Investigating the psychological effects of home quarantine during the early peaks of the COVID-19 pandemic: a survey from Iran

Abstract

We assessed psychological symptoms among individuals who were quarantined during early coronavirus disease 2019 (COVID-19) peaks. This cross-sectional study was performed during April–October 2020 in Iran. We surveyed 100 individuals with COVID-19 patients in their families and 100 others with health conditions associated with a higher risk of developing critical forms of COVID-19 infection, who have completed at least 14 days of home quarantine. Validated Persian versions of the 21-item Depression, Anxiety, and Stress and 22-item Impact of Event Scale-Revised were used to measure the symptoms of depression, anxiety, stress and distress. The rates of stress, anxiety, depression and quarantine-related distress were 46.5, 48.5, 57.0 and 80.5%, respectively; however, they were not significantly different between the contact and no-contact groups. Female sex and being unemployed were significantly associated with quarantine-related distress, P = 0.007 and P = 0.018, respectively. Independent risk factors for anxiety were a history of medical comorbidity (P = 0.025) and contact with COVID-19 patients (P = 0.007). Findings show high prevalence rates of psychological symptoms among quarantined individuals, regardless of whether they had contact with COVID-19 patients or not. Female sex and unemployment were risk factors for quarantine-related distress.

Introduction

The coronavirus disease 2019 (COVID-19) continues to infect the global population with over 500 million reported cases and 6 million deaths by the end of April 2022 (^{WHO, 2022}); it has led to a pandemic that drastically changed the way people live worldwide. These altered circumstances have attracted research attention to psychological problems which may have been imposed on communities. The main focus in the early phases of the pandemic was directed toward the physical disease burden; however, as over 2 years have passed since the emergence of COVID-19, more recent studies have demonstrated its implications on the mental health of various populations (^{Cullen et al., 2020; Wang et al., 2020}; Kumar and Nayak, 2021; ^{Roy et al., 2021}). Pandemics of highly contagious infectious diseases are indubitably associated with increased psychological morbidities for all populations, and COVID-19 is no exception (^{Rogers et al., 2020; Robinson et al., 2022; Xie et al., 2022}). There have been reports of anxiety, depression, sleep disturbance, and distress in populations affected by COVID-19 (^{Benke et al., 2020; Louvardi et al., 2020; López Steinmetz et al., 2022; Parvar et al., 2022}), even at the initial stages of the pandemic (^{Torrente et al., 2020; Mechili et al., 2021}) or during its early remission phase (^{Xie et al., 2022}). The psychological implications of COVID-19 will need to be outlined in different populations in order to deliver specific solutions, tailored to their needs.

Many may not be aware of the extent to which the pandemic is responsible for their negative emotions and psychological symptoms. This may lead to a phenomenon known as ‘displaced coping’, where seemingly unrelated decisions are influenced by ambiguous sources of anxiety and sadness (^{Raghunathan et al., 2006}). In the case of the COVID-19 pandemic, this could mean that individuals may have adopted altered decision-making strategies in order to cope with their unclear source of distress. In this regard, recognizing the COVID-19 pandemic as a source of mental distress, synergic with other drivers, can aid professionals in consulting communities and resolving potential problems. Many studies have already reported this psychological burden on populations such as frontline workers, affected patients, quarantined individuals, and those who have been in close contact with patients or have lost a loved one due to this disease.

Various communities have been psychologically affected by the COVID-19 pandemic; however, the main burden of the pandemic is on infected patients as they suffer from extreme psychological stress due to fear of death, loss of social contact, lack of emotional support, and societal discrimination; all of which can subsequently manifest as depression, anxiety, and distress (^{Guo et al., 2020; Varatharaj et al., 2020; Zhang et al., 2020; Şimşir et al., 2022}). Frontline healthcare providers have also been recognized as an unmistakably distressed population, since not only are they faced with an increased workload and lack of equipment, but they also encounter notable social discrimination and stigma due to their close contact with infected patients (^{Singh and Subedi, 2020}), ultimately causing psychological distress and burnout in this population (^{Cullen et al., 2020; Krishnamoorthy et al., 2020; Rajkumar, 2020; Zhang et al., 2020; Nyashanu et al., 2022; Pan et al., 2022}). Regardless of these population-specific issues, all communities have been influenced by self-quarantine and social distancing policies laid out by the government. These psychological symptoms are even more substantial in individuals who have been quarantined (^{Tang et al., 2020}).

Nation-wide disease control policies including social distancing were adopted in Iran in the early stages of the pandemic; these included the closure of schools, universities, malls, and the facilitation of remote working. A campaign was launched on June 2020 to encourage citizens to wear masks, and their use became compulsory in social situations. Further restriction measures were applied throughout 2020 as infection and mortality rates continued to grow with new peaks (^{Peyravi et al., 2022a, b}). In many instances, home quarantine was issued for durations of up to 3 weeks at a time. Following these restrictive measures, the general population was faced with the burden of social isolation and loneliness, which has been noted by many authors (^{Lin et al., 2010; Naeim et al., 2021; Asanjarani et al., 2022}). Quarantined individuals are highly likely to experience psychiatric symptoms even regardless of their health and physical wellbeing, which can be due to the social restrictions and the loneliness that accompanies isolation (^{Huremović, 2015; Jaspal and Breakwell, 2022}). Following the first peak of the COVID-19 pandemic and nationwide quarantine during the Persian New Year (20 March–4 April), high rates of depression, anxiety, and distress were observed among the Iranian population with rates of moderate to severe anxiety and depression being 38.5% and 22%, respectively (^{Reskati et al., 2021}). Large-scale quarantine has been shown to intensify anxiety, claustrophobia, and feelings of not being in control (^{Rubin and Wessely, 2020}) since people experience loss of freedom, separation from loved ones, uncertainty from disease conditions, and daily exposure to media reports on death tolls (^{Barbisch et al., 2015; Rubin and Wessely, 2020}). Even past experiences regarding large-scale quarantine, namely with the Middle Eastern respiratory syndrome, had shown significantly higher needs for psychiatric interventions among isolated or quarantined patients compared to those not in isolation or quarantine (^{Kim et al., 2018}). While nationwide quarantine aims to reduce the spread of disease in communities, restrictions can inadvertently impact many aspects of daily life and subsequently mental health (^{Jin et al., 2021}), and further data is still required on this matter.

The present study aimed to assess the psychological symptoms of anxiety, depression, and stress in an Iranian sample of individuals who were quarantined, either due to contact with infected patients (contact group) or because of their high risk for critical COVID-19 outcomes (^{Zhao et al., 2021}) (no-contact group). Furthermore, we compared this population based on their cause of quarantine (contact versus no-contact group), while also investigating potential risk factors for such psychological symptoms. The results of the current study can add to the current understanding of quarantine as a large-scale control measure in disease outbreaks, while also providing more specific knowledge on the psychological state of individuals after COVID-19-related quarantine, to aid professionals in their consultations and problem-solving efforts.

Methods

Ethical approval

We performed this study in complete accordance with the ethical codes of the Declaration of Helsinki. Our study has been reviewed and approved by the ethics committee of Mashhad University of Medical Sciences under the approval number IR.MUMS.REC.1399.046.

We obtained informed consent from all participants before they enter the study. Moreover, in order to assure the confidentiality of participants’ information, data were gathered anonymously using codes instead of individual identities.

Study design and settings

In this cross-sectional study, we surveyed individuals who had completed at least 14 days of home quarantine as recommended by the local health authorities in Mashhad, Iran during April–October 2020. This duration included the first peak of the COVID-19 pandemic during the Persian New Year (late March to mid-April) and continued through the second peak (early-July to early August) and the early stages of the third peak (late October). According to the national interim guidelines, individuals with COVID-19 patients in their family and close relatives or those with health conditions associated with a great risk of developing critical COVID-19 infection need to be quarantined to minimize the risks of contracting the disease and further complications (^{Wang et al., 2020}).

Study outcomes

We used the validated Persian versions of the following questionnaires via an online platform to measure the symptoms of depression, anxiety, stress, and quarantine-related distress in the quarantined participants.

The 22-item Impact of Event Scale-Revised

The 22-item Impact of Event Scale-Revised (IES-R) is an instrument comprising 22 items that together evaluate the subjective response to a specific traumatic event and the distress associated with it. Intrusion, hyperarousal, and avoidance are the three subscales of this tool in addition to a total subjective stress score, which evaluates the mental disability of respondents when encountering a specific traumatic event. Every item in the IES-R questionnaire can be scored based on the severity of symptoms during the past seven days on a Likert scale between 0 (asymptomatic) and 4 (extremely symptomatic). The sum of scores can range from 0 to 88, with 0–23 showing no distress, 24–32 meaning partial distress, 33–38 determining definitive distress, and ≥39 indicating severe distress associated to the specific event (^{Moradi et al., 2008}).

The 21-item Depression, Anxiety, and Stress Scale

The 21-item Depression, Anxiety, and Stress Scale (DASS-21) is the abridged version of the original 42-item DASS questionnaire. This quick tool is often used to evaluate psychological burden via evaluating the severity of symptoms including depression (seven items), anxiety (seven items), and stress (seven items). Each item can be scored on a four-point basis from 0 (none) to 3 (high). Total depression scores 0–9, 10–13, 14–20, 21–27, and 28–42 reflect the normal, mild, moderate, severe, and extremely severe intensity of this symptom, respectively. Total anxiety scores of 0–7, 8–9, 10–14, 15–19, and 20–42 represent normal, mild, moderate, severe, and extremely severe anxiety, respectively. Regarding stress, total scores ranging from 0–14, 15–18, 19–25, 26–33, and 34–42 indicate normal, mild, moderate, severe, and extremely severe stress intensities, respectively (Lovibond and ^{Lovibond, 1996; Sahebi et al., 2005}). We did not focus on the severity of these symptoms; therefore, participants were regarded as being affected by depression, anxiety, and stress symptoms based on the following cutoff points: 9 for depression, 7 for anxiety, and 14 for stress.

Participants

According to the reported 22.4% prevalence of depressive symptoms in quarantined people (^{Lei et al., 2020}), keeping an alpha error of 0.05 and considering d = 0.058, the sample size was calculated to be 200 participants in total. Therefore, 200 quarantined individuals whose health records were available in local health centers of Mashhad were included in this pilot study, using a non-probability convenience sampling technique. Overall, 100 individuals with COVID-19 patients in their family (Contact group) and 100 others having health conditions associated with a higher risk of developing critical COVID-19 infection (No-contact group) who had no history of contact with suspicious or definitive cases of COVID-19 were included.

Inclusion criteria were the education of at least middle school level, verbal informed consent for participation in the study, and having completed at least 14 days of home quarantine during the COVID-19 outbreak as recommended by local health authorities. Using a non-random available sampling method, the contact group was selected from family members of patients with a definitive diagnosis of COVID-19 (those with definitive diagnosis based on real-time PCR), who had health records in Imam Reza COVID-19 clinic and the local health centers of Mashhad, Iran. The second high-risk group was selected through a non-random purposive sampling method, from individuals whose health records indicated at least one of the following risk factors: age ≥65 years, history of chronic pulmonary diseases, severe cardiac diseases, immunocompromised patients, BMI ≥40 kg/m², diabetes mellitus, and renal or hepatic failure.

Individuals who did not fill the questionnaires thoroughly and those who showed any symptoms of the infection or were required to be hospitalized during their quarantine period were excluded from the study.

Data collection

Contact details for individuals who met our eligibility criteria were retrieved from the health records of Imam Reza COVID-19 clinic and local health centers in Mashhad. Eligible individuals were recruited through phone calls, in which the research was explained to them and they were requested to fill questionnaires via an online platform. Individuals who provided informed consent received a link to our online survey. The sampling was continued until we received complete survey response from 100 participants in each group.

Demographic data including age, sex, education, job status, marital status, and income, as well as anthropometric measures including weight and height were recorded. Other information such as psychiatric history, medical history, history of close contact with COVID-19 patients, loss of loved ones due to COVID-19, and quarantine duration were also asked. The participants were then directed to fill the online DASS-21 and IES-R questionnaires. BMI was calculated based on the reported values for height and weight.

Statistical analysis

Data were analyzed using SPSS version 23 (IBM Statistics, Chicago, Illinois, USA). Data normality was assessed using Kolmogorov–Smirnov test. Normally distributed data were presented in mean and SD, and non-normally distributed data were presented by median and interquartile range. Variables were compared between the two subgroups of participants using independent samples t-test, Mann–Whitney U test, and Chi-square test where appropriate. Spearman correlation test was used to assess the correlations between quantitative variables.

Multiple logistic regression models were used to determine the factors associated with the outcomes, that is, psychological symptoms. Sex, age, marital status, occupational status, education, income, medical history, psychiatric history, quarantine duration, contact with COVID-19 patients, and loss due to COVID-19 were included as covariates in all the regression models. Results of the regression analyses were reported as odds ratio (OR) and 95% confidence interval (95% CI). P < 0.05 was considered statistically significant.

Results

Overall, 200 quarantined people were studied, of whom 100 were in close contact with COVID-19 patients in their family (Contact group) and 100 had health conditions associated with higher risk of developing critical COVID-19 infection (No-contact group). Out of the 200 participants, 111 (55.5%) were female and the mean age was 41.87 ± 14.79 (range: 16–81) years. Mean duration of quarantine was 33.14 ± 21.50 (range: 14–95) days.

Demographic and personal characteristics of the participants are showed in Table 1. As the table implies, the no-contact group were significantly older than the contact group and had significantly higher prevalence of medical comorbidities (P < 0.001). The frequency of married status was also significantly higher in the no-contact group than in the contact group (P = 0.044). Moreover, the length of quarantine was significantly higher in the no-contact group compared to those having close contact with COVID-19 patients (P < 0.001).

Table 1 - Demographic and personal characteristics of participants

Variable	No-contact (N = 100)	Contact (N = 100)	P
Age (years)	50.50 (33.25–61.00)	35.86 (29.00–42.00)	<0.001*
Sex			0.064**
Female	62 (62.0%)	49 (49.0%)
Male	38 (38.0%)	51 (51.0%)
Marital status			0.044**
Married	77 (77.0%)	64 (64.0%)
Single	23 (23.0%)	36 (36.0%)
Education			0.244**
Primary	42 (42.0%)	34 (34.0%)
University	58 (58.0%)	66 (66.0%)
Job			0.203**
Employed	47 (47.0%)	56 (56.0%)
Unemployed	53 (53.0%)	44 (44.0%)
Income			0.130**
Insufficient	63 (63.0%)	73 (73.0%)
Sufficient	37 (37.0%)	27 (27.0%)
BMI (kg/m2)	25.93 ± 4.32	25.99 ± 4.02	0.113†
Medical comorbidity	100 (100.0%)	18 (18.0%)	<0.001**
Psychiatry visit	23 (23.0%)	26 (26.0%)	0.622**
Psychiatric drug use	12 (12.0%)	12 (12.0%)	–
Loss due to COVID-19	2 (2.0%)	6 (6.0%)	0.279‡
Quarantine duration (days)	40.00 (21.50–58.00)	20.00 (14.00–30.00)	<0.001*

^*Mann–Whitney test.

^**Chi-square test.

^†Independent samples t-test.

^‡Fisher’s exact test.

COVID-19, coronavirus disease 2019.

Regarding the main study outcomes, the overall prevalence of stress, anxiety, and depression among the quarantined participants were 46.5% (93 cases), 48.5% (97 cases), and 57.0% (86 cases), respectively. In addition, 161 participants (80.5%) experienced different severities of quarantine-related distress.

The contact and no-contact group had no significant difference regarding the distribution of symptom intensities and their mean scores in any of the main study outcomes, namely stress, anxiety, depression, and quarantine-related distress (Table 2).

Table 2 - Comparison of study outcomes between the study groups

Outcome	No-contact (N = 100)	Contact (N = 100)	P
Stress
Mean DASS-21 score	7.89 ± 5.87	8.32 ± 5.97	0.676*
Normal	52 (52.0%)	55 (55.0%)	0.409**
Mild	10 (10.0%)	7 (7.0%)
Moderate	12 (12.0%)	13 (13.0%)
Severe	18 (18.0%)	11 (11.0%)
Very severe	8 (8.0%)	14 (14.0%)
Anxiety
Mean DASS-21 score	4.25 ± 2.44	5.12 ± 3.87	0.143*
Normal	58 (58.0%)	45 (45.0%)	0.439**
Mild	5 (5.0%)	9 (9.0%)
Moderate	20 (20.0%)	25 (25.0%)
Severe	5 (5.0%)	6 (6.0%)
Very severe	12 (12.0%)	15 (15.0%)
Depression
Mean DASS-21 score	6.53 ± 3.67	7.15 ± 5.10	0.538*
Normal	45 (45.0%)	41 (41.0%)	0.950**
Mild	14 (14.0%)	15 (15.0%)
Moderate	17 (17.0%)	16 (16.0%)
Severe	9 (9.0%)	12 (12.0%)
Very severe	15 (15.0%)	16 (16.0%)
Quarantine-related distress
Mean IES-R score	34.73 ± 15.54	38.20 ± 15.05	0.110*
No distress	23 (23.0%)	16 (16.0%)	0.334**
Partial distress	22 (22.0%)	18 (18.0%)
Definitive distress	19 (19.0%)	18 (18.0%)
Severe distress	36 (36.0%)	48 (48.0%)

^*Independent samples T-test.

^**Chi-square test.

DASS-21; The 21-item Depression, Anxiety, and Stress Scale; IES-R score, Impact of Event Scale-Revised score.

Spearman correlation test showed no significant correlation between the length of quarantine and the measured scores for assessed symptoms of depression (r = 0.061; P = 0.391), anxiety (r = 0.058; P = 0.411), and stress (r = 0.028; P = 0.691) based on DASS-21 and quarantine-related distress based on IES-R (r = 0.038; P = 0.594).

Table 3 elaborates on the results of multiple binary logistic regression models used to find the factors associated with mental health outcomes in quarantined participants. As the table illustrates, female sex (OR, 3.440; 95% CI, 1.401–8.446) and being unemployed (OR, 3.090; 95% CI, 1.213–7.872) were the only factors independently associated with quarantine-related distress among participants. For stress, being single and history of psychiatric visits (OR, 0.838; 95% CI, 0.180–0.816 and OR, 0.327; 95% CI, 0.121–0.885, respectively) were the independent protective factors. While the history of medical comorbidity and contact with COVID-19 patients (OR, 5.471; 95% CI, 1.579–18.958 and OR, 3.468; 95% CI, 1.060–11.344, respectively) were the risk factors independently associated with stress. The independent risk factors for anxiety among our participants were history of medical comorbidity and contact with COVID-19 patients (OR, 4.279; 95% CI, 1.200–15.257 and OR, 5.446; 95% CI, 1.584–18.718, respectively). The sole risk factor for depression in our sample of quarantined individuals was higher (university) education (OR, 2.189; 95% CI, 1.124–4.263).

Table 3 - Findings of multiple logistic regression

Dependent variable covariates*	Odds ratio	95% CI		P
		Lower	Upper
Stress
Being single	0.383	0.180	0.816	0.013
History of psychiatric visit	0.327	0.121	0.885	0.028
History of medical comorbidity	5.471	1.579	18.958	0.007
Contact with COVID-19 patients	3.468	1.060	11.344	0.040
Anxiety
History of medical comorbidity	4.279	1.200	15.257	0.025
Contact with COVID-19 patients	5.446	1.584	18.718	0.007
Depression
Higher education	2.189	1.124	4.263	0.021
Quarantine-related distress
Female sex	3.440	1.401	8.446	0.007
Being unemployed	3.090	1.213	7.872	0.018

^*Sex, age, marital status, occupational status, education, income, medical history, psychiatric history, quarantine duration, contact with COVID-19 patients, and loss due to COVID-19 were included as covariates in all the regression models.

CI, confidence interval; COVID-19, coronavirus disease 2019.

Discussion

Quarantine during the outbreaks of contagious diseases has been traditionally linked to deleterious effects on mental health due to several reasons such as financial loss, lack of social contact, and boredom (^{DiGiovanni et al., 2004; Robertson et al., 2004; Maynard et al., 2017}); our study aimed to investigate the psychological symptoms of a sample of Iranian individuals who underwent quarantine during the COVID-19 pandemic. We compared people who had contact with COVID-19 patients (contact group), against people with no contact who were at high risk for severe COVID outcomes (no-contact group). We found the mean age, frequency of comorbidities, and frequency of married status to be higher in the no-contact group, which was quite expectable due to their selection criteria (having health conditions associated with a higher risk of critical COVID-19). The length of quarantine was also significantly higher in this group, compared to those with COVID-19 patients in their families. This might also be attributed to the fact that these individuals were older and of poorer prognosis in case they contract the disease; thus, they feared more and had higher persistence in their social distancing and quarantine. Moreover, their quarantine was more a result of their caution and care for themselves than extrinsic obligations, thus they would remain quarantined longer.

In our study, the prevalence of stress, anxiety, and depression among quarantined participants were 46.5%, 48.5%, and 57%, respectively. Moreover, 80.5% experienced quarantine-related distress. The length of quarantine was not associated with the severity of symptoms in the participants. The contact and no-contact groups showed no significant difference regarding the prevalence and severity of these psychological symptoms. Our multiple logistic regression analyses revealed female sex and being unemployed as risk factors for quarantine-related distress symptom among participants. Having higher education was the only risk factor for depression. Being single and history of psychiatric visit were independent protective factors for stress. This might be because single individuals have no worries regarding transmission of the virus to their spouse and children and they also experience no family conflicts, which is reported to be increased to some extent during the pandemic (^{Fegert et al., 2020}). The history of psychiatric visit can allude to the individual’s higher insight about their need for psychological interventions and their access to these services, which all can be protective against mental health threats that arise during quarantine. History of medical comorbidity and contact with COVID-19 patients were independent risk factors for both stress and anxiety.

A study on the general Libyan population following the COVID-19 pandemic reported a prevalence of 13.6% for severe depressive symptoms and a prevalence of 5.6% for severe anxiety symptoms based on the generalized anxiety disorder scale (GAD-7) (^{Elhadi et al., 2022}). These are considerably lower than our findings in quarantined individuals. In line with our results; however, they reported that anxiety and depressive symptoms are associated with female sex, older age, higher education, and contact with a family member hospitalized due to COVID-19. Unlike our population young age and being single was associated with depressive and anxiety symptoms, while in our study being single and having a history of the psychiatric visit was a protective factor against depression. These inconsistencies are likely due to the cultural and demographic differences between studied populations. Considering the extensive array of factors involved in the quality of interpersonal relationships, including cultural standards, sexual inequality, the average age of marriage, and the traditional customs of marriage, the exact determination of these discrepancies requires extensive and specific investigation. The authors speculate that a difference in the quality of interpersonal relationships between Iranian and Western societies may have had a role to play in this observation. For example, substantial rates of divorce have been observed over the last two decades in Iran, along with a large-scale transformation of the traditionally defined marital relationships (^{Aghajanian and Thompson, 2013}). As a result of this cultural change, uncertainties may have risen with regard to the roles that each individual has in a relationship; these could explain why being in such a relationship at a time of nation-wide distress, is a predictor of factor for psychological symptoms such as depression.

Louvardi et al. in their survey, assessed the effect of the quarantine duration on depression, anxiety, distress, and somatization in 1100 Greek healthy individuals and those with chronic diseases during the COVID-19 quarantine. In line with our findings, they found no statistically significant correlations between the quarantine duration and the scores for depression, anxiety, distress, and somatization symptoms (^{Louvardi et al., 2020}). This can be attributed to the sheer fact that these individuals remained in quarantine because of their own safety, not necessarily due to extrinsic obligatory motivation, and thus could bear longer periods of social restrictions while experiencing lower psychological distress. It is also worth mentioning that the time of symptom evaluation in a quarantined population is of cardinal importance and affects mental health outcomes. For instance, a longitudinal survey of the Argentinian general population which measured symptoms of anxiety and depression at the beginning of the mandatory quarantine and 2 weeks later, found an increase in depression, while the levels of anxiety decreased. Their findings indicate that the continuation of the lockdown can have a greater mental health effect in the long term (^{Canet-Juric et al., 2020}).

In a study by ^{Tang et al. (2020}) in China, 172 people quarantined in an affected area and 219 others quarantined in unaffected areas, as well as 769 non-quarantined people during the COVID-19 outbreak were assessed for symptoms of depression and anxiety using the center for epidemiological studies depression scale (CES-D-20) and the Goldberg anxiety or depression scale (GADS). They found the prevalence of depression and anxiety to be 36% and 76% among people quarantined in the affected area, while the respective figures were 55% and 84% in people quarantined in unaffected areas. Our findings regarding the prevalence of anxiety and depression contrast the reported values by Tang and colleagues; we found a higher rate of depression but the anxiety was less prevalent in our studied sample of quarantined people. They found male sex, having a college degree or above, living alone, and exercising at home with family to be independently associated with a higher risk of probable depression, while older age, higher income, and being screened for COVID-19 in the community were independently associated with lower risk of probable depression. For anxiety, they found being married and having a college degree or above to be the risk factors, while male sex, older age, and higher income were the protective factors against anxiety. Consistently, we found the female sex to be a risk factor for quarantine-related distress, while being married and having a university degree or above were risk factors for anxiety and depression in our quarantined sample, respectively. However, other factors reported in their study were not significantly associated with any psychiatric symptoms in our study. The inconsistencies between the results of our study and the one by Tang et al. could be attributed to their larger sample size, the difference in demographics and inclusion criteria for the population (they included non-quarantined people as well), and different assessment tools. The exact determination of where these cultural differences are rooted, especially regarding sex as a risk factor for psychological symptoms, requires a multidisciplinary approach that warrants future cross-cultural studies with more specific objectives. It could be argued that considering these discrepancies between the findings in different regions, quarantine policymakers will need to encourage studies targeting their specific populations, since risk factors of psychological symptoms appear variable across different cultures, especially in terms of sex and relationship status. Nonetheless, it has been previously shown that regardless of the country of origin there is a relationship between quarantine and anxiety, depression, and stress symptoms (^{Jin et al., 2021}).

Another Chinese study on 2237 quarantined general individuals found that depression, as measured by the Self-Rating Depression Scale (SDS) had a prevalence of around 6%, which is far below the rates found in our study. Peng et al. also found that depression was independently associated with younger age, lower education, and being single, which is in stark contrast with our results. They also found a significant association between SDS and IES-R scores, which was not found in the present study (^{Peng et al., 2020}). Lei et al. who used a self-rating anxiety scale (^{Varatharaj et al., 2020}) and SDS to measure symptoms, found the prevalence of anxiety and depression to be 12.9% and 22.4%, respectively, among the quarantined respondents to an online survey in China, which is relatively lower compared to our results. In line with what we found, Lei et al. found education, and self-perceived health conditions to be associated with depression and anxiety in the respondents affected by quarantine. However, their results showed income and financial loss to also be associated with psychological symptoms, which contradicts our findings (^{Lei et al., 2020}). The inconsistencies between the present work and the ones by Peng and Lei might be due to their different measurement scales, their notably larger sample sizes, the different duration of quarantine, and their different inclusion criteria i.e. not having chronic medical comorbidities or contact with COVID-19 patients.

In line with our findings regarding the factors associated with the symptoms of anxiety and stress, Mazza et al. reported having medical conditions and contact with infected COVID-19 patients in the family or acquaintance as two factors that were associated with higher levels of depression, anxiety, and stress. They reported female sex as another risk factor, which is consistent with our findings (^{Mazza et al., 2020}). ^{Fernández et al. (2020}) also surveyed 4408 online participants who had undergone two weeks of quarantine in Argentine and found lower prevalence rates of anxiety and depression (31.8% and 27.5%, respectively) than our study. It is worth noting that, unlike our study they used the Brief-Symptom Inventory-53, nonetheless, there seems to be a notable difference between the Iranian and Argentinian populations in terms of psychological symptoms associated with quarantine. Even so, the risk factors associated with these symptoms in the Argentinian population appear to closely resemble our Iranian sample; they reported that the female sex, being young, and having high levels of neuroticism are risk factors for psychological distress associated with quarantine. In our Iranian sample, female sex was a risk factor for quarantine-related distress, as well as unemployment. The financial state of Iran at the time of the pandemic along with the lack of governmental support in this area could have played a role in intensifying symptoms of distress in our population.

Several studies that included all population subgroups (not focusing on quarantined people) have reported fairly similar rates for symptoms of anxiety, depression, and distress, which are lower than the rates we found in this study. For instance, a study on 2331 people in China, using hospital anxiety and depression scale, found that 25.4% of the general population had symptoms of anxiety, 21.3% had depression, and 13.9% had both. These lower rates compared to our findings can be attributed to the difference in the target population, sample size, and measurement tools. They reported having cancer or other chronic diseases to be significantly associated with depressive or anxious symptoms, which is in line with `our results (^{Guo et al., 2020}). Similar results were obtained in another study in Italy, in which the prevalence of depression and anxiety symptoms in the general population during the COVID-19 pandemic were reported to be 24.7% and 23.2%, respectively (^{Gualano et al., 2020}). An online survey of 1041 Irish people during the first week of the nationwide quarantine also revealed anxiety, depression, or both in around 20%, 23%, and 27% of the population. ^{Hyland et al. (2020}) also reported factors such as having COVID-19 infection, younger age, female sex, financial loss due to COVID-19, and higher perceived risk of COVID-19 as correlates of anxiety or depression. Similarly, a report from China indicated 20.4% prevalence of anxiety or depression or both among the general population during the COVID-19 outbreak. They also interestingly found out that the daily time spent on COVID-19-related news was associated with the occurrence of anxiety and depression symptoms (^{Li et al., 2020}). Another online survey in Germany also yielded comparable findings regarding the rate of depression and anxiety in the community, using PHQ-9 and GAD-7 as the assessment tools. ^{Benke et al. (2020}) also found that higher restrictions due to lockdown measures (e.g. quarantine, stay-at-home order, and reduction of social contacts) were associated with higher impairment of mental health, which can be in line with our findings. Relatively similar rates of anxiety, depression, and post-traumatic stress were reported by ^{González-Sanguino et al. (2020}) from Spain, in an online survey of 3480 people. The authors reported female sex, history of neuropsychological disorders, and having COVID-19 patients in close relatives to be associated with higher levels of symptoms (^{González-Sanguino et al., 2020}). Consistently, we found history of medical comorbidities and close contact with COVID-19 patients (e.g. in the family) as risk factors for both stress and anxiety. In line with our findings, female sex has been reported as a risk factor associated with higher psychological symptoms in several studies (Bonati et al., 2022; ^{Gualano et al., 2020; González-Sanguino et al., 2020; Alzahrani et al., 2022; Jia et al., 2022}).

The contrasts between the results of the abovementioned studies and our findings are mainly because of the difference in target populations. They have studied a sample of all subgroups of the population, most of which were possibly healthy and did not need strict quarantine measures, so they only followed the general nationwide recommendations such as social distancing. Moreover, their surveys were performed during the first days of lockdown when the real effects of quarantine might not be evident yet. Besides, they used different tools to measure depression and anxiety.

Many studies on the European populations have reported that the female gender is commonly associated with psychological symptoms during the COVID-19 confinement (^{Bonati et al., 2021; Rettie and Daniels, 2021}), which is in line with our findings. However, unlike our population in which being single was a protective factor against stress, Italian and German studies have shown that having a partner during the lockdown predicts greater well-being and lower psychological impact (^{Balsamo and Carlucci, 2020; Saita et al., 2021}). This inconsistency may be due to the sociodemographic differences between the Iranian and European populations. One of the most evident consequences of lockdown, is social isolation; although, feelings of loneliness are directly and indirectly affected by emotional dysregulation, as adaptive emotion regulation strategies may ameliorate the negative impact of stress (^{Velotti et al., 2021}). Among other stressors during quarantine, we found unemployment to be associated with quarantine-related distress. As unemployed individuals are vulnerable to the economic effects of the pandemic, many studies have also indicated unemployment as a risk factor of anxiety and depression symptoms (^{Bonati et al., 2021; Paulino et al., 2021}).

There are several implications which need to be considered in terms of the high prevalence rates of depression, anxiety and stress, observed in quarantined individuals. Since isolation is the main hallmark among these individuals; psychological interventions are not readily available for people under quarantine. Not only are these interventions hard to access, but even the initial evaluation of these psychological symptoms is hindered. In other words, quarantined individuals are unlikely to be aware of the extent to which isolation has influenced their mental health status, while also being at a disadvantage regarding the availability of psychological healthcare. Symptoms such as anxiety are associated with aggression (^{Chung et al., 2019}), which when considered alongside the frustrations caused by the COVID-19 pandemic, may have been among the reasons for the increase of domestic and interpersonal violence during the pandemic (^{Mazza et al., 2020}). These all warrant psychiatric interventions such as telemedicine for individuals under quarantine. Public awareness of these mental health aspects will need to be elevated, especially during mass quarantine, and individuals will need to be provided with means to seek psychological help.

As restrictive measures have started to ease in many regions, post-pandemic psychological symptoms among previously quarantined individuals deserve significant attention as well. With people returning to their normal ways of life, the burden of these psychological symptoms may be carried over to their workplace, reducing their productivity and satisfaction, while also leading to career burnouts (^{Beck et al., 2011}). Arguably, the treatment measures needed to handle these symptoms are even financially justified considering the loss of workplace productivity (^{McTernan et al., 2013}) or even educational performance (^{Owens et al., 2012}). Post-pandemic psychological care must be made available to the returning workforce, especially those recognized as having a high-risk of developing psychological symptoms when quarantined. Periodic psychological screening plans may also aid in the global recovery from the pandemic by potentially reducing the financial loss associated with high rates of psychological symptoms.

The present study is not without limitations. Due to the nature of participant recruitment, there was significant heterogeneity between the contact and no-contact groups, which limited our ability to compare these groups; however, multiple logistic regression allowed us to draw more accurate conclusions despite the initial heterogeneity of the data.

Findings of the present study demonstrate a high prevalence of psychological symptoms including anxiety, depression, and stress among quarantined individuals. Female sex and being unemployed were risk factors for quarantine-related distress, contact with COVID-19 patients was a risk factor for anxiety and stress, while higher education was a risk factor for depression. Groups with a higher risk of psychological symptoms associated with quarantine should be prioritized in receiving psychiatric help, while the public awareness of the high risk in these populations should be elevated in order to increase the level of social support these groups receive. Psychiatric interventions such as telemedicine in people with risk factors for developing psychological symptoms during the quarantine could potentially be effective to reduce adverse outcomes.

Acknowledgements

The authors wish to thank Persian Cohort Study team, the Psychiatry and Behavioral Sciences Research Center and the Clinical Research Development Unit of Ghaem Hospital Affiliated to Mashhad University of Medical Sciences for their help in performing this study.

This work was supported by Mashhad University of Medical Sciences [grant number 990031].

Conflicts of interest

There are no conflicts of interest.