. Most of them were well educated, married, had no history of chronic physical illness, mental illness, psychiatric medication history, good moods at check-in, and fully understood the quarantine policies, as shown in Table 1.

The mean PHQ-9 score was 2.8 ± 4.8. 270 (19.9%) of them were screened for depression. The average score of GAD-7 was 2.2 ± 4.1, and 239 (17.6%) of them were screened for anxiety. The average ISI score was 4.7 ± 5.9. Ninety-seven of them (7.1%) with insomnia were screened out, as shown in Table 1.

Age, education, and whether quarantine alone showed no significant difference in the risk of depression, anxiety, and insomnia among people in quarantine (p > 0.05). Gender, history of physical illness, history of mental illness, history of psychiatric medication, moods at check-in, and attitude towards quarantine policies had statistical significance in the risk of depression, anxiety, and insomnia (p < 0.05). Marital status had no difference in the risk of anxiety and insomnia (p > 0.05) but had statistical significance in the risk of depression (p < 0.05). See Table 2.

Spearman correlation analysis showed that the PHQ-9 score was positively correlated with the GAD-7 score (r = 0.705, p < 0.0001). PHQ-9 score was positively correlated with the ISI score (r = 0.528, p < 0.0001). GAD-7 score was positively correlated with ISI score (r = 0.533, p < 0.0001).

Multivariate logistic regression analysis showed that being married (OR = 0.614, 95% CI = 0.450–0.915) was a protective factor for depression. Chronic disease (OR = 2.579, 95% CI = 1.416–4.698) was a risk factor for insomnia. No psychiatric medication history was a protective factor for depression (OR = 0.227, 95% CI = 0.068–0.757) and insomnia (OR = 0.240, 95% CI = 0.078–0.736). Female, history of mental illness, low moods at check-in, and partial/cannot understand the quarantine policies were risk factors for anxiety, depression, and insomnia. See Tables 3–5.

In COVID-19, the mental health study results varied significantly among countries [13]. A meta-analysis [21] indicates that depression, anxiety, and insomnia are prevalent among people in quarantine during the COVID-19 period. Those rates ranged from 0.9 to 48%, 0.7%–64%, and 0.9%–37.6%, respectively. Based on systematic reviews and meta-analyses [15] of coronaviruses (SARS, MERS, and SARS-CoV2), 14%–61% of those infected experience mental health complications, and 14.8%–76.9% experienced these problems after their illness. Our data showed a positive correlation between PHQ-9, GAD-7, and ISI scores. This positive correlation indicates that depression, anxiety, and insomnia were closely related and might have a high co-morbidity rate. Stress affects the internal state of balance between the body and mind, leading to physical and mental reactions such as tension, depression, anxiety, insomnia, or various physical problems [19]. Even if the pandemic is over, it to the general public health professionals and the influence on the vulnerable groups of mental health and well-being will continue for a long time. We want to be able to care for patients in a more personalized way [22].

This study suggests that being married was a protective factor for depression. Single people increased the risk of depression by 35.9% compared with married people, consistent with existing research. Wang et al. [23] found that unmarried people suffered higher psychological stress during the pandemic. Tian et al. [24] showed that married people showed fewer psychological symptoms than unmarried people. Others reported more somatization, paranoid ideas, obsessiveness, depression, anxiety, phobias, and psychosis than married people.

Chronic disease increases the risk of insomnia, and psychiatric medication history contributes to depression and insomnia. Additionally, the risk of anxiety, depression, and insomnia increases with a mental illness history. Available evidence suggests that some of the measures taken to contain the pandemic may harm vulnerable populations, including those with mental or physical health problems and those with mental illness [6, 25, 26]. Due to the loss of access to mental health support and active activities, COVID-19 isolation can increase anxiety and depression in people coping with it.

Female, lack of understanding of quarantine policies (including partial and cannot understand at all) and low moods (including general, bad, and very bad moods) at check-in would significantly increase the risk of depression, anxiety, and insomnia. This study showed that females were 1.484, 1.897, and 1.832 times more than males to experience the risk of depression, anxiety, and insomnia, respectively. When previous trends are considered, mental distress is higher than expected, especially among women. Consistent with many existing studies [27–29]. In an earlier cross-sectional study of COVID-19 in Spain (n = 3,480), the female was a strong predictor of anxiety. A meta-analysis [7] also noted a higher prevalence of anxiety among women during the pandemic. According to a Greek study [30], women were more likely to have sleep problems, reflecting an existing gender difference in anxiety and insomnia symptoms.

China continued to implement the most stringent prevention and control measures globally [31], including nucleic acid testing, tracking, self-isolation, quarantine, and broader population measures, including travel bans, school closures, assembly restrictions, curfews, and total lockdown. Entry quarantine was an essential means of preventing imported cases of COVID-19. However, some people still did not fully understand and agree with the quarantine policies. This study showed a positive correlation between the attitude towards quarantine policies and the moods at check-in (r = 0.361, p < 0.0001). The psychological resistance to quarantine policies would be reflected in the emotion during quarantine, resulting in adverse effects. More obvious negative emotions were observed for people in quarantine as a particular group. The low moods at check-in reflected the negative coping emotion of people in the face of the epidemic and quarantine policies. The psychological stress level of people with negative coping emotions was higher [23]. Negative coping emotions might be related to psychological stress or mental illness such as anxiety and depression [32, 33]. Considering the positive correlation between the attitude towards quarantine policies and the moods at check-in, people would have more pronounced negative emotions under quarantine measures [34]. In the study [35] of psychological interventions for people in quarantine, members of the psychological support team mentioned the experience of communicating with people in an equal manner, explaining the importance of quarantine measures, and acknowledging their contribution to social security. On 26 January 2020, China’s National Health Commission released guidelines for emergency psychological crisis intervention during the COVID-19 epidemic [36]. The notice provisions, psychological crisis intervention should be part of the public health response, by the municipal city and provincial levels organize joint prevention and control mechanism, interventions should be distinguished according to the group intervention staff. With the implementation of crisis interventions in China to reduce the negative psychosocial impacts on public mental health, there are challenges to mobilizing an effective mental health care system [22], which should provide which type of intervention, which groups need help, and in what way. In addition to emergency psychological crisis intervention measures afterward, this study suggests how to allow the public to understand different policies during different pandemic periods, which may significantly reduce various emotional and spiritual problems in quarantine. This involves official media must be authoritative and credible, rather than simply creating information panic to attract traffic. There is a risk that fake news can spread faster than the virus itself and create uncertainty and fear. This should be regulated through constant interaction with the media and national regulations.

This study is a single-center study without considering the impact of the difference in living environments and medical conditions from other quarantine sites on mental health problems. The research results might be biased. In addition, due to the limitation of conditions, there was no comparative investigation of mental health before and after quarantine and psychological intervention research.

People in quarantine had problems with depression, anxiety, and insomnia. Depression, anxiety, and insomnia are closely related and positively correlated. Being married was a protective factor for depression. Chronic disease was a risk factor for insomnia. No psychiatric medication history was a protective factor for depression and insomnia. Female, history of mental illness, low moods at check-in, and partial/cannot understand the quarantine policies were risk factors for anxiety, depression, and insomnia. Positive correlations were found between attitudes towards quarantine policies and moods at check-in. In the face of public emergencies such as COVID-19, the psychological status of quarantined people will constantly change with the epidemic’s progress. Quarantine is part of our public health, but side effects must be weighed against alternatives like every medical intervention [2]. For example, voluntary isolation may be associated with good compliance and less psychological impact, especially when clearly explained and advertised as altruism. Therefore, timely monitoring and intervention of psychological status should be carried out during quarantine, but how to make the public understand quarantine policies during different periods deserve further studies.