Impact of perceived severity on depression, anxiety, and insomnia among Chinese community residents during the COVID-19 lockdown

Abstract

BACKGROUND

Depression, anxiety, and insomnia were found out that were significant relevance to the mental health impact of the coronavirus disease 2019 (COVID-19) lockdown.

AIM

To examine the interrelationships among perceived severity, anxiety, depression, insomnia, and sense of security in Chinese community residents during the COVID-19 lockdown period.

METHODS

Participants were selected using simple random sampling from four large gated communities in Chengdu, China. All participants were invited to complete a survey that included the Perceived Severity Questionnaire, Security Questionnaire, Patient Health Questionnaire-9, Generalized Anxiety Disorder 7-item, and Insomnia Severity Index-7. In total, 568 valid questionnaires were gathered. Correlation analysis and structural equation models were used to explore the relationship between perceived severity, sense of security, depression, anxiety, and insomnia.

RESULTS

The observed prevalence rates of anxiety, depression, and insomnia among residents during lockdown were 27.5%, 17.6%, and 16.0%, respectively. Correlation analysis showed that both perceived severity and sense of security were positively correlated with anxiety, depression, and insomnia [Pearson’s r was perceived severity and anxiety r = 0.44 (P < 0.01); with depression r = 0.48 (P < 0.01); with insomnia r = 0.43 (P < 0.01); security with anxiety r = -0.65 (P < 0.01); with depression r = -0.65 (P < 0.01); with insomnia r = -0.53 (P < 0.01)]. Structural equation modeling and bootstrap tests revealed that sense of security acted as a significant mediator in the relationship between perceived severity and emotional and sleep disorders (anxiety, depression, and insomnia).

CONCLUSION

This study demonstrates that sense of security is a significant predictor of emotional and sleep disorders (namely, depression, anxiety, and insomnia) among residents during the COVID-19 lockdown, with sense of security acting as a mediating factor. These findings suggest that mental health interventions for Chinese community residents during lockdowns may benefit from developing community-based educational programs to reduce perceived severity and ensuring the stable supply of essential resources and promoting social support networks to enhance the sense of security.

Key Words: Perceived severity; Sense of security; Depression; Anxiety; Insomnia; Structural equation modeling; COVID-19

Core Tip: The observed prevalence rates of anxiety, depression, and insomnia among residents during lockdown were 27.5%, 17.6%, and 16.0%, respectively during the coronavirus disease 2019 lockdown. Perceived severity is a significant predictor of emotional and sleep disorders (namely, depression, anxiety, and insomnia) among residents. Structural equation modeling and bootstrap tests revealed that sense of security acted as a significant mediator in the relationship between perceived severity and emotional and sleep disorders (anxiety, depression, and insomnia).

INTRODUCTION

The global coronavirus disease 2019 (COVID-19) pandemic has exerted a profound influence on public health worldwide[1]. By the end of 2022, the Chinese government had instituted a “dynamic zero-COVID” policy to mitigate viral transmission[2]. This policy entailed the rapid identification and management of individuals who are infected or at risk of infection, thereby inhibiting the formation of transmission chains. Nonetheless, in regions with extensive pandemic spread, city-wide or partial lockdowns were implemented as a crucial measure to curb the virus’s transmission. These lockdowns confined residents to their homes for living and working, effectively reducing cross-infection risks[3]. These lockdowns confined residents to their homes for living and working, effectively reducing cross-infection risks[4]. Nevertheless, the prolonged isolation during lockdowns has been shown to have detrimental effects on mental health[5]. Wang et al[6] evaluated the mental health of Chinese college students during the pandemic and discovered a marked escalation in levels of anxiety and depression under lockdown circumstances. Likewise, Liu et al[7] reported a prevalence rate of 12.33% and 6.26% for depression and anxiety, respectively, among children and adolescents following the lockdown. Despite the fact that China reclassified COVID-19 as a common infectious disease in 2023, ceased isolation, contact tracing, and embargoes, and as of 2025, the quarantine measures have long been lifted, exploring the psychological state of the Chinese population during the COVID-19 quarantine remains highly significant. With purpose of understanding the psychological responses of the population during the quarantine period and offer insights into how individuals and communities cope with extreme stressors and major life disruptions, the current study aimed to explore the relationship between perceived severity on depression, anxiety, and insomnia among Chinese community residents during the COVID-19 lockdown, especially the mediating effect of sense of security between them.

Depression, anxiety, and insomnia were found out that were significant relevance to the mental health impact of the COVID-19 lockdown[4]. Depression reflects a negative emotional state with symptoms such as persistent sadness, loss of interest, and feelings of worthlessness. Anxiety represents excessive worry, restlessness, and fear, which are common reactions to the uncertainties of the pandemic. Insomnia, on the other hand, is a physiological manifestation that can be both a cause and a consequence of psychological distress, highlighting the complex interplay between mental and physical health during this period. The COVID-19 pandemic and its associated lockdown measures have brought about a series of stressors, such as social isolation, economic uncertainty, and fear of infection. Previous studies have shown that these stressors can lead to an increased risk of mental health problems. Depression, anxiety, and insomnia are among the most common psychological and physiological responses[8-10]. Research on previous public health emergencies has indicated that the disruption of normal life patterns, like those experienced during lockdowns, often leads to a spike in the prevalence of these conditions[11]. In the context of the COVID-19 lockdown, cognitive-behavioral framework point out that individuals’ negative thoughts and appraisals about the pandemic, such as believing that they are highly likely to get infected or that the situation will never improve, can trigger feelings of anxiety and depression[12,13]. These negative emotions then lead to changes in behavior, such as avoiding social contact (in the case of depression) or engaging in excessive worry-related behaviors (in the case of anxiety). Insomnia can be seen as a behavioral consequence of these emotional states. At the same time, insomnia can further distort cognitions, making negative thoughts more likely, and thus perpetuating the cycle of anxiety and depression. Stress-diathesis model also indicated that individuals with a pre-existing vulnerability (diathesis), such as a genetic predisposition to mental health problems or a history of previous psychological difficulties, are more likely to develop anxiety, depression, and insomnia in response to the significant stressor of COVID-19[14]. Moreover, the presence of one of these conditions (e.g., anxiety) can act as an additional stressor, increasing the risk of developing the other conditions (depression and insomnia), thereby creating a complex web of relationships.

Perceived severity is defined as an individual’s subjective evaluation and emotional response to the gravity of a particular disease or risk[15]. When an individual perceives a threat from the environment, it can initiate a stress response, leading to symptoms such as tension, anxiety, and depression[16]. Prolonged exposure to threatening situations can lead to various changes in an individual’s emotional, physiological, and behavioral responses, ultimately affecting their overall health. Moreover, such exposure can compromise immune system function, exacerbating risks to well-being[17]. The COVID-19 serves as an example of a threatening scenario[18]. A survey involving approximately 900 South Koreans[18] found that nearly 70% of respondents defined the severity of COVID-19 as “high” or “very high”. Studies have indicated that during the early stages of the pandemic, Chinese individuals’ perception of the severity of the situation was significantly linked with emotional and sleep disturbances[19]. Other research has reported that higher perceived severity is associated with greater risk of emotional problems such as anxiety and depression. Similarly, studies on the severe acute respiratory syndrome and Ebola epidemics have found a similar relationship[20-22]. While prior research has primarily examined the impact of epidemics on the general public, little attention has been given to the specific experiences of residents under lockdown. Furthermore, there is a dearth of research on the mechanisms underlying the effects of perceived severity on mental health. Individuals confined to their homes face an elevated risk of infection and may perceive a greater threat from the virus, as well as potential threats to their physical and economic well-being[23]. Building on prior research, we hypothesize that perceived severity serves as a significant predictor of mental health outcomes among residents under lockdown[24].

Maslow’s Hierarchy of Needs places the sense of security as the second lowest level of need. This is distinct from anxiety and fear and represents a subjective perception and response to personal safety[25]. A low sense of security is a significant predictor of mental health problems such as depression and anxiety[25,26]. Ecosystem theory holds that the development of individuals takes place in multiple nested ecosystems, including microsystems (e.g., families, communities), meso-systems (e.g., interactions between families and communities), and macrosystems (e.g., sociocultures, policies, etc.)[27]. During the epidemic lockdown, these systems have undergone tremendous changes. From a microsystem perspective, the interaction patterns within families may change due to home isolation, and the pressure between family members may increase, affecting individuals’ sense of security. At the meso-system level, community prevention and control measures, material availability, and neighborhood relationships all have an impact on individual security. In terms of the macro system, social and cultural attitudes towards the epidemic, government prevention and control policies, etc. will also shape the individual’s cognition and sense of security about the epidemic[13]. When an individual perceives that the severity of the epidemic is high, if the ecosystem in which they are located can provide stable support, such as harmonious family relations, sufficient community materials, and effective government prevention and control, the individual’s sense of security can be guaranteed, thereby reducing the occurrence of psychological problems; conversely, the instability of the ecosystem will weaken the sense of security and increase the risk of psychological problems[27]. Moreover, recent research suggests that the COVID-19 pandemic has adversely affected individuals’ sense of security, making it a key psychological response to the pandemic[28,29]. Individuals who perceive the pandemic as more severe tend to have a significantly lower sense of security[28]. The Cannon-Bard theory emphasizes that the generation of emotions is the result of the collaborative activity of the cerebral cortex and the thalamus. Physiological changes and emotional experiences occur simultaneously. In the context of the epidemic, the perception of the severity of the epidemic will simultaneously activate areas in the brain related to emotions and a sense of security. When the perceived severity is high, the area in the brain responsible for evaluating danger is activated, simultaneously affecting the sense of security and emotional response[11]. If the level of security is high, the brain will process danger signals more actively and suppress the generation of negative emotions; on the contrary, when there is insufficient security, negative emotions will dominate, triggering psychological problems such as anxiety and depression and interfering with normal sleep[30]. Based on these observations, our second hypothesis posits that the sense of security may mediate the relationship between perceived severity of the pandemic and the mental health of residents in lockdown situations. Overall, the current investigation focused on the impact of perceived severity on symptoms of anxiety, depression, and insomnia, using these symptoms as indicators of emotional and sleep disorders. Additionally, the study aimed to confirm the mediating role of the sense of security in this relationship.

MATERIALS AND METHODS

Participants

Data were recruited from four large gated communities in Chengdu via convenient sampling. To be eligible for this study, participants had to meet certain inclusion criteria: A negative result in nucleic acid testing, a minimum age of 22 years, sufficient cognitive capacity to respond to questionnaires, and the absence of hypertension, coronary heart disease, or other chronic conditions. Additionally, individuals who were taking medications for depression, anxiety, or insomnia were excluded. Participants completed a conventional paper-and-pencil test that was disseminated by trained volunteers. These volunteers provided participants with an overview of the survey after their nucleic acid tests. Participants were then instructed to take the questionnaires home, complete them, and return them the following day after undergoing another round of nucleic acid tests. The questionnaire commenced with a section for informed consent and contained items to screen for eligibility based on the inclusion criteria. Only individuals who provided signed informed consent and met the inclusion criteria were included in the study. In total, 602 questionnaires were disseminated, and 590 were returned. Of the returned questionnaires, 22 were excluded due to incomplete responses. As a token of appreciation for their time and effort, participants were given two boxes of milk, each valued at approximately one dollar.

Measurements

Perceived Severity Questionnaire: To assess the participants’ perception of COVID-19 severity, the Perceived Severity Questionnaire was employed[19]. This instrument consisted of five items evaluating the severity of various factors, including infection rate, morbidity, mortality, negative impact on social order, and negative influence on the economy. Participants were asked to rate each item on a five-point Likert scale, which ranged from 1 (“not severe”) to 5 (“extremely severe”). Higher scores denoted greater perceived severity. The questionnaire has demonstrated good internal consistency reliability in previous studies[19]. In the present study, the Cronbach’s α coefficient for the Perceived Severity Questionnaire was 0.84.

Security Questionnaire: The Sense of Security Scale, initially developed by Maslow et al[31] in 1945, was used to gauge participants’ sense of security[32]. The scale has been translated into Chinese and undergone revisions, maintaining strong reliability and validity. The questionnaire assessed two facets of security: Interpersonal security and control certainty. Each facet contained eight items. Participants’ responses were scored on a five-point Likert scale, ranging from 1 (“very consistent”) to 5 (“very inconsistent”). A higher aggregate scores signified a more robust sense of security. In this study, the internal consistency coefficients for the two subscales were 0.75 and 0.72, while the overall scale yielded a coefficient of 0.80, confirmed the instruments acceptable reliability.

Patient Health Questionnaire-9: In this study, the Patient Health Questionnaire-9 (PHQ-9) was employed to assess symptoms of depression. This tool was adapted from the depression module of the PHQ[33]. The PHQ-9 consists of nine items that evaluate the frequency of depressive symptoms over a two-week period. Sample items included statements such as “I experience poor appetite or overeating” and “I feel tired or have little energy”. Scores were calculated using a four-point Likert scale, which ranged from 0 (“not at all”) to 3 (“nearly every day”). The total score was calculated as the sum of all individual item scores, with higher scores indicative of more severe symptoms. In this study, a cut-off point of ≥ 10 was employed for positive screening. The internal consistency of the PHQ-9 was found to be high, as indicated by a Cronbach’s α coefficient of 0.91.

Generalized Anxiety Disorder 7-item: To assess anxiety symptoms, the Generalized Anxiety Disorder 7-item (GAD-7) questionnaire was used[34]. The GAD-7 consists of seven items that ask participants to rate their levels of anxiety on a 4-point Likert scale, ranging from 0 (“not at all”) to 3 (“nearly every day”). Examples of the items included statements like “I excessively worry about various things” and “I tend to get easily frustrated or irritated”. Anxiety severity was quantified by summing all the item scores, with higher scores representing more severe anxiety symptoms. For this study, a threshold score of ≥ 10 was established for positive screening. The internal consistency reliability of the GAD-7 was found to be 0.96, as measured by Cronbach’s α coefficient.

Insomnia Severity Index-7: To assess insomnia, we employed the Insomnia Severity Index-7, which consisted of seven items[35]. Participants were asked to evaluate their experiences based on statements such as “I have difficulty falling asleep” and “I wake up during the night”, using a 5-point Likert scale that ranged from 0 (“none”) to 4 (“very”). The total score was calculated by summing the individual item scores, and a cut-off score of ≥ 15 was applied to categorize participants as testing positive for insomnia. The internal consistency reliability of Insomnia Severity Index-7, as determined by Cronbach’s α coefficient, was found to be 0.83 in this study, demonstrated good reliability.

Statistical analysis

Data were analyzed using SPSS 20.0 and AMOS 20.0 software packages. Positive screening rates for depression, anxiety, and insomnia were assessed, and associated risk factors were examined using logistic regression analysis. Structural equation models were constructed using scores for depression, anxiety, and insomnia as dependent variables, perceived severity as the independent variable, and sense of security as the mediating factor. The significance of the mediation effect was assessed by estimating the 95% confidence intervals for both direct and indirect effects through bootstrap tests.

RESULTS

Common method bias test, descriptive statistics, logistics regression analysis, and correlation analysis

To assess the potential issue of common method variance, the Harman’s single-factor test was conducted. In the exploratory factor analysis, all items were included in the model. The results showed that 10 factors had eigenvalues greater than 1, and the first factor explained 26.79% of the variance, which was below the 40% critical threshold[36]. This indicated that common method bias was not a significant concern in this study. Meanwhile, subgroup analyses were conducted to assess the confounding effect in the study. The sample was divided into different subgroups based on demographic variables such as age, gender, and socioeconomic status, results showed the relationships between the main variables, such as perceived severity, sense of security, and anxiety, remain consistent across these groups which indicated that the confounding effect was not significant in the current study.

The observed positive screening rates for depression, anxiety, and insomnia were 27.5%, 17.6%, and 16.0%, respectively. Logistic regression analysis revealed that female participants had were significantly more likely to report depression [odds ratio (OR) = 1.82, P < 0.01] and anxiety (OR = 2.05, P < 0.01) compared with male participants. Similarly, participants without a stable income were more likely to report depression (OR = 1.91, P < 0.01), anxiety (OR = 1.90, P < 0.01), and insomnia (OR = 1.85, P < 0.01). Additionally, individuals aged 30-40 years exhibited an increased likelihood of experiencing depression, anxiety, and insomnia, as displayed in Table 1.

Table 1 Prevalence of depression, anxiety and insomnia according to the demographic variable.

	No. of cases (%)	Adjusted OR (95%CI)	P value	No. of cases (%)	Adjusted OR (95%CI)	P value	No. of cases (%)	Adjusted OR (95%CI)	P value
Depression	156 (27.5)
Anxiety				100 (17.6)
Insomnia							91 (16.0)
Gender
Male	73 (22.4)	1 (reference)		43 (13.2)	1 (reference)		44 (13.5)	1 (reference)
Female	83 (34.3)	1.81 (1.25-2.62)	0.01	57 (23.6)	1.31 (1.16-3.14)	< 0.01	47 (19.4)	1.55 (0.99-2.42)	0.06
Age
> 40	51 (22.2)	1 (reference)		31 (13.5)	1 (reference)		25 (10.9)	1 (reference)
30-40	99 (31.5)	1.62 (1.09-2.39)	0.02	67 (21.1)	1.74 (1.10-2.77)	0.02	63 (20.1)	2.06 (1.25-3.39)	0.01
< 30	6 (25.0)	1.17 (0.44-3.10)	0.75	2 (8.3)	0.58 (0.13-2.61)	0.48	3 (12.5)	1.17 (0.32-4.21)	0.81
SI
Yes	96 (23.8)	1 (reference)		60 (14.9)	1 (reference)		55 (13.6)	1 (reference)
No	60 (36.7)	1.83 (1.24-2.70)	< 0.01	40 (24.2)	1.83 (1.17-2.87)	0.01	36 (21.8)	1.77 (1.11-2.82)	0.02

OR: Odds ratio; CI: Confidence interval; SI: Stable income.

Table 2 outlines the outcomes of the correlation analysis conducted on the study’s observational variables. The data revealed a significant positive correlation between perceived severity and symptoms of anxiety, depression, and insomnia, thus supporting H1. Furthermore, perceived severity was found to be significantly and negatively correlated with both dimensions of sense of security. Sense of security also exhibited a significant negative correlation with anxiety, depression, and insomnia.

Table 2 Correlation analysis.

	mean ± SD	1	2	3	4	5
1. Perceived severity	17.75 ± 4.80
2. Interpersonal security	30.48 ± 7.10	-0.33b
3. Certainty in control	30.36 ± 7.94	-0.37b	0.87b
4. Depression	6.81 ± 5.74	0.48b	-0.65b	-0.73b
5. Anxiety	4.83 ± 5.31	0.44b	-0.65b	-0.73b	0.82b
6. Insomnia	8.88 ± 6.05	0.43b	-0.53b	-0.60b	0.71b	0.65b

^bP < 0.01.

Structural equation modeling analysis

The two-step procedure recommended by Anderson and Gerbing[37] was adopted to investigate the mediation effects. Firstly, the measurement model was tested to assess that each latent variable was adequately represented by its indicators. the measurement model included five latent constructs - perceived severity, sense of security, depression, anxiety, and insomnia - represented by 44 observed variables, initial assessment of the measurement model indicated a satisfactory fit: χ²/df = 3.83, comparative fit index = 0.94, standardized root mean square residual = 0.04, and root mean squared error of approximation = 0.07, confirmed the effective representation of the latent factors by their indicators, the validity of measurement instruments was further verified.

Then, structural equation modeling analysis was employed to examine the relationships among perceived severity, sense of security and depression. Specifically, we constructed a partially mediated structural model, designated as model a, wherein perceived severity served as the independent variable, sense of security functioned as the latent mediating variable, and depression acted as the dependent variable (Figure 1). The model demonstrated a high level of fit, as evidenced by the metrics presented in Table 3. Notably, all paths within the model yielded significant results (P < 0.01). Similarly configured structural models were also constructed for anxiety (model b) and insomnia (model c) as dependent variables, while retaining perceived severity as the independent variable and sense of security as the latent mediating variable, these models are shown in Figure 1, respectively.

Figure 1 The structural models. ^bP < 0.01.

Table 3 Structural models fitness index.

	χ2	df	RMSEA	SRMR	CFI
Model a	1.72	1	0.04	0.01	0.99
Model b	1.12	1	0.03	0.01	0.99
Model c	1.88	1	0.03	0.02	0.99

RMSEA: Root mean squared error of approximation; SRMR: Standardized root mean square residual; CFI: Comparative fit index.

Bootstrap tests

To evaluate the direct and indirect effects of perceived severity on depression, anxiety, and insomnia, bootstrap testing was performed (Table 3). The 95% confidence intervals for the indirect effects ranged from 0.17 to 0.30 for depression, 0.18 to 0.31 for anxiety, and 0.13 to 0.24 for insomnia. Notably, none of these intervals overlapped with zero, confirming the statistical significance of all mediating effects.

DISCUSSION

The present study investigated the effects of perceived severity on anxiety, depression, and insomnia among Chinese community residents during the COVID-19 lockdown. The results revealed that perceived severity was a significant predictor of anxiety, depression, and insomnia. Additionally, results showed that a sense of security partially mediated this relationship.

Consistent with prior studies, including Li et al’s research[15] on the initial phases of the COVID-19 pandemic and investigations into previous epidemics such as severe acute respiratory syndrome and Ebola[20-22,38], results of the current study showed that prevalence rates for anxiety, depression, and insomnia during the lockdown period were 27.5%, 17.6%, and 16.0%, respectively. At the same time, this study showed that vulnerability factors for emotional and sleep disorders included being female, having no stable income, and being between 30 and 40 years old during the quarantine period[23,39]. According to stress theory, when individuals encounter stressors like epidemic - related lockdowns and perceive a high - level threat from the disease, a stress response is triggered[40]. Those who appraise the illness as highly threatening and consider the various consequences of the pandemic are more likely to have an elevated perception of severity[15]. Cognitive behavioral theory also point out that the individual’s cognitive evaluation of the event (perceived severity) determines the degree of their stress response[41]. In the context of a lockdown, residents are more likely to perceive a high probability of infection, leading to higher levels of perceived severity. If an individual regards the epidemic as a serious threat, they will enter a state of stress, and there will be a series of physiological changes in the body, such as activation of the sympathetic nervous system, increased blood pressure, and accelerated heart rate[42]. Excessive apprehension about contracting COVID-19 and the subsequent consequences can result in chronic stress. This stress may manifest physiologically as activation of the sympathetic nervous system, elevated blood pressure, and increased heart rate, which could contribute to the onset of anxiety and depression and adversely affect sleep quality.

Through structural equation modeling and Bootstrap test, this study clearly reveals that sense of security plays a significant partial mediating role between perceived severity and emotional and sleep disorders (anxiety, depression, and insomnia). Cognitive behavioral theory states that cognition, emotion, and behavior interact and influence each other. Elevated levels of perceived severity are associated with increased negative effects, heightened risk, environmental uncertainty, diminished control, and elevated levels of panic and tension, which in turn undermine the sense of security[43]. Therefore, from a theoretical perspective, when an individual perceives a high degree of severity of the COVID-19 epidemic, this perception will break the original psychological balance state. Individuals often have a strong sense of uncertainty about their own safety, worrying about being infected by viruses, fearing that the order of life will be severely disrupted, and that the economic situation will be negatively affected. This uncertainty directly leads to a decrease in a sense of security. For example, during the epidemic lockdown period, residents may doubt the safety of their own environment due to factors such as inability to travel freely and short-term fluctuations in the supply of daily necessities, and then feel uneasy and fearful, and the sense of security will decrease accordingly[44]. As a fundamental basic psychological need and a safeguard for mental health[25], a sense of security plays a crucial role in maintaining mental health. When the sense of security decreases, an individual’s psychological defense mechanism will be unbalanced, a compromised sense of security predisposes individuals to emotional and sleep disorders, such as anxiety, depression, and insomnia. The psychological defense mechanism is a series of unconscious psychological strategies adopted by an individual to protect their psychological balance when facing pressure and threats. In the absence of a sense of security, these psychological strategies cannot function normally, and it is difficult for an individual to effectively deal with negative emotions and is more likely to fall into negative emotional states such as anxiety and depression[42]. At the same time, the accumulation of negative emotions will further interfere with an individual’s normal sleep. Anxious emotions may make it difficult for an individual to relax at night and repeatedly think about issues related to the epidemic, leading to difficulty in falling asleep; depressive emotions may affect an individual’s sleep cycle and cause sleep problems such as early awakening and multiple dreams[44]. Previous studies also point out that, during the epidemic, due to excessive worry about the epidemic, some residents will involuntarily think about whether they will be infected with the virus and when the epidemic will end every night when lying in bed, resulting in a serious decline in sleep quality and insomnia symptoms in the long run.

The results of this study had theoretical and practical significance. Theoretically, his study deeply analyzes the relationship between perceived severity, sense of security and mental health problems such as depression, anxiety and insomnia of residents during the COVID-19 lockdown period, and reveals the mediating role of sense of security in it[44]. This provides a new perspective and empirical basis for understanding the influence mechanism of mental health in special situations. Although previous studies have involved the impact of the epidemic on mental health, the exploration of the internal connections and action paths among different factors is not in-depth enough. Through methods such as structural equation modeling, this study clarifies the relationship between variables, enriches the theoretical system of the influence mechanism of mental health, and helps subsequent studies to more comprehensively understand the generation and development process of psychological problems[42]. Meanwhile, the current study further expanded stress and psychological adaptation theory. Based on stress theory, this further refines the specific process and influencing factors of individual psychological stress responses under special stressors such as epidemic lockdown. The study found that residents’ perception of the severity of the epidemic triggers stress responses, which in turn affect security and mental health[45]. This supplements the application details of stress theory in the context of specific public health events. At the same time, it also provides a new case for psychological adaptation theory, illustrating the dynamic change process of psychological adaptation when individuals face sudden and continuous external pressure, and how to promote psychological adaptation by regulating key factors such as security.

In practice, the research results have important implications for guiding mental health interventions during the epidemic and similar crises and optimizing the construction of mental health service system. The current study has been clarified that perceived severity and sense of security are key factors affecting residents’ mental health, providing direct targets for mental health interventions. Studies found that during the epidemic lockdown, online mental health lectures and one-on-one psychological counseling services organized by the community can effectively relieve residents’ anxiety and depression and enhance their sense of security[44,46]. Based on those, relevant institutions and communities can formulate more targeted intervention strategies. For example, through various channels such as official media and community publicity, in the form of easy-to-understand popular science articles and videos, scientific knowledge about the COVID-19 virus is spread to residents, including the virus’s transmission mechanism, symptoms after infection, and treatment effects, guiding residents to view the epidemic rationally and avoid excessive panic, thereby reducing their perceived severity of the epidemic[8]. While Shi et al[47] proposed that providing residents with a stable supply of daily necessities can not only meet basic living needs but also have a positive impact on enhancing residents’ sense of security and mental health level. The government and communities should ensure the timely and accurate disclosure of epidemic-related data and guarantee the sufficient supply and reasonable distribution of living materials. For instance, by establishing a complete material distribution system and regularly providing basic living materials to residents, this can enhance residents’ sense of security and then improve their mental health status[48]. The research results were also helpful to promote the optimization of the mental health service system in responding to public health events. Knowing that women, people without stable incomes, and residents aged 30-40 are more likely to have psychological problems during the epidemic period suggests that mental health service institutions should focus on resource allocation and service provision[45]. Special psychological counseling projects can be carried out for these high-risk groups, and professional psychological counselors can be trained to improve the accuracy and effectiveness of services. In addition, the research also provides a reference for the methods and contents of mental health services. For example, increasing the proportion of online mental health services and developing psychological intervention courses and tools suitable for different groups.

CONCLUSION

This study summarizes the relationship between perceived severity, sense of security and mental health problems. Based on these findings, we further consider how to provide guidance for specific intervention methods. When individuals perceive a high degree of severity of the COVID-19 epidemic, this perception will break the original balance state in their psychology. Individuals often have strong uncertainties about their own safety, worrying about being infected by the virus, being afraid that the order of life will be severely disrupted and the economic situation will be negatively affected[49]. This uncertainty will directly lead to a decrease in sense of security. For example, during the epidemic lockdown, residents may doubt the safety of their environment due to factors such as being unable to travel freely and short-term fluctuations in the supply of daily necessities, and then feel uneasy and fearful, and the sense of security will decrease accordingly[48]. Based on our results, we recommend several strategies to safeguard the mental well-being of the general population during pandemics. First, disseminating accurate information about the COVID-19 virus can mitigate the perceived severity of the situation. Therefore, offering guidance to the public on how to interpret the COVID-19 pandemic in a rational and optimistic manner is crucial. This should involve the dissemination of relevant information and the encouragement of adherence to pandemic prevention measures. Second, our data underscore the significance of preserving a sense of security as a defense against emotional and sleep disorders. To this end, ensuring the timely and accurate release of pandemic-related data and the adequate distribution of essential daily supplies like food and water to residents in lockdown situations can help to alleviate fear and bolster their sense of security. This study is subject to several limitations that warrant discussion. First, the sample was recruited through convenient sampling and exclusively consisted of participants from specific, large communities in Chengdu. Consequently, the findings may lack generalizability to populations with different educational and socioeconomic backgrounds. Subsequent studies should adopt probability sampling methods such as stratified sampling and multi-stage sampling to widely cover people from different regions (rural areas, remote areas, first-tier cities and small and medium-sized cities), different social and economic strata (high-income, middle-income, and low-income groups), and different educational levels (primary school and below, middle school, university and above). Second, the cross-sectional nature of this study precludes the establishment of intervariable causality; it can only suggest correlations. Therefore, the observed associations between perceived severity and emotional and sleep disorders should be interpreted with caution. Future research needs to employ a longitudinal research design to conduct long-term follow-up investigations on the same batch of research subjects at different stages of the epidemic (initial stage, peak stage, mitigation stage, normalized prevention and control stage, etc.). For example, a large number of representative samples are selected in the initial stage of the epidemic, and questionnaires and psychological evaluations are conducted regularly (such as monthly or quarterly) to continuously monitor the dynamic changes of variables such as perceived severity, sense of security, depression, anxiety, and insomnia. Several key variables should be identified and measured which may be the potential confounding variables, such as social support, personality traits, pre-existing mental health conditions, and cultural background, to addressing the issue of confounding variables in the future study. Third, data collection occurred on the 10^th day following the initiation of the lockdown, potentially limiting the representativeness of the findings for the entire lockdown period. To illustrate, Shanghai, China’s largest city, experienced a nearly two-month-long complete lockdown in April 2022, during which some areas experienced shortages of essential goods. The duration and conditions of a lockdown may exert different effects on mental health, and these aspects merit further exploration in future studies.

ACKNOWLEDGEMENTS

We would like to thank all the individuals who participated in the study.